Calibration of Simulation Models

C

Calibration of Simulation Models Nick Malleson School of Geography, University of Leeds, Leeds, UK

Overview Simulation models are becoming increasingly popular in criminology research. In order for researchers to have confidence in the results of simulation studies, it is essential to make sure that the models are properly evaluated. Calibration is a major element to this evaluation and refers to the estimation and adjustment of model parameters to improve the agreement between model output and a data set. This entry will discuss some different methods of model calibration as well as related quantitative methods of error assessment.

Introduction Simulation models are becoming increasingly popular in criminology research, both as tools for exploring real-world crime patterns (explanatory models) and for experimenting with underlying theory (conceptual models). In order for researchers to have confidence in the results of simulation studies, it is essential to make sure that the models are properly evaluated. The process of evaluation is commonly

divided into three activities, defined by Rykiel (1996) as follows: • Verification is a demonstration that the modeling formalism is correct. • Calibration is the estimation and adjustment of model parameters and constants to improve the agreement between model output and a data set. • Validation is a demonstration that a model within its domain of applicability possesses a satisfactory range of accuracy consistent with the intended application of the model. It is possible to calibrate a model quantitatively by assessing error using statistics or qualitatively by manually comparing model results and field data. Qualitative approaches are particularly well suited to spatial models where the researcher is able to compare maps. However, for complicated models with many parameters, a qualitative approach to calibration will probably be very time consuming and unlikely to reveal the optimal model configuration (which is the ultimate aim of calibration). Therefore, this article will focus on quantitative methods of error assessment (rather than relying on human objectivity) and automatic calibration routines that are able to explore a model’s parameter space and estimate error without human intervention. Also, in the context of environmental criminology, it is much more important to accurately reflect field conditions when working with explanatory models – i.e., those that simulate real-world conditions – so the discussion will focus on these in particular. Conceptual models,

G. Bruinsma, D. Weisburd (eds.), Encyclopedia of Criminology and Criminal Justice, DOI 10.1007/978-1-4614-5690-2, # Springer Science+Business Media New York 2014

C

244

Calibration of Simulation Models

on the other hand, will not necessarily attempt to replicate real environmental conditions so automatic calibration methods are less relevant. The article is organized as follows. Section outlines the main principles of calibration and Section follows with a more in-depth assessment of a number of methods for model calibration. Sections and complete the article by discussing the state of the art, literature controversies, and open questions.

Background Calibration refers to the process of configuring a model’s parameters to match some observed historical data. This usually consists of searching for a combination of parameter values that cause the model to produce data which are similar to that collected from the real system under investigation. In many cases, a single “fitness” value is sought which succinctly summarizes the correspondence between simulated data and field observations. Figure 1 illustrates the calibration process. The model is repeatedly reconfigured with the aim of reducing the error between the results and the field data. Once a predetermined error level has been reached, the process ends and the model can be considered calibrated,

i.e., configured in such a way that it is apparently able to simulate the real system effectively. The level of error deemed acceptable is subjective and depends on the individual study. It is also worth noting that simply matching data is not necessarily a sufficient criteria for establishing model correctness, which is a point Section will address in more detail. With simple models, the process of configuring parameters and calculating error is usually relatively simple. However, simulation models are often extremely complex and can contain a large number of configurable parameters. To confound the situation, model parameters often have nonlinear effects on the model’s behavior which makes it difficult to predict how the model will behave under new parameter configurations. Hence, the process of manipulating a model’s parameters to match some field conditions is often nontrivial. To being with, the means of evaluating the degree of similarity between model results and expected data can be troublesome in itself. Once a quantitative measure of similarity has been developed, there are numerous methods that can be used to explore the parameter space of the model in search of the optimal configuration. These are also known as optimization methods. The following section will first discuss the

Start

Run Model

Observed Data

Adjust model parameters

No

Genera te data

Compare Data and Calculate Error

Error sufficiently low? Yes Stop

Calibration of Simulation Models, Fig. 1 An overview of the process of calibration

Simulated Data


245

different methods that can be used to calculate error and follow with a review of a number of automatic calibration/optimization routines.

Using the same notation, R2 can be defined as P i R2 ¼ 1 P

Methods for Model Calibration

C

0

yi yi

2

ðyi yÞ2

(2)

i

C Assessing Goodness of Fit In the context of calibration, goodness of fit (GoF) is an important measure. It refers to the error between the results of a model and the data that it is trying to replicate (hereafter referred to as “observed data”). Commonly, GoF statistics are applied to tabular data, such as the number (or rate) of crimes against different types of people, within different spatial areas or during a particular time period. Knudsen and Fotheringham (1986) experimented with a number of goodness-of-fit statistics and found the standardized root mean square error (SRMSE) to be the best performing. A drawback with SRMSE, however, is that the value itself is difficult to understand. For example, with the SRMSE it is not possible to state what percentage of the variation in observed data can be accounted for by a model. An alternative statistic, R2 solves this problem because it represents the percentage of agreement between the model and the expected data. However, R2 is insensitive to the overall amount of error, predicting a good fit in some circumstances where the SRMSE would not (Harland 2008). The SRMSE can be defined as

SRMSE ¼ 0

rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ffi 2 0 S yi yi =n y

(1)

where yi is the predicted value at matrix point i, yi is the actual value at i, y¯ is the mean value of the predicted values (y0 ), and n is the total number of values. The lower limit of the statistic is 0 which indicates no difference between the predicted 0 values (yi ) and the observed values (yi). The upper limit is usually 1 (Knudsen and Fotheringham 1986) but can be greater, particularly when matrices are sparse (Harland 2008).

and a value of 1 indicates identical data sets. The lower limit of the statistic is 0. There are a variety of alternative GoF statistics that could be used to assess model error and the most appropriate will depend on the data and application area. For more information about the statistics discussed here, the interested reader can refer to the comprehensive assessment in Knudsen and Fotheringham (1986). Alternatively, there are a wide range of textbooks that define methods for both parametric and nonparametric data. In terms of calibration, the most important decision is to choose the GoF statistic that is appropriate to the study and the nature of the data so that a reliable assessment of error can be made. In addition, more than one statistic could be used simultaneously to provide a more comprehensive assessment of model error. Goodness of Fit for Spatial Data Many simulation models in environmental criminology explore the spatial distributions of crime (Liu et al. 2005; Groff 2007; Hayslett-McCall et al. 2008; Dray et al. 2008; Birks et al. 2012; Malleson et al. 2013). If the simulations work at an aggregate spatial scale (measuring crimes per area), then the procedure for assessing GoF is the same as that for nonspatial data. However, it is preferable for simulation models to use data on individual crime occurrences and, therefore, generate point pattern data. Common GoF measures cannot be used to compare point patterns directly as the data must be in the form of a table or matrix. One solution to this problem is to first aggregate the point data to commonly used areal boundaries such as the enumeration district or census tract. However, this process is far from ideal. Firstly, such aggregation will expose the results to the modifiable areal unit problem (MAUP: Openshaw 1984). Openshaw found

C

246


that changing the size and shape of the boundaries themselves can have a dramatic effect on the resulting spatial patterns and subsequent results. Secondly, the process of aggregating is likely to hide important patterns that are present at finer geographies. Andresen and Malleson (2011), for example, showed that there was considerable spatial heterogeneity in crime rates at the street level which would be hidden at larger spatial scales. Hence aggregation will inevitably introduce error. An alternative approach is to compare the point patterns directly without aggregating. Although no commonly accepted methods exist

for this purpose, such as the SRMSE or R2 there are a a number of useful spatial statistics that can describe the distributions of point patterns and can be used to make mathematical comparisons. These could replace, or compliment, traditional GoF statistics used to determine model error during calibration. Table 1 summarizes some of these statistics and illustrates the results of their application to three data sets: two similar point patterns produced by a simulation model (“Model1” and “Model2”) and a point pattern produced by a random process (“Random”). The functions outlined in Table 1 provide information about the degree of clustering in

Calibration of Simulation Models, Table 1 A summary of spatial statistics that can be used to describe and compare the spatial structure of point patterns (Malleson 2010) Pros/cons Gives a concise general picture of whether or not clustering is present (compared to random data) Usefil as a preliminary procedure (Bailey and Gatrell 1995) It is difficult to account for edge effects (there are some solutions to edge effect problems, such as circular or rectangular corrections (Levine 2006), but these are not ideal (Chainey and Ratcliffe 2005)) Too simplistic to be really useful

Usage with example data The NNI statistics suggests that all the data used in the following examples (Modell, Model2 and Random) are clustered which is to be expected (Malleson et al. 2010). Althought the statistic is not comprehensive enough to assess error in isolation, it could be useful as a preliminary measure of similarity as part of a larger error assessment during calibration

The G function, at a given distance, d, is the fraction of points, si, whose nearest neighbour is less than d away: ð4Þ G ¼ #ðdminnðsi Þ 0

Model1 Model2 Random d

point pattern data. Although these statistics have the benefit that they are not susceptible to the modifiable areal unit problem, they could not be used during calibration, in isolation, because it is possible that two different point patterns nevertheless have the same clustering properties. However, these could be used in conjunction with other methods (such as aggregating the points to area boundaries and applying a GoF test) to provide a more comprehensive error assessment. As well as comparing mathematical descriptions of clustering in point patterns, it is also possible to generate raster density maps from the point patterns and compare these mathematically. This approach is common in the field of spatial modeling for comparing simulated and real land use. For a review of recent approaches, the reader is directed to Kuhnert et al. (2005). Finally, it is also possible to aggregate the point patterns to a regular grid and then use traditional GoF statistics on the resulting matrix as discussed by Costanza (1989). This approach has the

advantage of reducing the effects of the modifiable areal unit problem because numerous regular grids can be applied to the point patterns at the same resolution. Exploring the Parameter Space Having determined an effective means of estimating the goodness of fit between simulated and observed data, calibration itself can begin. The task of model calibration is effectively a search through the model parameter space, assessing the accuracy of the model under different combinations of parameters. For simulation models with large numbers of continuous parameters, this search space will be extensive. This is confounded by the fact that simulation models are often nonlinear, so the effects of changing parameter values are not easy to predict. For example, Fig. 2 presents an example parameter space for a nonlinear model with two continuous parameters (A and B). Varying parameter combinations changes the accuracy of

C

C

248


Calibration of Simulation Models, Fig. 2 An example parameter space for a nonlinear model with two parameters

Global optima Local maxima

rB

Global minima Param

eter A

the model results; hence the aim of the calibration process is to find the optimal parameter configurations (“global optima”). However, even for a model with a low number of parameters, it becomes apparent that a manual search of the parameter space is unlikely to reveal global optima. Therefore calibration is often conducted by computer algorithms which are able to intelligently search the vast parameter space. Parameter Sweeps

A parameter sweep is a simple process of systematically varying the model parameters in sequence so that many possible combinations are explored. Each parameter to be tested has a start, end, and increment value which determines the range of possible values that the parameter will take. Again using Fig. 2 as an example, a parameter sweep with start ¼ 1, end ¼ 10, and increment ¼ 1 for both parameters A and B results in 100 possible parameter combinations. Although the method is simple, it has a number of drawbacks, namely: • The number of individual runs can be extensive because it increases exponentially with the number of parameters. For example, to conduct a parameter sweep on a model with five parameters, each of them integers between 1 and 10, a total of 105 ¼ 100,000 model runs will be required.

ete

m ara

P

• The sweep explores the entire parameter space equally. More advanced approaches are able to expend greater effort fine tuning already successful configurations and ignore parts of the parameter space that are extremely unlikely to reveal global optima. Hill Climbing

Hill climbing is a procedure that starts with an arbitrary model configuration and makes a small change to one of the parameters. If the change improves the performance of the model (reduces error), then it is retained. If the change increases error, then it is discarded. This process repeats until there are no possible parameter changes that improve the model. The main drawback with the approach is it is likely to get stuck on suboptimal configurations (e.g., the local maxima in Fig. 2), in which case it would be necessary to temporarily accept changes to the configuration that will actually increase the error between the model and the calibration data. Simulated Annealing

Annealing is the process used in metallurgy in which a material is heated and then cooled in order to alter its properties (strength, brittleness, etc.). The amount of time spent at a high temperature and the rate of cooling influence the resulting properties of the material. The


C

249 Criteria for acceptance

1.5 t=1 t=2

P

1.0

C 0.5

0.0 0

1

2

3

4

Change in fitness

Calibration of Simulation Models, Fig. 3 The effect of temperature on the probability of accepting a regressive move

simulated annealing optimization procedure takes its inspiration from an algorithm that simulates this annealing process (Metropolis et al. 1953). Originally formulated by Kirkpatrick et al. (1983), the procedure improves upon hill climbing algorithms by occasionally allowing choices that lower the fitness of the model. Whereas hill climbing will always choose the best move from those available, simulated annealing chooses a random move from the neighborhood. If the move improves the fitness, then it is always accepted, but if it does not, then there is still a possibility that it will be accepted. This helps the algorithm to climb out of local maxima. The criteria for accepting a lower fitness is given by P ¼ ec=t > Rð0; 1Þ

(7)

where c is the change in fitness (negative for an improvement, positive for deterioration), t is the temperature, and R(0,1) is a random number in the range 0–1. The temperature is used to reduce the chance of accepting poor moves over time so that the algorithm will converge. This is analogous to gradually reducing the temperature in the annealing process. If t ¼ 0 then only improvements to the fitness will be accepted which causes the algorithm to behave like a hill climbing

procedure. Figure 3 illustrates the change in P for a range of moves under two different temperatures. Note that if the change is positive (x < 0), then P > 1 so the move will always be accepted. As t decreases, so does the probability of accepting a move. The rate at which t drops, as well as its initial and final values, will determine how successfully the algorithm will run. Unfortunately there are no standard rules for determining these values, although a number of methods for estimating suitable values have been proposed (Reeves 1995). Genetic Algorithms

A Genetic Algorithm (GA) is a form of evolutionary algorithm, based on Darwin’s theory of natural selection (Darwin 1859). The algorithm works on the premise that small variations in organisms can accumulate if they induce an increase in the overall fitness and this improves the individuals’ ability to reproduce (Reeves and Rowe 2003). When describing GAs, the following concepts are important: • A gene is a single model parameter. • A chromosome is a combination of genes, i.e., a unique model configuration. • The population is the current set of chromosomes that the algorithm is using to explore the parameter space.

C

250


Randomly generate initial chromosomes

Optimal fitness reached?

No

Selection Select the best chromosomes to parent the new generation

Recombination Combine parent genes to create new chromosomes

Mutation Randomly mutate some genes on some chromosomes

Yes

Return the best chromosome(s)

Calibration of Simulation Models, Fig. 4 The process of running a genetic algorithm

• Selection is the process of choosing the fittest chromosomes. • Recombination is the process of combining the best chromosomes to initialize the new population. • Mutation is a means of exploring a wider area in parameter space by randomly varying some genes in a new population. The GA is an iterative process, as illustrated by Fig. 4. At each iteration, the algorithm finds the fittest chromosomes (the model configurations with lowest error) and uses these to create the population in the subsequent iteration. The algorithm runs for a set number of iterations or until an acceptable goodness-of-fit value has been found. Although the process of running a GA does not change substantially across implementations, the means of performing selection, recombination, and mutation can be adapted depending on the characteristics of the parameter space. See Reeves and Rowe (2003) for more information. The main advantage of a GA over the other procedures is that it is able to “home in” on the space with the fittest parameter configurations without devoting effort to exploring suboptimal areas. This makes it much more efficient. Also, mutation allows the algorithm to explore a wider search space which can prevent it from becoming trapped in local maxima.

Calibration of Criminology Models On the whole, very few simulation models in the field of criminology apply the techniques outlined here. This is partly because simulation modeling has been, until recently, relatively underused in criminology so there is only a limited literature base to begin with. Also, and perhaps more importantly, a large number of studies are conceptual rather than predictive. Conceptual models do not attempt to replicate real-world crime patterns and instead explore the dynamics of criminology theory in an abstract, artificial environment. Hence it is normally neither necessary nor possible to configure these models to simulate data from the real world. Predictive models, on the other hand, do attempt to replicate real-world patterns and therefore calibration should be an important part of the modeling process. However, of the limited relevant published studies, very few have applied the formal methods outlined here. For example, Malleson et al. (2013) implement a simulation model of residential burglary and, although automatic calibration is noted as advantageous, the authors result to manual parameter configuration because the extensive run time of the model leads to in insurmountable computational requirements for automated calibration algorithms (Malleson et al. 2012). Similarly, Groff (2007) discusses the


advantages of a “common sense” approach to model evaluation by examining factors such as the degree of clustering and the spatial dynamics of hotspots. Again, formal calibration methods such as those outlined here are not applied, although the authors stress that the field is “wide open” (Groff 2007, p. 99). There are numerous other predictive simulation models that would benefit from a rigorous approach to calibration but, for a number of reasons, calibration is not applied with the same rigor as with simulation models in other fields. Partly, this will be because the methods discussed here are not simple and require a degree of computer literacy. But, also, it is due to the novelty of the models to the field and, with time, it is extremely likely that calibration, as well as the other elements to model evaluation, will form a more substantial part of the modeling process.

Controversies and Open Questions This entry has made reference to a number of calibration methods that can be used to calibrate simulation models. Other methods are available, including simply adjusting parameters manually. But those discussed here are the most suitable for models of complex systems which often have a large number of parameters, behave nonlinearly, and consume/produce large amounts of data. However, unless reliable observed data can be gathered and methods developed to compare the data to simulation results, the process of calibration will not be able to improve model performance (at least in the sense that the model is a good representation of the real world). The first problem is how to obtain reliable, real-world data on which to build a picture of the underlying system. Police recorded crime data is a common source but has a number of drawbacks: • Discrepancies in police recording practices mean that some crimes, which are heavily underreported, will be misrepresented in police data – although this is somewhat mediated by the assertion that unreported crime clusters near reported crime (Chainey and Ratcliffe 2005).

251

C

• The temporal accuracy is often questionable because the actual time of the event is not always known. • Spatial accuracy can be variable if a location is hard to code accurately (e.g., somewhere in a park) or if human error corrupts the recorded location. • Offender data, which can be useful for the calibration of the spatial movements of offenders, by definition only provides information about people who have been in contact with the police and therefore misrepresents the actual population of offenders. Assuming good-quality crime data are available, it is still not certain that those data are the most suitable for use in calibration. Firstly, many models might be able to recreate the observed crime patterns, but this does not guarantee that any of them correctly represent the internal dynamics of the system. This is known as the identifiability problem and one that is common to all modeling approaches – see, for example, the discussion in Windrum et al. (2007). To determine which, if any, are “correct,” it might be necessary to simultaneously calibrate against various different data sources that capture elements of the system other than simply the crimes committed. Examples might include the use of social surveys to represent victim behavior, crowdsourced data to explore “normal” day-to-day behavior patterns or transport data to estimate the routes that people use to navigate cities. The “correct” model will fit the patterns illustrated by these data and closely approximate the observed crime data. Although there is huge scope for improving the calibration of criminology simulation models, relatively little has been done in practice. This is not, however, unexpected. The methods employed are still in their infancy, relative to their traditional mathematical counterparts at least, so it takes some considerable effort to develop a model in the first place. However, as the methods become more widely used and the tools to develop them become easier to manage, there is no reason that standard, widely adopted approaches to calibration cannot emerge.

C

C

252

Related Entries ▶ Agent-Based Models to Predict Crime at Places ▶ Spatial Models and Network Analysis

Recommended Reading and References Andresen M, Malleson N (2011) Testing the stability of crime patterns: implications for theory and policy. J Res Crime Delinq 48(1):58–82 Bailey T, Gatrell T (1995) Interactive spatial data analysis. Prentice Hall. http://www.amazon.com/InteractiveSpatial-Analysis-Trevor-Bailey/dp/0582244935 Birks D, Townsley M, Stewart A (2012) Generative explanations of crime: using simulation to test criminological theory. Criminology 50(1):221–254 Chainey S, Ratcliffe J (2005) GIS and crime mapping, 1st edn. Wiley, Chichester Clark PJ, Evans FC (1954) Distance to nearest neighbour as a measure of spatial relationships in populations. Ecology 35:445–453 Costanza R (1989) Model goodness of fit: a multiple resolution procedure. Ecol Model 47:199–215 Darwin C (1859) On the origin of species by means of natural selection, or the preservation of favoured races in the struggle for life. John Murray, London Dray A, Mazerolle L, Perez P, Ritter A (2008) Policing Australia’s ‘heroin drought’: using an agent-based model to simulate alternative outcomes. J Exp Criminol 4(3):267–287 Groff ER (2007) Simulation for theory testing and experimentation: an example using routine activity theory and street robbery. J Quant Criminol 23(2):75–103 Harland K (2008) Journey to learn: geographical mobility and education provision. PhD thesis, School of Geography, University of Leeds, Leeds Hayslett-McCall K, Qui F, Curtin KM, Chastain B, Schubert J, Carver V (2008) The simulation of the journey to residential burglary. In: Liu L, Eck J (eds) Artificial crime analysis systems: using computer simulations and geographic information systems. Information Science Reference, Hershey Kirkpatrick SS, Gelatt CDC, Vecchi MPM (1983, May) Optimization by simulated annealing. Science 220(4598):671–680 Knudsen DC, Fotheringham AS (1986) Matrix comparison, goodness-of-fit, and spatial interaction modeling. Int Reg Sci Rev 10:127–147 Kuhnert M, Voinov A, Seppelt R (2005, August) Comparing raster map comparison algorithms for spatial modeling and analysis. Photogramm Eng Remote Sens 71(8):975–984 Levine N (2006) Crime mapping and the crimestat program. Geogr Anal 38:41–56

Cambridge-Somerville Youth Experiment Liu L, Wang X, Eck J, Liang J (2005) Simulating crime events and crime patterns in RA/CA model. In: Wang F (ed) Geographic information systems and crime analysis. Idea Publishing, Reading, pp 197–213 Malleson N (2010) Agent-based modelling of burglary. PhD thesis, School of Geography, University of Leeds, Leeds Malleson N, Heppenstall A, Evans A, See L (2010) Evaluating an agent-based model of burglary. Working paper 10/1, School of Geography, University of Leeds, Leeds. http://www.geog.leeds.ac.uk/fileadmin/ downloads/school/research/wpapers/10_1.pdf Malleson N, See L, Evans A, Heppenstall A (2012) Implementing comprehensive offender behaviour in a realistic agent-based model of burglary. Simul Trans Soc Model Simul Int 88(1):50–71 Malleson N, Heppenstall A, See L, Evans A (2013) Using an agent-based crime simulation to predict the effects of urban regeneration on individual householdburglary risk. Environ Plan B Plan Des 1–28. http://www. envplan.com/abstract.cgi?id=b38057 Metropolis N, Rosenbluth AW, Teller E, Rosenbluth MN, Teller AH (1953) Equation of state calculations by fast computing machines. J Chem Physica 21(6):1087–1091 O’Sullivan D, Unwin DJ (2003) Geographic information analysis. Wiley, Hoboken, NJ Openshaw S (1984) The modifiable areal unit problem. Concepts and techniques in modern geography. Geo Books, Norwich Reeves C (1995) Modern heuristic techniques for combinatorial problems. McGraw-Hill, London Reeves CR, Rowe JE (2003) Genetic algorithms – principles and perspectives: a guide to ga theory. Kluwer, Dordrecht Rykiel EJJ (1996) Testing ecological models: the meaning of validation. Ecol Model 90:229–244 Windrum P, Fagiolo G, Moneta A (2007) Empirical validation of agent-based models: alternatives and prospects. J Artif Soc Soc Simul 10(2)

Cambridge-Somerville Youth Experiment Sonia Jain and Alison K. Cohen Health and Human Development Program, WestEd, Oakland, CA, USA

Synonyms Cambridge-Somerville youth study

Cambridge-Somerville Youth Experiment

Overview The Cambridge-Somerville Youth Experiment was a longitudinal study of 650 boys that began in 1935 in Cambridge and Somerville, two neighboring towns in the Boston, Massachusetts, metropolitan area. It was the first documented criminology experiment of that scale (Farrington 2003). A treatment group of at-risk “difficult” boys and control group were created, each with 325 boys assigned based on 160 matching characteristics. The treatment group received social welfare services including family guidance, medical and academic assistance, coordination of community agencies, and recreational services, provided by a friendly counselor sustained for almost 10 years; the control group received no special services. The goal of the study was to examine if delinquency could be prevented and if counseling was an effective approach to preventing delinquency (Powers and Witmer 1951). However, results were mixed, with no crime prevention benefits observed (McCord and McCord 1959), and it has largely been described as both well-intentioned and a failure (Dodge 2001).

Fundamentals Participants Boys attending public and parochial schools in Cambridge, Massachusetts, and public schools in Somerville, Massachusetts, who were age six or seven at enrollment were eligible for participation in the study (Powers and Witmer 1951). At the time, Cambridge and Somerville were primarily working-class, factory towns (McCord and McCord 1959). “Difficult” boys were identified by schools and social service agencies in the two towns and referred to the study team (Powers 1949). The selection process was lengthy; it involved interviewing hundreds of boys that had been identified as difficult, conducting medical and psychological exams and home visits, and completing a camping trip in which detailed notes were taken about each boy (Cabot 1940; McCord and McCord 1959). This initial process happened without potential participants’

253

C

knowledge, but upon being contacted, the majority were willing to participate (Powers 1949). Participants were solicited beginning in 1935, and participants were enrolled between 1937 and 1939 (Powers and Witmer 1951). Study Design The study was designed by Dr. Richard C. Cabot and funded by the Ella Lyman Cabot Foundation, which was created in honor of Dr. Cabot’s wife. The study was originally designed to be implemented over the course of 10 years, although this plan was curtailed by World War II and lasted for just over 8 years (Powers 1949; Powers and Witmer 1951). Additionally, some boys dropped out sooner, and other boys were intentionally dropped from the sample after 2–3 years if they were definitively classified as nondelinquent (Powers 1949). The study design was a randomized matched study: the two boys closest to each other in terms of approximately 160 background characteristics were paired, and then one of them was assigned to the intervention group, and the other was assigned to the control group on the basis of a coin flip (Cabot 1940; McCord and McCord 1959). These pairs were categorized as likely to be delinquent, not likely to be delinquent, or future delinquency trajectory uncertain (Cabot 1940). Intervention The structure of the intervention was to pair each boy in the treatment group with a trained counselor, who would then determine what services were most appropriate for each boy’s unique situation with an eye towards positive character development (Cabot 1940). The trained counselor used social welfare practices, including health, educational, family, and community assistance, guidance, and support; boys on average received five such years of counseling (McCord and McCord 1959). However, the quality of the intervention varied in terms of frequency of contact between the counselor and student, length of time the boy participated in the program, the gender of the counselor, and counselor turnover (McCord and McCord 1959). The most intensive intervention was defined as weekly participation for at least

C

C

254

2 years, in which the counselor and boy had a close relationship that included discussing personal issues; however, only 12 of the boys met these criteria, and so potential analyses were limited but illustrative that this more intense level of treatment may be effective (McCord and McCord 1959). Most boys had done recreational activities with their counselors, approximately half received tutoring, and medical care and summer camps were also common among boys in the intervention group (McCord 2003). Some boys in the intervention group were sent to foster homes and/or private schools per counselors’ recommendations (Powers 1949), although this was done after attempting more moderate services (McCord et al. 1960). In comparison, the control group received “the usual services of the community” (McCord and McCord 1959). Outcomes of Interest The primary outcome of interest was criminal behavior (as measured by criminal convictions) (McCord and McCord 1959), although additional outcomes were observed in follow-up studies. Other outcomes of interest included mental hospital visits, participation in alcoholic treatment, and vital statistics (e.g., marriage, death) (McCord 1978). Anecdotally, individuals and counselors credited the program with changing boys’ developmental trajectories (Powers 1949). Key Findings Although boys who participated in the intervention qualitatively credited the program to helping them in their development, there were null and/or harmful effects of the intervention for a range of quantitative outcomes, including criminal and health outcomes (McCord 1978, 2003). For example, there were no statistically significant differences between the intervention and control groups in criminal convictions (both ever being convicted and number of convictions) in their matched analyses and after controlling for additional potential confounders that had been identified more recently (McCord and McCord 1959). After a series of subgroup analyses, though, they found that boys who had met with their adult

Cambridge-Somerville Youth Experiment

counselor once a week on average and boys who were younger when they enrolled in the program had statistically significantly lower incidences of criminal convictions compared to their counterparts (McCord and McCord 1959). In later follow-up studies, boys who received the intervention were more likely to have been convicted of “serious street crimes,” died at an earlier age, and been diagnosed with a mental illness, compared to the controls (McCord 2003). Some of the programming characteristics that were later attributed to the study’s failure include that the boys were publicly identified as at risk and were given resources without being held accountable, and the boys also participated in activities together (e.g., summer camps) (Dodge 2001). Others noted that adverse effects associated with the intervention were most apparent for boys who participated in the intervention for the longest amount of time, with the most frequent visits, and with family cooperation (McCord 2003). Additional studies took advantage of the matched pairs to consider subgroups of boys. For example, the boys who were in the intervention group and placed in foster care were compared to their matched pairs, and harmful effects of foster care placement were observed (McCord et al. 1960). Since the Cambridge-Somerville Youth Experiment, several other programs have been rigorously evaluated, and other delinquency prevention program evaluations have also observed less-than-positive impacts, indicating that more research must be done and published in the academic literature, regardless of positive or negative results (McCord 2003). The CambridgeSomerville Youth Experiment has also inspired the development of new theories, like deviancy training, to explain what may have occurred then; these theories have since been applied to and tested in new contexts (Gottfredson 2010).

Conclusion The Cambridge-Somerville Youth Experiment is considered the first major randomized

Campus Crime

experiment in criminology. It is commonly used to exemplify the importance of rigorous evaluation of well-intentioned programs, including use of evaluation to determine if the positive results desired are actually observed.

255

C

Camera Surveillance ▶ CCTV and Crime Prevention

C Related Entries

Campus Crime ▶ Randomized Experiments in Criminology and Criminal Justice

Recommended Reading and References Cabot PSD (1940) A long-term study of children: the Cambridge-Somerville youth study. Child Dev 11(2):143–151 Dodge KA (2001) The science of youth violence prevention: progressing from developmental epidemiology to efficacy to effectiveness to public policy. Am J Prev Med 20(1S):63–70 Farrington DP (2003) A short history of randomized experiments in criminology: a meager feast. J Adolesc Res 27(3):218–227. doi:10.1177/ 0193841X03027003002 Gottfredson DC (2010) Deviancy training: understanding how preventive interventions harm. J Exp Criminol 6(3):229–243. doi:10.1007/s11292-010-9101-9, The Academy of Experimental Criminology 2009 Joan McCord Award Lecture McCord J (1978) A thirty-year follow-up of treatment effects. Am Psychol 33:284–289 McCord J (2003) Cures that harm: unanticipated outcomes of crime prevention programs. Ann Am Acad Polit Soc Sci 587(1):16–30. doi:10.1177/0002716202250781 McCord J, McCord W (1959) A follow-up report on the Cambridge-Somerville youth study. Ann Am Acad Polit Soc Sci 322(1):89–96. doi:10.1177/ 000271625932200112 McCord J, McCord W, Thurber E (1960) The effects of foster-home placement in the prevention of adult antisocial behavior. Soc Serv Rev 34(4):415–420 Powers E (1949) An experiment in prevention of delinquency. Ann Am Acad Polit Soc Sci 61:77–88 Powers E, Witmer H (1951) An experiment in the prevention of delinquency: the Cambridge-Somerville youth study. Columbia University Press, New York

Cambridge-Somerville Youth Study ▶ Cambridge-Somerville Youth Experiment

John J. Sloan III1 and Bonnie S. Fisher2 1 Department of Justice Sciences, The University of Alabama at Birmingham, Birmingham, AL, USA 2 School of Criminal Justice, University of Cincinnati, Cincinnati, OH, USA

Overview During the past two decades, campus crime has captured the attention of both the public and researchers who sought to unpack the causes, correlates, and consequences of various types of violent and property crime committed on college and university campuses. Much research to date has focused on identifying how students’ lifestyles and daily routines create opportunities for on-campus victimization. Other studies have examined college campuses as places and explored how their features create opportunities for crime. This entry examines both lines of research and synthesizes the extant literature to describe the dynamics of campus crime. The entry also presents some of the controversies within the study of campus crime, including measurement issues, overreliance upon case studies of single campuses, and the problem of underreporting of crimes occurring on college campuses. The entry concludes with a brief discussion of open questions in the study of campus crime, including the question of the extent of campus crime, how to better translate research into policies or programs aimed at reducing campus crime, and the ultimate utility of legislation at the federal and state level designed to reduce campus crime.

C

256

Introduction America’s colleges and universities present an apparent anomaly: physically attractive places fostering the intellectual development of young adults that may, in reality, be dangerous places beset by violence, vice, and victimization. More specifically, the seemingly safe appearance of college campuses has been coupled with a conventional view held by many campus administrators, students, and their parents that campus violence, vice, and victimization are merely “youthful indiscretions” resulting from students’ newfound personal freedoms, including sexual promiscuity, alcohol consumption, and recreational use of illegal drugs. These views, however, changed during the 1990s when four groups – a student advocacy group called Security On Campus, campus feminists, campus crime victims and their families, and public health researchers – made claims that crime on US college campuses was not only increasing but becoming more serious and deadly in its consequences. Collectively these groups convinced the public, Congress, and several state legislatures that the “dark side” of the ivory tower was a “dangerous place” rife with “heinous crimes” that not only undermined students’ safety and well-being but also threatened the core educational mission of higher education. As a result, during the early 1990s, Congress and multiple states passed statutes designed to address campus crime. For example, Congress passed the Student Right to Know and Campus Security Act of 1990 (20 USC 1092[f]) (now known as the Jeanne Clery Disclosure of Campus Security Policy and Campus Crime Statistics or the Clery Act) which mandates that all Title IV schools (those eligible to participate in federal financial aid) not only publicly report their crime statistics in an annual report but also to implement policies and programs designed to address campus crime, in particular sexual assault against college coeds (Sloan and Fisher 2011). Additionally, nearly one-third of the states passed campus crime-related legislation with wide-ranging requirements (Sloan and Shoemaker 2007).

Campus Crime

Looking at 2009 campus crime statistics compiled by the United States Department of Education (2011a) in fulfillment of the Clery Act, there were over 47,000 violent and property crimes reported on college campuses in the United States. The total included 31 murders and non-negligent manslaughters; over 3,300 forcible and non-forcible sex offenses; more than 4,600 robberies; approximately 4,900 aggravated assaults; more than 26,000 burglaries; over 7,200 motor vehicle thefts; and more than 700 incidents of arson. While the magnitude of these numbers is alarming, to put them into perspective, in 2009, over 20 million students were enrolled at 6,883 Title IV designated postsecondary institutions that were required to file annual campus crime reports with the federal government (United States Department of Education 2011b). As the public began perceiving campus crime as a “serious” problem, a number of researchers started exploring the patterns, correlates, and causes of campus crime. During the past 20 years, this body of research can be classified as having two different, yet complementary, foci: studies whose primary interest was describing and explaining patterns of victimization relating to individuals and studies whose interest was understanding how the characteristics of college campuses such as physical layout and geography explained victimization patterns. In the former instance, the studies examined students’ victimization patterns by different types of personal and property crime, including recent analyses of cyber-based victimizations suffered by students. Identifying individual-level behavior that apparently increased the risk for victimization led researchers to examine how students’ lifestyles and daily routines played a significant role in increasing (decreasing) their risk of victimization. In the latter instance, researchers adopted a perspective which emphasized the college campus as a specific place with distinctive physical and social characteristics that contributed to victimization. Social scientists – including sociologists, criminologists, and geographers – adopting this perspective argued that a criminology of place is not only possible, but is

Campus Crime

a valuable theoretical and practical lens through which researchers and crime prevention practitioners alike can examine and understand campus crime and, more importantly, take steps to prevent its occurrence (Eck and Weisburd 1995). This entry discusses the interrelationship between the characteristics of colleges and universities and students’ lifestyles and daily routines that provide ample opportunities for on-campus personal and property victimization. Campuses are unique places characterized by a distinctive, yet somewhat stereotypical, built environment that helps to structure the daily routines of places and ultimately influences students’ daily lifestyles and ensuing crime patterns. As discussed in more detail below, the patterns of the campus environment-student lifestyle nexus enhance opportunities for different types of crimes to occur more frequently than others. The entry begins by discussing the notion of the college campus as a “place.” Relevant literature is then presented that addresses how the characteristics and daily routines of places either may facilitate or retard crimes occurring at or near them. Next, what is known about the relationship between the characteristics of college campuses and crime occurring there is presented. The entry concludes by exploring important controversies and questions that remain concerning campus crime, especially those arising from the perspective of the criminology of place.

Background: The College Campus as a Place To draw a detailed portrait of offenses occurring on campuses, one must first understand that these events occur at a specific place. Further, one must understand that the characteristics and daily routines of places are not simply physical locations where illegal activities occur. Rather, one must appreciate that the characteristics and routines of a place provide important ingredients necessary to create opportunities for victimization. Geiryn (2000, pp. 464–465) described a place – such as a college campus – as possessing

257

C

several salient features. First, a place occupies geographic location (a “unique spot in the universe”) the size of which can vary tremendously. Many land-grant public universities occupy a dozen or more square miles of space, while so-called virtual universities exist in a single building whose carbon footprint is several 1,000 sq ft. Places, such as college campuses, also possess material form “[a] compilation of things or objects” (Gieryn 2000, p. 465). When people describe a particular college campus to others, they may do so by commenting on the particular architectural features the campus possesses such as a large bell tower atop the student union or in terms of natural features such as being adjacent to mountains or having a stream running through it. Places like college campuses also have physicality to them, including geographic markers (e.g., logo signage; staffed entry gates; the hallmark student union) that uniquely characterize a place’s physical space and functions, and signal to visitors that they have entered a place separate and distinct from the adjunct community. The physicality of a college or university campus is important because “. . . social processes happen through the material forms” that comprise the place (Gieryn 2000, p. 465). On campuses, daily interactions occurring among students, faculty, staff, and visitors take place both inside and outside of offices, classroom buildings, residence and dining halls, stadia, libraries, pathways, streets, parking structures, and green spaces. These interactions are the result of lifestyles or daily routines that each of these campus community members takes on as part of their respective purpose or role while on campus. These interactions and lifestyles – which are structured by physical spaces and their layout – create opportunities for a variety of violent (e.g., rapes, assaults) and property (e.g., automobile thefts, burglaries) victimizations to occur. Finally, a college campus is invested with meaning and value. When individuals walk by a set of buildings consisting of a large residence hall, a library, a classroom building, and green space where young adults are throwing Frisbees, reading books and connecting to the

C

C

258

internet, lounging in the sun, or playing touch football, most would not think “oh, that is a prison.” Some scholars have argued that the meanings created by and resulting from the material form of a place prompt humans to think of these places as communities: locations where people with similar interests purposely gather to interact with one another (e.g., Smith 2002). Annually, college campuses draw groups of individuals who share interests. Researchers are drawn to college campuses because they house needed facilities and resources. Faculty members are drawn to the campus as a place where they can teach young adults and not only help shape their minds, but train them to enter an increasingly complex and competitive workforce. Those seeking positions as support staff are drawn for the opportunities the campus presents to find meaningful employment. Ultimately, those drawn to the campus – despite disparate reasons for arriving there – interact with one another and create a community (Bromley 2007).

The Routines of Place and Their Patterns Places also have identifiable routines or rhythms that vary by time of day, day of the week, and season of the year which are shaped by both the built environment – manmade structures and features – and the natural environment. The routines of place include not only the movement of people into or away from it at regular intervals during the day but also the myriad of interactions that are facilitated by the features of the place. In effect, places have patterns associated with them. Brantingham et al. (1995) have suggested that the patterns or routines that characterize college campuses can be conceptualized as occurring at different levels of spatial analysis. At the macro level, campus routines include vehicular and pedestrian traffic coming onto or exiting the campus and students moving en mass across the campus as one set of classes end and another begins from early morning to late evening. At the meso level, a particular building – for example, the student union – has its own routines and patterns which vary by time of day, day of the

Campus Crime

week, and season of the year. For example, the student union may be overrun with students during meal times each day during an academic term, but relatively quiet at other times of day when students are studying in their residence halls or in the library. Students’ presence at the union on weekends may be reduced when visitors come to enjoy concerts, movies, or lectures. On some weekends, especially during final exams or the summer, the union may be essentially vacant of both students and visitors. Finally, at the micro level are the routines associated with a single room within a building such as “Dr. Jones’ laboratory” or “Ms. Smith’s dorm room.” Again, variation in the activities occurring in or around the lab or dorm room takes place according to time of day, day of the week, or academic term. The lab might be very busy in the morning but empty at night while just the opposite might be true of the dorm room. At the beginning (or end) of a term, the lab and the dorm room would be a beehive of activity, while during various breaks in the academic year, each may be occupied by only a handful of students. Conceptualizing the routines of place as occurring at different levels can be useful to understanding the crime patterns of a place, including college campuses. On a college campus, identifying macro level crime patterns helps to understand the “big picture” of crime at that particular place. What is the most commonly occurring offense on the campus? Are there particular areas of the campus where crime tends to be concentrated – socalled hot spots? If so, where are these located? At the meso level, crime patterns can also be identified for a much smaller part of the larger place. How many rapes or burglaries occurred in the undergraduate residence halls during the fall semester? Which parking garage had the most auto break-ins? Finally, at the micro level, one can again assess patterns that may be associated with a single residence hall room, laboratory, office, or classroom. For example, several years ago at UAB, a particular computer laboratory in one of the older buildings on campus was repeatedly broken into over the course of several months and most of the equipment was stolen. By analyzing evidence at the site of the break-ins, the

Campus Crime

campus police learned the thefts were generally confined to very early morning hours, that homeless people had been seen climbing through unlocked windows into the building, and that the security devices protecting the equipment had been disabled from the inside. The patterns associated with the routines of place have both theoretical and practical importance for understanding crime and developing sound situational interventions to address it. Research has consistently shown, for example, that crime is not randomly distributed either spatially or temporally but rather tends to cluster by place and in time (Eck and Weisburd 1995). When crime clusters at certain locations, these are known as “hot spots” (Sherman 1995). The most basic hot spot is a place or area that has a greater than average number of crime or disorder events, or where targets have higher than average risk of victimization. Places include specific addresses, street corners, or other small locations that can be seen by a person, such as a house, convenience store, or drug dealing location. As a result of the groundbreaking “hot spots” work by Sherman and others over the past several decades, combined with developments in spatial and temporal analysis of crime, criminologists developed a new theoretical orientation that instead of focusing on individuals and their behaviors focuses on places – the “criminology of place” – to underscore that opportunities for crime are shaped by the characteristics of places, individuals’ lifestyles and daily routines, and time of day. Further, places – like individual offenders – have “careers” associated with them including onset, continuance, specialization, and desistance (Sherman 1995). A particular hot spot was not always “hot.” It develops over time based on how its features – both physical and symbolic – and routines draw prospective offenders and targets to it (onset). The place continues as a hot spot for some period (continuance), which may be years, and may disproportionately involve certain crimes such as armed robbery (specialization). Finally, because of direct police intervention, neighborhood changes, or other factors, the hot spot desists as a place disproportionately responsible for crime (desistance).

259

C

On a practical level, understanding the dynamics of place including how they are organized for crime to occur and how the routines of place at the macro, meso, and micro level bring together would-be offenders and potential targets can inform primary prevention measures to reduce the probability of crime occurring. Strategies such as target hardening, changing the physical design features of a place, having individuals change their daily routines (e.g., not carrying large amounts of cash on their person; insuring valued goods stored in a vehicle are not left in the open), or altering their lifestyles (e.g., not getting drunk, not consuming illegal substances) when frequenting certain types of places (e.g., residence halls) can help to change the routines of a place and patterns of behavior and therefore might reduce the opportunity for victimization.

College Campuses and Opportunities for Crime: Institutional, Demographic, and Lifestyle-Routine Characteristics The institutional, demographic, and lifestyle characteristics of colleges and universities create untold opportunities for violent and property victimization for the nearly 20 million students and the millions of faculty and staff, and countless visitors to college campuses each year. Below, these characteristics and their contribution to crime on college campuses are examined.

Institutional and Demographic Characteristics Although the more than 6,000 Title IV 2- and 4-year post-secondary institutions differ widely across their institutional characteristics, researchers have identified a set of characteristics strongly correlated with official crime rates – crimes known to and recorded by the campus police or security (Fernandez and Lizotte 1995; Bromley 1994; Sloan 1994). Schools with large enrollments, for example, have higher rates of on-campus rape, assault, robbery, burglary, and

C

C

260

larceny than schools with smaller enrollments. Campuses with large student enrollments apparently provide more opportunities for victimization because they: (1) possess more targets for would-be offenders, and (2) allow for greater anonymity among potential offenders within the student body. Research also shows that campuses with a large percentage of full-time students have higher official rates of assault, robbery, burglary, and motor vehicle theft. One explanation for this is that full-time students tend to spend more time on campus than part-time students which increases their exposure to potential offenders and provides a larger number of targets from which offenders may choose. Residential campuses – schools that have a large number of students living in on-campus residence halls or apartments – have higher rates of rape and larceny whereas commuter schools have higher rates of motor vehicle theft and robbery. In contrast, rural campuses, on average, have lower rates of all forms of violent (rape, assault, and robbery) and property crime (burglary, motor vehicle theft, and larceny) than do urban campuses. One of the strongest and most consistent predictor of official rates of campus crime is the cost associated with room and board per academic year. Campuses with higher room and board costs generally report higher rates of crime whose motivation is economic, for example, robbery, burglary, motor vehicle theft, and larceny. Students enrolled at “costly” schools might also have plenty of attractive, high-quality targets (e.g., electronic devices and their accessories; designer clothing) to steal with relatively little risk for the offender. Further, research shows that most students while on campus do not routinely engage in proven crime prevention behaviors that reduce their chances of victimization (Fisher et al. 1997). Research further indicates there is a demonstrable relationship between students’ demographic characteristics and campus crime rates (Fernandez and Lizotte 1995; Bromley 1994; Sloan 1994). Since the 1940s, the demographic composition of colleges and universities

Campus Crime

has changed dramatically, with total student enrolled increasing annually. In particular, the number of women and minorities enrolled at post-secondary degree-granting institutions has seen tremendous increases, especially over the past three decades. In 2009, approximately 55 % of the 20 million students enrolled at all postsecondary institutions were women (United States Department of Education 2011c) while some 32 % of students enrolled at degreegranting post-secondary schools in 2009 were racial and/or ethnic minorities (Rigg 2010). Several studies find that the percentage of male students enrolled at a college or university consistently is linked with campus crime rates, such that the more males enrolled, the higher the overall crime rates (Fernandez and Lizotte 1995; Bromley 1994; Sloan 1994). Research also shows that colleges and universities with higher percentages of minority students also have higher rates of violence reported to officials. A similar relationship is found between the percentage of AfricanAmerican students enrolled and rates of campus assault, robbery, and burglary. Understanding the effects of institutional and students’ demographic characteristics allows post-secondary administrators to determine whether campus crime rates at their school are typical for or similar to comparable campuses or due to other factors, perhaps students’ lifestyles and routines. For example, some campuses with large student enrollments may have “naturally” high crime rates. A problem arises when campuses with high crime rates have institutional and demographic characteristics associated with low crime rates. This requires exploring how the physical design features of the campus as a place potentially interact with the lifestyleroutine activities of campus members to see how those interactions provide opportunities for victimization.

Campus Places College campuses occupy finite geographical areas that vary in size. Campuses typically are divided into smaller parcels designated for

Campus Crime

various uses including housing, instruction, research, common areas, entertainment and leisure, etc. These spaces also are usually arranged in a way such that the space utilized for one function does not mix with space designated for another. Collegiate residence halls or apartments for married students, for example, would ordinarily not be located directly adjacent to a football stadium or research laboratories where exotic and potentially lethal viruses are being studied. Rather, residence halls tend to be located near one another and away from other space clusters designed for different uses. Arranging common usage of space on the campus is conducive to creating daily interactions among faculty members, students, staff, and visitors. There may be several such segmented areas on a campus depending on its age and needs, or as new needs arise or priorities change to account for expansion of student and faculty needs. However, for reasons ranging from the utilitarian to the aesthetic, the general pattern is that similarly designated spaces tend to cluster together in space. This segmentation of a campus into areas of designated usage has implications for understanding the spatial and temporal distribution of campus crime. A college or university with several thousand students living on campus will have many residence halls and student apartments to accommodate housing demand. These, in turn, will likely be clustered together and several such clusters will be located around campus. Buildings that primarily contain classrooms or research laboratories will likewise be clustered together, rather than randomly placed, and again may be found in several locations around campus. Parking lots and decks are typically located on the edges of the campus, especially if the campus is “pedestrian friendly” and vehicular access is restricted to the fringes of the campus. Clusters of similarly utilized space provide differential opportunities for offenders. For example, a five-story parking deck is a far more likely candidate for automobile theft or automobile break-ins than is a building housing research laboratories where theft of property, such as laboratory equipment, would be likely more

261

C

common. Offenders may perceive that a cluster of dormitories is a good target for burglary, but less so for crimes against the person such as robbery. A small, isolated parking lot on the fringe of campus may be perceived as a good target for robberies but not for automobile theft, due to the limited number of vehicles actually parked there. Ultimately, spatial distribution and space utilization of a campus has significant implications for target selections by prospective offenders. Additionally, clusters of similarly utilized spaces have their own routines at various points during the day or evening which likely differs from the routines associated with other clusters. During a typical day, there will likely be specific times when clusters of buildings, parking lots and decks, residence halls, etc. are “busy” with students, faculty, or staff coming and going. There are specific times during the day or at night when such spaces are relatively quiet. These “down-periods” during the 24-h cycle of the day present opportunities to offenders since few people are present to serve as guardians for the places that prospective offenders may perceive as attractive targets. Large parking decks or lots while classes are in session likely mean few people would be present, which increases opportunities for vehicular thefts or break-ins. As a result, because of variation in the routines associated with clusters of differentially utilized spaces, prospective offenders are likely to perceive these clusters – and the individuals typically found there – as differently convenient targets. Thus, not only the spatial distribution of spaces and their usage but also the temporal patterns associated with those spaces matter in terms of understanding campus crime.

Campus Routines and Lifestyles The routine, everyday activities of students, faculty, staff, and visitors have significant influences on their chances of becoming a victim of crime while on campus. Although its population is moderately transitory, college campuses also have constant and familiar activities and routines.

C

C

262

The lifestyle-routine activity approach to understanding crime victimization describes how one’s daily activities and lifestyles converge in time and space with a would-be offender to create opportunities for criminal victimization (Cohen and Felson 1979). Four concepts are central to this approach for explaining how and why victimization can occur: attractiveness of the target; proximity of one or his/her property to a motivated offender; guardianship over a person or his/her property; and exposure of his/her property to a potential offender. Understanding how each of these concepts plays out on campus broadens one’s understanding of the mechanisms underlying the occurrence of crime on college campuses. Students and Victimization. Students’ pursuit of education and recreation has them on campus at all hours of the day, every day of the week. They also experience a variety of violent and property victimizations while on campus. Fisher et al. (1998) found that nearly one-fourth of the students in a national sample had been victimized at least once while on campus during the current academic year. Campus victimizations occur in a wide array of situational contexts that are unique to on-campus locations, compared to those that are off campus. For example, Hart and Miethe (2011) have shown that the specific attributes of “minor assaults of male victims in daytime hours” were far more likely to occur in on-campus contexts than off campus. Students who live on campus spend a considerable amount of time in their residence halls. These provide not only shelter and food, but they are also places where daily relationships and routines are established (and broken). Living in a residence hall, especially those that are high-density with little supervision as to who enters or exits, increases opportunities for victimization by increasing the number of suitable targets and exposure of them to motivated offenders, while reducing the level of guardianship over person and property. Fisher and her colleagues (1997) reported that of the student victims of crime who lived on campus in undergraduate or graduate housing, or in a fraternity or sorority, 86 % of them lived in an undergraduate

Campus Crime

residence hall. Identifying on-campus locations where incidents had occurred revealed that crimes of violence, vandalism, threats, and harassment occur most frequently in students’ living quarters. Students’ lifestyles and routines also lend themselves to opportunities for victimization every day of the week around the clock. Each year students bring attractive, portable, lightweight targets to campus with them: portable computers, electronic pads and tablets, cell and smart phones, calculators, backpacks, books, roller blades, skateboards, and bicycles. Students carry many of these items with them while they attend classes or social events. Many students drive motor vehicles to campus daily or park vehicles on or near campus during the duration of the academic term. Concurrently, they are all too often inattentive guardians of their property (or their person) and leave residence hall, office, or vehicle doors unlocked or propped open; leave bicycles unlocked; or just walk away from their property at the library, student union, or classroom for “just a few minutes.” As a result, dormitories, bicycle racks, libraries, student unions, or classrooms can all become “hot spots” of crime because students are typically poor guardians of their property. Moreover, students leave the campus en mass at the end of an academic term or for long periods during holiday breaks, which leaves the campus filled with many attractive targets without guardianship. The college years are also notorious for an active social life for many students. They may attend weekly parties; drink habitually and heavily or experiment with drugs; and attend entertainment or cultural events. All of this activity occurs in places on campuses where students naturally congregate. Such lifestyles and routines have, in turn, been linked to increased probabilities for on-campus victimizations. The Harvard Alcohol Study found that students who had been the victims of crime on campus reported significantly more frequent drug and alcohol use than students who had not been victimized (Dowdall 2009). The number of nights spent on campus partying and the likelihood of regularly taking recreational drugs during the past year also

Campus Crime

have been reported to be significant predictors of students’ on-campus violent victimization (Fisher et al. 1998). Given their 24–7 on-campus lifestyles, it is not surprising that a temporal pattern concerning on-campus student victimization shows that violent crimes were likely to have occurred in the early morning hours, in particular between 12:00 a.m. and 2:00 a.m. (Fisher et al. 1998). Faculty, Staff, and Victimization. Faculty members, like students, also have routine activities on college campuses that increase their risk for victimization. Faculty members, especially untenured assistant professors, are frequently on campus during the week and on weekends, sometimes spending long hours into the night and early morning working on research, grading papers, or writing manuscripts. Many spend considerable amount of time in their offices or laboratories, or walking to and from classes, meetings, the library, or a seminar. Most faculty members work alone in their office even when support staff have gone home for the day. Many faculty members also teach night courses. Because of their routines, many faculty members are exposed to a large number of students every academic term. Some faculty members even socialize with students outside of class in the student union or other establishments on or near campus. Faculty members also have valuable, attractive targets such as laptop computers, multimedia equipment, laser printers, laboratory equipment, secured data, books, motor vehicles, and other personal items. Faculty can also be poor guardians of their property. At times, they leave offices or laboratories unlocked or computers unsecured between classes or meetings, and may prop open doors to buildings, offices, or laboratories so as not to have to fetch keys or recall a lock combination. With the routine arrival of new faculty members each academic term, while others are on leave or sabbatical during the school year, access control to buildings and offices may be severely compromised on a daily basis. Wooldredge, Cullen, and Latessa (1995), in a study of faculty members’ victimization experiences at the University of Cincinnati West campus (all colleges except the medical and

263

C

nursing colleges), reported 27 % of faculty members surveyed reported they had experienced at least one property victimization on campus (burglary, stolen property, or damage to property) during the academic year, while 5 % reported they had experienced at least one personal victimization (robbery, assault, sexual assault, or assault with a deadly weapon). Those whose offices are not within shouting distance of their colleagues; who do not teach in buildings where their office is located; and whose office is in a comparatively non-secure building on campus are more likely to experience an on-campus property crime. Faculty members who spend more time on campus after hours and on Saturdays, who walk alone on campus more frequently (other than going to class), and who socialize with students outside class are also more likely to experience personal victimization while on campus. Staff members typically work 8-h shifts, primarily Monday–Friday and during daylight hours. Like faculty, they too possess attractive targets. To date, there are no published studies on the extent and nature of victimization among staff working on college campuses. Anecdotal evidence, however, suggests that their patterns of victimization may be similar to those of faculty. Visitors to Campus and Victimization. The routine activities of visitors to campus are somewhat different from those of students, faculty members, and staff. Every term, prospective students and their parents frequently meander around the campus. Students from other campuses visit with friends; sports fans attend athletic events or hold tailgate gatherings; and cultural buffs frequent movies, concerts, art galleries, and plays. Delivery trucks filled to the brim with products and packages are daily sights on campus. Visitors bring plenty of property with them that is attractive to would-be offenders – cash, portable electronic equipment, expensive jewelry and watches, motor vehicles. Like staff, no studies of campus visitors have been published but it is likely that students who visit friends and enjoy activities where alcohol and drugs are plentiful and visitors who attend sporting events and enjoy pregame, during-the-game, and

C

C

264

post-game celebrations might experience a pattern of victimization similar to students who actively engage in an active “party lifestyle.”

Tracking the Spatial and Temporal Distribution of Campus Crime Crime on campus, like crime in other locations, is not randomly distributed in space or time. Robinson and Roh (2007) illustrated this point in an analysis of crime on the campus of Appalachian State University (ASU) for the period 2004–2005 using calls-for-service data stored in campus police crime logs. Analyzing those data revealed that the four most commonly reported offenses on the ASU campus over the 2-year period of study included alcohol violations, theft of property, illegal drug possession, and vandalism (Robinson and Roh 2007, p. 239). More importantly, they determined that hot spots on the ASU campus included locations at or near: (1) student residence halls, (2) high-traffic areas between the main parts of the campus such as near classroom buildings and parking lots near pedestrian tunnels, and (3) specific locations for drug violations, alcohol violations, breaking and entering, and assaults/sexual assaults/forcible fondling. These “crime attractors” or “crime generators” remained relatively stable over the 2 years of study and possessed certain features. For example, one hot spot included two residence halls located within walking distance to the football stadium, its parking lots, and an open field which students use for recreational activities including consuming alcohol, playing sports, and hosting parties. Robinson and Roh also reported that residence halls that were “hot spots” for various crimes tended to be older residences located in the heart of the campus. Compared to these “older” residence halls, newer residence halls’ distance from the center of the campus made them both more physically and socially isolated and apparently less vulnerable to opportunities for victimization. In a similar study, Resler (n.d.) examined the spatial distribution of crime on the campus of the University of Texas at Austin using police

Campus Crime

calls-for-service data from 1998. Using geographic information system (GIS) crime mapping software and spatial analysis models, Resler (n.d.) found that “. . . concentrations of crime occur[ed] predominantly around the periphery of campus [and] the characteristics of these high crime areas, such as location and density, influence victimization.” Resler concluded that use of GIS software to map crime distributions and concentrations was an effective tool, and its use opened new avenues for research. Brower and Carol (2007) examined the temporal and spatial distribution of reported incidents of liquor law violations, assaults and batteries, vandalism, and noise complaints during 2003 on the campus of the University of Wisconsin-Madison and the City of Madison. They also used GIS crime mapping technology and tracked the movement of different crimes through the city of Madison by time of day, and investigated the relationship between these spatial and temporal movements of crime and the proximities of various student and nonstudent neighborhoods to each other and to high-density bar areas. They found different categories of crime presented different temporal and spatial patterns. For example, serious crimes such as assaults tended to peak between 2:00 a.m. and 3:00 a.m. which coincided with bar closing time. Less serious crimes, such as noise violations, peaked between 11:00 p.m. and 12:00 a.m. Vandalism tended to peak during the late morning and early afternoon. “Hot spots” for the different crimes moved throughout the downtown by time of day. Results of the study were then used by University of Wisconsin and City of Madison officials to implement changes to address highrisk drinking behavior.

Controversies Because the study of crime on college and university campuses is, relatively speaking, still in its infancy, controversies over study results are bound to arise. Below, a few of the more important questions that have been raised with these studies are presented.

Campus Crime

One significant controversy in the literature concerns wide-ranging estimates that have been produced concerning the extent college women experience various forms of sexual violence during their college tenure. Part of the explanation for disparate estimates of college women’s victimization experiences has to do with how questions about rape, sexual assault, stalking, and other forms of violence are framed. Generally, the broader the question about sexual victimization asked of college women, the larger are the estimates of sexual violence. Fisher and her colleagues (2000) were among the first to document how specific, behaviorally based questions relating to sexual violence against college women significantly affected estimates for victimization. As a result of their work, studies now are more apt to use narrow, behaviorally based questions in surveys of college women to explore the extent of sexual victimization perpetrated against them. A second controversy is the predominance of research using case studies of crime occurring on a single campus. Typically, one or more researchers will survey students (or faculty or staff) enrolled at a single school or use official data compiled by a single school to examine the causes, correlates, and consequences of crime occurring on campus. While many of these studies have contributed insight into the dynamics of campus crime, the problem is that results from such case studies cannot be generalized to colleges and universities more generally. The works of Fisher and her colleagues (1998, 2000) are notable exceptions; both projects involved national samples of college students. A final major controversy revolves around the utility of officially compiled data collected by Title IV eligible post-secondary institutions in compliance with the mandates of legislation such as the federal Clery Act. Among other requirements, the legislation mandates that colleges and universities compile statistics on crimes reported to campus authorities for the following offenses: murder and non-negligent homicide; rape; robbery; assault; burglary; automobile theft; and arson. One weakness is that the Clery Act fails to include what is generally regarded (based on victimization

265

C

surveys) as the most commonly occurring crime on campus, namely, theft or larceny. A second problem, again based on victimization surveys conducted on college students, is that large numbers of offenses occurring on campus go unreported to campus law enforcement or other authorities. As a result, those offenses are never included in the crime statistics published as part of Clery Act requirements, which means Clery Act statistics grossly underestimate the volume of crime occurring on the nation’s college campuses (Fisher et al. 1998). Unless and until these controversies are settled, contributions from studies of campus crime will likely not have the impact they might, were these issues to be addressed. In particular, campus administrators who may be relying upon extant literature to help them craft responses to crime occurring on their campuses may not develop as effective a response as they might were the controversies settled.

Open Questions Given the above controversies, several questions remain open to exploration, comment, and debate. The first is rather basic: How much crime occurs on college campuses each year? Currently, the answer to that question depends on where one looks: case studies (from which national-level extrapolations might be developed); “official” data (such as those compiled in compliance with the Clery Act but which, as discussed above omit a major type of crime and suffer from underreporting); or national-level studies (some of which are more than 15 years old). A second open question is how to better translate the results of academic research on campus crime into practical policies and programs to address the problem and then to evaluate their impact on both crime and fear of crime. Assuming study results used to develop specific programs or policies to reduce campus crime are sound, there is little evidence that programs based on these results are ever evaluated. Further, policies or programs aimed at reducing campus crime too often fail to use sound scientific designs

C

C

266

that allow for such necessities as control groups, or designs which rule out rival hypotheses. Criminologists are increasingly advocating the need to use controlled experiments to examine the effectiveness of interventions designed to reduce crime; the same should be true of efforts occurring on college campuses. Finally, but certainly of no less importance is the ultimate utility of legislation like the Clery Act that is designed to reduce/eliminate campus crime. There are actually several problems with campus crime reporting legislation. First, because reliable baseline statistics for the years prior to the passage of Clery do not exist, it is difficult to know if the legislation has had a positive, negative, or neutral impact on campus crime. Second, post-secondary institutional compliance with Clery has been less than total for a number of reasons, none the least is the amount and type of resources necessary to compile and report the data required by the law. Finally, there is enough evidence to indicate that the very consumers whom the law was passed to serve (particularly prospective college students and their parents) are either unaware of the law and the various reports that are to be generated by colleges and universities in compliance with the law’s mandates, or if they are aware of these reports, few actually consult them (Gregory and Janosik 2007).

Conclusions Campuses are neither hot beds of criminal victimization nor are they ivory towers immune from the grim reality of crime. They are places whose built and physical environments, institutional and demographic characteristics, space distributions, functional usage, temporal patterns, lifestyles and routines create opportunities for victimization. By developing an orientation that focuses on the criminology of place and on identifying campus characteristics and lifestyles/routines that facilitate opportunities for crime, researchers now better understand how and why campus crime happens. Through identification of crime patterns, researchers also have informed interventions

Campus Crime

(e.g., banning alcohol on campus or enhancing campus lighting) intended to disrupt the rhythms and routines associated with campus crime. This work has clear implications for campus policing and campus security, because it is these agencies that will benefit the most from research focusing on the relationship between campus places and campus crime (Ratcliffe 2002). It is these agencies that routinely collect data that can then be used to identify hot spots and adapt personnel and tactics according to changing conditions. These agencies, through place-based analyses of campus crime, also can work with campus administrators to develop and better enforce policies relating to parking, consumption of alcohol, noise ordinance violations, and other problems. There is reason for optimism that as research devoted to college campuses as places matures, nagging questions about campus crime will be answered. There is also reason to hope that the answers uncovered will, in turn, inspire the development and evaluation of evidence-based policies and programs to reduce or even eliminate campus crime.

Related Entries ▶ Criminology of Place ▶ Hot Spots and Place-based Policing ▶ Lifestyle Theory ▶ Routine Activities Approach ▶ Situational Crime Prevention

Recommended Reading and References Brantingham P, Brantingham P, Seagrave J (1995) Crime and fear at a Canadian university. In: Fisher BS, Sloan JJ (eds) Campus crime: legal, social and policy perspectives. Charles C. Thomas, Springfield, pp 123–155 Bromley ML (1994) Correlates of campus crime: a nationwide exploratory study of large universities. J Secur Admin 17:37–52 Bromley ML (2007) The evolution of campus policing: different models for different eras. In: Fisher BS, Sloan JJ (eds) Campus crime: legal, social, and policy perspectives, 2nd edn. Charles C. Thomas, Springfield, pp 280–303

Capture Recapture to Estimate Criminal Populations Brower AM, Carroll L (2007) Spatial and temporal aspects of alcohol-related crime in a college town. J Am Coll Health 55:267–75 Cohen L, Felson M (1979) Social change and crime rate trends: a routine activity approach. Am Sociol Rev 44:588–605 Dowdall GW (2009) College drinking: reframing a social problem. Praeger, Westport Eck JE, Weisburd D (1995) Crime places in crime theory. In: Eck JE, David W (eds) Crime and place, vol 4. The Police Executive Research Forum, Washington, DC, pp 1–34 Fernandez A, Lizotte AJ (1995) An analysis of the relationship between campus crime and community crime: reciprocal effects? In: Fisher BS, Sloan JJ (eds) Campus crime: legal, social, and policy perspectives. Charles C. Thomas, Springfield, pp 79–102 Fisher BS, Sloan JJ, Cullen FT, Chunmeng L (1997) The on-campus victimization patterns of students: implications for crime prevention by students and post-secondary institutions. In: Lab SP (ed) Crime prevention at a crossroads. Anderson Publishing Co, Cincinnati, pp 101–126 Fisher BS, Sloan JJ, Cullen FT, Chenmung L (1998) Crime in the ivory tower: the level and sources of student victimization. Criminology 36:671–710 Fisher BS, Cullen FT, Turner MG (2000) The sexual victimization of college women. http://www. ncjrs.gov/pdffiles1/nij/182369.pdf. Accessed 4 Dec 2012 Gieryn TF (2000) A space for place in sociology. Annu Rev Soc 26:463–496 Gregory DE, Janosik SM (2007) Research on the Clery Act and its impact on higher education administrative practice. In: Fisher BS, Sloan JJ (eds) Campus crime: legal, social, and policy perspectives, 2nd edn. Charles C. Thomas, Springfield, pp 45–64 Hart TC, Miethe TD (2011) Violence against college students and its situational contexts: prevalence, patterns, and policy implications. Victims Offenders 6:157–180 Mustaine E, Tewksbury R (2007) The routine activities and criminal victimization of students: lifestyle and related factors. In: Fisher BS, Sloan JJ (eds) Campus crime: legal, social, and policy perspectives, 2nd edn. Charles C. Thomas, Springfield, pp 147–166 Ratcliffe J (2002) Damned if you don’t and damned if you do: crime mapping and its implications in the real world. Policing Soc 12:211–225 Resler L (n.d) “Campus crime at the university of Texas at Austin: a spatial and organizational approach using GIS.” http://www.utexas.edu/depts/grg/hudson/ grg394k/studentprojects/resler/resler.html. Accessed 4 Dec 2012 Rigg L (2010) “College student enrollments: facts and figures.” http://www.chillicious.com/finance/collegestudent-enrollment-facts-and-figures/. Accessed 4 Dec 2012

267

C

Robinson M, Roh S (2007) Crime on campus: spatial aspects of crime at a regional comprehensive university. In: Fisher BS, Sloan JJ (eds) Campus crime: legal, social, and policy perspectives, 2nd edn. Charles Thomas, Springfield, pp 231–255 Sherman L (1995) Hot spots of crime and criminal careers of places. In: Eck JE, David W (eds) Crime and place. Criminal Justice Press, New York, pp 35–52 Sloan JJ (1994) The correlates of campus crime: an analysis of crimes known to campus police and security. J Crim Just 22:51–61 Sloan JJ, Fisher BS (2011) The dark side of the ivory tower: campus crime as a problem. Cambridge University Press, New York Sloan JJ, Shoemaker J (2007) State level clery Act initiatives: symbolic politics or substantive policy? In: Fisher BS, Sloan JJ (eds) Campus crime: legal, social, and policy perspectives, 2nd edn. Charles C. Thomas, Springfield, pp 102–124 Smith MK (2002) “Community.” http://www.infed.org/ community/community.htm. Accessed 4 Dec 2012 United States Department of Education (2011a) http:// www2.ed.gov/admins/lead/safety/campus.html#data. Accessed 4 Dec 2012 United States Department of Education (2011b) http:// nces.ed.gov/fastfacts/display.asp?id¼372. Accessed 4 Dec 2012 United States Department of Education (2011c) http:// nces.ed.gov/Programs/digest/d08/tables/dt08_190.asp. Accessed 4 Dec 2012 Wooldredge J, Cullen F, Latessa E (1995) Predicting the likelihood of faculty victimization: Individual demographics and routine activities. In: Fisher BS, Sloan JJ (eds) Campus crime: legal, social, and policy perspectives, Charles C. Thomas, Springfield, pp 122–133

Capture Recapture to Estimate Criminal Populations Peter G. M. van der Heijden1,2, Maarten Cruyff1 and Dankmar Bo¨hning3 1 Department of Methodology and Statistics, Utrecht University, Utrecht, The Netherlands 2 University of Southampton, Southampton, UK 3 Statistical Sciences Research Institute, University of Southampton, Southampton, UK

Synonyms Population size estimation

C

C

268

Capture Recapture to Estimate Criminal Populations

Overview

offenses are infrequent. For a more elaborate comparison and an overview of the literature on police registers, victim surveys, and self-report studies, the reader is referred to Wittebrood and Junger (2002). The methodology presented here makes use of a single register. There is a large literature on capture recapture making use of two registers. Estimation based on a single register has two important advantages: first, it does not require the unverifiable assumption that the two sources are statistically independent, and, second, it does not require the elaborate process of linking databases that is often troubled by privacy regulations. Also, often there are technical problems in making correct linkages and in avoiding incorrect linkages. A single register that contains (re)captures circumvents these problems. In this entry, a way to estimate the number of offenders from police data is discussed. The data are from the Dutch police register system. Offenses committed by a known offender are registered in this system. So for each specific offense, like, for example, illegally owning a gun, an offender-based data set can be constructed that shows the number of illegal gun owners apprehended once, twice, three times, and so on. Note that illegal gun owners who were not apprehended are not part of this offender-based data set. Yet, if their number could be estimated, this would yield an estimate of the total number of illegal gun owners (compare van der Heijden et al. 2003a). The aim is to estimate the number of offenders never apprehended, using the data about offenders apprehended at least once. These estimates are derived under two assumptions. First, the number of apprehensions is a realization of a Poisson distribution. Second, the logarithm of the Poisson parameter for an offender is a linear function of his covariates. These assumptions are discussed in greater detail at the end of the introduction. At this point, it is indicated how, using these assumptions, the size of the population never apprehended can be estimated. Consider an offender with a Poisson parameter that gives him a probability of.25 to be apprehended at

Methodology is presented that allows to estimate the size of population from a single register, such as a police register of offenders. A capturerecapture variable is constructed from Dutch police records and is a count of the police contacts for a violation. A population size estimate is derived assuming that each count is a realization of a Poisson distribution and that the Poisson parameters are related to covariates through the truncated Poisson regression model or variants of this model. As an example, estimates for perpetrators of domestic violence are presented. It is concluded that the methodology is useful, provided it is used with care.

Fundamentals of Capture Recapture Introduction For many policy reasons, it is important to know the size of specific delinquent populations. One reason is that it provides insight into the threat these populations may pose on society. Another reason is that it gives an estimate of the workload of the police. However, estimating the size of a delinquent population may be problematic for various reasons. Counting the number of crimes from police records may lead to a dark number problem. It may be that the crime is registered but the offender is not known or, as is often the case with victimless crimes, the crime is not registered at all. In victim surveys, people report the number of times they have been the victim of a particular crime, like robbery or burglary. Based on that information, an estimate can be obtained of the total number of these crimes. However, victim surveys do not provide an estimate of the number of offenders, since they usually are unknown to the victim, nor do they provide insight into victimless offenses. Self-report studies can potentially estimate the size of a delinquent population since people are simply asked whether they are member of this type of population. Problems related to self-report studies are the difficulty of obtaining a representative sample, the risk of socially desirable answers, and the need for large samples if


least once. Suppose that this offender is indeed apprehended, then there are three other offenders with the same Poisson parameter who have not been apprehended. By performing this trick for every offender who is apprehended and adding up all individual estimates, an estimate is obtained of the number of offenders who are not apprehended, and this solves the problem. The methods employed in this entry originate from the field of biology, where they are used to estimate animal abundance. In these applications, the data are collected at specific time points, and for each animal that is seen at least once, there is a capture history. For example, if there are five capture times, a history could be 01101 if the animal is seen at captures 2, 3, and 5 and not seen at captures 1 and 4. In the present methodology, however, the data are collected in continuous time, so only the total number that someone is captured is used. Typically, in the biological application area, covariate information is not available or not used, leading to a basic model in which the Poisson parameters are assumed to be homogenous over the animals. For an overview of this area, the reader is referred to Seber (1982, Chap. 4), Chao (1988), and Zelterman (1988). In the statistical literature, this problem is also known as the estimation of the number of (unseen) species (Bunge and Fitzpatrick 1993). In criminology, there are some early studies by Greene and Stollmack (1981) who use arrest data to estimate the number of adults committing felonies and misdemeanors in Washington D.C. in 1974/1975; Rossmo and Routledge (1990) who estimate migrating (or fleeing) fugitives in 1984 and prostitutes in 1986/1987, both in Vancouver; and Collins and Wilson (1990) who use arrest data to estimate the number of adult and juvenile car thiefs in the Australian capital territory in 1987. In the field of drug research, a one-source capture-recapture analysis has also been applied to estimate, for example, the size of the marijuana cultivation industry in Quebec (Bouchard 2007) and young drug users in Italy (Mascioli and Rossi 2008). These studies do not devote systematic attention to covariate information on the apprehended individuals.

269

C

The methodology reviewed here makes use of covariates that are available in police registers, such as age, gender, and so on. The methodology yields the following results: (1) the hidden number of offenders and a 95 % confidence interval for this hidden number, (2) a distribution of these hidden numbers over covariates, and (3) insight into which part of the hidden number is visible in the register and which part is missed, stratified by the levels of the covariates. Covariate information is incorporated by using a regression model that makes use of truncated Poisson distributions, such as the truncated Poisson regression model and the truncated negative binomial regression model, that are well known in econometrics (e.g., Cameron and Trivedi 1998, Chap. 4). These models are elaborated so that a frequency can be estimated for the zero count as well as a confidence interval for this point estimate (see Van der Heijden et al. 2003a, 2003b; Cruyff and van der Heijden 2008; Bo¨hning and van der Heijden 2009, that also show examples on undocumented immigrants, illegally owned firearms, and drunk driving). The methodology will be illustrated for perpetrators of domestic violence (van der Heijden et al. 2009). Assumptions As the methodology originates from the field of biology and it is used in the field of criminology, the assumptions of the methodology are discussed here in greater detail. Of course, assumptions that are realistic for animals will not always be realistic for human offenders. The first assumption is that the number of times an individual is apprehended is a realization of a Poisson distribution (equations will be provided later). Johnson et al. (1993) discuss the genesis of the Poisson distribution and state that it was originally derived by Poisson as the limit of a binomial distribution with success probability p and N realizations, where N tends to infinity and p tends to zero, while Np remains finite and equal to l. It turns out that even for N ¼ 3, the Poisson distribution approximates the binomial distribution reasonably well if p is sufficiently small. Johnson et al. (1993) also note

C

C

270

that the probability of success p does not have to be constant for the Poisson limit to hold. This implies that an individual’s count still follows a Poisson distribution, even if this individual’s Poisson parameter has changed during the period of observation. It follows for the type of applications being discussed that individuals do not need to have a constant probability to be apprehended, but it suffices if they can be apprehended a number of times (see van der Heijden et al. 2003a). It is important to note that the Poisson assumption is only valid if a change in the individual Poisson parameter is unrelated to any prior apprehensions or non-apprehensions. This follows from the independence of subsequent trials in a binomial distribution. In the biostatistical literature, this problem is known as positive contagion (if the probability increases) or negative contagion (if the probability decreases). Closely related to the contagion issue, there is the problem of an open or closed population. A population is closed if the number of offenders is constant over the period of data collection and is open if offenders may enter or leave the population during this period. Given what has been noted above, it is clear that the population may be open as long as entering or leaving it is not related to apprehension or non-apprehension. For example, detention following an apprehension removes the person from the population and excludes the possibility of any subsequent apprehensions and can therefore be seen as an extreme case of negative contagion. So far the Poisson assumption pertaining to an individual count is discussed. The second assumption follows from using a regression model, in which the logarithm of the Poisson parameters is a linear function of covariates. In the regression model, the Poisson parameters are still assumed to be homogeneous for individuals with identical values on the covariates, but they are allowed to be heterogeneous for individuals with different values. Since here the differences in Poisson parameters are determined by the observed covariates, this is referred to as observed heterogeneity. If, in addition to observed heterogeneity, there are differences in the Poisson parameters that cannot be explained


by the observed covariates, we speak of unobserved heterogeneity. If the Poisson regression model does not fit due to unobserved heterogeneity, this is referred to as overdispersion. In conclusion, the most important violations of the Poisson assumptions in criminological applications are contagion and overdispersion. The contagion problem may be larger for some offenses than for others, and an indication of its importance can be obtained by studying the behavior of offenders as well as police officers (e.g., by doing qualitative research). If no additional information is available on their behavior, it seems best to interpret the results with caution. Overdispersion can be assessed in the data as a result of the analysis, and this will be discussed below. Models An informal definition of a Poisson distribution is as follows. The Poisson distribution is characterized by a Poisson parameter denoted by l. This parameter l expresses the probability of a given number of events (i.e., the count) under two assumptions, namely, that events occur: 1. With an average rate in a fixed interval of time 2. Independent of the time since the last event The probability that the count Y generated by a Poisson distribution with Poisson parameter l is equal to j is pj ¼ PðY ¼ jÞ ¼

el lj : j!

(1)

Three models for count data are discussed, namely, (1) the truncated Poisson regression model, (2) the truncated negative binomial regression model, and (3) the Zelterman regression model. First consider Eq. 1. Two examples of a Poisson distribution are provided in Table 1. In the first example, an individual has a Poisson parameter l ¼ .5. Then his or her probability to be seen zero times is .607, the probability to be seen once is .303, twice is .076, and so on. These probabilities add up to 1. In the second example, it is assumed that there is an individual with Poisson parameter 1. His or her probability not

Capture Recapture to Estimate Criminal Populations Capture Recapture to Estimate Criminal Populations, Table 1 Two examples of Poisson distributions (line 1 and 2) and their corresponding truncated Poisson distributions. l ¼ .5 l¼1 l ¼ .5 l¼1

0 .607 .368 – –

1 .303 .368 .771 .582

2 .076 .184 .193 .291

3 .013 .061 .032 .097

4 .002 .015 .004 .024

5 .000 .003 .000 .005

6 .000 .001 .000 .001

Total 1.000 1.000 1.000 1.000

Notes: The columns show the number of times an individual is seen. The cells give the probabilities

to be seen is .368, the probability to be seen once is .368, twice is .184, and so on. Note that the individual with Poisson parameter l ¼ 1 has a larger probability to be observed at least once, namely, (1 .368) ¼ .632, whereas this probability for the individual Poisson parameter l ¼ .5 is only (1 .607) ¼ .393. It follows that the individual with Poisson parameter l ¼ 1 has a larger probability to be seen. Model 1: The Truncated Poisson Regression Model

In single-register capture-recapture data the observed individuals each have a count larger than zero. Since for the observed individuals the count cannot be zero, these individuals have the so-called truncated Poisson distributions. For the data in the first and second row of Table 1, the truncated distributions for y > 0 are obtained by dividing the probabilities by 1 P(y ¼ 0|l), that is, for the first example we divide by 1 .607 ¼ .393 and for the second example we divide by 1 .368 ¼ .632; see rows three and four of Table 1. As a first step towards developing the truncated Poisson regression model, assume that there are no covariates. This implies the homogeneity assumption, so that only a single Poisson parameter needs to estimated. This Poisson parameter is then used to obtain the probability of not being registered given that the Poisson parameter equals the value l, as denoted P(y ¼ 0|l). Using P(y ¼ 0|l), the part of the population is estimated that we did not see. A small example will illustrate this. Assume that n ¼ 250 individuals are observed with a Poisson parameter for which P(y ¼ 0|l) ¼ .667.

271

C

This would mean that P(y > 0|l) ¼ .333, that is, only one-third of the population is observed so our n refers to one-third of the population size. The missed part of the population size is (.667/.333) * 250 ¼ 500. The estimated population size would then be equal to the observed part plus the missed part, that is, N ¼ 250 + 500 ¼ 750. In the statistical literature, this is known as the Horvitz-Thompson estimator of the population size (see Van der Heijden et al. 2003b). Secondly, the Poisson parameter for individual i is related to the covariate values x1i , x2i , . . . of individual i by a log-linear model, that is, Log li ¼ b0 þ b1 x1i þ b2 x2i þ

(2)

This equation explains the term “regression” in the name “truncated Poisson regression model.” Once the model is estimated and the parameters are known, the Horvitz-Thompson method can be applied on an individual level (see Van der Heijden et al. 2003b). Using the earlier example, for each of the 250 individuals in the data, there is an estimated li-parameter. For each individual i separately, P(y ¼ 0|li)/P(y > 0|li) can be calculated, and this yields the number of missed individuals with covariate values identical to individual i. If these numbers of missed individuals over i are summed and 250 is added, the estimated population size is found derived under the truncated Poisson regression model. Summarizing, the homogeneity assumption is exchanged for a heterogeneity assumption that allows individuals to be different with regard to their covariates. It is important to consider the case where heterogeneity is completely or partly ignored. Van der Heijden et al. (2003a) show that ignoring heterogeneity leads to an estimated population size that is too low. Ignoring heterogeneity may happen if important covariates are not used in Eq. 2. It is possible to investigate whether there is ignored heterogeneity by using a test presented by Gurmu (1991). If this test is significant, then there is evidence for additional heterogeneity (i.e., additional with regard to the heterogeneity that is already taken into account by the covariates), and the estimated population size is

C

C

272


to be interpreted as a lower bound for the true population size. Thus, the Gurmu test can be used to investigate the fit of the model. Another way of investigating the validity of the Poisson assumption is the ratio plot (Bo¨hning and Del Rio Vilas 2008). According to the previous section, rj ¼ ( j + 1)pj+1/pj ¼ l, a constant, if the pj follows a Poisson distribution. Departure from this constant indicated unobserved heterogeneity. Model 2: The Truncated Negative Binomial Regression Model

The distinction between the truncated negative binomial regression model and the truncated Poisson regression model is that the truncated negative binomial regression model allows for additional heterogeneity of Poisson parameters after the covariates have been taken into account. Equations for the negative binomial model are complicated; see Hilbe (2011) for details. Cruyff and Van der Heijden (2008) discuss the application of the truncated negative binomial regression model for population size estimation. It turns out that this model is often difficult to fit numerically, in particular when Poisson parameters are small. For details, see Cruyff and Van der Heijden (2008) where this is shown using a simulation study. Unfortunately there are also numerical problems for the example discussed below. For an example of opiate users where fitting the truncated negative binomial regression truncated Poisson regression model did not lead to numerical problems, see Cruyff and Van der Heijden (2008). Model 3: The Zelterman Regression Model

The Poisson distribution Eq. 1 has the property that, for any j, P( j + 1|l)/P( j|l) ¼ l/( j + 1). This can be rewritten as l ¼ ( j + 1) P( j + 1|l)/P( j|l). Zelterman (1988) uses this property to propose local estimators of the Poisson parameter by plugging in observed frequencies fj of count j for P( j|l) and P( j + 1|l). In particular, if the frequencies f1 and f2 are plugged in, an estimate of the Poisson parameter 2f2 ^ l¼ f1

is found, and this estimate of l can then be used to estimate P(0) ¼ 1 exp(l) and hence the Zelterman population size estimator NZ ¼ n/(1 exp(l)). This estimator is closely related to Chao’s estimator NC ¼ n + f12/2f2, that is used for similar purposes (Bo¨hning 2010). Zelterman’s proposal has the advantage that it is robust against violations of the Poisson assumption such as unobserved heterogeneity. Also, when interest goes out to the frequency of the missed count f0, then it makes sense to use the information that is the most close to f0, that is, f1 and f2, because individuals that make up f1 and f2 will be most similar to the individuals that make up f0. Probably for this reason and because the estimator is easy to understand, Zelterman’s estimator is quite popular in estimates of drug using populations (see, Van Hest et al. 2007). Recently Bo¨hning and Van der Heijden (2009) generalized the Zelterman estimator so that it can take covariates into account as in Eq. 2. As the Zelterman estimator is a useful competitor of the Poisson estimator if there is heterogeneity of the Poisson parameters, the Zelterman regression model is a useful competitor of the truncated Poisson regression model when the homogeneity assumption of the Poisson parameters (conditional on the covariates) is violated. This follows from the robustness property of the Zelterman estimator. Model Choice

In the truncated Poisson regression model, the presence of unobserved heterogeneity in the data can be investigated with a test proposed by Gurmu (1991) and ratio plots (Bo¨hning and Del Rio Vilas 2008). If unobserved heterogeneity can be ignored, the truncated Poisson regression model is the model of choice. If unobserved heterogeneity cannot be ignored, the population size estimate given by the truncated Poisson regression model is to be interpreted as a lower bound for the true population size. In this situation, the truncated negative binomial regression model can be tried. However, regularly numerical problems are encountered in the estimation of the model. If there is doubt about


273

C

Capture Recapture to Estimate Criminal Populations, Table 2 Observed and fitted frequencies of counts 0, 1, 2, . . . for domestic violence data under truncated Poisson regression model Observed Estimated

0 – 51,629

1 15,169 14,814

2 1,957 2,494

3 393 317

4 99 33

5 28 3

>5 16 0

Total 17,662 69,290a

a

95 % confidence interval is 66,242–72,338

C the Poisson assumption for individuals having counts higher than 2, then the Zelterman regression model provides a robust estimate.

Example In the context of domestic violence policy, it is important to have reliable estimates of the scale of the phenomenon. In 2009, a study was conducted to supplement a victim study and perpetrator study (van der Heijden et al. 2009). Here the perpetrator study is reported. cthe prevalence of domestic violence, particularly because so few victims are willing to file charges, which leads to underreporting, that is, the dark number in registers. The aim of the capture-recapture methods used in this study is to estimate the size of the underreporting. Adding up the reporting and underreporting then yields an estimate of the total population of offenders. The estimates have been made using data from incidents that were reported and where charges were filed. The data represent the Netherlands except the police region for The Hague. The estimates presented have been calculated using the Poisson regression model and the Zelterman regression model. The negative binomial regression model had numerical estimation problems. The estimates are presented for a year ranging from mid-2006 to mid-2007. Distinctions were also drawn as regards specific features of the estimated populations, that is, the sex and age of the individual suspect, the type of violence used, the type of victim of domestic violence, and the ethnic background of the suspect. The first line of Table 2 shows the observed distribution of the counts. A total of 17,662 perpetrators are observed in the year ranging from

mid-2006 to mid-2007, of whom 15,169 were observed once, 1,957 twice, and so on. The second line of Table 2 shows the fitted values under the truncated Poisson regression model. The estimated population size is 69,290 (confidence interval is 66,242–72,338) of whom 51,629 perpetrators are not found in the police register. The fit of the model is not good, as is revealed by comparing the observed frequency of the counts and the estimated frequency of the counts: The Gurmu test on overdispersion gives a chi-squared distributed test value of 30.04 for 1 of freedom (p < .0001), showing that the departure from the truncated Poisson regression model is significant. Similarly, Fig. 1 shows a ratio plot (Bo¨hning and Del Rio Vilas 2008) where these ratios are monotonically increasing, indicating strong unobserved heterogeneity. For this reason, further results for the truncated Poisson regression model are not presented, but the focus will be on the Zelterman regression model. The Zelterman regression model yields a population size estimate of 84,862 (confidence interval is 80,293–89,431). Columns 1–3 of Table 3 provide parameter estimates, standard errors, and p-values for a Wald test. Gender is significant, and the fitted conditional Poisson parameter for males is exp (.49) ¼ 1.63 times as large as the fitted conditional Poisson parameter for females. For type of domestic violence, sexual and physical assaults have a lower fitted Poisson parameter than threat. In comparison to an (ex-) partner, a family friend and “other” have lower fitted Poisson parameters. Ethnic background (first or second generation) is not significant, except for “not registered” in the official register, which are undocumented foreigners, who have a lower fitted Poisson parameter than the native Dutch population.

C

274


Capture Recapture to Estimate Criminal Populations, Fig. 1 Ratio plot for the observed frequencies in Table 2

1.4 1.2

ratio

1.0 0.8 0.6 0.4 0.2 1

Columns 4 and 5 of Table 3 provide observed univariate frequencies and fitted frequencies. For example, 14,827 males are found in the police register, and the estimate of the number of males is 68,048; thus, it can be concluded that 14,827/68,048 * 100 ¼ 22 % of the male perpetrators is observed. For females, this percentage is 2,653/16,813 * 100 ¼ 16 %. The other percentage can be derived in an identical way.

Future Directions It is shown how police records can be used to estimate the size of criminal populations. These estimates can be used to evaluate the effectiveness of the police forces and grant insight into differential arrest rates (Collins and Wilson, 1990) for different groups. Even though the definition of the data is straightforward, however, the data are sometimes contaminated with errors. The reason is that the police does not collect these data for the purpose of conducting statistical analyses, they do so to facilitate the process of apprehending offenders. Therefore, the register is not always as careful as it should be. For example, data cleaning was required to minimize the likelihood of incorrect double counts (i.e., the same apprehension appears twice in the

2

count j

3

4

system). Clearly, an incorrect double count incorrectly decreases fk by 1 and increases fk+1 by 1, so that there appear to be more recaptures than there actually are. The result is that the estimated zero count is too low. Although careful attention has been devoted to the appearance of incorrect double counts, it is possible that there are still a few in the data. Conversely, suppose that an individual has been apprehended several times, but this has not been recognized so that this individual has been entered several times as several single persons. This will lead to the result that f1 is too large whereas fj with j > 1 is too small, leading to an overestimation. Apart from problems in the correctness of the data, it may also be that data do not follow the assumptions of the model. This will lead to biased estimates. Above we discussed the problem of positive contagion (leading to estimates that are too low) and negative contagion (leading to estimates that are too high). Another problem discussed above is unobserved heterogeneity, leading to estimates that are too low. It may very well be that unobserved heterogeneity always plays a role when the data are derived from a police registration. As regards the meaning of our population size estimate, one might wonder what it stands for?


275

C

Capture Recapture to Estimate Criminal Populations, Table 3 Parameter estimates, observed and estimated subpopulation sizes under Zelterman regression model. Estimated total population size is 84,862 (95 % CI is 80,293–89,431) Constant Male Female Age/10 Threat Waylay Physical Other viol. Psychological Sexual (Ex-)partner Friend family Child Elderly Parents Other family Other Native Dutch Morocco (2) Morocco (1) Turkey (2) Turkey (1) Suriname (2) Suriname (1) Dutch Antilles (2) Dutch Antilles (1) Not western (2) Not western (1) Western (2) Western (1) Not registered

Par 2.07 .49 .00 .02 .00 .20 .38 .20 .13 .42 .00 .40 .03 .05 .16 .11 .79 00 .10 .03 .03 .25 .05 .16 .64 .76 .28 .16 .08 .02 .54

SE .13 .09

p .00 .00

.02

.26

.13 .06 .11 .18 .15

.12 .00 .08 .46 .01

.16 .11 .26 .09 .09 .19

.01 .78 .84 .08 .18 .00

.18 .21 .18 .22 .15 .19 .39 .41 .34 .35 .12 .17 .11

.57 .89 .89 .26 .72 .40 .10 .07 .41 .65 .49 92 .00

Basically, the apprehended individuals can only be generalized to similar individuals who are not apprehended (but who are in principle apprehensible and where charges could be filed). Thus, the population size estimate does not stand for the total number of perpetrators of domestic violence; it stands for the apprehensible ones. The population estimate is still useful, since it represents the number of individuals who pose a threat to society and is thus an indication for the police potential workload.

Observed

Estimated

14,827 2,653

68,048 16,813

3,822 556 11,234 1,063 298 686 12,130 616 1,271 168 1,301 1,611 561 9,642 345 730 318 687 489 1,368 109 735 108 1,146 830 593 1,268

14,442 1,830 59,046 4,859 1,067 3,607 56,113 4,178 6,366 811 4,890 7,027 5,467 42,253 1,597 3,267 1,451 3,811 2,297 7,214 927 2,460 641 5,242 3,869 2,807 9,325

Related Entries ▶ Domestic Violence ▶ Feminist Theory and Domestic Violence ▶ History of the Dutch Crime Victimization Survey(s) ▶ History of the Self-Report Delinquency Surveys ▶ History of the Statistics of Crime and Criminal Justice ▶ International Crime Victimization Survey

C

C

276

▶ National Victimization Surveys ▶ Police Family Violence Services ▶ Repeat Victimization ▶ Self-Reported Offending: Reliability and Validity ▶ Surveys on Violence Against Women ▶ Understanding Victimization Frequency

Recommended Reading and References Bo¨hning D (2010) Some general comparative points on Chao’s and Zelterman’s estimators of the population size. Scand J Stat 37:221–236 Bo¨hning D, Del Rio Vilas VJ (2008) Estimating the hidden number of scrapie affected holdings in Great Britain using a simple, truncated count model allowing for heterogeneity. J Agric Biol Environ Stat 13:1–22 Bo¨hning D, van der Heijden PGM (2009) A covariate adjustment for zero-truncated approaches to estimating the size of hidden and elusive populations. Ann Appl Stat 3(2):595–610 Bouchard M (2007) A capture-recapture model to estimate the size of criminal populations and the risks of detection in a Marijuana cultivation industry. J Quant Criminol 23(3):221–241 Bunge J, Fitzpatrick M (1993) Estimating the number of species: a review. J Am Stat Assoc 88:364–373 Cameron AC, Trivedi PK (1998) Regression analysis of count data, vol 36, Econometric society monographs no. 30. Cambridge University Press, Cambridge Chao A (1988) Estimating animal abundance with capture frequency data. J Wildl Manag 52:295–300 Collins MF, Wilson RM (1990) Automobile theft : estimating the size of the criminal population. J Quant Criminol 6(4):395–409. Cruyff MJLF, van der Heijden PGM (2008) Point and interval estimation of the population size using a zero-truncated negative Binomial regression model. Biom J 50(6):1035–1050 Greene MA, Stollmack S (1981) Estimating the number of criminals. In: Fox JA (ed) Models in quantitative criminology. Academic, New York, pp 1–24 Gurmu S (1991) Tests for detecting overdispersion in the positive Poisson regression model. J Bus Econ Stat 9:215–222 Hilbe JM (2011) Negative Binomial regression. Cambridge University Press, Cambridge Johnson NL, Kotz S, Kemp AW (1993) Univariate discrete distributions, 2nd edn. Wiley, New York Mascioli F, Rossi C (2008) Capture-recapture methods to estimate prevalence indicators for the evaluation of drug policies. Bull Narc LX(1/2):5–25 Rossmo DK, Routledge R (1990) Estimating the size of criminal populations. J Quant Criminol 6:293–314 Seber GAF (1982) The estimation of animal abundance, 2nd edn. Griffin, London

Car Theft Van der Heijden PGM, Cruyff MJLF, van Houwelingen HC (2003a) Estimating the size of a criminal population from police registrations using the truncated Poisson regression model. Stat Neerl 57:289–304 Van der Heijden PGM, Bustami R, Cruyff MJLF, Engbersen G, van Houwelingen HC (2003b) Point and interval estimation of the truncated Poisson regression model. Stat Model 3:305–322 Van der Heijden PGM, Cruyff MJLF, Van Gils GHC (2009) Omvang van huiselijk geweld in Nederland. Ministerie van Justitie, Den Haag, WODC (33 pagina’s) Van Hest NHA, Grant AD, Smit F, Story A, Richardus JH (2007) Estimating infectious diseases incidence: validity of capture–recapture analysis and truncated models for incomplete count data. Epidemiol Infect 136:14–22 Wittebrood K, Junger M (2002) Trends in violent crime: a comparison between police statistics and victimization surveys. Soc Indic Res 59:153–173 Zelterman D (1988) Robust estimation in truncated discrete distributions with application to capturerecapture experiments. J Stat Plan Inference 18:225–237

Car Theft ▶ Motor Vehicle Theft

Career Criminals and Criminological Theory Raymond Paternoster Department of Criminology and Criminal Justice, University of Maryland, College Park, MD, USA

Synonyms Chronic offenders; Persistent offenders

Overview In 1986, the National Academies Press published a two-volume compendium entitled “Criminal Careers and ‘Career Criminals’”

Career Criminals and Criminological Theory

(Blumstein et al. 1986) that gave expression to the growing interest in both the field of criminology and among policy makers in the career criminal. Although there is no exact agreement on what a career criminal is, in the literature it has generally referred to persons who commit many crimes beginning at an early age and who persist in offending over the life course. Among other things, then, the career criminal is a habitual, persistent, or “chronic” criminal offender, committing criminal acts at every stage of the life course. Blumstein and colleagues’ report was not, however, the first instance of academic interest in the career criminal. From the 1930s to 1950s, Sheldon and Eleanor Glueck assembled several large longitudinal data sets that tracked offending from a young age to adulthood. Most notably, the Unraveling Delinquency Study (Glueck and Glueck 1950) which included 500 boys selected from two youth correctional centers in Massachusetts – the Lyman School for Boys and the Industrial School for Boys. Initial data were collected when the average age of the boys was 14, and they were followed up at age 25 and again at age 32. The presence of a persistent or career offender could be deduced from the fact that 55 % of these seriously delinquent boys experienced an arrest at some time between the ages 32 and 45 (Sampson and Laub 1993: 30). Additional interest in the notion of career criminals was sparked by Wolfgang and colleagues’ Philadelphia cohort studies. In the 1945 cohort, they classified youth with five or more police contacts as chronic or habitual offenders (Wolfgang et al. 1972). While these chronic offenders comprised only 6 % of the total birth cohort of nearly 10,000, they had 52 % of all police contacts and committed 63 % of all recorded index crimes, 71 % of all homicides, 73 % of all the rapes, and 82 % of the robberies. In a subsequent work, they followed a sample of this cohort up to age 30 and identified what they called “persistent” offenders as those who committed offenses both as juveniles and adults. Of the 975 boys who were followed up, 459 had at least one official arrest, and nearly 39 % of these were persistent offenders who had

277

C

an arrest as juveniles and adults. Further, these persistent offenders committed both more frequent and more serious offenses than those who committed crimes only as juveniles or only as adults (Wolfgang et al. 1987: 24).

C Key Issues and Themes Interest in the career criminal persists today among scholars and policy makers. With respect to the former, Laub and Sampson (2003) have obtained arrest histories for the Glueck boys up to age 70 and identified a small group (less than 5 % of the total) of chronic offenders who were experiencing arrests into their 60s. In a Dutch sample, Blokland (2005; Blokland and Nieuwbeerta 2005; Blokland et al. 2005) identified a small group of offenders (also less than 5 % of the total) who experienced criminal convictions from their teens to their 60s. Finally, Sohoni et al. (2013) found a group of offenders (about 4 % of the total) in the South London data of Farrington and West who experienced convictions from their teens to age 50. With respect to public policy interests, most states have habitual offender statutes that provide for enhanced penalties for offenders with several prior convictions as well as “three-strike” laws that provide for life imprisonment for those repeatedly convicted of serious felonies. Given the very strong interest in identifying the career criminal, it might perhaps be surprising that an equivalent amount of effort and attention has not been devoted to understanding the cause of persistent offending. The Glueck’s, for example, were loath to provide a comprehensive theoretical account either of delinquents in general or those who persisted in crime throughout their lives arguing that it was “premature and misleading to give exclusive or even primary significance to any one of the avenues of interpretation” (Glueck and Glueck 1950: 281). Instead, they put forward an atheoretical multifactored explanation that simply consisted of a list of the risk factors they empirically found to be related to crime. Wolfgang and his colleagues were silent with respect to how the

C

278

small group of chronic and habitual offenders they identified could be explained. It was really not until the mid-1980s rebirth of empirical interest in the career criminal that some attention was beginning to be devoted to understanding and explaining the high-rate offender who persists over time. In fact, the empirical work of career criminal researchers was in part responsible for the emergence of developmental and life-course theories of offending, theories that among other things were devised to explain why it is that offending takes place throughout the life course.

Theoretical Explanations for the Career Criminal There really is no theory of the career criminal in the sense of a separate theory of crime to explain this phenomenon. Rather, there are theories of persistent offending, without speaking to the career criminal, and there are general theories of offending that can be adapted to explain the habitual or persistent offender whose long-term pattern of offending can be described as a career in crime. Terrie Moffitt (1993) has presented a typological theory of criminal offending that provides some theoretical light on the career criminal. She argues for a distinction between two qualitatively distinct types of offenders: the adolescent-limited and the life-coursepersistent offender. The life-course-persistent offender would qualify as a career criminal since they are described as committing crimes at a high rate over a protracted period of time: as children, adolescents, and adults. They are a small group, comprising approximately of 5 % of any sample of youth, consistent with the Glueck’s high-rate offender and the chronic and habitual offender of Wolfgang et al. In accounting for the life-course-persistent offender, Moffitt appeals to an integration of constitutional and environmental factors. The life-course-persistent offender suffers initially from neuropsychological deficits that are either prenatal or neonatal in origin and are manifested


in profound cognitive and executive function difficulties (impulsivity and poor verbal skills are particularly important). Since infants with these kinds of neuropsychological deficits are frequently born in disadvantaged and criminogenic environments, these deficits frequently go uncorrected and are more often made worse. In a nutshell, a difficult child often overwhelms distressed and poorly resourced parents unable to cope with the challenge who frequently respond with erratic and harsh discipline. Such responses by caregivers only exacerbate the child’s initial difficulties, and as a result of this destructive interaction, the child fails to develop behavioral restraints – they are undersocialized. This negative interaction continues with development, and the poorly controlled toddler becomes the poorly controlled child who becomes the poorly controlled adolescent on into adulthood. As described by Moffitt (1993: 12) the life-course-persistent child becomes ensnared in an ongoing sequence of failed development: “personal characteristics such as poor self-control, impulsivity, and inability to delay gratification increase the risk that antisocial youngsters will make irrevocable decisions that close the doors of opportunity.” A related theory of the persistent offender, related because it appeals to the same set of empirical facts and develops a similar theoretical argument, was presented by Patterson and colleagues (1990). Like Moffitt, Patterson et al. define two etiologically distinct trajectories of offending. One starts early (the early-onset offender) and persists over the life course into adulthood in a way that would be harmonious with the idea of a career criminal. The behavioral trajectory of the early-starting offender is similar to Moffitt’s life-course-persistent offender. The second starts offending in mid- to late adolescence and ceases offending as they enter adulthood. The behavioral trajectory of the latestarting offender is similar to Moffitt’s adolescent-limited offender. In Patterson’s typology, the cause of the earlystarting and late-persisting offender is a combination of disruptive family practices (harsh discipline, emotional coldness and little


intimacy, little parental involvement with the child, and poor monitoring and supervision). The immediate result of these inept child-rearing practices is that the child learns to be coercive and aggressive both directly from parents (modeling their physical violence) or indirectly (parents tolerate or support the child’s violence) which over time culminates in an aggressive poorly controlled child. Although perhaps adaptive in the short term in the home, such verbally and physically aggressive behavior results in both academic and social failure in the school as well as rejection by normal peers and acceptance by equally disruptive peers. Finally, in a process that is reminiscent of Moffitt’s “ensnarement,” these aggressive adolescents are at a great risk for committing serious delinquent offenses as adolescents and criminal offenses as adults. There is an important common causal thread that runs through the theories of the persistent or career criminal in both Moffitt and Patterson’s work. While Moffitt’s theory admittedly has a neuropsychological component that is lacking in Patterson, both appeal to a disruption in family processes, particularly inadequate training or socialization by parents or caregivers. Caregivers who for various reasons are emotionally uninvolved with their children and therefore inadequately monitor and supervise their conduct create children who are impulsive, poorly controlled, and act in ways that may be in their immediate interest (ex: snatching a toy away from another child, stealing another child’s bike) but is destructive in the long term. Such poorly socialized children are not easily able to make up for past deficits and in fact misplay prosocial opportunities and experiences they confront. In other words, they never catch up and in fact fall further behind developmentally in diverse areas, with the ultimate outcome that they are at high risk for frequent adult criminal offending and other self-destructive behaviors such as drug use, alcohol abuse, long bouts of unemployment, and unhealthy social relationships. The preceding paragraphs may be very familiar to many criminologists since it describes a causal explanation for repeated offending that

279

C

is not tied directly to the career criminal paradigm, in fact one that has been hostile to that paradigm – Gottfredson and Hirschi’s (1990) general theory of crime. There are in fact several general theories of crime that provide an adequate theoretical explanation for the persistent offending that characterizes a criminal career. In 1990, Michael Gottfredson and Travis Hirschi (1990: 117) published what was to be a highly influential book titled A General Theory of Crime in which they presented a theory which could explain “all crime, at all times” and that could also explain behaviors that are as self-destructive in the long term as crime. The key theoretical construct they premised their theory on was the notion of low self-control which, while its precise meaning has changed somewhat over time, is manifested in a person’s inability to consider both the short- and long-term negative consequences of their actions. The person low in self-control, then, is attracted to behaviors which while they may provide gratification in the short term, such as crime and drug and alcohol abuse, have detrimental long-term consequences, and since the costs of actions are generally delayed relative to their benefits, they play a much less consequential role. Very similar to Moffitt’s and Patterson’s theories, Gottfredson and Hirschi trace the origins of low self-control to family disruption. In fact, there is a very close correspondence in the causal mechanisms underlying the person with low self-control and the life-course-persistent and early-starting offender. In each theory, an emotional connection or relationship between caregiver and child is absent, and this affective distance results in inept socialization. Caregivers who are not emotionally attached to the child are unlikely to monitor or supervise the child’s conduct properly and unlikely to correct inappropriate and reward appropriate behavior. Another important piece of common ground among the theories is that the initial failure to socialize has reverberations down the life course. Moffitt (1993), for instance, speaks about poorly socialized youth becoming “ensnared in a deviant life style by crime’s consequences” because they “will make irrevocable

C

C

280

decisions that close the doors of opportunity.” She links poor childhood socialization to teenage pregnancy, school dropout, drug addiction, unemployment, and incarceration such that “the behavior of life-course-persistent antisocial persons is increasingly maintained and supported by narrowing options for conventional behavior.” Patterson (1990: 266) in a process that parallels the notion of “ensnarement” argues that the child who is poorly socialized becomes aggressive and predatory at an early age and that such behavior leads to a lack of success in school and ostracism from normal peers in a causal spiral that over time leads to “school dropout, uneven employment histories, substance abuse, marital difficulties, multiple offenses, incarceration and institutionalization.” Similar to both of these, Gottfredson and Hirschi (1990: 96) account for the persistent offender by noting that because they are relatively unaffected by the negative consequences of their actions, the poorly socialized person with low self-control fails to take advantage of opportunities to change throughout life, and they too become ensnared in a cycle of repeated failure and persistent criminality: “The traits composing low self-control are also not conducive to the achievement of long-term individual goals. On the contrary, they impede educational and occupational achievement, destroy interpersonal relations, and undermine physical health and economic well-being.” There are, however, some important distinctions among the three theories. First, the theory of self-control argues that low self-control and therefore the propensity to become a habitual or career offender lies on a continuum rather than being a typology. They argued (1990: 95) that “though there will be little variability among people in their ability to see the pleasures of crime, there will be considerable variability in their ability to calculate potential pains” (emphasis added). Second, unlike Moffitt’s theory that relies on antecedent neuropsychological mechanisms or Patterson’s that relies on the mediating causal effect of deviant peers, Gottfredson and Hirschi locate the source of poor self-control solely in the family, and involvement with deviant peers has no causal


effect on persistent offending but is simply another manifestation of low self-control. Another general theory of crime that can offer us an explanatory account of the persistent offender is Sampson and Laub’s (1993, 2003) age-graded informal theory of social control. In discussing this theory, one cannot help but be struck with the substantial common ingredients in all of these theories of the persistent offender in spite of their supposed, much ballyhooed differences. Acknowledging the existence of a persistent criminal offender whose numbers may be small but whose maladaptive behavior is quite broad to cover marital problems, unemployment, drug and alcohol addiction, and the corporal punishment of children (heterotypic continuity), Sampson and Laub (1993: 123–138) appeal to a causal process that should by now be familiar to the reader. Children who are poorly socialized and antisocial are maintained or ensnared in their behavior through two causal mechanisms. First, antisocial behavior in children calls forth destructive and maladaptive responses by caregivers and others in their social environment, and such responses fuel further antisocial behavior by children, a process termed “interactional continuity” (Sampson and Laub 1993: 124). Second, the antisocial behavior of young children and the delinquent acts of adolescents have “negative structural consequences” for subsequent life chances that “may lead to the ‘closing of doors’ as far as opportunities go (for example, school failure, unemployment),” in a process termed “cumulative continuity.” In their 2003 book, Shared Beginnings, Laub and Sampson (2003: 150–195) discuss the persistent or habitual offender at some length and make a concerted effort to draw a distinction between their theoretical account and Moffitt’s. Their view of Moffitt’s theory is that it places great weight on early childhood factors (perhaps ignoring Moffit’s explanation of being ensnared in a life of crime over time), and they find little evidence either in the 2003 or earlier 1993 book that childhood risk factors have an effect on adult criminality. Instead, Sampson and Laub put the majority of their


eggs in the “adult life events” basket, arguing that persistence in adult crime is due to a weakness in adult social bonds. In point of fact, however, in further differentiating their position from so-called propensity theorists like Moffitt, Sampson and Laub also place great weight on a labeling process to account for the persistent or career offender. In Crime in the Making (1993: 137), they refer to a “structural version of labeling theory and state dependence” in which early arrest and incarceration experiences cut persons off from conventional opportunities: “the connection between official childhood misbehavior and adult outcomes may be accounted for in part by the structural disadvantages and diminished life chances accorded institutionalized and stigmatized youth.” In fact, in an earlier paper, Sampson and Laub (1997) described the criminal career as a “stable pattern of deviant behavior that is sustained by the labeling process.” In other words, contact with both the juvenile and adult criminal justice system creates problems of adjustment (Lemert 1967: 63) for people because it “denies them the ordinary means of carrying on the routines of everyday life open to most people” (Becker 1963: 35). In a reversal of the desistance process they describe, those with histories of imprisonment are “knifed off” from normal lives because of weakening social bonds. Sampson and Laub (2005: 171) in fact subsequently concede that “explanations of desistance from crime and persistent offending are two sides of the same coin.” Just as a strengthening of adult social bonds may knife off a criminal offender from a previous life of crime and criminal associates, so may criminal justice intervention knife them off from a conventional. It is important at this time to reiterate the similarities among all these theories of the persistent offender. In Moffitt’s, Patterson’s, Gottfredson and Hirschi’s, and Sampson and Laub’s theory, some childhood deviants grow up to be persistent adult offenders because they are all too frequently cut off from many conventional opportunities and fail at virtually every opportunity they are afforded to do good.

281

C

In addition, events that occur in childhood are important in initiating that causal process. As such, all of the theories we have discussed thus far reflect Hagan’s (1993) notion of the consequences of “criminal embeddedness.” Hagan (1993: 487) argues that youth crime and contacts with the justice system cut adolescents off from sources of social capital and other social networks that connect to conventional lives. He further asserts that youths in trouble instead acquire “criminal capital,” that is, they have social networks that connect them to illegal labor opportunities such as theft and drug dealing which embed them in a criminal lifestyle: “criminal youths are embedded in contexts that isolate them from the likelihood of legitimate adult employment.” Other general theories of crime offer their own theoretical account of the persistent, career criminal that are variations of the ones just discussed (with, of course, their own nuances). Farrington (2005) presented what he called an Integrated Cognitive Antisocial Potential (ICAP) theory that is a very eclectic account of criminal offending over the life course, including both desistance from and persistence in crime. According to ICAP, initiation into offending is due to a mixture of short-term and long-term antisocial potential (AP, which is defined simply as the potential to commit criminal acts) interacting with a person’s social environment, while persistence in criminal offending is due to the fact that an initial distribution of long-term AP is relatively stable over time. Reminiscent of Gottfredson and Hirschi, then, persistence in offending in Farrington’s theory is brought about because of the relative stability of an initial distribution of a time-stable individual trait, longterm AP, which consists of impulsiveness, the experience of strain, criminal modeling, socialization, and criminogenic life events. Thornberry (Thornberry and Krohn 2005) applies his general Interactional Theory of crime to account for the persistent, career offender. In this theory, long-term persistence in crime is accounted for by two developmental processes: (1) the causal factors that produce an introduction into crime are themselves stable

C

C

282

over time, and (2) there is a constancy in “negative temperamental traits” that are stable from childhood to adulthood. In a quasi-typological approach, Thornberry and Krohn (2005) suggest that for those who commit their first offense at an early age, the developmental deficits (poor parental supervision and monitoring, weak parental attachments, poor school performance and adjustment) are more extreme than for those whose onset into crime comes later, and since they are more extreme, they are also more stable over time. In addition to this, they describe a state-dependence-like process whereby early offending triggers hostile and maladaptive responses from the environment such as caregivers, teachers, and conventional peers. Essentially, early conduct problems ensnare youths by limiting conventional opportunities and social networks (i.e., they become embedded in a criminal lifestyle), restricting the development of social, personal, and human capital. In their own words, but words that seem to just as easily could have come from Moffitt, Patterson, Hagan, or Sampson and Laub, Thornberry and Krohn (2005: 198) posit that: “individuals who initiate antisocial behavior at very young ages are more likely than average to persist because the causal factors are likely to remain in place (both environmental and personality/trait factors) and because early involvement in antisocial behavior generates cumulative and cascading consequences in the person’s life course. All of this reduces the formation of social bonds and social capital and increases embeddedness in deviant networks and belief systems, foreclosing conventional lifestyles and entrapping the individual in deviant lifestyles.” This account for persistence in offending for those who onset early is not, however, likely to hold for those whose initiation into offending is more “on time” and occurs during adolescence. Nevertheless, some of these on-time onsetters can also persist in offending in their adult years, and they do so primarily because their involvement with crime puts them at higher risk for school failure, unstable and unsatisfying social relationships, and problems at home and in the workplace and disrupts their successful transition


into adult roles. Thornberry and Krohn (2005: 200) place great emphasis (as does Moffitt and Patterson for the adolescent-limited and late-onset offender) on the detrimental consequences of becoming embedded in deviant social networks for the youth who commits his first crime later rather than earlier. Bouffard and Piquero (2010) recently offered a theory of the career criminal offending that appeals to Sherman’s (1993) notion of defiance as an emotional reaction to labeling. Sherman had argued that when a poorly bonded person responds to the imposition of what they perceive to be an unfair sanction with anger rather than shame and an acknowledgement of harm, they become defiant and resentful and as a result are likely to commit further acts of crime rather than be deterred. Bouffard and Piquero (2010: 233) take this general theory of crime and apply it to the study of desistance and persistence arguing with respect to the latter that persistent offending is a “defiant response of a poorly bonded offender who defines their sanction as unfair and stigmatizing and refuses to acknowledge the shame they feel. These individuals may continue or escalate their offending.” What happens when a poorly bonded person gets labeled, then, is that they react in such a way (with defiance, resentment, and rejection) that their already weak bonds become even weaker in a process that by now should be very familiar. The ingredient that Bouffard and Piquero add to this mix is the role of emotions, a welcome addition since emotions are a neglected factor in criminological research. Wikstro¨m and Treiber (2009) presented a theory of the persistent or career offender that is a derivative of Wikstro¨m’s (2005) Situational Action Theory (SAT). According to SAT, all criminal offending is moral rule breaking with the explicit implication that a key explanatory variable is an individual’s moral values as well as the interaction between an individual’s personal characteristics (such as their moral values) and environmental factors. An individual with a strong moral position against committing a criminal act will not view crime as a viable course of action and will not, therefore, be


affected by contingencies in the environment, such as opportunities or the perceived costs and benefits of offending. It is also reasonable to conclude that such individual factors as an individual’s propensity to commit crimes as manifested in their self-control are activated only when crime is perceived to be a morally available course of action. In SAT, then, the interaction between a person’s moral values and their propensity to rule breaking will strongly influence what courses of action are available in a given situation and, therefore, what choices are available to be selected. With respect to the career criminal, Wikstro¨m and Treiber (2009: 411) make a distinction between perception, choice, and action in a given setting that is “either habitual or deliberate, depending on the actor’s familiarity with the setting” (emphasis is original). The more familiar an actor is to a given setting, the more their perceptions and behavior are likely to be guided by habit rather than deliberation. Actors who repeatedly find themselves in particular environments, such as one of extreme deprivation or abundant criminal opportunities with low moral restraints and little risk, will respond habitually to such contexts, and their behavior will tend to be highly stable and persistent over time. This constancy of the person-environment interaction and resultant habituation led Wikstro¨m and Treiber (2009: 411) to conclude that “there may be strong habitual elements in what drives much of persistent offenders’ criminality.” Finally, it must be noted that there is more than one biologically based theory of the persistent, career criminal, though the common ground seems to be that they all link habitual offending to an early onset of offending and the interaction between biological and environmental factors (see Savage 2009). One such example is Tibbetts (2009: 183) whose initial premise is that “early onset is the most important factor in predicting which individuals are most likely to become chronic, habitual offenders.” The secret to unlocking the origins of the career criminal, therefore, is to understand who onsets their offending early. In explaining the early-onsetting

283

C

offender, Tibbetts’ account relies on the interaction between a cluster of biological risk factors such as pre- or perinatal brain trauma, lower heart rate, arousal, slower brain wave patterns, deficits in pre- and perinatal nutrition, personality attributes (impulsivity), and a cluster of environmental deficits such as poverty, single-parent families, family size, and a criminogenic neighborhood. Whatever their particular composition, Tibbetts’ and other biologically inclined theories of persistent offending are clear in stating that it is not biological factors nor environmental factors that are themselves responsible for the persistent offending of the career criminal but the interaction between the two. Career criminals, offenders with both a high frequency and long duration of offending, have been the topic of theoretical discussion in criminology for over 20 years. One might get the impression from that fact that there is a great diversity of opinion with respect to the causes of career criminal offending. If so, one is in for a disappointment, for while there are numerous attempts to explain the persistent or habitual offender, for the most part the explanatory schemes are surprisingly similar, despite the many disciplines that have participated. While some theorists do place more of their eggs in an initial biological/psychological basket – those who eventually become persistent offenders have various constitutional, neuropsychological, or personality deficits – and others less, the natural progression to persistent offending is oddly very similar across many theories. The explanatory schemes involve notions of a state-dependent or dynamic process whereby initial developmental failures (hostile relationships with inept caregivers, failure at school, social rejection by normal peers) have reverberations that are felt down throughout the life course. It is not an oversimplification to say that toddlers who get into trouble with parents turn into adolescents who get into trouble with teachers and “good peers” who get into trouble with employers and spouses/partners who then get into trouble with the criminal justice system. In many if not all of these theories, there is a strong labeling component whereby initial

C

C

284

antisocial behavior that is responded to informally leads to formal criminal justice system intervention which then leads to a “knifing off” of legitimate and conventional routines of life. The repeat adolescent offender finds it hard in Lofland’s (1969) terms to have access to normal places, others, and hardware; as a result of which, they are only able to secure “criminal capital” and become more and more embedded in a criminal lifestyle (Hagan 1993). While one may lament the absence of theoretical variety in theories of the persistent offender (many in fact argue that persistence is simply the flip side of desistance), the brute fact is we know very little about how adequate these explanations really are. The extensive empirical work to determine what factors best explain persistent offending or the career criminal has hardly begun. While we know something about the process by which offenders drop out of crime, we know far less about the causal mechanisms behind repeated, habitual criminal offending.

Related Entries ▶ Age-Crime Curve ▶ Criminal Careers ▶ Criminal Careers and Public Policy Responses ▶ Desistance from Crime ▶ General Strain Theory ▶ General Theory of Crime ▶ History of Criminological Theories: Causes of Crime ▶ Individual Characteristics and General Strain Theory ▶ Integrating Rational Choice and Other Theories ▶ Labeling Theory ▶ Labelling, Deviance, and Media ▶ Lifestyle Theory ▶ Longitudinal Studies in Criminology ▶ Monetary Strain and Individual Offending ▶ Optimizing Longitudinal Studies in Offending ▶ Pathways to Delinquency ▶ Psychopathy and Offending ▶ Risk Factors for Adolescent Sexual Offending


Recommended Reading and References Becker HS (1963) Outsiders. Free Press, New York Blokland A (2005) Crime over the life span: trajectories of criminal behavior in Dutch offenders. Netherlands Institute for the Study of Crime and Law Enforcement, Leiden Blokland A, Nieuwbeerta P (2005) The effects of life circumstances on longitudinal trajectories of offending. Criminology 43:1203–1240 Blokland A, Nagin D, Nieuwbeerta P (2005) Life span offending trajectories of a Dutch conviction cohort. Criminology 43:919–954 Blumstein A, Cohen J, Roth JA, Visher CA (1986) Criminal careers and career criminals. National Academies Press, Washington, DC Bouffard LA, Piquero NL (2010) Defiance theory and life course explanations of persistent offending. Crime Delinq 56:227–252 Farrington DP (2005) The integrated cognitive antisocial potential (ICAP) theory. In: Farrington DP (ed) Integrated developmental and life course theories of offending. Transaction, New Brunswick, pp 73–92 Glueck S, Glueck E (1950) Unraveling juvenile delinquency. Commonwealth Fund, New York Gottfredson MR, Hirschi T (1990) A general theory of crime. Stanford University Press, Stanford Hagan J (1993) The social embeddedness of crime and unemployment. Criminology 31:465–491 Laub JH, Sampson RJ (2003) Shared beginnings, divergent lives: delinquent boys to age 70. Harvard University Press, Cambridge Lemert EM (1967) Human deviance, social problems, and social control. Prentice Hall, Englewood Cliff Lofland J (1969) Deviance and identity. Prentice-Hall, Englewood Cliffs Moffitt T (1993) Adolescence-limited and life-coursepersistent antisocial behavior: a developmental taxonomy. Psychol Rev 100:674–701 Nagin D, Paternoster R (1991) On the relationship of past and future participation in delinquency. Criminology 29:163–190 Sohoni T, Paternoster R, McGloin J, Bachman R (2013) Hen’s teeth and horses’ toes: the adult onsetter in contemporary criminology. Journal of Crime and Justice (forthcoming). Patterson G, DeBaryshe B, Ramsey E (1990) A developmental perspective on antisocial behavior. Am Psychol 44:329–335 Sampson RJ, Laub JH (1993) Crime in the making: pathways and turning points through life. Harvard University Press, Cambridge Sampson RH, Laub JH (1997) Al life-course theory of cumulative disadvantage and the stability of delinquency. In: Thornberry T (ed.) Developmental theories of crime and delinquency: Advances in criminological theory. Transaction Publishers, New Brunswick, NJ, pp 133–161

Careers in Organized Crime Sampson RJ, Laub JH (2005) A general age-graded theory of crime: lessons learned and the future of life-course criminology. In: Farrington DP (ed) Integrated developmental and life course theories of offending. Transaction, New Brunswick, pp 165–181 Savage J (2009) The development of persistent criminality. Oxford University Press, New York Sherman LW (1993) Defiance, deterrence, and irrelevance: a theory of the criminal sanction. J Res Crime Delinq 30:445–473 Thornberry TP, Krohn MD (2005) Applying interactional theory to the explanation of continuity and change in antisocial behavior. In: Farrington DP (ed) Integrated developmental and life course theories of offending. Transaction, New Brunswick, pp 183–209 Tibbetts SG (2009) Perinatal and developmental determinants of early onset of offending: a biosocial approach for explaining the two peaks of early antisocial behavior. In: Savage J (ed) The development of persistent criminality. Oxford University Press, pp 179–201 Wikstro¨m PH (2005) The social origins of pathways in crime: Towards a developmental ecological action theory of crime involvement and its changes. In: Farrington DP (ed) Integrated developmental and life-course theories of offending, vol 14, Advances in criminological theory. Transaction, New Brunswick, pp 211–246 Wikstro¨m P-OH, Treiber K (2009) What drives persistent offending? The neglected and unexplored role of the social environment. In: Savage J (ed) The development of persistent criminality. Oxford University Press, Oxford, pp 389–422 Wolfgang ME, Figlio RM, Sellin T (1972) Delinquency in a birth cohort. University of Chicago Press, Chicago Wolfgang ME, Figlio RM, Sellin T, Thornberry TP, Figlio RM (1987) From boy to man, from delinquency to crime. University of Chicago Press, Chicago

Careers in Organized Crime Edward R. Kleemans and Vere M. van Koppen VU University Amsterdam, Amsterdam, The Netherlands

Overview This entry elaborates upon theoretical and empirical insights into criminal careers in organized crime. The first section provides an overview of life-course criminology, older offenders, and adult onset. The second section briefly reviews

285

C

literature on criminal careers in white-collar crime and concludes that situational context is a key concept in explaining a large proportion of involvement in white-collar crime. The third section explores involvement mechanisms in organized crime: recruitment, social ties, work ties, leisure activities and sidelines, and life events (including financial setbacks). The fourth section reviews major studies on criminal careers in organized crime and describes how the findings of these studies differ from the main findings of developmental and life-course criminology. The last section comprises the main conclusions and suggestions for further research. Traditional research on criminal careers has mainly focused on juveniles, adolescents, and high-volume crime. Although much empirical progress has been made in developmental and life-course criminology (e.g., Farrington 2005), this research tradition tends to ignore certain kinds of offenders, particularly adult offenders, and certain types of crime, particularly organized crime, white-collar crime, and terrorism (Weisburd and Waring 2001; Leeper Piquero and Benson 2004; Steffensmeier and Ulmer 2005). Mainstream research focuses on property crimes and violent crimes, crimes for which opportunities are ubiquitous and open to anyone. Furthermore, many empirical longitudinal studies focus on a relatively early and short lifespan (Piquero 2008). These studies are therefore unable, almost by definition, to identify offenders who become engaged in crime only later on in life. There are several reasons why criminal careers in organized crime may be different from highvolume crime. Anybody is able to steal something or commit a violent act, but things are somewhat more complex in organized crime and particularly in cases of cross-border crime or “transit crime.” Nowadays, many activities of organized crime groups boil down to international smuggling activities such as drug trafficking, smuggling illegal immigrants, human trafficking for sexual exploitation, arms trafficking, trafficking in stolen vehicles, and other transnational illegal activities, such as money laundering and evasion of taxes (e.g., cigarette smuggling, value-added tax fraud, and

C

C

286

European Community fraud) (Kleemans 2007). Such activities pose different requirements than traditional high-volume crime (Kleemans and De Poot 2008). The first distinct feature is the greater importance of social relations in organized crime. It is simply impossible to get started without access to suppliers, clients, co-offenders, and profitable criminal opportunities. Trust is also important, as the financial stakes are high and the rules and mechanisms that make transactions in the legal world so much easier are absent: entering into contracts, paying via the official banking system, and – in case of disagreement – the availability of mediation or the judiciary (Reuter 1983). For this reason, existing social ties are used or illegal business relationships have to be built up. Not everyone has suitable ties that provide access to these profitable illegal opportunities, whereas building up such relationships takes time and energy. The second distinct feature of organized crime is the transnational character of many of these criminal activities. Many types of organized crime are based on international smuggling activities. Yet many people lack international contacts that may connect them to these illegal opportunities, as social networks are typically constrained by laws of geographical and social distance (e.g., Feld 1981). Thus, not all offenders have access to these transnational contacts and some only later on in life. A third common feature is that the activities are logistically considerably more complex and call for more co-offenders and specific expertise (e.g., Cornish and Clarke 2002). As more cooffenders are generally required for the successful commission of these crimes, seeking and finding suitable co-offenders is important (see, e.g., Tremblay 1993). Reliance on co-offenders from within one’s own social circle is not always sufficient, as they may not possess the necessary capabilities. Contacts with the legal world are also salient for transport, money transactions, and shielding activities from the authorities. Many people lack these contacts and expertise, and some acquire them only later on in life, e.g., through their professional activities and contacts.

Careers in Organized Crime

This entry elaborates upon theoretical and empirical insights into criminal careers in organized crime. The first section provides an overview of life-course criminology, older offenders, and adult onset. The second section briefly reviews literature on criminal careers in whitecollar crime and concludes that situational context is a key concept in explaining a large proportion of involvement in white-collar crime. The third section explores involvement mechanisms in organized crime: recruitment, social ties, work ties, leisure activities and sidelines, and life events (including financial setbacks). The fourth section reviews major studies on criminal careers in organized crime and describes how the findings of these studies differ from the main findings of developmental and life-course criminology. The last section comprises the main conclusions and suggestions for further research.

Life-Course Criminology, Older Offenders, and Adult Onset Developmental and life-course criminology place a strong emphasis on individual characteristics and long-term risk factors when explaining differences in criminal careers (see, e.g., Moffitt 1993). For example, Moffitt’s well-known “dual taxonomy” distinguishes between a large group of people who engage in crime and antisocial behavior mostly during adolescence (“adolescence-limited” offenders) and a small group of people who are antisocial from an early age, engage in crime in adolescence, and remain active in crime and other forms of antisocial behavior throughout their lives (“life-course persistent”). The explanation for “life-course-persistent” criminal behavior is generally sought in fairly stable biological or psychological characteristics – often deficiencies – such as low intelligence, high impulsivity, or low self-control. Moffitt argues that early problem behavior is an indicator that a person belongs to the life-course-persistent group and that antisocial and criminal behavior is likely to continue into later life. Gottfredson and Hirschi (1990) follow a similar route in their “general theory of


crime.” Although they deny the very existence of different “types” of offenders, their basic approach is quite comparable: people develop self-control early in life, and those who lack sufficient self-control engage in criminal acts and similar behavior – with short-term benefits and long-term costs such as drinking, smoking, and adultery – at higher rates than others throughout their lives. Many developmental and life-course theories of offending, though often more nuanced and empirically grounded, share this strong emphasis on individual characteristics and long-term risk factors and treat antisocial behavior in children as essentially similar to offending in adults (Farrington 2005). In this light, many theories assume that onset occurs early in life and that adult offending requires childhood antisocial behavior (see, e.g., Robins 1978). From this perspective, adult-onset offending is a rare phenomenon that needs no serious research effort (e.g., Gottfredson and Hirschi 1990). As a result, adult onset is an anomaly in developmental and life-course criminology and is treated in the same way as anomalies are generally treated in “normal science” (e.g., Kuhn 1962). Anomalies are discovered now and then but are often brushed aside empirically or theoretically (for an overview on adult onset, see Van Koppen et al. 2010). The general neglect of adult-onset offending can be attributed to two major factors. First, almost all research on criminal careers is concerned with high-volume crime, in which adult-onset offending seems relatively scarce. Second, the majority of trajectory research focuses on a relatively small (and mostly young) age frame and is therefore unable to identify adult-onset offenders by definition. Piquero (2008) provides an overview of all research using trajectory methodology in studying criminal activity over the life course between 1993 and 2006. These 90 studies reflect about 50 different samples. Only half of the samples include individuals after age 18, and only a fifth have data available after age 30. Birth cohort studies that do follow individuals long enough into adulthood systematically identity a significant share of adult-onset offenders. A quarter up to half of the

287

C

criminal samples of cohorts drawn in Racine, Stockholm, and Philadelphia turn out to start offending in adulthood (e.g., Eggleston and Laub 2002). Hence, the dominant way in which criminal careers are studied – longitudinally, over a short (adolescent) life-span, using self-reports with limited follow-ups, and mainly focused on highvolume crime and (re)traceable young offenders – reinforces received wisdoms that might be challenged by other types of research, focusing on other types of crime, such as white-collar crime and organized crime.

Criminal Careers in White-Collar Crime Over the past years, a number of studies have examined criminal careers in white-collar crime. Although there is a difference between whitecollar crime and organized crime, both concern more uncommon types of crime, and both may reveal different patterns than the ones discovered in more traditional research. Until 1980, most of what was known concerning white-collar offenders and their crimes was extracted from case studies. From 1980 on, datasets were constructed containing white-collar crimes committed in the United States (Leeper Piquero and Benson 2004; Weisburd and Waring 2001). These new efforts immediately refuted a common belief about white-collar offenders. They are not all one-time offenders, as had been stated before (see, e.g., Edelhertz and Overcast 1982). Almost half of the white-collar offenders have had at least one other official contact with the criminal justice system (Benson and Moore 1992). Furthermore, these criminal careers turn out to differ substantially from careers of highvolume crime offenders on several aspects. Or as noted by Leeper Piquero and Benson (2004), “the typical white-collar offender greatly differs from the typical street offender and does not appear to fit into the proposed explanations of life-courseoffending patterns.” Based on their findings on white-collar criminals, Leeper Piquero and Benson (2004) propose a theory on career patterns in white-collar

C

C

288

crime called punctuated situationally dependent offending. It assumes that white-collar criminals are criminally active during adolescence (just like high-volume crime offenders), then desist for a while, and begin to offend again when they reach their 30s or 40s. External factors are used to explain this revival later in life. One explanation involves a personal or occupational crisis the offender is going through. A second explanation relates to opportunities presented when a certain occupational status has been reached. Situational context is a key concept in explaining a large proportion of involvement in white-collar crime (see for a review, e.g., Benson and Simpson 2009). Results of a study by Weisburd and Waring (2001) corroborate this idea: the white-collar offenders studied differ on various career aspects from offenders of high-volume crime, the age of onset was relatively late, and they were criminally active for a substantially longer period and were arrested for fewer offenses throughout their criminal career. These lines of research show the importance of incorporating new groups of offenders.

Organized Crime: Involvement Mechanisms Before contemporary research on criminal careers in organized crime will be discussed further below, several involvement mechanisms are presented, explaining processes and situations that provide opportunities for organized crime. How do people get involved in organized crime? Recruitment According to the standard idea of recruitment, criminal “organizations” recruit outsiders who may move up in the hierarchy by proving their suitability. They start by doing dirty jobs and are ultimately rewarded. Behind this idea lies the assumption of a more or less static criminal organization (cf., e.g., Cressey 1969), while research shows that, much more often, the world of organized crime consists of criminal networks in which offenders may cooperate with one another


in changing constellations (for a review, see, e.g., Morselli 2009). Within these criminal networks, there often are nodal actors, on whom many other offenders depend because of their money, knowledge, and contacts. These nodal actors resurface time and again, in different criminal investigations and different criminal collaborations. Still, other offenders may gradually become less dependent on these nodal actors because they accumulate money, knowledge, and contacts for themselves and subsequently develop increasingly more criminal activities of their own. Social Ties and the Social Snowball Effect Social ties are very important in organized crime. Opposite to the standard view of recruitment, social ties create a “social snowball effect”: people get involved in organized crime through their family, friends, and acquaintances; as they go along, their dependency on the resources of other people (such as money, knowledge, and contacts) gradually declines; and they generate new criminal collaborations. They involve people from their own environment in their activities, keeping the snowball in motion (Kleemans and Van de Bunt 1999: 29–34). Examples comprise partners, girlfriends, spouses, sons, daughters, brothers, sisters, and half-brothers, who get involved in criminal activities; a father who starts helping his son in drug trafficking; or a smuggled Iranian who ultimately becomes responsible for human smuggling operations himself. Hence, the choices people make are prestructured, and the opportunities that present themselves are not presented to anyone. Similar prestructured choices relate to three other involvement mechanisms, which are encountered when analyzing starters in organized crime. Within the context of the Dutch Organized Crime Monitor, 120 major investigations into organized crime groups have been analyzed systematically (see for more information: Kleemans 2007). In 2008, the criminal careers of 92 “starters” were analyzed, suspects that did not have any other judicial contacts before and who had not “progressed” from high-volume crime into organized crime (Kleemans and De Poot 2008). Next to deliberate recruitment, emphasis


was placed upon social ties and the social snowball effect, work ties, leisure activities and sidelines, and life events (including financial setbacks). Work Ties Generally, people associate occupations with white-collar crime, yet organized crime is equally dependent on contacts with the licit world. Therefore, occupations are also important in explaining involvement in organized crime. Just like social ties cross boundaries between legality and illegality, so do some work settings in which people spend a considerable share of their daily life. Several occupational characteristics may help in providing opportunities for organized crime (Kleemans and Van de Bunt 2008). First, the social nature of work activities may create opportunities; individuals who often interact with others can meet with potential co-offenders. Second, international contacts and travel movements may create such opportunities. Contacts with other countries and other social groups enable someone to discover and effectuate certain possibilities to commit (transnational) criminal activities. Examples are occupations involving mobility, transport, and logistics. It is the story of the fruit trader who also uses his contacts and knowledge for the import and export of drugs; airline personnel who travel on a regular basis, and who know how to evade baggage checks in airports; and an international trader in clothes who also engages in human trafficking and underground banking. Third, the individual freedom of action of particular occupations is important, or in other words: autonomy. This explains the involvement of directors of (usually) small businesses, independent professionals, and relatively autonomous persons employed by larger organizations, such as corporations and banks. It also explains the relative absence of professionals in the salaried employment of the government or large organizations. Similar findings on occupations that emphasize the assets of international social contacts and autonomy follow from interviews with incarcerated offenders, most notably highlighting the salience of people working in logistically important

289

C

sectors such as boating, fishing, import and export, transport, and the airline sector (e.g., Dorn et al. 2005). Leisure Activities and Sidelines The chances of getting into contact with organized crime differ from person to person and from occupation to occupation. Yet people from different social worlds, involved in disconnected networks, may also run into each other through leisure activities and sidelines. Local bars, nightlife, parties, after parties, drug joints, brothels, shooting clubs, or motor clubs, these are all places where all sorts and conditions of people can meet. These meeting places are especially important for forging contact between representatives of the “underworld” and the “licit world,” who would not meet each other so easily in everyday life. This is why these places also play a vital role in the coming about of involvement in organized crime. Life Events and Financial Setbacks A final involvement mechanism relates to life events. In developmental and life-course criminology, the emphasis is on the positive effects of life events such as getting a permanent job, a stable relationship, or having children (Laub and Sampson 2003). These positive life events make problems like crime disappear like snow in summer. Involvement in organized crime, however, results from the negative sides of life events, particularly among men in middle age: divorce, illness, disability, unemployment, and bankruptcies; these are all moments when people are suddenly confronted with setbacks, socially and financially. They are down and out, bankrupt or deeply in debt, and subsequently some of them weigh pros and cons differently when generous financial investors appear on the scene or lucrative illegal opportunities present themselves. What these five involvement mechanisms have in common is that the opportunities presenting themselves to people are prestructured in a way which those people do not necessarily have to realize. Options differ from person to person and are different at the beginning of one’s life course compared to later on in life.

C

C

290

This is also one of the reasons why criminal careers in organized crime may evolve differently than traditional high-volume crime careers, as the options are not open to anyone and generally only provide themselves later on in life for a selected group of criminals and noncriminals.

Criminal Careers in Organized Crime Empirical research into criminal careers in organized crime is scarce, yet recent studies have challenged several basic facts and received wisdoms of developmental and life-course criminology. For one, the age-crime curve describing that in general the prevalence of offending rises steeply in the early teenage years and reaches a peak between the ages of 15 and 17, followed by a decline over the rest of the life course. For another, the dominance of individual differences (risk factors) explaining persistence in offending for a very small group of offenders (“life course persistent offenders”) as opposed to the majority of “adolescence-limited offenders” (e.g., Moffitt 1993). First, juveniles are almost absent in organized crime, and most offenders are quite old, in their 30s, 40s, 50s, or even older. Second, a large share of organized crime offenders get involved in crime only at a later age. Late starters are not exceptional, as in traditional research on criminal careers, yet comprise a substantial group of offenders (Kleemans and De Poot 2008). This finding of a substantial presence of “late-onset” offenders is robust across several different criminal activities (drugs, fraud, and other activities) and different roles in criminal groups (leaders, coordinators, and lower-level suspects) (Van Koppen et al. 2010). These findings contradict the general idea that crimes at a later age should be preceded by crimes in adolescence and that individual characteristics and long-term risk factors are the major explanation for a life in crime, let alone a life in serious and organized crime. The scarce literature on organized crime offenders often resonates the preoccupation with


special characteristics of offenders. For instance, Bovenkerk (2000) states that certain characteristics that are conducive to a career in business (based on the “Big Five” from test psychology) are also conducive to a career in organized crime. Basically, the personality traits of successful legal entrepreneurs would be mirrored in the personality traits of illegal entrepreneurs: extraversion, controlled impulsiveness, a sense of adventure, megalomania, and narcissistic personality disorder. This line of reasoning is based upon ideas of “the survival of the fittest” in which only offenders with certain characteristics end up in organized crime or in certain leadership positions. However, there are important lines of research that raise objections against this traditional focus on individual characteristics. Some offenders of organized crime are quite normal and well adapted in many respects, even though they are involved in serious forms of crime. A significant amount of offenders have legitimate jobs and sources of income. Moreover, not all crimes are the same nor just symptoms of latent characteristics such as low self-control and a variety of people can become involved in organized crime (e.g., Steffensmeier and Ulmer 2005). Next to individual personality differences, these studies draw attention to situational factors that may be important to explain involvement in organized crime. Most of what is known on organized crime offenders, comes from (auto)biographies and case studies of mafia groups (e.g., Paoli 2003; Steffensmeier and Ulmer 2005). These studies, based on memoirs and testimonies, illustrate the complexity of a criminal life course, including contacts, situational context, and processes such as persistence and desistance. Morselli (2005) describes in detail the careers of Howard Marks and “Sammy The Bull” Gravano. Based upon Burt’s work on brokerage (Burt 2005), he has put forward the idea that “brokers do better” and that advancements in criminal careers can be explained by deliberate choices people make in investing in certain relationships. Based upon Marks’ autobiography, Morselli focuses on the personal network that explains why Marks was


able to be successful in the cannabis trade. Central issues are brokerage (linking disconnected networks), the number of (non-redundant) contacts, and social network dynamics. One of the main problems of case studies is that one may wonder whether these case studies reflect the average or exceptional offender in organized crime and whether these case studies are able to capture the full range of different pathways in organized crime. Case studies may be biased toward more interesting, more talkative or defective, atypical offenders. Moreover, autobiographies often focus on people with long criminal careers instead of “late-onset” offenders. Two studies, however, comprise a broader, comparative view on several criminal careers in organized crime. The first study is based upon quantitative and qualitative information about the criminal careers of 979 suspects who were involved in 79 different organized crime cases that have been analyzed in the context of the Dutch Organized Crime Monitor (Kleemans and De Poot 2008). First, key figures of these criminal careers are compared to careers of the total offender population. Offenders in the studied organized crime cases are mostly men between 30 and 50 years old. They are generally older than the total offender population, and the studied group of suspects includes no juveniles at all. The age distribution seems to reflect the fact that it is a highly specific group that does not come within the focus of traditional research into criminal careers. Kleemans and De Poot also analyzed criminal careers of 92 “starters,” suspects who did not have any other judicial contacts before and who had not “progressed” from high-volume crime into organized crime. Five involvement mechanisms were described in more detail: deliberate recruitment by criminal groups, social ties and the social snowball effect, work ties, leisure activities and sidelines, and life events (including financial setbacks) (see above). Finally, they analyzed the careers of 66 (ring) leaders and “nodal” offenders in more detail to find out how their criminal careers had developed. Some (ring) leaders had a long criminal career: the versatile but regionally constrained “local hero”

291

C

(versatility as a result of both opportunities and constraints of the local context); progression through an increase in scale (specialization, particularly in international drug trafficking); progression through capital accumulation (investing in drug transports (while refraining from hands-on activities) and becoming a “background operator”; and expertise, contacts, and network formation (people getting a central position in criminal networks because many offenders and criminal groups need their contacts and expertise). Surprisingly, a significant share of (ring) leaders could be typified as “late-onset” offenders, who had made a switch from a legal occupational background to organized crime. Among these 32 offenders, a distinction could be made between a group of 19 with a background in legal trade (including import and export) and a group of 13 with other types of occupation – people from the business sector, the construction industry, assembly, hotels and catering, financial services, or government. Three different types of “late onset” were described: criminal activities extending from legal activities (opportunities that arise during day-to-day work, particularly in fraud cases); people who obviously switch careers, from legal to illegal commodities, motivated by the big profits that can be obtained in the trade in prohibited commodities such as narcotics; and finally, people seizing upon criminal opportunities only later on in life after specific, significant life events (including financial setbacks and problematic debt situations). Both the research on “starters” and on “lateonset (ring) leaders” illuminate the mechanisms behind the “anomaly” of “adult onset.” In essence, the story is that explaining involvement and success in organized crime requires more than a focus on individual differences and longterm risk factors. Kleemans and De Poot (2008) explain their findings by social opportunity structure – social ties providing access to profitable criminal opportunities. Social opportunity structure is unequally distributed across the population and across age. Certain kinds of people have social ties that give them direct access to illegal opportunities. Major examples are migrants, who

C

C

292

connect source countries of narcotics to Western consumer markets, and people with occupations that are connected to mobility, transport, and logistics. Furthermore, social opportunity structure is unequally distributed across age. Some offenders lack necessary social ties at a younger age, which explains why some offenders become involved in organized crime only at a later age. It also explains the interesting phenomenon of “late starters” – people without any appreciable criminal history – and legally employed people switching careers. The second study that takes a comparative view on careers in organized crime was a follow-up study by Van Koppen et al. (2010). They elaborated upon this research and investigated criminal trajectories of 854 suspects, based upon quantitative and qualitative data from the Dutch Organized Crime Monitor. A semiparametric group-model was used to cluster 854 individuals into groups with similar developmental trajectories. The most important finding of this study relates to the substantial group of adult-onset offenders (40 %) and a group without any previous criminal records (19 %), next to a group of early starters (11 %) and a group of persisters (30 %). Up till then, no trajectory study had ever discovered such a vast share of adult-onset offenders. Furthermore, the findings turn out to be quite robust across different kinds of criminal activities (drugs, fraud, and other criminal activities) and different roles in criminal groups (leaders, coordinators, and lower-level suspects). Adult-onset offenders emerge in several kinds of criminal activities and in several kinds of roles in criminal groups. Hence, it is not true that the amount of fraud cases in the sample would explain the substantial share of late-onset offenders; late onset is equally important in other types of criminal activities such as drug trafficking. Neither could roles in criminal groups explain “late onset”: late onset seems to be equally present in lower-level suspects as in coordinators and (ring) leaders. The mechanisms described in the study by Kleemans and De Poot (2008) may explain these robust findings, as these mechanisms may equally apply to different kinds of criminal


activities and to people with different types of roles in criminal groups.

Conclusion and Challenges for Future Research Differences between organized crime and highvolume crime may explain several interesting findings that challenge basic findings from developmental and life-course criminology. First, juveniles are almost absent in organized crime, and most offenders are quite old, in their 30s, 40s, 50s, or even older. Second, a large share of offenders get involved in organized crime only at a later age. Late starters are not exceptional, as in traditional research on criminal careers, yet comprise a substantial group of offenders. This finding of a substantial presence of “late-onset” offenders is robust across several different criminal activities (drugs, fraud, and other activities) and different roles in criminal groups (leaders, coordinators, and lower-level suspects). These findings contradict the general idea that crimes at a later age should be preceded by crimes in adolescence and that individual characteristics and long-term risk factors are the major explanation for a life in crime, let alone a life in serious and organized crime. The findings can also be explained by different involvement mechanisms that play a role in organized crime. Next to deliberate recruitment, four other involvement mechanisms play a key role: social ties and the social snowball effect, work ties, leisure activities and sidelines, and life events (including financial setbacks). What these involvement mechanisms have in common is that the opportunities presenting themselves to people are prestructured. Options differ from person to person and are different at the beginning of one’s life course compared to later on in life. This is also one of the reasons why criminal careers in organized crime may evolve differently than traditional high-volume crime careers, as the options are not open to anyone and generally only provide themselves later on in life for a selected group of criminals and non-criminals.


Kleemans and De Poot (2008) use the theoretical concept of social opportunity structure – social ties providing access to profitable criminal opportunities. Social opportunity structure is unequally distributed across the population and across age. Certain kinds of people have social ties that give them direct access to illegal opportunities. Major examples are migrants, who connect source countries of narcotics to Western consumer markets, and people with occupations that are connected to mobility, transport, and logistics. Furthermore, social opportunity structure is unequally distributed across age. Some offenders lack necessary social ties at a younger age, which explains why some offenders become involved in organized crime only at a later age. It also explains the interesting phenomenon of “late starters” – people without any appreciable criminal history – and legally employed people switching careers. An important new insight from the literature is that occupations do not only play a major role in white-collar and corporate crime but in organized crime as well. Many studies on organized crime tend to draw a sharp distinction between licit activities and illicit activities and between the licit world and the underworld. Some studies focus on the infiltration or extortion of licit economic sectors by organized crime, while other studies tend to emphasize symbiosis and “interfaces” between legal and illegal actors. Empirical research, however, indicates that both worlds may converge. Occupations, work relations, and work settings may provide the breeding ground for organized crime activities, particularly transit crime. Furthermore, occupations offer work relations and work settings that are directly relevant for providing opportunities to search for suitable co-offenders, to gain trust, and to exercise control. Social ties cross boundaries between legality and illegality, as do some work settings in which people spend a considerable share of their daily life. Empirical research highlights the salience of social relations, work relations, and work settings for criminal activities, also in cases of organized crime. From a theoretical point of view, these findings are important in bridging ideas from social network theory and opportunity theory.

293

C

Social relations are influenced by “foci of interaction” (Feld 1981). Social relations do not happen at random but often obey the laws of social and geographical distance, producing a certain kind of clustering based on geographical distance, ethnicity, education, age, etc. (e.g., Burt 2005). For many people, work is also an important focus of interaction. Second, social relations and work relations often coincide, either because people find a job through social relations or because working relationships develop into more or less close social relations. Third, the blending of work relations and social relations may also contribute to the effective concealment of illegal activities. For a variety of reasons, potential witnesses may remain silent, look the other way, or feel reluctant to blow the whistle. Effective concealment does not require secret societies. “Walls of silence” may also emerge in licit worlds that are mobile, dynamic, and social (Van de Bunt 2010). Studies on criminal careers in organized crime pose several challenges for criminal career research. First, these studies might refocus attention to social context and co-offenders promoting certain pathways in crime. Co-offending is still an understudied aspect in criminal career research. Second, these findings challenge the traditional boundaries between legality and illegality. Social ties cross boundaries between legality and illegality, as do some people’s activities at particular times in their lives. Third, the studies on organized crime careers suggest that criminal career research should focus not only on adolescence but also on later stages in life, as important changes in criminal careers occur with age. Fourth, studying these stages in life should go beyond a preoccupation with desistance. Steffensmeier and Ulmer (2005) state that desistance cannot be explained by stakes in conformity alone; another important factor is a lack of profitable criminal opportunities or the reverse. One could argue that this is particularly salient during stages in life when making money is more important than in adolescence. It could also provide an answer to the question why successful offenders in organized crime never seem to “retire” and continue

C

C

294

along the path of crime. Finally, some people seize upon illegal opportunities, while others do not. Clearly, there is both “agency” and “opportunity” (Laub and Sampson 2003). The question why, when confronted with opportunities, some people turn to crime and others do not, and under which conditions, is still unresolved and needs further exploration.

Related Entries ▶ Age-Crime Curve ▶ Career Criminals and Criminological Theory ▶ Co-offending ▶ Criminal Careers ▶ Criminal Careers and Public Policy Responses ▶ Moffitt’s Developmental Taxonomy of Antisocial Behavior ▶ Organized Crime, Types of ▶ Pathways to Delinquency ▶ Social Network Analysis of Organized Criminal Groups

Recommended Reading and References Benson ML, Moore E (1992) Are white-collar and common offenders the same? An empirical and theoretical critique of a recently proposed general theory of crime. J Res Crime Delinq 29:251–272 Benson M, Simpson S (2009) White collar crime: an opportunity perspective. Routledge, New York Bovenkerk F (2000) Wanted: Mafia boss: essay on the personology of organized crime. Crime Law Soc Change 33:225–242 Burt RS (2005) Brokerage & closure: an introduction to social capital. Oxford University Press, Oxford Cornish DB, Clarke RV (2002) Analyzing organized crime. In: Piquero R, Tibbetts SG (eds) Rational choice and criminal behavior: recent research and future challenges. Garland, New York, pp 41–62 Cressey DR (1969) Theft of the nation: the structure and operations of organized crime in America. Harper & Row, New York Dorn N, Levi M, and King L (2005) Literature review on upper level drug trafficking. Home Office Online Report 22/05 Edelhertz H, Overcast TD (1982) White collar crime: an agenda for research. Lexington Books, Lexington Eggleston EP, Laub JH (2002) The onset of adult offending: a neglected dimension of the criminal career. J Crime Justice 30:603–622

Careers in Organized Crime Farrington DP (2005) Integrated developmental and life-course theories of offending: advances in criminological theory 14. Transaction Publishers, New Brunswick Feld SL (1981) The focused organization of social ties. Am J Sociol 86(5):1015–1035 Gottfredson MR, Hirschi T (1990) A general theory of crime. Stanford University Press, Stanford Kleemans ER (2007) Organized crime, transit crime, and racketeering. In: Tonry M, Bijleveld C (eds) Crime and justice in The Netherlands: crime and justice: a review of research, vol 35. University of Chicago Press, Chicago, pp 163–215 Kleemans ER, De Poot CJ (2008) Criminal careers in organized crime and social opportunity structure. Eur J Criminol 5(1):69–98 Kleemans ER, Van de Bunt HG (1999) The social embeddedness of organized crime. Trans Organized Crime 5:19–36 Kleemans ER, Van de Bunt HG (2008) Organised crime, occupations, and opportunity. Global Crime 9(3):185–197 Kuhn TS (1962) The structure of scientific revolutions. University of Chicago Press, Chicago Laub JH, Sampson RJ (2003) Shared beginnings, divergent lives: delinquent boys to age 70. Harvard University Press, Cambridge Leeper Piquero N, Benson ML (2004) White collar crime and criminal careers: specifying a trajectory of punctuated situational offending. J Contemp Crim Justice 20:148–165 Moffitt TE (1993) Adolescence-limited and life-course persistent anti-social behavior: a developmental taxonomy. Psychol Rev 100:674–701 Morselli C (2005) Contacts, opportunities, and criminal enterprise. University of Toronto Press, Toronto Morselli C (2009) Inside criminal networks. Springer, New York Paoli L (2003) Mafia brotherhoods: organized crime, Italian style. Oxford University Press, New York Piquero AR (2008) Taking stock of developmental trajectories of criminal activity over the life course. In: Liberman AM (ed) The long view of crime: a synthesis of longitudinal research. Springer, New York Reuter P (1983) Disorganized crime: illegal markets and the Mafia. MIT Press, Cambridge Robins LN (1978) Sturdy childhood predictors of adult antisocial behaviour: replications from longitudinal studies. Psychol Med 8:611–622 Steffensmeier D, Ulmer JT (2005) Confessions of a dying thief: understanding criminal careers and criminal enterprise. Transaction Aldine, New Brunswick Tremblay P (1993) Searching for suitable co-offenders. In: Clarke RV, Felson M (eds) Routine activity and rational choice: advances in criminological theory, vol 5. Transaction Publishers, New Brunswick, pp 17–36

Causes of Police Legitimacy Van de Bunt H (2010) Walls of secrecy and silence. The Madoff case and cartels in the construction industry. Criminol Public Policy 9(3):435–453 Van Koppen MV, de Poot CJ, Kleemans ER, Nieuwbeerta P (2010) Criminal trajectories in organized crime. Br J Criminol 50(1):102–123 Weisburd D, Waring E (2001) White collar crime and criminal careers. Cambridge University Press, Cambridge

Case Management ▶ Behavioral Management in Probation

Causal Processes ▶ Situational Action Theory

Cause of Police Legitimacy ▶ Democratic Policing

Causes ▶ Establishing Causes of Offending in Longitudinal and Experimental Studies

Causes of Police Legitimacy Tammy Rinehart Kochel Southern Illinois University, Carbondale, IL, USA

Overview Police legitimacy plays a crucial role in promoting a variety of crucial, pro-social outcomes. Research demonstrates the capacity of police legitimacy to promote law-abiding behaviors, cooperation with police, reporting crime and suspicious behaviors,

295

C

support for police, and public collaboration with police (Fagan and Tyler 2004; Fagan and Tyler 2005; Gau 2011; Kochel et al. 2011; Reisig et al. 2007; Sunshine and Tyler 2003; Tankebe 2009; Tyler 1990; Tyler 2006; Tyler and Fagan 2008). These outcomes are particularly important to the efficiency and effectiveness of police within democratic contexts. Police rely on public support for resources and authorization, and even more importantly, police are expected to generate voluntary compliance to the law and to police directives by a majority of citizens, rather than using force and fear to coerce cooperation (Tyler 2004). Jackson and colleagues (2012) suggest that normative-based compliance (such as that motivated by perceptions of police legitimacy) is more stable than instrumentally motivated compliance. Police cannot be at all places at all times, thus minimizing the potential effectiveness of a deterrence-approach for securing order. Instead, police in democratic contexts rely on a social culture supportive of law-abiding behavior and order. Police legitimacy serves as a basis for this social culture (see “▶ Procedural Justice, Legitimacy, and Policing” and “▶ Police Legitimacy and Police Encounters” in this volume to learn about the consequences of police legitimacy). Views about police legitimacy and other judgments about police tend to be fairly stable in the United States and elsewhere, but scholars have reported declines in police legitimacy in recent decades LaFree 1998; Tyler et al. (1997). Thus, understanding the causes of or contributors to police legitimacy and whether patterns of police legitimacy and the factors that promote it vary under different contexts or across different characteristics of people has important practical implications for police, the public, and democracies in general. Tyler (1990) suggests that policing style and the process of policing may influence individuals’ judgments about police legitimacy. This encyclopedia entry describes, based on past theoretical and empirical research, three factors that may contribute to police legitimacy – procedural justice, crime control effectiveness, and distributive justice. The entry describes the three models, their relative impacts, and whether these effects vary under different conditions.

C

C

296

Causes of Police Legitimacy

Background: Defining Legitimacy

Key Issues

Police legitimacy is a normative perspective that police authority is valid and ought to be obeyed and respected. Police have legitimacy when members of the public believe police have a right to their authority and thus that citizens have a moral obligation and civic duty to honor, accept, and adhere to that authority (Tyler and Huo 2002). The concept of institutional legitimacy originates with Max Weber (1947). He first explained that citizens who view an authority as legitimate feel a personal sense of responsibility to act consistently with that authority’s expectations, even in the absence of that authority and even when the compliant behavior contradicts self-interests. Although this definition of legitimacy is widely acknowledged, research examining the correlates and consequences of police legitimacy has suffered from construct validity problems (Gau 2011; Reisig et al. 2007). Scholars have sometimes operationalized legitimacy to include the anticipated consequences of police legitimacy, such as compliance with the law or cooperation with police, and at other times, legitimacy measures incorporate theorized antecedents, especially elements of procedural justice. Additionally, while most studies of police legitimacy incorporate into the measure a perceived obligation to obey, some of Tom Tyler’s studies (the present day scholar who refocused the criminologists and criminal justice practitioners on this issue with a series of survey studies) and other studies incorporate moral trust or institutional trustworthiness as a component to legitimacy. This component is not part of Weber’s initial construct, and recent psychometric studies have suggested that trust does not combine with views about the obligation to obey to form a legitimacy measure (Gau 2011; Reisig, et al. 2007). Thus, this entry aims to provide the status of current research on police legitimacy conceptualized as the perceived moral obligation to voluntarily obey police, while avoiding some of the construct validity pitfalls experienced by it.

One way individuals form opinions about whether and the degree to which police authority is legitimate is based on personal or vicarious experiences with police. While Tyler (1990) found that only about four percent of the variance of legitimacy in a Chicago sample was explained by individual personal experiences with police, Tyler and Huo (2002) and Tyler (2001) saw a much stronger impact of personal experience on legitimacy judgments, explaining up to 34 % of the variation in legitimacy. Other sources include stories and media coverage about police behaviors and accomplishments, peer assessments of police, generalized understandings about the institution of police, and demographic and cultural characteristics of the individual and context that may color individuals’ perceptions. Past scholarship has focused on how three aspects of these experiences and attitudes may contribute to police legitimacy. Procedural justice, distributive justice, and police effectiveness are the three antecedents expected to promote police legitimacy.

The Process-Based Model/Procedural Justice Model Procedural justice refers to an impartial process that is not affected by race, gender, age, or other demographic characteristics: one in which people have a voice, one where different people in similar situations are treated consistently, a process whereby decisions are based on facts, one in which people are treated with dignity and respect, and a process protected by accountability to higher authorities (Leventhal 1976). The process-based model or procedural justice model purports that when members of the public perceive that police behave fairly and respectfully toward citizens, this generates a perception that the police and the public share similar values and beliefs. Lipset (1959), writing about the time of Weber, suggests that social institutions acquire legitimacy based on whether the institution’s values align with its constituents.


Gau (2011) explains that being treated well by police may indicate to people that they are a valued member of society, affirming the individual’s status as a member of the group, which includes police. This belief in a common moral perspective with police or common group membership encourages acceptance of a moral obligation to obey police authority and promotes a cultural value supportive of lawfulness (Tyler and Huo 2002). Tyler (1997) links this process back to social identity theory, whereby individuals gain information about their identity and status by how authorities treat them. When individuals feel that they have been treated respectfully by an authority, this action signifies their accepted group status and they respond with deference. Of course, not all members of the public experience personal contact with the police from which to assess procedural justice. Tyler and others assert that residents can draw on their general perceptions about the character of police to form opinions about procedural justice and thus legitimacy. Additionally, the process-based model does not require an objective assessment of the fairness of police procedures. Rather, what matters are individuals’ subjective opinions about police fairness, which derive from a variety of perspectives and experiences. The types of measures that are typically used to operationalize procedural justice include items asking whether police treat people fairly or with dignity and respect, make decisions based upon facts, explain their actions, listen to people before making decisions, respect people’s rights, try to help people solve problems, or have the best interest of a group or area at heart. Some studies also include whether police fairly decide whom to stop and question or arrest or whether police apply too much force under certain circumstances.

297

C

demonstrate prolonged effectiveness. Lipset suggests that both shared values and effectiveness can provide the source of legitimacy, but prolonged effectiveness in the absence of shared values offers only an unstable form of legitimacy for a social institution. The instrumental or performance-based model of legitimacy suggests that residents accept and conform to police authority when they perceive that doing so will produce beneficial outcomes or avoid negative ones. To gain benefits, residents must see police as a useful resource. Residents who perceive that police are competent and capable of reducing crime, bringing offenders to justice, creating and maintaining order, reducing fear of victimization, and improving neighborhood conditions will grant them legitimacy and subsequently provide their cooperation and adherence to police authority because they believe that doing so will achieve anticipated goals of better safety and less crime. In other words, they see police as an authority with the capacity to secure compliance and thus grant them legitimacy. Thus, when members of the public experience positive interactions with police or perceive these favorable outcomes are being and can be obtained, the performancebased model expects higher police legitimacy. However, when residents doubt the capacity of police to produce order, control crime, and hold criminals accountable, an instrumental model predicts that residents will be less inclined to acknowledge or feel obligated to conform to police authority, because they have little to gain. The types of questions that are typically used to measure police performance address assessments about response time, the effectiveness of police at addressing social disorder, crime, or citizen fear, as well as general assessments about the safety of a neighborhood and other neighborhood conditions.

The Instrumental/Performance-Based Model

The Distributive Justice Model

Lipset (1959), who advocated the importance of the process-based model, also contends that social institutions gain legitimacy when they can

The distributive justice model suggests that residents are more apt to view police as legitimate when they perceive that police services are

C

C

298

equitably distributed and not motivated by characteristics about people or places (e.g., wealth, social class, race/ethnicity, political affiliation, geographic location) that are unrelated to the police mission. This is not to suggest that police attention would have to be evenly or equally allocated, rather that the allocation of resources is perceived as appropriate. The distributive justice model suggests that police will have greater legitimacy when residents perceive everyone is getting suitable attention and services from the police. Judgments about what is equitable are subjective and, like the other models, do not require personal interactions with police in order to form them. The types of questions that are typically used to measure distributive justice generally ask whether groups are treated equally based on social characteristics such as race or wealth, or whether a certain group is treated better by police or receives better quality outcomes from police.

Weight of the Evidence Much of the evidence about the antecedents of police legitimacy is based on citizen surveys across cities in the United States. Tom Tyler conducted a series of these studies in Oakland, California, Los Angeles, New York City, and Chicago (Sunshine and Tyler 2003; Tyler 1990; Tyler 2001; Tyler and Huo 2002; Tyler and Fagan 2008). Since this initial contribution, other scholars have also contributed, examining the antecedents of legitimacy in other US cities (Gau 2011) and in other developed and developing countries such as Australia (Hinds 2007; Hinds and Murphy 2007; Murphy et al. 2008), England (Hough et al. 2010; Jackson et al. 2012), Ghana (Tankebe 2008), Israel (Jonathan-Zamir and Weisburd 2011), Jamaica (Reisig and Lloyd 2009), and Trinidad and Tobago (Kochel 2012; Kochel et al. 2011). Past research has shown some support for each of the three antecedents of legitimacy, suggesting that the pathway to legitimacy is not one dimensional. For example, Papachristos and colleagues (2009) show that a measure combining elements


of police performance, procedural justice, and distributive justice was a strong positive predictor of legitimacy. Hinds and Murphy (2007) as well as Murphy et al. (2008) found a positive association among Australians between legitimacy and each of procedural justice, police performance, and distributive fairness. Tyler’s (2002) New York sample showed the same result. Generally, however, these predictors do not have equal influence. Past research strongly supports procedural justice as the most consistent and influential antecedent: the primary mechanism for promoting police legitimacy. Studies conducted by Sunshine and Tyler (2003); Fagan and Tyler (2004); Tyler et al. (2010), as well as Van der Toorn et al. (2011) in New York, Tyler (1990) in Chicago, Tyler and Huo (2002) in Los Angeles and Oakland, by Hinds (2007), Hinds and Murphy (2007), and Murphy et al. (2008) in Australia, by Hough and colleagues (2010) as well as Jackson and colleagues (2012) in England and Wales, Kochel (2012) and Kochel et al. (2011) in Trinidad and Tobago, and by Reisig and Lloyd (2009) in Jamaica have all found support that when individuals perceive the police as being more procedurally just, they also view them with higher levels of legitimacy and that procedural justice is a stronger predictor than distributive justice or perceived effectiveness. Kochel (2012) further demonstrates that these relationships carry through to the neighborhood level. In Trinidad and Tobago neighborhoods, when residents perceived higher levels of procedural injustice, a smaller portion of residents reported normative obligations to accept police decisions. Other studies demonstrating the important positive influence of procedural justice on legitimacy, but without accounting for either distributive justice or performance include Fagan and Tyler (2005), Tyler and Fagan (2008), and Gau (2011). Limited past research also indicates a positive relationship between distributive justice judgments and perceptions about police legitimacy; however, the evidence is not nearly as clear or consistent as that for procedural justice. Tyler and Fagan (2008), Tyler and Huo (2002), Murphy and colleagues (2008), and Hinds and Murphy (2007)


show a positive association between distributive justice and police legitimacy. However, Reisig and Lloyd (2009) and Tyler et al. (2007) found no significant effect. No studies of police legitimacy have reported a significant negative effect of distributive justice. Similarly, past research shows mixed findings for the instrumental model; the influence of performance effectiveness in fighting crime, improving neighborhood conditions, and reducing fear of victimization on police legitimacy is not clear. However, Sunshine and Tyler (2003) (see the pre-crisis sample) and Van der Toorn and colleagues (2011) suggest that while performance effectiveness is less important than procedural justice, it can play a substantive role in forming legitimacy judgments. In particular, Van der Toorn and colleagues (2011) show that perceptions about police performance are nearly as important to their sample as procedural fairness in predicting police legitimacy. (Although Hinds and Murphy (2007) and Murphy and colleagues (2008) also support this association and suggest that in their study in Australia, police performance is nearly as influential as procedural justice on perceptions of police legitimacy, their measure of legitimacy is not well aligned with how it is defined in this encyclopedia entry as it includes components of satisfaction and performance and confidence in police.) Fagan and Tyler (2004), Tyler and Fagan (2008), and Tyler and colleagues (2010) did not show a significant effect of police performance on legitimacy judgments in New York among residents overall, or Muslim-Americans specifically. Hough and colleagues (2010) and Jackson and colleagues (2012) did not find a significant influence of effectiveness on legitimacy in a survey of British residents. No studies of police legitimacy have reported a significant negative effect of performance effectiveness. Panel studies that have examined views about legitimacy over time suggest that individuals have fairly stable opinions about police legitimacy – their views about police legitimacy at the earlier timeframe are the best predictor of the present views about legitimacy. Often, more recent assessments about the fairness of police procedures also

299

C

matter (Murphy et al. 2008; Tyler 1990; Tyler and Fagan 2008). In a few of these studies, some of the other models are also supported. Tyler and Fagan (2008) and Murphy and colleagues (2008) show support for performance effectiveness, while Murphy and colleagues also support the role of recent distributive justice assessments. Tyler and Fagan (2008), who conducted a panel study in New York City, reported that legitimacy judgments at time two were influenced by their initial views of police legitimacy at time one, as well as both procedural justice components and distributive justice opinions at time two, but not by police performance. Hinds (2007) conducted a study among Australian youth and found that procedural justice considerations outweighed the other significant predictors including police performance, youth-police relationships, expectations of police, negative contact with police, and fear of property crime. Among the three models purported to influence police legitimacy, the process-based model is the clear champion. More research is needed to better understand the nature of the relationship between legitimacy and both distributive justice and police performance. In addition to these factors, favorable outcomes and prior positive interactions with police also have positive relationships with legal institution legitimacy (Tyler 1990; Tyler and Huo 2002; Sunshine and Tyler 2003).

Demographic Characteristics and Police Legitimacy Individuals with different cultural, demographic, and experiential backgrounds often differ in their views about police legitimacy. For example, recent studies by Hough and colleagues (2010) and by Jackson and colleagues (2012) found a significant relationship between London residents’ moral values (a cultural measure of the degree to which the individual viewed certain behaviors as morally wrong) and their sense of duty to obey police. Independent of perceptions of procedural justice or police effectiveness, individuals with a stronger sense of

C

C

300

personal morality had a greater sense of obligation to obey police. Other personal characteristics, discussed below, have shown similar independent effects. Cherney and Murphy (2011) caution, however, that background characteristics may not only be associated with different levels of legitimacy, but in forming legitimacy judgments differently. They assert that when individuals have diminished legitimacy about the law, procedural justice qualities of police may not promote police legitimacy and in fact may have a backfire effect (They base this on their study, reported in Murphy et al. (2009), that found that among ethnic minorities who have reduced legitimacy for the law, procedural justice had a negative relationship on willingness to cooperate with police – an anticipated outcome of legitimacy and often used proxy for legitimacy). Conversely, Tyler and Huo (2002) have suggested that people form opinions about legitimacy in the same way across different background characteristics. They theorize that even if people vary in the level of legitimacy that they ascribe to police, they tend to follow the same psychological process in coming to that conclusion. Several studies have found relationships between views about police legitimacy and demographic characteristics. For example, a series of studies have shown a positive link between being Caucasian and higher levels of police legitimacy. Sunshine and Tyler (2003) found that white residents viewed police with more legitimacy than African-American and Hispanic residents in New York. Kirk and Papachristos (2011) found similar results in Chicago. Gau (2011) found that White residents feel a stronger obligation to obey police than Asian individuals. In spite of these differences, Tyler (2001), Tyler and Huo (2002), and Sunshine and Tyler (2003) found evidence that residents of different racial and ethnic backgrounds all rely primarily on procedural justice judgments to inform their legitimacy assessments. In these studies, different ethnic groups relied on similar psychological processes to form legitimacy judgments.


Additionally, older individuals tend to perceive the police with more legitimacy than younger people (Murphy et al. 2008; Papachristos et al. 2009; Tyler 1990; Tyler and Fagan 2008). Yet, Fagan and Tyler (2005) report that the adolescents they sampled relied on procedural justice to form views about legitimacy, as did adults in the many surveys conducted by Tyler across cities in the United States. Individuals of different ages rely on a similar process, even if they grant police different levels of legitimacy. Past research by Tyler (1988) found that residents in Chicago who had recent experiences with police but with different demographic backgrounds (age, gender, race, income, education, and political ideology) used the same criteria when they assessed the fairness of their experience with police. Also in the New York sample, residents who had citizen-initiated contact with police and residents who had police-initiated contact, both reported that the quality of the treatment they received had a significant influence on their perceptions of police legitimacy, when accounting for other demographic characteristics (Tyler and Fagan 2008). Furthermore, a sample of Muslim-Americans relied on procedural justice judgments when forming opinions about legitimacy, just as has been found in samples that do not distinguish religious affiliation (Tyler et al. 2010). Finally, a sample of law enforcement officers and a sample of soldiers each also relied on the procedural justice model to form their views about legitimacy (Tyler et al. 2007). Thus, based on available research, preliminary indications favor Tyler and Huo (2002), suggesting that while some demographic groups have significantly higher or lower opinions about police legitimacy, different groups form their opinions relying primarily on procedural justice judgments.

Context and Police Legitimacy Under certain contextual conditions, the strength of the factors predicting legitimacy or what is most important may vary. Prior research


addressing a variety of individuals’ attitudes toward police (e.g., satisfaction with police, support for police, confidence in police, views about fairness and effectiveness) suggests that neighborhood characteristics such as concentrated disadvantage, community disorder, collective efficacy, and the violent crime rate are important factors affecting individuals’ opinions (Bradford and Jackson 2010a; Cao et al. 1996; Dunham and Alpert 1988; and Kirk and Papachristos 2011; Kochel 2012; Reisig and Parks 2000; Sampson and Bartusch 1998; Sun et al. 2004). So, it may be reasonable to expect that perceptions about legitimacy will also be impacted by these contextual factors and the realities and feelings that they generate, and that different antecedents to legitimacy may matter more in one context than another (Hough et al. 2010). Dunham and Alpert (1988) found that generally, attitudes about police practices were more similar within neighborhoods than across neighborhoods. Kochel and Tankebe (forthcoming) find support that this holds true for police legitimacy. A variety of neighborhood qualities may explain this, including shared experiences with poverty, victimization risk, police contact, and sanctions, which may affect residents’ fears about safety, feelings of marginalization or acceptance, desires for police assistance, and other attitudes that may be important in forming legitimacy opinions. Supporting this notion, Kochel (2012) found that neighborhoods with a higher proportion of victims, fewer negative police contacts, as well as fewer extremely wealthy residents had more residents who viewed police with legitimacy. Sun, et al. (2004) reported that police legitimacy is lower in areas of concentrated disadvantage. Other scholars have looked at the related measure of legal cynicism within a neighborhood context. Sampson and Bartusch (1998) and Kirk and Papachristos (2011) found that people living in neighborhoods with concentrated disadvantage were significantly more likely to express legal cynicism. In Chicago, Kirk and Papachristos (2011) also identified high proportions of youth and residential stability as predictors of legal cynicism.

301

C

However, none of these studies examined whether residents of different kinds of neighborhoods or other varied contexts not only hold different views, but also rely on different processes in making legitimacy judgments. Two past studies offer the only indications about this issue. Both studies attempt to determine whether different legitimation processes are used when security is threatened. Sunshine and Tyler (2003) conducted a survey of New York City residents immediately prior to the terrorist crisis on September 11, 2001, and then shortly after the attacks. They found that under the different security threat circumstances, the relative importance of the three models differed somewhat. The first survey revealed that procedural fairness and to a lesser degree performance effectiveness were the antecedents to police legitimacy. In this survey, the effect of procedural justice was nearly five times stronger than performance. Subsequently, residents who viewed police with more legitimacy also were willing to empower police to undertake a variety of actions, including actions that limit the freedoms of citizens. Shortly after the terrorist attack, a second survey found that the influence of distributive justice on legitimacy was statistically significant and was about half as important as procedural fairness to predict perceptions of legitimacy. Performance effectiveness during this time of crisis was not a significant antecedent of legitimacy; instead, it was directly related to residents’ willingness to empower police to perform a variety of actions, some intrusive, including stopping and questioning people on the street and searching homes without permission from a judge. Empowerment was also still predicted by legitimacy, but in this context, performance drove empowerment to action, not perceptions of legitimacy (Similar results are reported in Ghana by Tankebe 2009). Jonathan-Zamir and Weisburd (2011) studied a community in Israel that was located in close proximity to the Gaza Strip and subject to severe threat for missile attacks compared to other communities in Israel, who lacked similar threats. In both communities, police legitimacy (actually a measure of trust in the police) was positively

C

C

302

predicted first by procedural justice and second by performance. The difference was that within non-threatened communities, but not in the community undergoing crisis, negative encounters with police and offenders’ risk of getting caught also impacted their legitimacy measure. Thus, early indications are that the process of forming legitimacy assessments may not drastically vary across social contexts. Procedural justice judgments are extremely important under a variety of situations and appear to be the primary influence on legitimacy. However, more research is needed on the process of forming legitimacy judgments within different contexts.

Conclusions The important outcomes of police legitimacy substantiate careful consideration of the factors that promote it. Three models dominate theory: the process-based model, the instrumental model, and the distributive justice model. Available research strongly supports the consistent and strong influence of fair procedures in forming positive views about police legitimacy. This model is supported across individuals with a variety of different background characteristics as well as across social contexts, including in developed and developing countries, across a variety of cities in the United States and in places of severe security threats and those lacking such threats. Mixed evidence on the role of police performance effectiveness and distributive justice on legitimacy assessments speak to the need for further research in those areas.

Recommended Reading and References Bradford B, Jackson J (2010a) Cooperating with the police: social control and the reproduction of police legitimacy. Available at SSRN: http://ssrn.com/ abstract¼1640958. Retrieved 20 July 2012 Cao L, Frank J, Cullen FT (1996) Race, community context and confidence in the police. Am J Police 15(1):3–22 Cherney A, Murphy K (2011) Understanding the contingency of procedural justice outcomes. Policing 5(3):228–235

Causes of Police Legitimacy Dunham RG, Alpert GP (1988) Neighborhood differences in attitudes toward policing: evidence for a mixedstrategy model of policing in a multi-ethnic setting. Criminology 79(2):504–523 Fagan J, Tyler T (2004) Policing, order maintenance, and legitimacy. In: Mesko G, Pagon M, Dobovsek B (eds) Policing in central and eastern Europe: dilemmas of contemporary criminal justice. Faculty of Criminal Justice, University of Maribor, Ljubljana, Slovenia, pp 38–48 Fagan J, Tyler T (2005) Legal socialization of children and adolescents. Soc Just Res 18(3):217–242 Gau JM (2011) The convergent and discriminant validity of procedural justice and police legitimacy: an empirical test of core theoretical propositions. J Crim Just 39:489–498 Hinds L (2007) Building police-youth relationships: the importance of procedural justice. Youth Justice 7(3):195–209 Hinds L, Murphy K (2007) Public satisfaction with police: using procedural justice to implement police legitimacy. Aust NZ J Crim 40:27–42 Hough M, Jackson J, Bradford B, Myhill A, Quinton P (2010) Procedural justice, trust and institutional legitimacy. Policing 4(3):203–210 Jackson J, Bradford B, Hough M, Myhill A, Quinton P, Tyler TR (2012) Why do people comply with the law? Legitimacy and the Influence of Legal Institutions. Brit J Criminol 9:2012. doi:10.1093/bjc/azs032, Published online: July Jonathan-Zamir T, Weisburd D (2011) The effects of security threats on antecedents of police legitimacy: findings from a quasi-experiment in Israel. J Res Crime Delinq. doi:10.1177/00224227811418002, Available online DOI Kirk DS, Papachristos AV (2011) The structural and cultural dynamics of neighborhood violence. U.S. Department of Justice, Washington, DC Kochel TR (2012) Can police legitimacy promote collective efficacy? Justice Q 29(3):384–419 Kochel TR, Tankebe J (forthcoming) Legitimacy as a neighborhood-level construct Kochel TR, Parks RB, Mastrofski SM (2011) Examining police effectiveness as a precursor to legitimacy and cooperation with police. Justice Q. doi:10.1080/ 07418825.2011.633544, Available online November 23, 2011 LaFree GD (1998) Losing legitimacy. Westview Press, Boulder Leventhal GS (1976) Fairness in social relationships. In: Thibaut J, Spense JT, Carson RC (eds) Contemporary topics in social psychology. General Learning Press, Morristown Lipset SM (1959) Some social requisites of democracy: economic development and political legitimacy. Am Pol Sci Rev 53(1):69–105 Murphy K, Hinds L, Fleming J (2008) Encouraging public cooperation and support for police. Policing and Society 18(2):136–155

Causes of Wrongful Convictions Murphy K, Tyler TR, Curtis A (2009) Nurturing regulatory compliance: is procedural justice effective when people question the legitimacy of the law? Regul Gov 3:1–26 Papachristos AV, Meares TL, Fagan J (2009) Why docriminals obey the law? The influence of legitimacy and social networks on active gun offenders. Yale Law & Economics Research Paper No. 373; Columbia Public Law Research Paper No. 09–199. Available at SSRN: http://ssrn.com/abstract¼1326631 Reisig M, Lloyd C (2009) Procedural justice, police legitimacy, and helping the police fight crime: results from a survey of Jamaican adolescents. Police Quart 12:42–62 Reisig M, Parks R (2000) Experience, quality of life, and neighborhood context: a hierarchical analysis of satisfaction with police. Justice Quarterly 7(3):607–630 Reisig MD, Bratton J, Gertz MG (2007) The construct validity and refinement of process-based policing measures. Crim Justice Behav 34:1005–1028 Sampson R, Bartusch D (1998) Legal cynicism and (subcultural?) tolerance of deviance: the neighborhood context of racial differences. Law Soc Rev 32(4):777–804 Sun IY, Triplett RA, Gainey RR (2004) Social disorganization, legitimacy of social institutions and neighborhood crime: an exploratory study of perceptions of the police and local government. J Crim Just 27(1):33–60 Sunshine J, Tyler T (2003) The role of procedural justice and legitimacy in shaping public support for policing. Law Soc Rev 37(3):513–547 Tankebe J (2008) Police effectiveness and police trustworthiness in Ghana: an empirical appraisal. Criminol Crim Just 8:185–202 Tankebe J (2009) Public cooperation with the police in Ghana: does procedural fairness matter. Criminology 47(4):1265–1293 Tyler TR (1988) What is procedural justice? Criteria used by citizens to assess the fairness of legal procedures. Law Soc Rev 22:301–355 Tyler TR (1990) Why people obey the law: procedural justice, legitimacy, and compliance. Yale University Press, New Haven Tyler TR (1997) The psychology of legitimacy: a relational perspective on voluntary deference to authorities. Pers Soc Psychol Rev 1(4):323–345 Tyler TR (2001) Public trust and confidence in legal authorities: what do majority and minority group members want from law and legal institutions? Behav Sci Law 19:215–235 Tyler TR (2004) Enhancing police legitimacy. Ann Am Acad Pol Soc Sci 593:84–99 Tyler TR (2006) Psychological perspectives on legitimacy and legitimation. Annu Rev Psychol 57:3 Tyler TR, Fagan J (2008) Legitimacy and cooperation: why do people help the police fight crime in their communities? Ohio St J Crim L 6:231–276 Tyler TR, Huo Y (2002) Trust in the law: encouraging public cooperation with the police and courts. RussellSage, New York

303

C

Tyler TR, Boeckmann RJ, Smith HJ, Huo YJ (1997) Social justice in a diverse society. Westview Press, Boulder Tyler TR, Callahan PE, Frost J (2007) Armed and dangerous (?): motivating rule adherence among agents of social control. Law Soc Rev 41(2):457–492 Tyler TR, Schulhofer S, Huq AZ (2010) Legitimacy and deterrence effects in counterterrorism policing: a study of Muslim Americans. Law Soc Rev 44(2):365–402 Van der Toorn J, Tyler TR, Jost JT (2011) More than fair: outcome dependence, system justification, and the perceived legitimacy of authority figures. J Exp Psychol 47:127–138 Weber M (1947) Max Weber: the theory of social and economic organization (trans: Henderson AM, Parsons T). New York: The Free Press

Causes of Wrongful Convictions Shawn Armbrust and Susan Friedman Mid-Atlantic Innocence Project, American University, Washington, DC, USA

Overview In any system of justice, there is a risk of convicting innocent people. The United States is no exception, and innocent people have been convicted throughout American history, a phenomenon that was first documented in 1923 (Borchard 1923). In the past two decades, however, more attention has been paid to this problem because, for the first time, post-conviction DNA testing has allowed us to determine with scientific certainty that innocent people have been convicted and often can point to the real perpetrator. The first such exoneration occurred in 1989, and as of 2012, the Innocence Project reports the exoneration of 297 individuals by post-conviction DNA testing. The National Registry of Exonerations identified hundreds more who have been exonerated using nonbiological evidence of innocence. As more of these exonerations occurred, they sparked interest in what can go wrong during an investigation and trial that can lead to such troubling systemic failures. Thus, advocates, academics, and policymakers have invested significant

C

C

304

time and resources into identifying isolating the causes of wrongful convictions. In 2008, in the most thorough study to date, University of Virginia law professor Brandon Garrett exhaustively studied the first 200 DNA exonerations, examining how they were handled from investigation through the trial, appellate, and post-conviction process. Garrett and others have demonstrated that there are six major causes of wrongful convictions: eyewitness testimony, unreliable forensic science evidence, false confessions, incentivized witnesses, government misconduct, and inadequate defense counsel. This entry will summarize how each of these problems can cause wrongful convictions and what can be done to ensure that these problems are addressed so that they do not lead to wrongful convictions in the future.

Eyewitness Identification To date, eyewitness error is the number one cause of documented wrongful convictions in DNA cases and also is quite prevalent in non-DNA exoneration cases (Garrett 2008; Gross 2005). Professor Garrett found that in 158 of the first 200 DNA exonerations – 79 % – eyewitness error led to the wrongful conviction, and Professor Gross found that 219 of the 340 DNA and non-DNA wrongful convictions between 1989 and 2003 involved eyewitness error. Thus, it is clear that eyewitness fallibility is a significant problem. The simple reason for this is that human memory for faces is far less accurate than most people – particularly participants in the criminal justice system – have long believed. In addition to being inaccurate, both human memory and confidence about the accuracy of one’s memory are quite malleable (Wells et al. 1998). Moreover, even when problems with an eyewitness’s memory can be exposed in the courtroom by pointing out discrepancies in the witness’s description or problems with their ability to clearly see the perpetrator of a crime, research has shown that if an eyewitness is confident, that is the number one factor considered by jurors and

Causes of Wrongful Convictions

courts in determining the accuracy of an eyewitness (Wells et al. 1998). This is true even though confidence at trial does not have a correlation with accuracy (although confidence at the time of the initial identification has a slightly higher correlation with accuracy) (Wells et al. 1998). Unfortunately, unintended manipulation of human memory and confidence through the traditional identification process – in which an investigator on the case in question shows the witness several photos or individuals at the same time – exacerbates these problems (Wells et al. 1998). Social psychologists have been studying this issue for decades and have identified several key problems in traditional eyewitness identification procedures that exacerbate the fallibility and malleability of human memory. Those problems include: • Feedback (often unintentional) to witnesses by law enforcement officers who are implementing a lineup procedure, which can improve a witness’s confidence in the accuracy of the identification • The use of “relative judgment,” in which a witness looking at a group of photos or individuals at the same time chooses the individual who looks most like the perpetrator, even when the actual perpetrator is not in the lineup • Repeated identification procedures (photo array, live lineup, preliminary hearing, etc.) that can increase a witness’s confidence in the accuracy of an identification (Wells et al. 1998) The same social scientists have recommended a series of reforms that have been shown to make eyewitness identifications more accurate and capture the confidence of a witness’s identification at the time it was made, instead of relying on inflated confidence estimates at the time of trial. Those reforms include: • Double-blind administration, a procedure in which neither the witness nor the officer administering the photo array or lineup knows who the suspect is • Confidence statements that would require the witness to immediately make a statement about his or her confidence after making an identification


• Sequential lineups in which photos or individuals are shown to a witness one at a time, requiring the witness to use absolute judgment instead of relative judgment, comparing her memory to one individual in a lineup or one photo at a time • Videotaping the identification procedures, which will allow defense counsel to identify any potential issues with the identification procedures (Wells et al. 1998) According to a recent review of reforms that have been implemented (Norris et al. 2010), ten states have taken steps to improve eyewitness identification procedures by either recommending or requiring some or all of these recommendations. Those states are Georgia, Maryland, New Jersey, North Carolina, Ohio, Rhode Island, Vermont, Virginia, West Virginia, and Wisconsin.

Unreliable Forensic Science Evidence With its purported ability to scientifically link a defendant to a crime scene, forensic science plays a critically important role in criminal trials. For decades, forensic science and forensic scientists were seen as infallible. With the development of DNA testing, however, flaws in many types of forensic science have been exposed. In fact, forensic science evidence is the second most common type of evidence used to secure a wrongful conviction. One hundred thirteen of the first 200 DNA exonerations involved cases in which some kind of forensic science evidence either linked the defendant to the crime scene or placed the defendant within the population of individuals who might have committed the crime (Garrett 2008). In those cases where analysts said that forensic science linked the defendant to the crime, DNA testing proved that this testimony and those sciences were wrong. The most common forensic evidence involved serology (conventional blood typing), followed by hair evidence, soil comparison, DNA tests, bite-mark comparisons, fingerprint evidence, dog scent identification, spectrographic voice evidence, shoe prints, and fiber comparison.

305

C

Most forensic science used in courtrooms – both before and after the advent of DNA testing – involves the attempt to determine whether a suspect can be tied to a crime scene. This can happen in several ways, including the following: (1) because a suspect leaves part of himself, such as hair, semen, saliva, sweat, or blood, at a crime scene; (2) because a suspect leaves something else, such as fibers, soil, or footprints, at a crime scene; or (3) because the weapon used to injure the victim has markings that can be compared to an item (usually a gun or knife) owned by the suspect. Oftentimes, forensic scientists testified that a suspect or items in possession of the suspect “matched” evidence found at a crime scene and that this match was conclusive proof that the defendant had, in fact, been at the crime scene. This testimony often went unchallenged by defense lawyers in the era before DNA testing. However, the explosion of DNA exonerations in cases where at trial other forensic sciences conclusively linked a defendant to a piece of evidence at a crime scene proved that in many cases, what people believed to be infallible science was actually quite fallible and possibly not even science. Thus, in 2009, a committee of the National Academy of Sciences (“NAS”) issued a report commissioned by Congress that was designed to evaluate the reliability of forensic science techniques. The report identified several flaws with the way that forensic science has been developed, regulated, and practiced. It expressed particular concern with microscopic hair comparison, fingerprint comparison, gunshot residue analysis, firearm and toolmark analysis, shoe prints, and soil and fiber comparisons. In each of those sciences, the report found that the conclusions reached by experts (that samples from crime scenes could be matched to defendants or items in the possession of defendants) were not grounded in any scientific research and were not validated by any empirical data. In addition, the lack of accreditation and certification among crime laboratories means that there frequently are no standards to determine what constitutes a match – examiners are allowed to determine for themselves, based on their experience, whether items match. Finally,

C

C

306

even where there are standards, the lack of proficiency testing means that unqualified examiners are allowed to continue working and testifying without sufficient oversight. In recent years, states have made some effort to reform crime laboratories and the use of forensic evidence in criminal trials. By statute, 13 states enacted some type of forensic oversight, and in two states, the state’s attorney general established oversight entities. In addition to ensuring that crime laboratories are accredited and follow best practices, some oversight commissions are also charged with investigating forensic misconduct. In 2004, Congress passed the Justice for All Act, which included a provision that links the receipt of grants by state crime laboratories under the Paul Coverdell Forensic Science Improvement Grants Program to the establishment of guidelines for handling allegations of negligence or misconduct. In the wake of the NAS report, various advocates also are working to create an entity within the federal government that would foster real scientific research into various areas of forensic science, as well as set standards and licensing requirements that would help create some uniformity within the various forensic science disciplines.

False Confessions Prior to the advent of DNA testing, it was difficult for most actors in the criminal justice system – including defense lawyers – to believe that an innocent person would confess to a crime that he or she did not commit. Although it was easy to imagine that such confessions would occur in an environment with physical coercion, most American police departments long ago abandoned those tactics and focus instead on persuasive psychological tactics. However, in Professor Garrett’s study of the first 200 DNA exonerations, he found that 31 cases involved false confessions. Professors Steve Drizin and Richard Leo documented in a 2004 study published in the University of North Carolina Law Review 125 recent provable false confessions in the United States.


False confessions are particularly problematic because the phenomenon is so hard for most people to understand. Once an individual confesses to a crime, a conviction is almost assured. The police stop investigating, prosecutors ask fewer questions, defense lawyers are more likely to encourage a guilty plea, and juries are unlikely to believe protestations of innocence. This is true regardless of how reliable the confession appears. In many exoneration cases, the confessors were wrong about basic facts like the race of the victim, the location of the crime, and the weapon used, but they were convicted nonetheless. Quite simply, very few people believe that anyone could confess to something they did not do. Because these false confessions are such a perplexing problem, they have been the subject of significant research by both psychologists and legal scholars who have sought to examine what factors – both individual and systemic – might lead someone to falsely confess to a crime. That research has determined that individuals who are “unusually suggestible and compliant” – such as juveniles and the mentally retarded – are more likely to provide authorities with a false confession (Drizin and Leo 2004). Of the 31 false confessions documented by Professor Garrett, 11 involved mentally retarded defendants, and 12 involved juvenile defendants. Juveniles are likely to falsely confess because they are more likely to make impulsive decisions, engage in risky behavior, and be more susceptible to suggestions. Individuals with mental retardation may be more susceptible to suggestions because they want to please authorities, want others to believe they are competent, often cannot control their impulses, and are willing to accept blame. The idea that inherently suggestible individuals are more likely to falsely confess to crimes is not terribly surprising; it is far more surprising that a significant number of the false confessions documented by Professor Garrett and by Professors Drizin and Leo involved individuals who are not inherently suggestible. Thus, despite that lack of inherent suggestibility, these individuals made the choice to confess to a crime that they did not commit. Legal scholars and social scientists examining the reasons for this have determined


that modern interrogations involve multiple tactics that – particularly when combined – can help cause false confessions. First, the purpose of interrogations is to elicit confessions and to ignore protestations of innocence (Drizin and Leo 2004) (“police are trained to interrogate only those suspects whose guilt they presume or believe they have already established”). Second, recognizing the inherent difficulty of getting anyone – guilty or innocent – to confess to a crime, interrogations are designed to convince the suspect that confessing is the logical choice. To do this, investigators rely upon a technique known as “minimization and maximization” (Drizin and Leo 2004), in which investigators maximize the extent of the evidence against the defendant to intimidate them but then offer them the opportunity to minimize their involvement (e.g., by claiming self-defense in a murder or consent in a rape) and thus avoid harsher criminal penalties. Third, the police are allowed to lie to suspects about the evidence against them (Drizin and Leo 2004). Fourth, during the course of an interrogation, police may – often inadvertently – provide the suspect with information about the crime that can be used to validate a false confession later. And fifth, interrogations often are quite long, which has been shown to increase an individual’s likelihood of falsely confessing (Kassin 2010). One study found that in cases where a false confession was made, the interrogation lasted an average of 16.3 h (Kassin 2010). The same study also found that 34 % of interrogations lasted 6–12 h, and 39 % lasted 12–24 h (Kassin 2010). Solving the problem of false confessions is achievable through reforms that often are beneficial to both law enforcement and to innocent defendants. The most common reform recommended to combat false confessions is the videotaping of all interrogations, from start to finish. Although this cannot prevent false confessions, it can increase the likelihood that false confessions are caught before they lead to conviction. More than 500 jurisdictions now do just that. In addition, some have recommended a reformation of interrogation practices, advocating for investigative interviews of

307

C

suspects instead of interrogations that are designed only to elicit confessions (Drizin and Leo 2004). Thus, while false confessions are one of the most perplexing causes of wrongful convictions, they also are one of the easiest problems to solve.

C Government Informants: Snitches Investigating crime is difficult for many reasons, and one of those reasons is that relevant witnesses often do not want to cooperate with the police, often because of fear or bias against law enforcement or because it is against their interest. Thus, the government is permitted to incentivize testimony by offering reduced sentences, reduced charges, favorable treatment in prison, or – in some cases – reward money. Such deals can be offered to codefendants, reluctant witnesses, or – in the most dangerous form – jailhouse informants who allegedly heard the defendant confess to the crime in a jail cell. While such deals are a necessary evil, the potential for abuse is obvious. Wrongful conviction cases prove this point again and again. In 35 of the first 200 DNA exonerations, Professor Garrett found that some type of incentivized testimony was used from an informant, a jailhouse informant, or an alleged co-perpetrator. In 2005, the Center on Wrongful Convictions released a study on the 111 death row exonerations since the death penalty was reinstated in this country in the 1970s. The study found that of those exonerations, 51 individuals were wrongfully convicted based on informant testimony, making incentivized testifying the leading cause of wrongful convictions in capital cases. The reason for this is that the incentives, combined with the ease of making up seemingly credible information, create a world in which informants are able and willing to lie in ways that are difficult to detect. In the most problematic cases, informants make statements that the defendant has confessed to the informant – either out on the street or in a jail cell. Those statements often contain details that seemingly would be

C

308

difficult to obtain from anyone other than the true perpetrator of a crime. However, when an informant has access to the Internet, media, friends, and – in some cases – the court file that the defendant has left in his jail cell, he or she can obtain information about the case that is used to create a plausible story that is difficult to either verify or disprove. Indeed, in a 60-min segment, Leslie Vernon White, who admitted to lying on many occasions as a jailhouse informant, demonstrated the ease with which he could obtain information about a crime and make up a confession. These problems are exacerbated when either the deals provided to an informant or an informant’s full history of cooperating with the government is not fully disclosed to the defense. If a jury knows that an informant has received some benefit for his testimony or that he has testified as an informant on multiple occasions, the jury at the very least has the benefit of more balanced information about the informant. In many cases, however, that information is not disclosed, which takes testimony that is inherently suspect and inappropriately bolsters it in the minds of the jury. Despite its inherent flaws, banning informant testimony is not a tenable solution. Without the ability to incentivize witnesses, the government would have a difficult time convincing witnesses to talk. Thus, any solution to the problem of informants involves both appropriately verifying facts in the informant’s testimony and properly disclosing an informant’s history and benefits he has received in exchange for his testimony. Recommendations on how to handle incentivized testimony include having informants wear wires, recording informant conversations with authorities and providing the recording to defense counsel, disclosing the benefit that the informant will receive in exchange for testifying, and keeping a database of informants so that law enforcement and the defense know who may be manipulating the system to repeatedly receive benefits from the government. In Illinois, the state legislature requires significant disclosures, pretrial reliability hearings, corroboration, and cautionary jury instructions in any capital case involving a jailhouse informant.


Government Misconduct In many cases, wrongful convictions are caused by an unfortunate series of honest mistakes – eyewitnesses who have erred, scientific testimony that was unintentionally overstated, and confessions that, while false, were obtained using what police believed to be appropriate tactics. In other cases, however, police and prosecutors who are under pressure to get convictions sometimes lose sight of their obligation to do justice and – whether intentionally or not – commit misconduct by, among other things, failing to turn over exculpatory evidence, improperly characterizing evidence at trial, engaging in improper interrogation tactics, or inappropriately attempting to influence an eyewitness. Such misconduct is far more widespread than anyone might have believed before the advent of DNA testing. Although it is impossible to quantify exactly how many exonerations involved such misconduct because it is so difficult to uncover, it clearly has been a significant factor in many wrongful conviction cases (Garrett 2008). In 2003, the Center for Public Integrity found that since 1970 prosecutorial misconduct was a factor in dismissed charges, reversed convictions, or reduced sentences in at least 2,012 cases of the 11,452 cases in which claims of prosecutorial misconduct were reviewed by courts (The Justice Project 2009). Unfortunately, studies also have shown that there are few incentives for prosecutors, in particular, to avoid committing misconduct (Ridolfi and Possley 2010). Thus, this is an area that is ripe for reform. The reforms suggested in the sections on eyewitness error and false confessions also would address government misconduct in these areas, but they would not address the problem of failing to turn over exculpatory evidence. In civil cases, both sides are required to liberally exchange information in lengthy and complicated discovery procedures. In criminal cases, however, discovery is much more limited. The Supreme Court found in Brady v. Maryland that the Constitution requires the government to disclose exculpatory and impeachment information to the defense, but the


Constitution requires nothing else. Some states, such as North Carolina, require the government to provide open-file discovery, but that is not the norm. In the federal system and the District of Columbia, for example, criminal defendants are entitled to only what is constitutionally required, their own statements, and expert notice. Unless a defendant has a good investigator, he may not even know who the witnesses against him are until those witnesses testify at trial. Because so many states have such limited discovery, the onus usually is on the prosecutor to determine what information needs to be disclosed. Prosecutors are trusted to do this because they have two roles in the criminal justice system – to zealously prosecute crimes and to do justice. Unfortunately, the DNA exonerations and other exonerations have shown that prosecutors often fail to disclose information that could have been helpful to the defense. Sometimes, this is deliberate. In other cases, however, prosecutors who are working to zealously prosecute a defendant simply do not see certain information as exculpatory and therefore do not turn it over to the defense. Various entities looking at this problem have proposed various solutions. Those solutions include open-file discovery and more meaningful sanctions for prosecutors who do break the rules. In addition, the Innocence Project has proposed criminal justice reform commissions that can bring all parties – including police and prosecutors – to the table to come to a consensus on how to handle the actors in the criminal justice system who do not follow the rules.

Inadequate Defense Counsel The Sixth Amendment of the United States Constitution guarantees criminal defendants the right to counsel, and the Supreme Court in Gideon v. Wainwright found that this requires the state to provide indigent defendants with a lawyer who is financed by the state. While Gideon was a tremendous milestone for indigent defendants, it did not specify how states were required to make attorneys available to indigent defendants. Thus, states responded to Gideon in a variety of ways,

309

C

including statewide public defender systems, contract systems in which independent agencies bid for court appointments and are awarded contracts based on a variety of factors, contract systems in which contracts go to the lowest bidder, and ad hoc court appointments in which judges appoint lawyers on a case-by-case basis. Predictably, this has resulted in a huge disparity between jurisdictions, with some committing to the provision of high-quality indigent defense lawyers despite the unpopularity of doing so and others providing defendants with inexperienced, poorly paid attorneys who are overburdened with extraordinary caseloads and denied access to investigators and necessary experts. There is little incentive for states to provide anything other than the bare minimum. Although the Supreme Court found in Strickland v. Washington that defendants have a right to effective counsel, a defendant challenging his conviction based upon ineffective assistance of counsel must prove that his attorney was defective and that if his attorney had acted properly, the result of the proceeding would have been different. In any event, challenging a conviction based on ineffective assistance of counsel requires competent counsel, something many defendants simply do not have, particularly after they have been convicted and through the appellate process, when the right to counsel ends. Unsurprisingly, the lack of quality defense lawyers leads to wrongful convictions. When an innocent defendant lacks an advocate who can investigate and meaningfully challenge the government’s case, it is to be expected that a wrongful conviction might occur. According to the Innocence Project, many wrongful convictions in DNA cases involved lawyers who were incompetent, ineffective, overworked, or all three. Among the errors noted in these cases are lawyers who slept in the courtroom during trial, failed to investigate alibis, failed to consult experts or adequately challenge forensic evidence, and failed to show up for hearings. This problem is difficult to solve largely because it requires the investment of scarce government resources in something that is considered to be unpopular, the representation of

C

C

310

accused criminals. However, a commitment of sufficient resources to indigent defense systems would solve most, if not all of the problems that cause wrongful convictions. While it would not directly solve the problem of unethical or lazy attorneys, the proper investment of resources would help decrease caseloads, allow for sufficient investigative funds, provide funding for experts, provide training that is sorely lacking, and help reduce the disparity of resources between indigent defense systems and prosecutors. While such funding might not be popular, most prosecutors and other actors in the criminal justice system recognize that a well-funded indigent defense system helps make the system function properly and can help prevent the conviction of innocent people.

Conclusion Every wrongful conviction is a tragic example of how the criminal justice system can fail at its most critical function: the reliable conviction of the guilty and exoneration of the innocent. While these cases are tragic, they also provide us with an unprecedented opportunity to learn where our criminal justice system is fallible and how it can be improved. It is known that wrongful convictions have exposed problems with eyewitness testimony, forensic science evidence, false confessions, government informants, government misconduct, and ineffective assistance of counsel. It is important to continue to study wrongful convictions so that best practices can be determined for how to handle these issues. With continued study and reform, steps can be taken to correct problems with these factors in order to prevent the conviction of the innocent in the future.

Recommended Reading and References Borchard E (1923) Convicting the innocent. Yale University Press, New Haven Committee on Identifying the Needs of the Forensic Science Community of the National Research Council (2009) Strengthening forensic science in the United States: a path forward. The National Academies Press, Washington, DC

CCTV Drizin SA, Leo RA (2004) The problem of false confessions in the post-DNA world. North Carol Law Rev 82:891 Garret B (2011) Convicting the innocent. Harvard Press, Cambridge Garrett B (2008) Judging innocence. Columbia Law Rev 108:55 Gideon v. Wainwright (1963) 372 U.S. 335 Gross SR et al (2005) Exonerations in the United States 1989 through 2003. J Crim Law Criminol 95:523 InnocenceProject.org Kassin SM et al (2010) Police-induced confessions: risk factors and recommendations: looking ahead. Law Hum Behav 34:49 Norris R, Bonventre C, Redlich AD, Acker JR (2010/2011) Than that one innocent suffer: evaluating safeguards against wrongful convictions. Albany Law Rev 74:1301 Ridolfi KM, Possley M (2010) Preventable error: a report on prosecutorial misconduct in California 1997–2009 Strickland v. Washington (1984) 466 U.S. 668 The Justice Project (2009) Improving prosecutorial accountability: a policy review. Justice Project, Washington, DC United States Constitution Amendment VI Wells G et al (1998) Eyewitness identification procedures: recommendations for lineups and photospreads. Law Hum Behav 22(6):603–647

CCTV ▶ Identification Technologies in Policing and Proof

CCTV and Crime Prevention Brandon C. Welsh1,2 and David P. Farrington3 1 School of Criminology and Criminal Justice, Northeastern University, Boston, MA, USA 2 Netherlands Institute for the Study of Crime and Law Enforcement (NCSR), Amsterdam, The Netherlands 3 Institute of Criminology, University of Cambridge, Cambridge, UK

Synonyms Camera surveillance; Closed-circuit television; Video surveillance

CCTV and Crime Prevention

Overview Western countries are experiencing a substantial increase in the use of closed-circuit television (CCTV) surveillance cameras to prevent crime in public places. Amid this expansion and the associated public expenditure, as well as concerns about their effectiveness and social costs, there is an increasing need for an evidence-based approach to inform CCTV policy and practice. This entry reports on an updated Campbell Collaboration systematic review and meta-analysis of the effects of CCTV on crime in public places. The results suggest that CCTV has a modest but significant effect on reducing crime. This overall result was largely driven by the effectiveness of CCTV schemes in car parks, which were targeted at vehicle crimes and included other interventions such as improved lighting, fencing, and security guards. Nonsignificant effects on crime were observed in the other public settings in which CCTV schemes were evaluated: city and town centers, public housing communities, and public transportation facilities. CCTV has become a widely popular and commonplace technology that is being deployed to help make public (and private) places safer from crime. Across many cities and towns, it is touted as a panacea to crime. But it is not without its controversies, and many questions remain about its ability to reduce crime. This entry reports on the results of a systematic review – incorporating meta-analytic techniques – to assess the scientific evidence on the effectiveness of CCTV to prevent crime in public places. In recent years, there has been a marked and sustained growth in the use of CCTV surveillance cameras to prevent crime in public places in many Western nations. The United Kingdom in particular is on the cusp of becoming, in the words of some, a “surveillance society,” with upward of a million or more cameras in public places (Norris 2007). There are no national estimates as yet on the number of CCTV cameras in the United States, but local accounts indicate that they are being installed at an unprecedented rate and their popularity is not limited to large urban centers. Some

311

C

of this increased use in the United States has come about in an effort to aid the police in the detection and prevention of terrorist activities, especially in New York City and other metropolises. However, the prevention of crime remains an important aim of these CCTV systems. Similar claims about the purpose of public CCTV have been made in the United Kingdom. There are also signs that other countries are increasingly experimenting with CCTV to prevent crime in public places. Evaluation studies of public CCTV schemes have been carried out recently in a number of European countries, including Germany, Norway, and Sweden, as well as in Australia, Canada, and Japan. Many of these countries have not previously used CCTV in public places, let alone evaluated its effects on crime. The growth in CCTV has come with a huge price tag. In the United Kingdom, CCTV has been and continues to be the single most heavily funded crime prevention measure operating outside of the criminal justice system. In the United States, estimates suggest that public expenditure on CCTV may be as much as $100 million each year (Savage 2007). There has been much debate about the effectiveness of CCTV in preventing crime and hence on the wisdom of spending such large sums of money. A key issue is to what extent funding for CCTV, especially in the United Kingdom and the United States, has been based on high quality scientific evidence demonstrating its effectiveness in preventing crime (Welsh and Farrington 2009a, b).

Methods The methodology employed in our systematic review follows the conventions set out by the Campbell Collaboration. Evaluations were included in the systematic review if they met a number of criteria, including if CCTV was the main intervention, if there was an outcome measure of crime, and if the evaluation design was of high methodological quality. At a minimum, high quality evaluation designs involve before and after

C

C

312


measures of crime in experimental and comparable control areas. Control areas are needed to assess what would have happened in the absence of CCTV and to counter threats to internal validity. Extensive search strategies were employed to locate studies meeting the criteria for inclusion, including searches of electronic bibliographic databases, searches of literature reviews on the effectiveness of CCTV on crime, and contacts with leading researchers. Altogether, 44 studies were found that met our inclusion criteria. In order to carry out a meta-analysis, a comparable measure of effect size and an estimate of its variance are needed in each program evaluation (Lipsey and Wilson 2001). In the case of CCTV evaluations, the measure of effect size had to be based on the number of crimes in the experimental and control areas before and after the intervention. This is because this is the only information that was regularly provided in these evaluations. The “relative effect size” or RES is used to measure effect size. The RES is intuitively meaningful because it indicates the relative change in crimes in the control area compared with the experimental area. For example, RES ¼ 2 indicates that the change in the control areas is twice as large as the experimental areas. This value could be obtained, for example, if crimes doubled in the control area and stayed constant in the experimental area, or if crimes decreased by half in the experimental area and stayed constant in the control area, or in numerous other ways. The RES is computed from the pre- and post-CCTV crime counts in the control and experimental areas. The equation is as follows: RES ¼ ða dÞ=ðb cÞ where a and b are the pre- and post-CCTV crime counts for the experimental areas, respectively, and c and d are the pre- and post-CCTV crime counts for the control areas, respectively. The meta-analysis was performed on the logged values of RES with the final results converted back into the original scale by taking the antilog. In order to perform the meta-analysis, the variance of each logged RES is needed. If we

assume that the crime counts are Poisson distributed, then the variance is 1/a + 1/b + 1/c + 1/d. This assumption is plausible because 30 years of mathematical models of criminal careers have been dominated by the assumption that crimes can be accurately modeled by a Poisson process (Piquero et al. 2007). In a Poisson process, the variance of the number of crimes is the same as the number of crimes (i.e., v ¼ n). However, the large number of changing extraneous factors that influence the number of crimes may cause overdispersion; that is, where the variance of the number of crimes exceeds the number of crimes. Farrington et al. (2007) estimated the overdispersion (D) in crime counts as follows: D ¼ 0:0008 N þ 1:2 D increased linearly with N and was correlated 0.77 with N. The mean number of crimes in an area in the CCTV studies was about 760, suggesting that the mean value of D was about 2. However, this is an overestimate because the monthly variance is inflated by seasonal variations, which do not apply to N and VAR. Nevertheless, in order to obtain a conservative estimate, V(LRES), which is the variance of LRES (and is distinguished from Va, which is the variance of a), calculated from the usual formula above was multiplied by D (estimated from the above equation) in all cases. Specifically, VðLRESÞ ¼ Va a2 þ Vb b2 þ Vc c2 þ Vd d2 where Va/a ¼ 0.0008 a + 1.2 This is our best available estimate of the degree of overdispersion in area-based crime prevention studies. Forty-one of the 44 studies could be used in the meta-analysis. RES effect sizes could not be calculated for three studies because numbers of crimes were not reported in the city and town center schemes in Ilford or (for the control area) Sutton or for the public housing scheme in Brooklyn. (Throughout this entry, the included studies are identified by the name of the city or town where they were implemented. Full references


can be obtained from the authors’ Campbell Collaboration systematic review, which is available at: http://www.campbellcollaboration. org/crime_and_justice/index.php.)

Findings Setting City and town centers. Twenty-two evaluations were carried out in city and town centers. Seventeen of these were carried out in the United Kingdom, three in the United States, one in Sweden, and one in Norway. Only some of the studies reported the coverage of the CCTV cameras. For example, in the Newcastle-uponTyne and Malmo¨ studies, the camera coverage of the target or experimental area was 100 %. Many more studies reported the number of cameras used and their features (e.g., pan, tilt, zoom). Information on camera coverage is important because if a large enough section of the target area or even high crime locations in the target area are not under surveillance, the impact of CCTV may be reduced. Most of the evaluations that reported information on the monitoring of the cameras used active monitoring, which means that an operator watched monitors linked to the cameras in real time. Passive monitoring involves watching tape recordings of camera footage at a later time. In some of the schemes, such as Newcastle and Birmingham, police carried out active monitoring. But more often it was carried out by security personnel who had some form of communication link with police (e.g., by a one-way radio or direct line telephone). On average, the follow-up period in the 22 evaluations was 15 months, ranging from a low of 3 months to a high of 60 months. Six programs included other interventions in addition to the main intervention of CCTV. For example, in the Doncaster program, 47 “help-points” were established within the target area to aid the public in contacting the main CCTV control room. Four other studies used notices of CCTV to inform the public that they were under surveillance, but CCTV notices do not necessarily constitute

313

C

a secondary intervention. A couple of the evaluations used multiple experimental areas (e.g., police beats), meaning that the CCTV intervention was quite extensive in the city or town center. Multiple control areas (e.g., adjacent police beats, the remainder of the city) were used in many more of the evaluations. The most comparable control areas were chosen for analysis. Where control and adjacent areas were used, control areas were analyzed. The city and town center CCTV evaluations showed mixed results in their effectiveness in reducing crime. Ten of the 22 evaluations were considered to have a desirable effect on crime, 5 were considered to have an undesirable effect, and 1, the multisite British evaluation by Sivarajasingam et al. (2003), was considered to have both (desirable effects according to emergency department admissions and undesirable effects according to police records). The remaining six evaluations were considered to have a null (n ¼ 5) or uncertain (n ¼ 1) effect on crime. Schemes usually showed evidence of no crime displacement rather than displacement or diffusion of benefits. As an example, in the program evaluated by Armitage et al. (1999), an unknown number of cameras were installed in the town center of Burnley, England. The experimental area consisted of police beats in the town center with CCTV coverage. Two control areas were used. The first comprised those police beats that shared a common boundary with the beats covered by CCTV. The second control area consisted of other police beats in the police division. The first control area was more comparable to the experimental area and was the one used in our analysis. After 12 months, the experimental area, compared with the two control areas, showed substantial reductions in violent crime, burglary, vehicle crime, and total crime. For example, total incidents of crime fell by 28 % in the experimental area compared with a slight decline of 1 % in the first control area and an increase of 10 % in the second control area. The authors found evidence of diffusion of benefits for the categories of total crime, violent crime, and vehicle crime, and evidence of territorial displacement for burglary.

C

C

314

In pooling the data from the 20 studies for which effect sizes could be calculated, there was evidence that CCTV led to a small but nonsignificant reduction in crime in city and town centers. The weighted mean effect size was a RES of 1.08 (CI 0.97–1.20), which corresponds to a 7 % reduction in crimes in experimental areas compared with control areas. However, when these 20 studies were disaggregated by country, the 15 British studies showed a slightly larger effect on crime (a 10 % decrease; RES ¼ 1.11, CI 0.98–1.27, ns), while the five others showed no effect on crime. An analysis of heterogeneity showed that the 20 effect sizes were significantly heterogeneous (Q ¼ 143.9, df ¼ 19, p < 0.0001). This means that the differences were not simply a matter of sampling error. The 15 UK studies were also significantly heterogeneous (Q ¼ 118.6, df ¼ 14, p < 0.0001), as were the five other studies (Q ¼ 14.02, df ¼ 4, p ¼ 0.007). Therefore, random effects models that assume study-level variability were used in calculating weighted mean effect sizes. Public housing. Nine evaluations were carried out in public housing. Seven were carried out in the United Kingdom and two in the United States. Camera coverage ranged from a low of 9 % (in Dual Estate) to a high of 87 % (in Northern Estate) in the six evaluations that reported this information. Active monitoring was used in all of the schemes, with monitoring in the Brooklyn evaluation conducted by police. In the six British schemes evaluated by Gill and Spriggs (2005), security personnel who monitored the cameras had some form of communication link with police (e.g., a one-way or two-way radio). On average, the follow-up period in the nine evaluations was 12 months, ranging from a low of 3 months to a high of 18 months. Only three schemes included other interventions in addition to the main intervention of CCTV. These involved improved lighting and youth inclusion projects. The public housing CCTV evaluations showed mixed results in their effectiveness in reducing crime. Three of the nine evaluations were considered to have a desirable effect on


crime, two had an undesirable effect, three had an uncertain effect, and one had a null effect. Only five schemes measured diffusion of benefits or crime displacement, and in each case it was reported that displacement and diffusion did not occur. In pooling the data from the eight studies for which effect sizes could be calculated, there was evidence that CCTV led to a small but nonsignificant reduction in crime in public housing. The weighted mean effect size was a RES of 1.07 (CI 0.83–1.39), which corresponds to a 7 % reduction in crimes in experimental areas compared with control areas. The eight effect sizes were significantly heterogeneous (Q ¼ 47.94, df ¼ 7, p < 0.0001). The evaluation by Williamson and McLafferty (2000) in Brooklyn, New York, the only one that could not be included in the metaanalysis, is somewhat representative of CCTV’s rather negligible effect on crime in public housing. The housing community that received the intervention (Albany project) did not show any change in the total number of police-recorded crimes, either inside the project or inside a 0.1 mile buffer zone (established to measure crime displacement or diffusion of benefits), while total crime in the control community (Roosevelt project) dropped by 5 % inside the project and 4 % inside the 0.1 mile buffer zone. When total crime was disaggregated, a desirable program effect was observed for major felonies in both experimental and control projects. However, the authors note that, “the substantial decrease in major felonies around both public housing projects seems to be part of a larger downward trend that was occurring not only in Brooklyn but across New York City in the late 1990s” (Williamson and McLafferty 2000, p. 7). Furthermore, the authors’ investigation of the occurrence of crime displacement or diffusion of benefits concluded that there was “no clear evidence” of either, “as the change in crime around the two housing projects does not vary predictably with distance” (p. 7). Public transport. Four evaluations were carried out in public transportation systems. All of them were conducted in underground railway


systems: three in the London Underground and one in the Montreal Metro. None of the studies reported on the percentage of the target areas covered by the cameras, but most did provide information on the number of cameras used. For example, in the Montreal program a total of 130 cameras (approximately 10 per station) were installed in the experimental stations. Each of the schemes involved active monitoring on the part of police; in the London Underground, this meant the British Transport Police. With the exception of the Montreal program, each evaluation included other interventions in addition to CCTV. In the first Underground scheme, special police patrols were in operation prior to the installation of CCTV. For the two other Underground schemes, some of the other interventions included passenger alarms, kiosks to monitor CCTV, and mirrors. For each of these three Underground schemes, CCTV was, however, the main intervention. The follow-up periods ranged from a low of 12 months to a high of 32 months. Overall, CCTV programs in public transportation systems present conflicting evidence of effectiveness: two had a desirable effect, one had no effect, and one had an undesirable effect on crime. However, for the two effective programs in the London Underground (southern sector and northern line), the use of other interventions makes it difficult to say with certainty that it was CCTV that caused the observed crime reductions, although in the first of these programs CCTV was more than likely the cause. Only two of the studies measured diffusion of benefits or crime displacement, with one showing evidence of diffusion and the other showing evidence of displacement. In pooling the data from the four studies, there was evidence that CCTV led to a sizable but nonsignificant reduction in crime in public transport. The weighted mean effect size was a RES of 1.30 (CI 0.87–1.94), which corresponds to a 23 % reduction in crimes in experimental areas compared with control areas. The substantial reduction in robberies and thefts in the first Underground evaluation (an overall 61 % decrease) was the main reason for this large

315

C

average effect size over all four studies. The four effect sizes were significantly heterogeneous (Q ¼ 30.94, df ¼ 3, p < 0.0001). Car parks. Six CCTV evaluations met the criteria for inclusion and were conducted in car parks (parking lots). All of the programs were implemented in the United Kingdom between the early 1980s and early 2000s. Camera coverage was near 100 % in the two schemes that reported on it. All of the schemes, with the exception of one that did not provide data, involved active monitoring on the part of security staff. The large-scale, multisite Hawkeye scheme evaluated by Gill and Spriggs (2005) also included a radio link with the British Transport Police. Each of the programs supplemented CCTV with other interventions, such as improved lighting, painting, fencing, payment schemes, and security personnel. In Coventry, for example, improved lighting, painting, and fencing were part of the package of measures implemented to reduce vehicle crimes. In each program, however, CCTV was the main intervention. The follow-up periods ranged from a low of 10 months to a high of 24 months. Five of the car park programs had a desirable effect and one had an undesirable effect on crime, with vehicle crimes being the exclusive focus of five of these evaluations. For example, Tilley (1993) evaluated three CCTV programs in car parks in the following UK cities: Hartlepool, Bradford, and Coventry. Each scheme was part of the British Government’s Safer Cities Programme, a large-scale crime prevention initiative that operated from the late 1980s to mid-1990s. In Hartlepool, CCTV cameras were installed in a number of covered car parks and the control area included a number of non-CCTV-covered car parks. Security personnel, notices of CCTV, and payment schemes were also part of the package of measures employed to reduce vehicle crimes. Twenty-four months after the program began thefts of and from vehicles had been substantially reduced in the experimental car parks compared with the control car parks. A 59 % reduction in thefts of vehicles was observed in the experimental car parks compared with a 16 % reduction in the control car parks. Tilley (1993, p. 9) concluded

C

C

316

that, “The marked relative advantage of CCTV covered parks in relation to theft of cars clearly declines over time and there are signs that the underlying local trends [an increase in car thefts] begin to be resumed.” The author suggested that the displacement of vehicle thefts from covered to non-covered car parks occurred. Most studies did not measure either diffusion of benefits or crime displacement. The RESs showed a significant and desirable effect of CCTV for five of the schemes. In the other scheme (Guildford), the effect was undesirable, but the small number of crimes measured in the before and after periods meant that the RES was not significant. When all six effect sizes were combined, the overall RES was 2.03 (CI 1.39–2.96, p ¼ 0.0003), meaning that crime decreased by half (51 %) in experimental areas compared with control areas. This indicates a very large and highly significant desirable effect of CCTV on vehicle crimes in car parks. The six effect sizes were significantly heterogeneous (Q ¼ 31.93, df ¼ 5, p < 0.0001). Other settings. Three of the 44 evaluations took place in other public settings: two in residential areas and one in a hospital. It was considered necessary to categorize these three schemes separately from the others because of the differences in the settings in which these three schemes were implemented as well as their small numbers. There were some notable differences between the two residential schemes. The City Outskirts scheme was implemented in an economically depressed area on the outskirts of a Midlands city, while the Borough scheme was implemented throughout a southern borough of mixed affluence. Camera coverage was quite good in City Outskirts (68 %), but in Borough it was considered low. Gill and Spriggs (2005) noted that this was due in large measure to the use of redeployable cameras in Borough, while fixed cameras were used in City Outskirts. Other interventions were used in City Outskirts, but not in Borough. Evaluations of the two schemes also found contrasting effects on crime: a significant desirable effect in City Outskirts (a 25 % decrease) and a nearly significant undesirable effect in Borough (a 25 % increase).


The one evaluation of CCTV implemented in a city hospital showed that it produced a desirable but nonsignificant effect on crime (RES ¼ 1.38, CI 0.80–2.40), corresponding to a 28 % decrease in crime in the experimental area compared with the control area. Among some of the scheme’s distinguishing features, camera coverage was high (76 %), active monitoring was used, there was a direct line between the camera operators and police, and other interventions were implemented, including improved lighting and police operations. Crime Type The major crime types that were reported were violence (including robbery) and vehicle crimes (including thefts of and from vehicles). Violence was reported in 23 evaluations, but CCTV had a desirable effect in reducing violence in only three cases (Airdrie, Malmo¨, and Shire Town). Overall, there was no effect of CCTV on violence (RES ¼ 1.03, CI 0.96–1.10, ns). The 23 effect sizes were not significantly heterogeneous (Q ¼ 30.87, df ¼ 22, n.s.). Vehicle crimes were reported in 22 evaluations, and CCTV had a desirable effect in reducing them in ten cases: in five of the six car park evaluations (all except Guildford), in three city or town center evaluations (Burnley, Gillingham, and South City), and in City Outskirts and City Hospital. Over all 22 evaluations, CCTV reduced vehicle crimes by 26 % (RES ¼ 1.35, CI 1.10–1.66, p ¼ 0.004). The 22 effect sizes were significantly heterogeneous (Q ¼ 115.1, df ¼ 21, p < 0.0001). The greatest effect was in the largescale, multisite Hawkeye study, but there was a significant effect even if this study was excluded (RES ¼ 1.28, corresponding to a 22 % decrease in crimes). Country Comparison Of the 41 evaluations that were included in the meta-analysis, the overwhelming majority of them were carried out in the United Kingdom (n ¼ 34). Four were from the United States and one each from Canada, Norway, and Sweden. When the pooled meta-analysis results were disaggregated by country, there was evidence


that the use of CCTV to prevent crime was more effective in the United Kingdom than in other countries. In the British studies, CCTV had a significant desirable effect, with an overall 19 % reduction in crime (RES ¼ 1.24, CI 1.10–1.39, p ¼ 0.0005). The British studies were significantly heterogeneous (Q ¼ 350.5, df ¼ 33, p < 0.0001). In the other studies, CCTV showed no desirable effect on crime (RES ¼ 0.97, CI 0.86–1.09, ns). The other studies were also significantly heterogeneous (Q ¼ 14.51, df ¼ 6, p ¼ 0.024). Importantly, the significant results for the British studies were largely driven by the effective programs in car parks.

Key Issues/Controversies Von Hirsch (2000) argues that there are two major issues that confront the “proper uses and limits” of surveillance for crime prevention in public places. The first issue pertains to privacy concerns. This can be expanded to include other social costs that may infringe on public interests or violate legal or constitutional protections. The second issue concerns the matter of the “legitimising role of crime prevention,” or as von Hirsch (2000, p. 61) posits, “To what extent does crime prevention legitimise impinging on any interests of privacy or anonymity in public space?” In car parks, there may be little resistance to the installation of CCTV cameras. In part, this is because the public space is utilized for one rather inconsequential purpose – parking vehicles. It is also the case that a car park is a well-defined and clearly marked physical space, meaning that individuals know that it is a car park and can choose to park their vehicle there or not (providing there are other alternatives). These points stand in sharp contrast to how individuals come into contact with CCTV in other public settings. CCTV in other public settings such as city and town centers, public housing communities, and transportation facilities evokes more resistance on the basis of threats to privacy and other civil liberties, and is associated with a larger number

317

C

of social harms, including the reinforcement of the notion of a fortress society and the social exclusion of marginalized populations (Clarke 2000). Indeed, it is often these settings that are at the center of the debate over how best to strike a balance between the potential crime reduction benefits and social costs associated with CCTV. In these other public settings, CCTV may also result in the social exclusion of vulnerable or marginalized populations such as unemployed youths and the homeless. The fear is that, instead of providing assistance for these groups to get off the street so to speak, CCTV, among other interventions like police and security guards, may push them further away from available services and cause increased harm in the form of crime, victimization, or both. Efforts to reduce any social exclusion effect associated with CCTV in these settings would need to involve other services.

Future Directions Advancing knowledge about the crime prevention benefits of CCTV schemes should begin with attention to the methodological rigor of the evaluation designs. The use of a reasonably comparable control group by all of the 44 included evaluations went some way toward ruling out some of the major threats to internal validity, such as selection, maturation, history, and instrumentation. The effect of CCTV on crime can also be investigated after controlling (e.g., in a regression equation) not only for prior crime but also for other community-level factors that influence crime, such as neighborhood poverty and poor housing. Another possible research design is to match two areas and then to choose one at random to be the experimental area. Of course, several pairs of areas would be better than only one pair. Also important is attention to methodological problems or to changes in programs that take place during and after implementation. Some of these implementation issues include: statistical conclusion validity (adequacy of statistical analyses); construct validity (fidelity); and statistical power (to detect change). For some of the

C

C

318

included evaluations, small numbers of crimes made it difficult to determine whether or not the program had an effect on crime. It is essential to carry out statistical power analyses before embarking on evaluation studies. Few studies attempted to control for regression to the mean, which happens if an intervention is implemented just after an unusually high crime rate period. A long time series of observations is needed to investigate this. The contamination of control areas (i.e., by the CCTV intervention) was another, albeit less common, problem that faced the evaluations. There is also the need for longer follow-up periods to see how far the effects persist. Of the 44 included schemes, many were in operation for 12 months or less prior to being evaluated. This is a very short time to assess a program’s impact on crime or any other outcome measure, and for these programs the question can be asked: Was the intervention in place long enough to provide an accurate estimate of its observed effects on crime? Ideally, time series designs are needed with a long series of crime rates in experimental and control conditions before and after the introduction of CCTV. In the situational crime prevention literature, brief follow-up periods are the norm, but “it is now recognized that more information is needed about the longer-term effects of situational prevention” (Clarke 2001, p. 29). Ideally, the same time periods should be used in before and after measures of crime. Research is also needed to help identify the active ingredients of effective CCTV programs and the causal mechanisms linking CCTV to reductions in crime. Forty-three percent (19 out of 44) of the included programs involved interventions in addition to CCTV, and this makes it difficult to isolate the independent effects of the different components, including the unique effect of CCTV, and the interaction effects of CCTV in combination with other measures. Future experiments are needed that attempt to disentangle elements of effective programs. Also, future experiments need to measure the intensity of the CCTV dose and the dose-response relationship, and need to include alternative methods of measuring crime (surveys as well as police records).


In order to investigate displacement of crime and diffusion of crime prevention benefits, the minimum design should involve one experimental area, one adjacent area, and one nonadjacent comparable control area. If crime decreased in the experimental area, increased in the adjacent area, and stayed constant in the control area, this might be evidence of displacement. If crime decreased in the experimental and adjacent areas and stayed constant or increased in the control area, this might be evidence of diffusion of benefits. Unfortunately, few CCTV studies used this minimum design. Instead, most had an adjacent control area and the remainder of the city as another (noncomparable) control area. Because of this, any conclusions about displacement or diffusion effects of CCTV seem premature at this point in time. This systematic review focused on CCTV’s effects on crime, but it would also be desirable to investigate effects on detection, prosecution, and conviction. Exactly what the optimal circumstances are for effective use of CCTV schemes is not entirely clear at present, and this needs to be established by future evaluation research. But it is important to note that the success of the CCTV schemes in car parks was mostly limited to a reduction in vehicle crimes (the only crime type measured in five of the six schemes) and camera coverage was high for those evaluations that reported on it. In the national British evaluation of the effectiveness of CCTV, Farrington et al. (2007) found that effectiveness was significantly correlated with the degree of coverage of the CCTV cameras, which was greatest in car parks. Furthermore, all six car park schemes included other interventions, such as improved lighting and security guards. It is plausible to suggest that CCTV schemes with high coverage and other interventions and targeted on vehicle crimes are effective. Conversely, the evaluations of CCTV schemes in city and town centers and public housing measured a much larger range of crime types and only a small number of studies involved other interventions. These CCTV schemes, as well as those focused on public transport, did not have a significant effect on crime.

Certainty, Severity, and Their Deterrent Effects

Crime policy is rarely formulated on the basis of hard evidence, and the use of CCTV is no exception. Campbell Collaboration systematic reviews, like the one presented here, provide the most rigorous source for scientific evidence on the leading criminological interventions. It is high time that the evidence is put at center stage in political and policy decisions about preventing crime. Acknowledgments We wish to thank David Wilson for helpful comments on an earlier draft of this entry.

Related Entries ▶ Situational Crime Prevention

319

C

Sivarajasingam V, Shepherd JP, Matthews K (2003) Effect of urban closed circuit television on assault injury and violence detection. Inj Prev 9:312–316 Tilley N (1993) Understanding car parks, crime and CCTV: evaluation lessons from safer cities, Crime prevention unit series paper, No. 42. Home Office, London von Hirsch A (2000) The ethics of public television surveillance. In: von Hirsch A, Garland D, Wakefield A (eds) Ethical and social perspectives on situational crime prevention. Hart, Oxford, UK Welsh BC, Farrington DP (2009a) Making public places safer: surveillance and crime prevention. Oxford University Press, New York Welsh BC, Farrington DP (2009b) Public area CCTV and crime prevention: an updated systematic review and meta-analysis. Justice Q 26:716–745 Williamson D, McLafferty S (2000) The effects of CCTV on crime in public housing: an application of GIS and spatial statistics. Paper presented at the annual meeting of the American Society of Criminology, San Francisco

Recommended Reading and References Armitage R, Smyth G, Pease K (1999) Burnley CCTV evaluation. In: Painter K, Tilley N (eds) Surveillance of public space: CCTV, street lighting and crime prevention, vol 10, Crime prevention studies. Criminal Justice Press, Monsey Clarke RV (2000) Situational prevention, criminology, and social values. In: von Hirsch A, Garland D, Wakefield A (eds) Ethical and social perspectives on situational crime prevention. Hart, Oxford, UK Clarke RV (2001) Effective crime prevention: keeping pace with new developments. Forum Crime Soc 1(1):17–33 Farrington DP, Gill M, Waples SJ, Argomaniz J (2007) The effects of closed-circuit television on crime: metaanalysis of an English national quasi-experimental multi-site evaluation. J Exp Criminol 3:21–38 Gill M, Spriggs A (2005) Assessing the impact of CCTV, Home Office Research Study No. 292. Home Office, London Hier SP (2010) Panoptic dreams: streetscape and video surveillance in Canada. University of British Columbia Press, Vancouver Lipsey MW, Wilson DB (2001) Practical meta-analysis. Sage, Thousand Oaks Norris C (2007) The intensification and bifurcation of surveillance in British criminal justice policy. Eur J Crim Policy Res 13:139–158 Piquero AR, Farrington DP, Blumstein A (2007) Key issues in criminal career research: new analyses of the Cambridge study in delinquent development. Cambridge University Press, New York Savage C (2007) US doles out millions for street cameras: local efforts raise privacy concerns. Boston Globe, 12 Aug 2007. www.boston.com

Centre for Children Committing Offences (CCCO) ▶ Stop Now and Plan (SNAP ®) Model

Certainty, Severity, and Their Deterrent Effects Horst Entorf Department of Economics and Business Administration, Goethe University Frankfurt, Frankfurt, Germany

Overview Severity of punishment without the credible threat of being detected and convicted “. . . is the sound of one hand clapping” (original quote: “Mediation without the credible threat of judicial determination is the sound of one hand clapping,” Genn 2010, p. 125). Theoretical and empirical analyses of general deterrence need to consider both certainty and severity of sanctions, that is, the intertwining activities of police, public

C

C

320

prosecution, courts, and prison conditions. Many studies only rely on the probability of detection, that is, on police activity or efficiency, when discussing the role of expected sanctions. Other studies, in particular articles dealing with US data and thus motivated by the highly persistent upward trend in prisoner population, focus on the imprisonment rate and the severity of sanctions as crucial factors of deterrence (see the survey by Donohue 2009). The (brief) survey at hand summarizes the classical rational-choice fundamentals of modern deterrence theory and covers major theoretical and empirical findings on the interplay of certainty and severity of punishment as well as important underlying methodological problems since the early 1970s of the last century. Any survey of the economics of crime would be incomplete without reminiscence of recent developments in behavioral economics of crime. The last few years seem to have witnessed a change in mainstream economics of crime. Rational-choice models are often criticized because they ignore that cognitive restrictions and emotional factors such as time pressure, peer group influence, or anger restrict the long-run optimality of individual decisions. Simon (1957) was the first to point out that the complexity of situations and limitations of both available information and cognitive capacity would lead to decisions under bounded rationality. Several recent papers focus on shortcomings and necessary extensions of the classical notion of certainty and severity of sanctions. This contribution discusses significant insights stemming from behavioral economics such as impulsiveness (myopia), prospect theory, and anger. Moreover, stigma effects (dynamic deterrence) and the distinction between objective and perceived threats of potential punishment will be addressed. This survey contains the following subchapters: fundamentals of general deterrence theory, theoretical and empirical results concerning (a) the certainty of punishment and (b) the severity of sentences, limitations and extensions of the classical deterrence model, and, finally, the threat of punishment when trust in criminal law is


absent. The entry concludes with remarks on future directions of research in deterrence.

Fundamentals of General Deterrence Theory General deterrence is the avoidance of crime by (deterrable) potential offenders through the credible threat of punishment. This definition entails (a) there is a threat in terms of a judicial system of appropriate sanctions and (b) this threat is credible, that is, there is a legal enforcement system of police, prosecutors, and judges which has the capability to realize a perceptibly high (strictly positive) probability of detection and conviction. The definition also requires that there is a nonempty group of compliers, that is, a deterrable subgroup of the population which can be correctly described as potential offenders because otherwise the theory of deterrence would be futile. The identification of deterrable and non-deterrable subpopulations is important but a rather neglected field in criminology. The theoretical foundations of general deterrence are usually ascribed to Gary Becker’s (1968) seminal article, although the philosophical foundation of deterrence dates back to eighteenth century (Bentham (1781)). In a broader sense, the idea of general deterrence is related to rational-choice theory, which assumes that all individuals, irrespective of being criminal or not, respond to incentives, or as Becker puts it, “Some people become ‘criminals’ not because their basic motivation differs from that of other persons, but because their benefits and costs differ” (Becker 1968, p. 176). Though often disputed and criticized (see, among many others, McAdams and Ulen 2009), the importance of rational choice and general deterrence has also been acknowledged by criminologists and sociologists (as discussed in Rupp 2008, p. 6). Becker’s theory on the supply of offenses is based on the comparison of (uncertain) expected utility from criminal activities to the (relatively certain) utility from not committing crimes. Thus, increasing the individual costs of crime by increasing the expected sentence would


321

lead to a reduction of criminal behavior for rational offenders. Assuming additivity, the theory predicts that if the expected utility from committing a crime, EC ¼ ð1 pÞUC þ pðUC CC Þ;

(1)

exceeds the expected utility from obeying the societal legal norms, ENC , a crime will be committed; otherwise individuals refrain from wrongdoing. Here p denotes the probability that the illegal act will be detected and punished, ð1 pÞ is the probability of getting away with it, UC represents utility from crime, ENC is utility from noncrime, and CC captures disutility (cost) from being punished. Obviously, to make punishment effective, CC needs to be higher than UC . This condition fits Bentham’s (1781) “Principles of Morals and Legislation”: According to Rule 1 of “Of the Proportion Between Punishments and Offences,” “The value of the punishment must not be less in any case than what is sufficient to outweigh that of the profit of the offence” (Bentham 1781, p. 141). Equation 1 can be expressed as the difference between the benefit from committing a crime and the expected sentence in case of detection, that is, EC ¼ UC pCC :

(2)

Equation 2 can be considered as the most parsimonious representation of severity and certainty of conviction. It is obvious that credibility of the threat of sanctions requires the expected sentence, that is, product pCC ; to be positive. As an example of the deterrence rationale, consider the choice between paying a certain amount of money, say, 3 Euros, for a short-term city parking space and not paying the amount and taking the risk of a fine of, say, 20 Euros. If on average every 10th free parking is detected, that is, p ¼ 0.1, then the utility from the illegal action, UC , would be 3 Euros, whereas the expected fine was 2 Euros. Hence the risk-neutral rational “offender” would decide against putting a coin into the parking meter and in favor of the

C

illegal alternative. Of course, such reasoning may be ambiguous in case of risk preference, and as mentioned above, it requires that a nonempty subgroup is complying according to rational-choice rules, that is, a significant number of people belong neither to the group of always strictly norm-abiding citizens nor to the group of never law-abiding individuals. Thus, summarizing the standard rational-choice reasoning covered in Eq. 2, deterrence would not work either when the certainty of a sanction is zero or when sanctions lack severity. In both cases, the product pCC would be zero, that is, there would be no credible threat from expected sanctions for wrong doing. Equation 2 also represents the basic foundation of Becker’s so-called supply of offenses, which is grounded on first-order conditions of the difference between illegal and legal sources of expected utility, EC ENC . Taking derivatives with respect to the crucial impacting factors p, CC and ENC yield the behavioral equation C ¼ Cð p ; CC ; ENC ; XÞ: ðÞ ðÞ

ðÞ

(3)

Thus, under standard neoclassical assumptions (see Becker 1968 for details), crime (C) would fall when the probability of detection and punishment increases or when potential offenders face the risk of higher costs from committing crimes. These disincentives to crime change in response to longer or tougher prison sentences, for example, in terms of tough-on-crime policies such as the Californian three-strikes law or the application of adult criminal law instead of juvenile criminal law (as discussed in Entorf 2012). Crime would also decrease when utility from noncriminal activities, ENC , increases, for instance, because of lower unemployment risk, or improving current and future legal income opportunities, which include the broad and lasting impact of education on crime. Of course, empirical models on crime would fail if only p, CC , and ENC were included as explanatory factors of crime. Thus, X represents further factors of which age,

C

C

322

gender, alcohol and drug addiction, migration background, race, peer influence, and family background belong to the (incomplete) list of variables often (but not regularly) found in crime studies. An indeterminacy of statistical tests of deterrence models is the rather arbitrary choice of X. Indeed, a meta-study by Do¨lling et al. (2009, Table 5) reveals that the choice of control variables has a high influence on finding significance or nonsignificance of deterrence indicators in empirical studies: Dependent on the choice of the control variable, reported average t-values on deterrence variables range from 0.5 to 2.3. The problem of ad hoc specifications is discussed in more detail by Durlauf and Nagin (2010).

On the Impact of Certainty Early Contributions. The first empirical tests of Becker’s supply of offenses model have been conducted in the late 1960s and early 1970s of the last century. Early studies such as Tittle (1969) have found significant crime reducing effects in response to increasing certainty (p). However, these contributions were flawed by endogeneity problems and ignorance of incapacitation effects. Although follow-up papers such as Ehrlich (1973) used two-stage least squares techniques to account for simultaneity problems, identifications strategies such as use of lagged endogenous variables would be considered problematic from the current viewpoint of modern econometrics (see also Durlauf and Nagin 2010). Ehrlich’s (1973) results were intensively discussed, and not all studies agreed on the evidence in favor of the deterrence hypothesis. Many studies argue that the model is too simplistic and empirical findings are not reliable enough to draw any conclusion in favor or against the deterrence hypothesis. According to Rupp (2008), this criticism seems to be exaggerated, but the ambiguity of Ehrlich’s findings was and still is perhaps the main reason for the lasting debate on Ehrlich’s problematic contribution to the deterrence literature.


Ehrlich’s (1973) article also received prominent attention as a significant contribution to the theoretical literature. He modified Becker’s theoretical model by using a time allocation model. This framework widens the perspective because it includes leisure time as a source of utility besides utility from time spent on legal and illegal activities. This makes theoretical predictions less clear than in Becker (1968), where utility is achieved either from illegal or from legal activities. Ehrlich (1973, p. 530, footnote 13) draws attention to the point that the unambiguous negative sign of deterrence effects only holds when individuals are risk neutral or risk averse. Survey of Surveys and Meta-studies. In the first three decades after Becker’s (1968) seminal article, several surveys on the economics of crime have been published. Rupp (2008) summarizes them in more detail. During the last few years, the frequency and number of surveys seems to accelerate. Surveys by Rupp (2008), Donohue (2009), Durlauf and Nagin (2010), and Ritchie (2011), to name only a selection of remarkable new publications, confirm what has been found by earlier studies, that is, that the deterrent effect of the certainty of sanctions far outweighs the severity of punishment. Eide et al. (1994) is a good starting point in order to provide a sample of typical estimates of the deterrent effect of the certainty of sanctions. The authors summarize 20 international cross-sectional studies based on a variety of model specifications, types of data, and regressions. They find the median value of the 118 elasticity estimates of crime rates with respect to various measures of the probability of punishment to be about 0.7. The median of the somewhat fewer severity elasticities is about 0.4. Donohue (2009) focuses on the effect of the imprisonment rate in six studies based on aggregate data and finds that most studies show a negative effect of incarceration rates on crime. However, the estimates of the elasticity of crime range considerably between 0.70 for robbery and results on all index crimes which suggest that marginal imprisonments would


even increase crime. Donohue (2009, p. 17) admits that his best guess for the elasticity is “highly uncertain”, most likely being between 0.10 and 0.15 but “conceivably within the broader interval between 0.05 and 0.40”. Durlauf and Nagin (2010) discuss the papers of Donohue’s survey at some length. They criticize the statistical methodology employed in these studies and dismiss Levitt (1996) as the sole author of the survey who convincingly addressed the simultaneous interdependencies between crime and imprisonment rates. A further problem of studies relying on the imprisonment rate as the sole indicator of p is that it covers compound effects which might cause an omitted variable bias (see the subsequent subchapter on Decomposing the Certainty Effect). Durlauf and Nagin (2010) consider studies based on police manpower more persuasive than those based on imprisonment rates. Their sympathy seems to be affected by the highly influential paper by Levitt (1997) and its critics on the choice of meaningful instruments. Durlauf and Nagin (2010) summarize the findings from these studies and further replication studies and mention an average elasticity of 0.3 for most estimates relating total crimes and police presence. However, given the high dependency of results on the type of crime under consideration (see below), such typical results covering the whole range of crime categories can hardly be considered useful for practical public policy purposes (which is in line with Donohue’s “uncertainty” about his best guess). A recent extensive meta-analysis based on 700 empirical studies with 7,822 estimates of the crime-preventing effect of general deterrence (certainty or severity of sanctions) by Rupp (2008; see also Do¨lling et al. 2009) reveals a large variation of effects depending on the choice of the deterrence variable. For instance, using the indicator ratio of convictions to reported crimes produces a highly statistical evidence in favor of the certainty of sanctions (median t-value ¼ 3.5), whereas using the clearance rate would produce less significant results (median t-value ¼ 1.9; Do¨lling et al. 2009, Table 3). Results also differ with respect

323

C

to the crime category under consideration. Rupp (2008), Table 3.43, p. 126 finds that among the offenses which have been in best accordance with the deterrence hypothesis are speeding (median t-value ¼ 2.2), tax evasion (2.1), severe theft (2.1), and fraud (2.0), whereas sexual assault (0.5), manslaughter (0.0), and drug dealing (0.0) were only rarely found to be consistent with the deterrence hypothesis. Thus, very prominent in the group of consistent types of crimes are nonviolent crimes, while the discordant part seems to incorporate violent crimes and drug-related offenses. Decomposing the Certainty Effect. Testing the theoretical prediction of p, that is, the certainty of sanction, requires empirical measurement of the joint probability of detection and subsequent punishment. Many empirical models are restricted to only one indicator such as clearance rates (Entorf and Spengler 2000) or number of police officers (Levitt 1996). However, variations in clearance rates may not change the threat of expected sentences when prior to judicial decisions cases are dismissed or discharged by the public prosecutor. Moreover, expected sentences differ in response to court decisions, which can be dismissal, unconditional imprisonment, probation, (financial) fine, or other sanctions such as educational measures. Thus, to improve the consistency of empirical models with their theoretical counterparts and in order to analyze the effectiveness of the various sources of deterrence, it is necessary to decompose the certainty effect p into its components. In its simplest way, this can be done by distinguishing between the probability of detection (clearance rate), pcl , and the conditional probability that a suspect is sentenced, psjcl (probability of a sentence, conditional on detection), that is, p ¼ pcl psjcl :

(4)

The variation of certainty effects becomes more transparent when changes in clearance or arrest rates are separated from the ones of courts

C

C

324

and sentences. To fully understand such variations, changes in psjcl can be further decomposed into changes of the indictment rate (driven by public prosecution), pcourtjcl (i.e., the probability that a suspect is brought to court), and those stemming from court decisions, psjcourt (i.e., the probability that a suspect is convicted, given he or she is brought to court), eventually leading to psjcl ¼ pcourtjcl psjcourt . In both the US and European legal systems, the (high) discretionary power of the prosecutor is to determine which case should be disposed of before trial either by dismissal of the charges or by imposing certain obligations on suspects in exchange for laying the file aside. The picture is still incomplete unless the risk of the most severe outcome conditional on detection has been covered, that is, the certainty of a (unconditional) prison sentence. After including pprisjs , that is, the probability of imprisonment given a conviction, the full decomposition of the probability of imprisonment (conditional on detection) looks as follows: pprisjcl ¼ p pprisjs ¼ pcl pcourtjcl psjcourt pprisjs : (5) Not surprisingly, also empirical indicators of pprisjcl vary across time and space (see Entorf and Spengler 2013) Thus, to summarize, the probability of a sanction depends on the interplay of many factors, and all of them are far from being constant parameters. Reasonable public policy analysis and recommendations should take the role of all key players and their decisions, that is, interplay of police, prosecutors, and courts, into account. Only few studies also cover the risk of convictions, for example, by the ratio of convictions to arrests. Cornwell and Trumbull (1994), who were among the first applying panel econometrics for testing general deterrence models, present exceptional work because of their comprehensive list of law enforcement variables containing the probabilities of arrest, conviction (conditional on arrest), and imprisonment (conditional on conviction) as well as the severity of sanctions. However,


the overall impression is that only few studies are taking account of a comprehensive list of factors. In particular, the interplay of conviction rates and sentence lengths, that is, the essential components of certainty and expected severity of punishment, has been neglected. However, it is obvious that components of general deterrence do not work independently of each other. Missing factors of the law enforcement system might cause severe omitted variable biases (Entorf and Spengler 2013).

On the Impact of Severity Death Penalty. The severity of punishment is the fundamental idea of capital punishment: “. . . if rational people fear death more than other punishment, the death penalty should have the greatest deterrent effect” (Ehrlich 1977, quotation found in Rupp 2008, p. 22). Although this statement is questionable, either because the expected sentence would be nil when there would be no (perceived) risk of detection or because “rational” suicide bombers expect going to paradise with the infamous 72 virgins, the focus of most studies in the aftermath of Ehrlich (1975) lies on the absolute deterrent effect of capital punishment. Most publications refer to US evidence, as can be seen from the fact that 71 out of 82 studies evaluated by Rupp (2008), p. 22, use US data and 5 are based on Canadian data. Ehrlich’s (1975, 1977) articles have fueled an ongoing debate about the effectiveness of the death penalty. His results have been criticized for data errors, misspecification problems, and other methodological problems. Numerous subsequent studies have rejected but also confirmed significance of the death penalty. Donohue and Wolfers (2005) summarize several studies on the death penalty in the United States. They conclude that all outcomes are too fragile and that the number of executions is too low to draw any noteworthy and robust conclusion. This result is in line with the meta-study on the significance of death penalties by Do¨lling et al. (2009, Table 10), where median t-values of


effects range between 0.5 and +0.1, that is, below usual significance levels. This survey is not the right place to summarize all facets of the complex debate on the death penalty which is interesting for its methodological aspects, but most probably the large majority of scientists from industrialized countries outside the USA would not consider it a practical and ethically tolerable alternative to lifelong sentences. Evidence on the Length of the Prison Sentence and Prison Conditions. According to the aforementioned meta-study (Do¨lling et al. 2009, Table 3), most results using the average length of served prison sentences as an indicator of the severity of sanctions do not show statistical significance (median t-value ¼ 0.6). This result is in line with the tenor of previous and recent surveys such as Eide et al. (1994), Durlauf and Nagin (2010), or Ritchie (2011). However, Durlauf and Nagin (2010) correctly state that most effects are measured as marginal effects in addition to already existing long sentences. Thus, more reliable evidence would be based on discontinuous jumps of severity. Levitt (1998) has found a significant drop in the offending of young adults when they reach the age of 18, that is, the age of jurisdiction for adult courts in Florida. Other abrupt and unexpected changes of deterrence come from natural experiments. Maurin and Ouss (2009) and Drago et al. (2009) study the effect of external variations arising in response to collective pardons in France and Italy, respectively. Maurin and Ouss (2009) show that 5 years after release, those who have received a reduced sentence as a consequence of the pardon had a 12 % higher rate of recidivism than those who had received no reduction. In the case of the Italian clemency, inmates received a conditional reduction of prison sentences. In case of reoffending, they had to serve the remaining amount of their sentence (in addition to the new sentence). Drago et al. (2009) find that the threat of increased sanction continued after release: For every month the former prisoner would have to serve if convicted, there was a 1.2 % reduction in the propensity to recommit crimes. Both studies touch the problem of specific

C

325

deterrence, that is, whether individuals would avoid future imprisonment because they are deterred by their own previous sentencing experience. However, in particular the study by Maurin and Ouss (2009) also tests the deterrent effects of longer prison sentences, though based on a subgroup of former inmates. In the case of Drago et al. (2009), the effect is a combination of certainty and severity because the threat of serving the remaining sentence also depends on the probability of detection and conviction.

Limitations and Extensions of the Classical Deterrence Model Predictions of the fundamental model (as presented in Section 2) are based on a static model under regular neoclassical assumptions. The classical rational-choice model needs to be reconsidered in the face of challenging recent findings of behavioral economics. In particular, the (objective) detection probability and the disutility from punishment need to be adjusted given insights from prospect theory, myopia, and dynamic deterrence: tþt Ct ¼ CðpðpÞ; btþt c CC:tþt ; bnc ENC:tþt ; ENC:t XÞ: ðÞ

ðÞ

ðÞ

ðÞ

(6) The new elements of a behavioral supply of offenses model (Eq. 6) are to be introduced: Criminological research suggests that offenders have unusually high discount rates, that is, they place high value on immediate utility gains, whereas future events are strongly discounted, that is, they consider them as less important for current decisions than noncriminals. As consequences of criminal activities, if any, would have to be faced at some unknown time in the future, this so-called present bias (also called impulsiveness or myopia) leads to underestimation of expected future costs from punishment, CC:tþt . In particular, adolescence, alcohol, or drug misuse (not to mention mental impairment) may lead to a very small personal weight of future consequences, that is, bc 90 % (2011a), Fagan based adherence/year to objectives, core et al. (2008), prevention components, and dosage) of 17 Fagan et al. programs different evidence-based school- (2009) based, after-school, and parenting programs Implementation Were more likely to monitor Fagan et al. monitoring prevention program (2008) implementation quality Prevention program reach

Delinquency initiation

Implemented more evidenceFagan et al. based prevention programs and (2011a) reached more children and parents Youth risk and behavior outcomesa Eighth grade – randomized Related controlled trial phase references Youth exposed to CTC: Reported significantly lower Hawkins et al. growth in risk factors targeted for (2008a, 2009) preventive intervention Were 24 % less likely to initiate Hawkins et al. delinquent behavior (2008a, 2009)

Alcohol use initiation

Were 32 % less likely to initiate alcohol use

Hawkins et al. (2009)

Tobacco use initiation

Were 33 % less likely to initiate cigarette smoking


Prevalence of delinquency, violence, and substance use

Were significantly less likely to use alcohol in the past month, binge drink in the past 2 weeks, commit fewer delinquent behaviors in the past year


Risk factors

Sustainability phaseb CTC communities: Eleven surviving coalitions implemented significantly more milestones and benchmarks than controls 20 months after support ended Sustained significantly greater adoption 18 months after support ended

Related references Gloppen et al. (2012)

Rhew et al. (2013)

Sustained high-fidelity Fagan implementation 2 years after the et al. randomized controlled trial ended (2011a)

Continued to be more likely to monitor prevention program implementation quality 2 years after support ended Sustained greater implementation and reach 2 years after support ended

Fagan et al. (2011a) Fagan et al. (2011a)

Tenth grade – sustainability phase Related references Youth exposed to CTC: Reported significantly lower Hawkins levels of targeted risk et al. (2012) Were 18 % less likely to initiate Hawkins delinquent behavior et al. (2012) Were 29 % less likely to initiate Hawkins alcohol use et al. (2012) Were 28 % less likely to initiate Hawkins cigarette smoking et al. (2012) Were significantly less likely to Hawkins commit delinquent or violent et al. behavior in the past year (2012)

(continued)

C

C

402

Communities That Care

Communities That Care, Table 1 (continued)

Universal effects

Benefit-cost analysis

Prevention system outcomesa Randomized controlled trial phase Generally benefited equally from CTC, regardless of differences in gender, baseline risk, early substance use, or early engagement in delinquency CTC was estimated to return $4,259 per youth in net present benefit or $5.30 per $1.00 invested

Related references Oesterle et al. (2010)

Sustainability phaseb

Related references

Kuklinski et al. (2012)

a

Communities were matched in pairs within state on population, racial and ethnic diversity, economic indicators, and crime rates. They were randomly assigned by coin toss to the CTC or control condition. At baseline, CTC and control communities did not differ in their knowledge and use of science-based prevention. CTC and control community youth did not differ with respect to levels of risk, protection, and key outcome variables including substance use and delinquency. Youth risk and behavior outcomes were from a panel of 4,407 youth initially constituted in Grade 5 and surveyed annually through Grade 10 b Prevention system sustainability phase outcomes are reported at latest measurement

School-Based Programs

After-School Programs

Parent Training Programs

Total

100%

Percentage Actually Delivered

80%

60%

40%

20%

0%

Percentage of Required Material or Components Actually Delivered Percentage of Required Number, Length, and Frequency of Sessions Actually Delivered * No session fidelity forms submitted.

Communities That Care, Fig. 5 CTC communities implemented prevention programs with fidelity (average results over four intervention years)


leaders from control communities (Brown et al. 2011). Adoption of a science-based approach to prevention is hypothesized to be a key to achieving better youth outcomes through CTC. Key leaders in CTC communities also were willing to allocate a greater percentage of funding for prevention programs than key leaders in control communities, providing some evidence of increased community support for prevention in CTC communities. Although levels of collaboration did not differ in CTC and control communities at the third year of intervention, high levels of collaboration were observed in both. Effects on adoption of a sciencebased approach to prevention and desired prevention funding were sustained 18 months after the intervention phase of the Community Youth Development Study ended, and, at that point, key community leaders in CTC communities reported stronger community norms against adolescent drug use than did key leaders in control communities (Rhew et al. 2013).

CTC Leads to Sustained Prevention of Youth Risk, Delinquency, and Substance Use CTC communities each prioritized 2–5 risk factors to target with tested and effective prevention programs, as shown in Table 2. Effects on youth risk and problem behaviors were first observed in Grade 7, after less than 2 full years of intervention and earlier than CTC’s theory of change suggested. Effects have been sustained through Grade 10, 1 year after the trial’s intervention phase ended. Effects on Risk Factor Exposure. Longitudinal panel youth in CTC and control communities reported similar levels of targeted risk in Grade 5, when the intervention began (Hawkins et al. 2008a), but targeted risk exposure grew more slowly for youth in CTC communities between Grade 5 and Grade 10 (Hawkins et al. 2012). Significantly lower levels of targeted risk were first reported by CTC panel youth 1.67 years into the intervention, in Grade 7, (Hawkins et al. 2008a) and have been sustained in CTC panel youth reports through Grade 10.

403

C

Effects on the Initiation of Delinquency and Substance Use. Panel youth from CTC and control communities also reported similar levels of delinquency, alcohol use, and cigarette smoking at Grade 5 baseline. However, between Grades 5 and 10, CTC had significant effects on the initiation of these behaviors. Significant reductions in the initiation of delinquency were first observed in the spring of Grade 7 (Hawkins et al. 2008a). Panel youth from CTC communities were 25 % less likely than panel youth from control communities to initiate delinquent behavior, and they remained 24 % less likely to do so in Grade 8 (Hawkins et al. 2009). Significantly lower delinquency initiation rates were sustained through Grade 10 (Hawkins et al. 2012), when panel youth from CTC communities were 18 % less likely to initiate delinquency than panel youth from control communities. Preventive effects on alcohol use and cigarette use were first observed in the spring of Grade 8, 2.67 years after intervention programs were implemented. Grade 8 youth from CTC communities were 32 % less likely to initiate alcohol use and 33 % less likely to initiate cigarette smoking than Grade 8 youth from control communities (Hawkins et al. 2009). Preventive effects were sustained through Grade 10 (Hawkins et al. 2012), when CTC panel youth were 29 % less likely to initiate alcohol use and 28 % less likely to initiate cigarette smoking than panel youth from control communities. As shown in Fig. 6, differences in the initiation of delinquency, alcohol use, and cigarette smoking from Grade 5 through Grade 10 led to cumulatively lower rates of initiation over time: 62 % of 10th-grade youth in the panel from CTC communities had engaged in delinquent behavior compared with 70 % of 10th-grade youth in the panel from control communities; 67 % versus 75 % had initiated alcohol use; and 44 % versus 52 % had smoked cigarettes. Reductions in the Prevalence of Delinquency, Violence, and Substance Use. CTC also significantly reduced the prevalence of youth problem behaviors in Grades 8 and 10. In Grade 8, the prevalence of alcohol use in the past month, binge drinking (five or more drinks in

C

C

404


Communities That Care, Table 2 Risk factors targeted for prevention by communities in the CYDS Risk factor Community Laws and norms favorable to drug use School Low commitment to school Academic failure Family Family conflict Poor family management Parental attitudes favorable to problem behavior Peer Antisocial friends Peer rewards for antisocial behavior Individual Attitudes favorable to antisocial behavior Rebelliousness Low perceived risk of drug use

CTC community 1 2 3 4

5

✔

✔

✔

✔ ✔

8

✔

✔

✔ ✔

✔

✔

9

10

11

12

✔ ✔

✔

✔

✔

✔ ✔

✔

✔

✔

✔ ✔ ✔

✔

✔

✔

✔

✔

✔

✔

✔ ✔

✔

✔ ✔ ✔

Delinquency

Alcohol Use

✔ ✔

Cigarette Smoking 75%

70%

Cumulative Initiation Rate

7

✔

80% 67%

65% 59%

57%

60%

62% 48%

67% 61%

47%

57%

52% 44%

53%

49%

36%

40%

38%

37% 44%

41%

23%

23%

26%

33%

30%

20%

6

18%

20%

20%

9%

44% 39%

32%

24%

17%

7%

0% 5th

6th

7th

8th

9th

10th 5th

6th

7th CTC

8th

9th

10th 5th

6th

7th

8th

9th

10th

Grade Control

Communities That Care, Fig. 6 CTC leads to significant reductions in the cumulative initiation of delinquency and substance use

a row) in the past 2 weeks, and the variety of delinquent behaviors committed in the past year were all significantly lower in CTC panel youth compared to control community panel youth (Hawkins et al. 2009). A subset of the delinquency items was used to create a measure of violent behavior. The CYDS found significant effects of CTC in reducing the prevalence of delinquent behavior and violence in the past year in the spring of Grade 10 (Hawkins et al. 2012).

Tenth-grade students in CTC communities had 17 % lower odds of reporting any delinquent behavior in the past year (t (9) ¼ 2.33; p ¼ .04; AOR ¼ .83) and 25 % lower odds of reporting any violent behavior in the past year (t (9) ¼ 2.51; p ¼ .03; AOR ¼ .75) compared to students in control communities. CTC’s Effects Are Universal. CTC was designed as a universal intervention, and the universality of its effects on the prevalence of


405

substance use and delinquent behavior was tested based on outcomes observed in panel youth in the spring of eighth grade (Oesterle et al. 2010). Four groups of students were established using baseline Grade 5 data: boys and girls, high baseline risk versus all others, early substance users versus nonusers, and early engagers in delinquency vs. non-engagers. Results generally supported the universal effectiveness of the CTC intervention with respect to the prevalence of delinquent behavior and substance use within groups. There were two exceptions. The effect of CTC on reducing the prevalence of substance use in eighth grade was stronger for boys than girls, and the effect on the prevalence of delinquency was stronger for students who were non-delinquent at baseline.

Communities That Care, cost-beneficial intervention

CTC is a Cost-Beneficial Intervention

communities need to overcome five types of challenges to successfully own and operate CTC. Be Ready to Implement a Comprehensive Approach to Prevention. Communities must be ready to initiate comprehensive evidence-based prevention in order to fully implement CTC. This requires readiness along several key dimensions. Differing views among stakeholders about the most effective approach to addressing youth problems must be considered. If particular groups advocate only exclusionary approaches such as arrest and incarceration, or exclusive reliance on treatment approaches, they may need further education about the potential effectiveness of prevention prior to engaging in the CTC process. If communities have an unsuccessful history of collaboration, characterized by conflict and competition for scarce resources, building trust through collaborative efforts to achieve smaller specific goals, like closing a crack house or getting a crosswalk installed on a busy street, may be necessary prior to initiating CTC. Recruit and Engage Key Leaders. If CTC is to succeed, key community leaders who control resources and shape opinions, like the mayor, school superintendent, and police chief, must decide to bring the CTC system into their community. Their recruitment and engagement is essential to CTC’s success. Without ongoing

A cost-benefit analysis was undertaken to determine whether CTC is a sound investment of public dollars based on significant preventive effects on cigarette smoking and delinquency initiation found in Grade 8 (Kuklinski et al. 2012). Findings are summarized in Table 3. CTC’s long-term financial benefits from reduced initiation were compared to a conservative CTC implementation cost of $991 per youth over 5 years. Most communities spent significantly less, roughly $513 per youth. Approximately 37 % of the total went toward a CTC coordinator and the coalition; 35 % to prevention programming; 23 % to CTC training, technical assistance, and monitoring; and 4 % to miscellaneous expenses. Implementation costs were more than offset by CTC’s estimated financial benefits of $5,250 per youth. The benefit-cost ratio indicates $5.30 is returned to society for every $1.00 invested, evidence that CTC is a cost-beneficial investment.

Challenges to Implementing CTC Like other community mobilization strategies, CTC can be challenging for communities to implement and sustain. In our experience,

Table

3 CTC

C is

a

Stakeholders Participants Taxpayers Others Total Per-youth benefitsa Smoking Delinquency Total benefits Per-youth costsa Per-youth net present benefita Benefit-cost ratio

$671 $671

$140 $2,033 $2,173

$812 $2,405 $4,438 $2,405 $5,250 $991 $4,259

$5.30

a

All benefits and costs are expressed in discounted 2004 dollars

C

C

406

“buy-in” from key leaders and influential champions who have access to resources, CTC will not be viable because local resources will need to be committed and often redirected to prevention. Evidence-Based Prevention Often Competes with Usual Practice. The board needs to advocate for a risk- and protection-focused approach to preventing problem behavior and for the use of tested and effective preventive interventions to address local priorities. At times, this can mean giving up programs that are rooted in community tradition or are local favorites if they lack research evidence of effectiveness. Faithfully Implement and Evaluate. Achieving prevention goals requires high-quality implementation of the CTC system and prevention programs and monitoring outcomes over time. Monitoring community-wide implementation and making needed adjustments can be a challenge when programs are implemented in multiple classrooms, agencies, and community settings. Evaluation of changes in risk, protection, and outcomes at the community level is also a challenge. At the local level, it may be several years before community-wide reductions are achieved, requiring long-term commitment to the CTC approach. Ensure Adequate Resources. CTC coalitions and community coordinators represent critical human resources needed to run CTC. Maintaining board engagement and recruiting and orienting new members to CTC and the coalition’s progress to date are essential to long-term sustainability. Securing ongoing funding for coalitions and coordinators, prevention programs, and training and technical assistance can be a challenge for coalitions. In the CYDS and in a second quasiexperimental study of CTC in Pennsylvania (Feinberg et al. 2008), coalitions were aided in initial implementation efforts by seed money for coalition organizing and training and implementing chosen effective prevention approaches. Many coalitions in both studies have demonstrated success in generating ongoing resources for CTC. Those most likely to obtain funding were well functioning and planned for financial sustainability as part of their regular operations (Feinberg et al. 2008).


Steps to Broader Dissemination of CTC All manuals and materials needed to implement CTC have been placed in the public domain by the US Substance Abuse and Mental Health Services Administration of the Department of Health and Human Services and are freely available at http://www.communitiesthatcare.net. CTC is currently being implemented in communities across the United States and in countries including Australia, Canada, Croatia, Germany, the Netherlands, and the United Kingdom. The CTC tests in both the Community Youth Development Study and the Pennsylvania quasi-experimental study were conducted with certified providers of training and technical assistance, as well as monitoring and oversight of implementation, including reviewing and providing feedback on annual plans. State and federal governments, local philanthropic organizations, the United Way, and other local funders could assist in broader dissemination of CTC by supporting training, technical assistance, and monitoring of processes and outcomes. Efficiencies can be achieved by offering CTC training through state governments, who can then train state prevention workers and contractors to deliver CTC training and technical assistance to local communities. A national CTC training and certification center could support the training of state and community personnel in CTC and share lessons learned from the implementation of CTC.

Conclusion Policymakers and researchers increasingly recognize the need for clear and compelling evidence about how best to prevent prevalent and costly problem behaviors in youth. Delinquency and substance abuse are problems that begin in late childhood and adolescence, often with lifelong consequences. The results of the Community Youth Development Study show that Communities That Care, a coalition-based prevention system employing a risk- and protection-focused approach, is a cost-beneficial, effective way to prevent delinquency, violence, and tobacco and alcohol use in youth. Intervention communities were


able to faithfully implement the CTC prevention system and evidence-based programs designed to address community-specific prevention priorities, and they sustained this success after the intervention phase of the Community Youth Development Study was over. These efforts were related to significant, meaningful, and sustained reductions community wide in the percentage of youth in intervention communities initiating delinquent behavior, alcohol use, and cigarette smoking, and in the prevalence of current involvement in smoking, delinquency, and violence when compared to youth from control communities. Acknowledgments The Community Youth Development Study was supported by research grant R01 DA015183-03 from the National Institute on Drug Abuse, with co-funding from the National Institute of Child Health and Human Development, the National Cancer Institute, the National Institute of Mental Health, the Center for Substance Abuse Prevention, and the National Institute on Alcohol Abuse and Alcoholism. The content of this entry is solely the responsibility of the authors and does not necessarily represent the official views of the funding agencies.

Related Entries ▶ Drug Abuse and Alcohol Dependence Among Inmates ▶ Poverty, Inequality, and Area Differences in Crime ▶ Randomized Block Designs ▶ Randomized Experiments in Criminology and Criminal Justice ▶ School-Based Interventions for Aggressive and Disruptive Behavior: A Meta-Analysis ▶ School Bullying as a Risk Factor for Later Criminal Offending ▶ Tobacco Abuse in Correctional Settings

Recommended Reading and References Arthur MW, Briney JS, Hawkins JD, Abbott RD, BrookeWeiss BL, Catalano RF (2007) Measuring risk and protection in communities using the Communities That Care Youth Survey. Eval Program Plann 30:197–211 Arthur MW, Hawkins JD, Brown EC, Briney JS, Oesterle S, Abbott RD (2010) Implementation of the Communities

407

C

That Care prevention system by coalitions in the Community Youth Development Study. J Community Psychol 38:245–258 Brown EC, Graham JW, Hawkins JD, Arthur MW, Baldwin MM, Oesterle S, Briney JS, Catalano RF, Abbott RD (2009) Design and analysis of the Community Youth Development study longitudinal cohort sample. Eval Rev 33:311–334 Brown EC, Hawkins JD, Arthur MW, Briney JS, Fagan AA (2011) Prevention service system transformation using Communities That Care. J Community Psychol 39:183–201 Catalano RF, Hawkins JD (1996) The social development model: a theory of antisocial behavior. In: Hawkins JD (ed) Delinquency and crime: current theories. Cambridge University Press, New York, pp 149–197 Catalano RF, Haggerty KP, Hawkins JD, Elgin J (2011) Prevention of substance use and substance use disorders: the role of risk and protective factors. In: Kaminer Y, Winters KC (eds) Clinical manual of adolescent substance abuse treatment. American Psychiatric Publishing, Washington, DC, pp 25–63 David-Ferdon C, Hammond WR (2008) Community mobilization to prevent youth violence and to create safer communities. Am J Prev Med 34:S1–S2 Fagan AA, Hanson K, Hawkins JD, Arthur MW (2008) Implementing effective community-based prevention programs in the Community Youth Development Study. Youth Violence Juvenile Justice 6:256–278 Fagan AA, Hanson K, Hawkins JD, Arthur MW (2009) Translational research in action: implementation of the Communities That Care prevention system in 12 communities. J Community Psychol 37:809–829 Fagan AA, Arthur MW, Hanson K, Briney JS, Hawkins JD (2011a) Effects of Communities That Care on the adoption and implementation fidelity of evidence-based prevention programs in communities: results from a randomized controlled trial. Prev Sci 12:223–234 Fagan AA, Hawkins JD, Catalano RF (2011b) Engaging communities to prevent underage drinking. Alcohol Res Health 34:167–174 Feinberg ME, Bontempo DE, Greenberg MT (2008) Predictors and level of sustainability of community prevention coalitions. Am J Prev Med 34:495–501 Gloppen KM, Arthur, MW, Hawkins, JD, Shapiro, VB (2012) Sustainability of the Communities That Care prevention system by coalitions participating in the Community Youth Development Study. J Adolesc Health 51:259–264 Hallfors D, Cho H, Livert D, Kadushin C (2002) Fighting back against substance abuse: are community coalitions winning? Am J Prev Med 23:237–245 Hawkins JD, Catalano RF Jr et al (1992a) Communities That Care: action for drug abuse prevention, 1st edn. Jossey-Bass, San Francisco Hawkins JD, Catalano RF, Miller JY (1992b) Risk and protective factors for alcohol and other drug problems in adolescence and early adulthood: implications for substance-abuse prevention. Psychol Bull 112:64–105

C

C

408

Hawkins JD, Brown EC, Oesterle S, Arthur MW, Abbott RD, Catalano RF (2008a) Early effects of Communities That Care on targeted risks and initiation of delinquent behavior and substance use. J Adolesc Health 43:15–22 Hawkins JD, Catalano RF, Arthur MW, Egan E, Brown EC, Abbott RD, Murray DM (2008b) Testing Communities That Care: the rationale, design and behavioral baseline equivalence of the Community Youth Development Study. Prev Sci 9:178–190 Hawkins JD, Oesterle S, Brown EC, Arthur MW, Abbott RD, Fagan AA, Catalano RF (2009) Results of a type 2 translational research trial to prevent adolescent drug use and delinquency: a test of Communities That Care. Arch Pediatr Adolesc Med 163:789–798 Hawkins JD, Shapiro VB, Fagan AA (2010) Disseminating effective community prevention practices: opportunities for social work education. Res Soc Work Pract 20:518–527 Hawkins JD, Oesterle S, Brown EC, Monahan KC, Abbott RD, Arthur MW, Catalano RF (2012) Sustained decreases in risk exposure and youth problem behaviors after installation of the Communities That Care prevention system in a randomized trial. Arch Pediatr Adolesc Med 166:141–148 Johnston LD, O’Malley, PM, Bachman, JG (2000) The Monitoring the Future national survey results on adolescent drug use: overview of key findings, 1999 (NIH Publication No. 00-4690). National Institute on Drug Abuse, Bethesda Kuklinski MR, Briney JS, Hawkins JD, Catalano RF (2012) Cost-benefit analysis of Communities That Care outcomes at eighth grade. Prev Sci 13:150–161 Merzel C, D’Affliti J (2003) Reconsidering communitybased health promotion: promise, performance, and potential. Public Health Matters 93:557–574 Oesterle S, Hawkins JD, Fagan AA, Abbott RD, Catalano RF (2010) Testing the universality of the effects of the Communities That Care prevention system for preventing adolescent drug use and delinquency. Prev Sci 11:411–424 Patton GC, Coffey C, Sawyer SM, Viner RM, Haller DM, Bose K, Vos T, Ferguson J, Mathers CD (2009) Global patterns of mortality in young people: a systematic analysis of population health data. Lancet 374:881–892 Quinby RK, Fagan AA, Hanson K, Brooke-Weiss B, Arthur MW, Hawkins JD (2008) Installing the Communities That Care prevention system: implementation progress and fidelity in a randomized controlled trial. J Community Psychol 36:313–332 Rhew IC, Brown EC, Hawkins JD, Briney JS (2013) Sustained effects of Communities That Care on prevention service system transformation. Am J Publ Health 103:529–535 Ringwalt C, Vincus AA, Hanley S, Ennett ST, Bowling JM, Rohrbach LA (2009) The prevalence of evidencebased drug use prevention curricula in U.S. middle schools in 2005. Prev Sci 10:33–40

Community Control

Community Control: Criminality, Community Social Organization, Cultural Organization, Deviant Opportunities, Directs Controls ▶ Control Theory

Community Courts Rachel Swaner Center for Court Innovation, New York, NY, USA

Overview A community court, often called a neighborhood or community justice center, is a neighborhoodfocused court that applies a problem-solving approach to local crime and safety concerns. Emphasizing improved outcomes for offenders (e.g., lower recidivism) and for communities (e.g., safer neighborhoods), community courts strive to prevent crime by addressing its underlying causes (Berman and Feinblatt 2005; Rottman and Casey 1999). Most community courts handle low-level criminal cases arising in a set of target neighborhoods. Some community courts are multi-jurisdictional, handling housing disputes, juvenile delinquency matters, or other case types in the family court system as well. Community courts are typically located in a separate courthouse situated within the targeted neighborhoods, although some community courts do operate out of a centralized “downtown” court. In 1993, the first community court was started in Manhattan, New York City. It was opened to address quality-of-life crimes such as prostitution, illegal vending, vandalism, and minor drug possession in the central business district of Midtown, which included Times Square, then the home of much low-level crime. Since then, over 60 community court projects have opened worldwide. In the United States alone, there about 40,

Community Courts

while there are at least 17 in South Africa, 13 in England and Wales, and one each in Australia and Canada (Henry and Kralstein 2011). These courts seek to bring a neighborhood-focused, problem-solving approach to local crime, social disorder, and safety concerns, focusing on alternatives to jail and fines, instead giving sanctions such as community service, social service, and treatment for drug addiction. This entry outlines the principles and theories of community courts and summarizes the current research regarding the effectiveness of community courts on expanding sentencing options, improving compliance rates for court mandates, lowering recidivism, and reducing costs. As will be discussed in further detail, the evaluation literature shows that community courts do indeed dispense more types of sanctions (e.g., community and social services) than regular courts; though jail is still used, it is mostly as a resentencing option. Mandate completion rates have shown to be higher than in more traditional courts, though more research may be needed in this area. Findings on differences in recidivism rates have been mixed; while some studies have shown lower recidivism rates for community court participants, other studies have failed to show an effect, which may in part be due to small sample size or no true comparison group. Finally, studies have shown mixed results on cost savings as well.

Background Description One of the main components of community courts is the use of alternative sanctions, the goals of which are twofold: first, to pay back the community that may have been harmed by the offense through visible community service work projects in the neighborhood, such as cleaning local parks and painting over graffiti. Second, social services such as drug treatment and job training are assigned to link defendants with programs that address the problems that may have played a role in leading them to commit crimes. This combination of accountability and help seeks to reduce chronic offending. In order to

409

C

understand local problems and priorities, community court staff also commonly venture out into local neighborhoods, attending community meetings and serving as a convener and problemsolver between criminal justice agencies (e.g., police, prosecutors, or probation) and the citizens they serve. However, there is no one concrete model for community courts, which makes it difficult to develop one standard definition. But as Berman (2010) outlined, community courts are usually guided by the following six key principles that differentiate them from more traditional courts: 1. Enhanced Information – Community courts seek to make as much information as possible available about the individual defendant and the community context of the crime at the defendant’s first court appearance. This allows for improved decision-making by judges, attorneys, and other justice officials, enabling them to match the defendant to appropriate sanctions and services. 2. Community Engagement – Community courts seek to assign offenders to community service projects in places where neighbors can see what they’re doing, welcome observers and visitors, make information about courtroom activities easily accessible to the public, and involve the community in addressing crimerelated problems. 3. Collaboration – Government agencies, social service agencies, and other community organizations work together under a single roof, where the physical proximity facilitates closer and more coordinated working relationships. 4. Individualized Justice – Community courts link defendants to individually tailored, community-based sanctions such as community service, job or educational training, and drug treatment. 5. Accountability – Community courts insist on regular and strict compliance monitoring, with clear consequences for noncompliance. Additionally, community courts often escort offenders immediately to the place where they will receive their community or social service assignment, greatly increasing the likelihood that they will comply with their

C

C

410

sanction. These mechanisms work to improve the accountability of the offenders. 6. Outcomes – Data collection and analysis are important tools at community courts. Data help to answer questions related to case processing (how many and how quickly), status and compliance, as well as to perceptions of the court, procedural fairness, and local quality of life. Expanding on the sixth point above, community courts usually follow one or more of the following theories of change for attempting to achieve these outcomes, as outlined by Lee et al. (2013). Procedural Justice and the Social Construction of Legitimacy As Tyler (2001) suggested, public trust and confidence in police and courts is not related to performance or outcomes but to how fairly people feel they were treated. The idea of fairness vis-a`-vis justice was clearly defined by John Rawls (1971) in his A Theory of Justice. He said that perfect procedural justice had two characteristics: (1) an independent criterion for what constitutes a fair or just outcome of the procedure and (2) a procedure that guarantees that the fair outcome will be achieved. Tyler (1990) put forth that citizens generally hold favorable views towards institutions that are perceived as unbiased, while holding negative views of those that are believed to be partisan or discriminatory. Procedural fairness is present when people perceive that they are experiencing the following in their interaction with judges (Tyler 2004: 443–47): • Respect: People are treated with dignity and their rights are respected. • Neutrality: Honest and impartial decisionmakers base their decisions on facts. • Participation: Each party has an opportunity to express his or her viewpoint to the decisionmaker, and conversely, decision-makers clearly articulate the basis of their decisions in ways that litigants can understand (without jargon). • Trustworthiness: Decision-makers appear benevolent, caring, motivated to treat parties fairly, and sincerely concerned about people.

Community Courts

Research has endeavored to assess the meaning of procedural justice for those who come in contact with the criminal justice system. As Wissler (1995), Lind and Tyler (1998), and Paternoster et al. (1997) have demonstrated, people are more willing to accept decisions when they think criminal justice officials or legal institutions are acting fairly. Similarly, Tyler (1990) has argued that citizens generally hold favorable views towards institutions that are perceived as unbiased, while holding negative views of those that are believed to be partisan or discriminatory. Elsewhere, Tyler (2001) has suggested that public trust and confidence in police and courts is related to how fair people feel they were treated; performance and outcomes are secondary factors. Research has also attempted to understand the relationship between procedural justice and the public’s law-related behavior, as there is growing concern that perceived injustice itself causes criminal behavior (LaFree 1998; Mann 1993; Russell 1998; Tyler 1990). Tyler and Huo (2002) have proffered that when citizens perceive justice system agencies to be fair, they are more likely to comply with the law, legal authorities, and court mandates, increasing institutional confidence. Compliance, Tyler (2004: 307) further explains, is linked to legitimacy – “the property that a rule or an authority has when others feel obligated to defer voluntarily” – and that to the extent that people regard the courts as legitimate, they are more willing to accept the directives of the courts. To put it another way, legitimacy develops out of the use of fair procedures and the provision of respectful treatment. Procedural justice, then, is the key antecedent to legitimacy, voluntary compliance, and lower recidivism: the greater the legitimacy, the less the likelihood of defiance, hostility, and resistance to laws, legal authorities, and legal institutions. Community Involvement In addition to building legitimacy through procedural fairness, community courts seek to strengthen ties to the community so that residents and merchants see themselves as allies of the court and are therefore more likely to reinforce

Community Courts

social norms that the court espouses. When the whole community comes together to reinforce norms that do not accept crime and disorder, neighborhoods will become safer as crime decreases. When community courts actively involve the community in decision-making and give back to the community in some tangible way, there will be improved perceptions of the court. Community courts engage the community in four primary ways: • Visible community service: By sanctioning offenders to community service work in places where neighbors can identify them as offenders and see what they are doing, community members will better understand the court’s role in restoring the community that had been harmed. These community service projects usually involve work such as graffiti removal or trash cleanup, serving as a tangible reminder of the offender’s membership in and responsibility to the group (see Herrschaft 2012). • Giving the community voice: By regularly asking the community what their needs and concerns are – through neighborhood surveys (see Swaner 2010), advisory boards, and community meetings – community courts are better able to offer programs and services that are relevant and address the desires of local residents and merchants. • Collaboration: Allowing community-based organizations to have space to run programming in the community court building, supporting community events organized by local organizations, and partnering with local groups to offer comprehensive programming and co-sponsoring events helps develop a strong neighborhood identity as well as community-based informal social controls that research links to crime reduction. • Voluntary programming: Community courts offer voluntary programs for all persons living within the court’s jurisdiction – regardless of whether or not they have a court case. These programs can include after-school programs for local youth, job training, educational classes, an open computer lab, health and mental health services, and mediation sessions. Offering

411

C

these programs helps the court to be seen as an asset to all members of the community, not just a place for those who commit crimes. Deterrence In 1982, Kelling and Wilson’s now famous “broken windows” theory was introduced to the criminology world, and it is the theoretical underpinning of many community courts. The basic premise behind this theory is that when the physical and visible conditions of a neighborhood – graffiti, illegal dumping, public intoxication – are not controlled and monitored, it produces more social disorder, and, eventually, these neighborhood problems escalate to more severe crimes. Community courts focus on trying to fix the problems when they are small so that further petty crime and antisocial behavior will be deterred and, hence, will not lead to more severe crimes. In this way, community courts tend to focus on cleaning up minor “quality-oflife” crimes under the assumption that this will lead to reductions in other types of crime as well. Focusing on “broken windows” also helps to reinforce that a particular community does not accept crime. Looking at the surrounding environment – whether it is clean or dirty – gives cues to individuals as to the social norms of the community. If it is a place that has broken windows and other visible signs of disorder, this is a signal to potential lawbreakers that any crime they commit is likely to be overlooked. If, on the other hand, it is a place that does not have broken windows, this is a signal that the community does not accept crime, and there would be certain punishment for committing a criminal act, serving as a deterrence to violating social norms. This means that in order to design an effective deterrent, community courts must consider the certainty of punishment. Research indicates that the certainty of punishment is the most important element in the potential criminal’s decision whether or not to commit a crime (Nagin and Pogarsky 2001). Criminal behavior is not deterred if lawbreakers do not think they will be caught or given meaningful punishment if they are. Accordingly, community courts replace sentences that lack ongoing obligations of any

C

C

412

kind with meaningful sanctions for even the most minor of offenses. They have an intense focus on misdemeanor offenses that create visible signs of disorder and also focus on sanctioning offenders to community service that repairs those conditions of disorder. Intervention The intervention theory of change assumes that sanctioning offenders to social service interventions will help them address the underlying problems that may have caused them to commit crimes (or have other legal problems) in the first place. For example, drug offenders may have an underlying addiction that causes them to offend. Assigning them to a drug treatment modality such as outpatient/detox or residential treatment may help them overcome the addiction that led them to commit a crime. Other offenses such as public intoxication, driving under the influence, and drug possession also frequently result from addiction to alcohol or drugs. Providing people with appropriate interventions such as short- or long-term treatment, anger management workshops, job training programs, or GED programs may help them break cycles of recurrent criminal behavior caused by social dislocations such as drug addiction, lack of employment opportunities, and educational disadvantages. The same principles animate community court practices when handling non-criminal cases. For instance, landlord-tenant disputes that escalate to a case in housing court may reflect the tenant’s need for employment assistance or for help accessing public benefits. Thus, community courts that handle housing matters commonly provide tenants with resources and referrals that may help them meet their payment obligations.

State of Art Due in part to the recent proliferation of community courts, there is scant research literature on their impact. Comprehensive impact evaluations have been carried out in only Midtown, Manhattan (Hakuta et al. 2008; Sviridoff et al. 2000a, b, 2001); Hennepin County, MN (Eckberg 2001;

Community Courts

Weidner and Davis 2000); Philadelphia, PA (Cheesman et al. 2009, 2010); Yarra, Australia (Ross et al. 2009); and Red Hook, Brooklyn (Lee et al. 2013). Methodologies applied in these sites included process evaluations; quasi-experimental studies testing impacts on case processing, outcomes, offender compliance, and/or recidivism; cost-benefit analyses; community surveys; and ethnographic observations and interviews with local offenders. The results of these impact evaluations and other smaller-scale evaluations are summarized by key themes below. Expanded Sentencing Options Confirming the central role of diversified sentencing options, a global survey of community courts (Karafin 2008) found that 92 % of community courts routinely mandate defendants to community service, and 84 % mandate defendants to social services, including treatment readiness classes (64 %), individual counseling (64 %), job skills (64 %), life skills (56 %), anger management (52 %), and substance abuse treatment (48 %). Two separate evaluations of the Midtown Community Court – one focusing on its early years and the other on recent impacts – both found that the court made significantly greater use of “alternative” sentences than the centralized Manhattan court. These studies also found that Midtown made less use of jail and less use of “walks,” defined as sentences such as fines or time served that lack any ongoing obligation (Hakuta et al. 2008; Sviridoff et al. 2001). Yet, despite sentencing a lower percentage of its defendants to jail, both studies found that when Midtown did use jail, the resulting sentences were longer on average. In addition, one evaluation found that Midtown was more likely to impose meaningful jail time as a “secondary sanction” due to noncompliance with what was initially an alternative sentence (Sviridoff et al. 2001). These dynamics meant that Midtown did not ultimately produce a significant net reduction in jail days served by its defendants. Similar results were found in an evaluation of the Red Hook Community Justice Center (RHCJC) (Lee et al. 2013). Red Hook grants fewer “walks” than the downtown court.

Community Courts

Convicted RHCJC defendants are much less likely than downtown defendants to receive a conditional discharge with no real conditions other than staying out of trouble (15 % of convictions vs. 26 %) or a sentence of time served (3 % of convictions vs. 32 %). Red Hook defendants are also less likely than downtown defendants to have their cases dismissed (17 % of cases vs. 21 %) or to receive an adjournment in contemplation of dismissal (which involves a promise of dismissal within 6 months) (ACD) without community service or social service conditions attached (31 % of ACDs vs. 76 %). Of those defendants who do receive a sanction, however, RHCJC sends a larger share to community service and social service programs and a smaller proportion to jail – largely using jail as a resentence option with those defendants who are initially noncompliant. Though only 1 % of Red Hook defendants receive jail as an initial sentence, when used as a resentence, the average number of days is longer than downtown (75.1 days, compared with 3.88 days). The net effect of this difference in the use of “secondary jail” sanctions is that fewer RHCJC cases ultimately receive a jail sentence (7 % vs. 17 %), but RHCJC cases ultimately average more time sentenced to jail (4.75 vs. 3.06 days among all defendants or 82.3 vs. 39.9 days among just those defendants who receive a jail sentence of any length). Compliance with Court Conditions As compared with the local centralized courts, the Midtown and Hennepin County community courts produced significant increases in offender compliance with community service mandates, from 50 % to 75 % in Midtown (Sviridoff et al. 2000a, b, Sviridoff et al. 2001) and 29–54 % in Hennepin (Weidner and Davis 2000). At the Red Hook Community Justice Center, between 2000 and 2009, 80 % of defendants mandated to community service and 69 % of defendants assigned to social service sanctions other than long-term drug treatment successfully completed their mandates. At the Neighbourhood Justice Centre (NJC) in Yarra, successful completion of communitybased orders (CBOs) over the course of 1 year

413

C

(September 2008–August 2009) at the NJC was 75 %, compared to a 65 % successful completion rate for all CBOs statewide. For community service by offenders, the completion rate for assigned hours is marginally higher than the state average (88 % vs. 85 %), and each NJC offender performs an average of 105 h of unpaid work compared with 68 h for Victorian offenders in general (Ross et al. 2009). An evaluation of the East of the River Community Court (ERCC) in Washington, D.C., found that between 2007 and 2009, 60 % of defendants who stayed on the ERCC calendar successfully completed the ERCC program (Westat 2012). Effects on Recidivism An evaluation of the Midtown Community Court failed to detect an effect on reoffending by individual offenders but did detect a drop in prostitution and illegal vending crime in the Midtown neighborhood, perhaps due to a displacement effect (Sviridoff et al. 2001). A study of recidivism of misdemeanants convicted in the Philadelphia Community Court (Allen and Schulman 2009) was conducted on offenders who completed all of the conditions of their sentence in 2007, reporting an overall 9 % recidivism rate, with lower rates reported for offenders that were required to complete a program. Cases removed from the Philadelphia Community Court before completing all of the terms of their sentence reported a higher 17 % recidivism rate. No recidivism data on a counterfactual comparison group, however, was conducted in that study. Two evaluations, respectively of the Seattle (WA) and Liverpool (England) community courts, showed mixed results. In Seattle, there was not a significant difference in the probability of rearrest, but there was a smaller average number of rearrests among those processed in the community court than among those processed in a centralized court during a pre-implementation period (Nugent-Borakove 2009). Similarly, in Liverpool, there was not an impact on the reconviction rate, but the community court produced a small reduction in the total number of reoffenses (falling just short of statistical significance; see Jolliffe and Farrington 2009).

C

C

414

A recidivism study for the community court in Yarra showed that offenders sentenced by the Neighbourhood Justice Centre who received services at the center showed a lower rate of reoffending than a comparison sample of offenders sentenced at other courts (34 % vs. 41 %). However, this estimate was based on a relatively small sample of cases (100) and the difference between the NJC and comparison group was not statistically significant. In addition it was not possible to take into account any variations in the risk of reoffending associated with the NJC and comparison offenders (Ross et al. 2009). At the East of the River Community Court, a survival analysis showed that recidivism was significantly lower among the ERCC diversion program and treatment court defendants than among the Metropolitan Police Department 5th District defendants. However, these findings only included data from the D.C. CourtView online records system. Analyses were conducted on the recidivism rates of only those defendants that participated in an ERCC diversion program compared to a matched group of defendants arrested in the MPD 5th District. Reoffending was 60 % lower among ERCC defendants than among MPD 5th District defendants 360 days after the case filing date. During the second follow-up period, recidivism was 42 % lower among ERCC defendants than among MPD 5th District defendants 360 days after the trigger case disposition date (Westat 2012). A quasi-experimental evaluation of the Red Hook Community Justice Center showed that over a 1 year period, RHCJC defendants appeared less likely to be rearrested than the downtown sample (28 % vs. 31 %) and were also arrested fewer times on average (0.57 vs. 0.66). Over a 2 year period, differences between the RHCJC and the downtown court were somewhat greater. RHCJC defendants were significantly less likely than downtown defendants to be rearrested (36 % vs. 40 %) and averaged significantly fewer rearrests (0.95 vs. 1.16). An additional survival analysis demonstrated that the lower rate of recidivism for RHCJC defendants persisted over more than a 4 year maximum tracking period (Lee et al. 2013).

Community Courts

Cost Savings Evaluations in Midtown, Hennepin, Yarra, and Red Hook included cost studies, but their findings differed. In Midtown, the evaluation identified approximately $1.3 million in annual savings based on a reduction in pre-arraignment detention time, reduced jail sentences on shoplifting cases (jail time was not reduced on other cases), and reduced prostitution arrests in the Midtown neighborhood (Sviridoff et al. 2001). The evaluation of the Neighbourhood Justice Center in Yarra also found that the court saved money; for every Australian dollar invested, the expected return was AUS$1.09–1.23. For the activities included in the NJC evaluation to return a positive net benefit, a change in reoffending behavior needs to be maintained for just over 4 years (Ross et al. 2009). The evaluation of the Red Hook Community Justice Center included a cost-efficiency analysis that found that the community court was cost-effective relative to the downtown court, but its cost-efficiency is reduced when it serves a small number of defendants (Lee et al. 2013). The Hennepin study, on the other hand, found that the community court was more expensive than regular case processing (Weidner and Davis 2000).

Possible Controversies While many of the evaluations have shown positive results, it is important to understand some of the critiques that have been made of community courts. First, community courts involve criminal justice agents and social service professionals working together in a coordinated way to best serve the offender and the community. But as Berman (2010) discussed, the guiding philosophies of criminal justice agencies and social service providers often differ in significant ways. Whereas law enforcement officials tend to promote harsher sanctions for defendants when they fail to comply, social workers, who have a better understanding of mental health and addiction, believe in giving second chances when an offender relapses. This tension plays out in other ways as well. While community court judges have greater

Community Courts

sanctions at their disposal, this also gives them control over assigning social service interventions to offenders. While judges are seen as being experts in law and the administration of justice, they may not have the social work training necessary to know which mental health interventions are most appropriate – and for how long – to meet offenders’ needs. Community courts, then, may inappropriately extend the reach of the court into people’s lives, transforming the judge into a powerful figure that transcends that of a traditional judge. Moreover, while some social service sanctions seek to address the underlying problems that caused someone to commit a crime, many times larger socioeconomic factors – not individual pathology – are what contribute to crime. While community courts may offer job training or resume-building workshops, when there is a bad economy and a lack of available jobs, such social service sanctions will not be effective prevention measures. Another critique is that one of the underlying program theories of community courts – “broken windows” – itself is flawed. This theory links disorder and crime prevention by using coercive social control to manage social disorder. However, just because disorder and crime are correlated, it does not mean that one causes the other. Moreover, research has shown that the link between the two is very weak. Broken window policing policy, then, tends to unfairly criminalize the poor, minorities, and homeless, and, as Malkin (2003) points out, “quality of life policing is not necessarily what residents envision as a response to their quality of life concerns” (1585). Finally, money may be a source of controversy. Community courts may be competing with community-based social service agencies and other local organizations for limited public and private resources. Given their potentially higher level of political and cultural capital, community courts may be taking funding away from community groups who may be better suited to efficiently and effectively run the programming (see Malkin 2003). Community courts also require an initial influx of money to start-up – money to open a new building or renovate an existing one, hire new staff, and obtain necessary technology. Additionally, there are significant

415

C

costs to run the social service programs that are essential to community courts. Critics might suggest that most of the key elements of community courts could be done more efficiently and with lower case management, staffing, and program costs within a centralized courthouse than a new, separate court.

Related Entries ▶ Drug Courts ▶ Mental Health Courts ▶ Problem-Solving Courts ▶ Reentry Courts ▶ Therapeutic Jurisprudence

Recommended Reading and References Allen S, Schulman S (2009) Recidivism of Philadelphia community court misdemeanor cases closed in 2007. Personal communication with community court coordinator, Bill Babcock Berman G (2010) Principles of community justice: a guide for community court planners. Center for Court Innovation, New York Berman G, Feinblatt J (2005) Good courts. The New Press, New York Cheesman F, Durkin M, Maggard S, Rottman D, Sohoni T, Rubio D (2009) Process evaluation of the Philadelphia community court. National Center for State Courts, Williamsburg Cheesman F, Rottman D, Gibson S, Durkin M, Maggard S, Sohoni T, Rubio D, Curtis R (2010) Philadelphia community court evaluation final report: outcome/impact analysis and update on process evaluation. National Center for State Courts, Williamsburg Eckberg D (2001) Hennepin county community justice project: summary report of short-term evaluation. Hennepin County District Court Research Department, Hennepin County Hakuta J, Soroushian V, Kralstein D (2008) Testing the impact of the midtown community court: updating outcomes a decade later. Center for Court Innovation, New York Henry K, Kralstein D (2011) Community courts: the research literature. Center for Court Innovation, New York Herrschaft BA (2012) New York City community cleanup: the impact of a program for low-level offenders. Center for Court Innovation, New York Jolliffe D, Farrington DP (2009) The effects on offending of the community justice initiatives in Liverpool and Salford. London. Available on request from the Ministry of Justice

C

C

416

Karafin DL (2008) Community courts across the globe: a survey of goals, performance measures and operations. Center for Court Innovation, New York. Report submitted to the Open Society Foundation for South Africa Kelling GL, Wilson JQ (1982) Broken windows: the police and neighborhood safety. The Atlantic March:29–38 Lee C, Cheesman F, Rottman D, Swaner R, Lambson SH, Rempel M, Curtis R (2013) A comprehensive evaluation of the red hook community justice center. National Center for State Courts, Williamsburg Lind E, Tyler T (1998) The social psychology of procedural justice. New York: Plenum Llewellyn-Thomas S, Prior G (2007) North Liverpool community justice centre: survey of local residents. Ministry of Justice, London. Ministry of Justice Research Series 13/7 LaFree G (1998) Losing legitimacy: street crime and the decline of social institutions in America. Westview, Boulder Malkin V (2003) Community courts and the process of accountability: consensus and conflict at the red hook community justice center. A Crim Law Rev 40:1573–1593 Mann CR (1993) Unequal justice: a question of color. Indiana University Press, Bloomington Nagin DS, Pogarsky G (2001) Integrating celerity, impulsivity, and extralegal sanction threats into a model of general deterrence: theory and evidence. Criminology 39(4):865–892 Nugent-Borakove E (2009) Seattle municipal community court: outcome evaluation final report. Justice Management Institute, Denver Rawls J (1971) A theory of justice. Harvard Press, Cambridge Ross S, Halsey M, Bamford D, Cameron N, and King A (2009) Evaluation of the Neighbourhood Justice Centre, City of Yarra. Final report. University of Melbourne, Melbourne Rottman D, Casey P (1999) Therapeutic jurisprudence and the emergence of problem-solving courts. Nat Inst Justice J 240:12–19. Reprinted in Palacios W et al. (eds) (2002) Crime and justice in America: present realities and future prospects. Prentice-Hall Russell K (1998) The color of crime: racial hoaxes, white fear, black protectionism, police harassment, and other macro-aggressions. New York University Press, New York Sviridoff M, Rottman D, Ostrom B, Curtis R (2000) Dispensing justice locally: the implementation and effects of the midtown community court. Center for Court Innovation, New York Sviridoff M, Rottman D, Weidner R, Cheesman F, Curtis R, Hansen R, Ostrom B (2001) Dispensing justice locally: the impacts, cost and benefits of the midtown community court. Center for Court Innovation, New York

Community Organization Swaner R (2010) Community perceptions of red hook, Brooklyn: views of quality of life, safety, and services. Center for Court Innovation, New York Tyler T (1990) Why people obey the law. Yale University Press, New Haven Tyler T (2001) Public trust and confidence in legal authorities: what do majority and minority group members want from the law and legal institutions? Behav Sci Law 19:215–235 Tyler T (2004) Procedural justice. In: Sarat A (ed) The Blackwell companion to law and society. Blackwell, Malden, pp 435–452 Tyler T, Huo YJ (2002) Trust in the law. Russell Sage, New York Weidner R, Davis C (2000) Benefits and costs of the Hennepin county community court – a preliminary analysis. Institute on Criminal Justice, University of Minnesota Law School, Minneapolis Westat (2012) East of the river community court evaluation, final report. Westat, Rockville Wissler R (1995) Mediation and adjudication in small claims court. Law Soc Rev 29:323–358

Community Organization ▶ Neighborhood Effects and Social Networks

Community Policing Jan Terpstra1, Bas van Stokkom2,3 and Kees van der Vijver4 1 Institute of Criminology and Criminal Justice, Radboud University Nijmegen, Nijmegen, The Netherlands 2 Radboud University Nijmegen, Nijmegen, The Netherlands 3 VU University Amsterdam, Amsterdam, The Netherlands 4 University of Twente, Enschede, The Netherlands

Synonyms Community-oriented policing; Neighborhood policing; Personalized policing; Proximity policing

Community Policing

Overview This entry deals with the characteristics of community policing, discusses its varieties and its problems and possibilities. Subsequently, the following aspects are examined: historical backgrounds, related police models (such as problem-oriented policing, disorder policing and reassurance policing), the Chicago Alternative Police Strategy, implementation, governance and accountability, and the effectiveness of community policing.

Introduction Community (or community-oriented) policing has been a very popular term during the past few decades in many parts of the world. It has even been described as the new organizing paradigm in public policing, a revolution in police work, or even a new orthodoxy (Bayley and Shearing 1996). However, despite its popularity, community policing has a multitude of meanings (Rosenbaum 1998; Ponsaers 2001). This has given some authors occasion to describe the concept as vague or as nebulous, one that seems to justify more or less any police program or activity. Brogden and Nijhar (2005) describe community policing as a buzzword that comes in all shapes and sizes. Three factors are relevant to understanding why it proves so difficult to have an unequivocal definition of community policing, one that is generally recognized and accepted. First, the vagueness of the concept is probably one of the factors that contributes to its popularity. As Rosenbaum (1998) said, community policing seems to be something that everyone can identify with. Who would be against “community”? Secondly, in practice, the activities known as community policing do not rest on one single model of policing, but on a multitude of them (Ponsaers 2001). Finally, it is often assumed that community policing does not represent a more or less coherent or identifiable set of practices and activities; rather, it is a more general organizational strategy for the police.

417

C

In line with this, it is seen more as a sort of police philosophy, which cannot be reduced to certain specific activities. In Skogan’s (2006) terminology, community policing is more a process than a product. Nevertheless, there have been many attempts to define community policing. One of the best known is the one by Trojanowicz and Bucqueroux (1990): “Community policing is both a philosophy and organizational strategy to allow community residents and police to work together in new ways to solve problems of crime, physical and social disorder and neighborhood decay.” In addition to this definition, there have been many other endeavors to describe what community policing is and what should be seen as its main ingredients. Many of these broader definitions, however, lack a certain coherence, for example, because they not only concern police strategies, but also the organization of the police force. For instance, both elements can be found in the well-known definition by Skogan and Hartnett (1997), who make a distinction between five central elements of community policing: organizational decentralization, a two-way communication between the police and the public, a broadly focused, problem-oriented policing strategy, responsiveness to citizens’ demands, and a commitment to helping neighborhoods solve crime problems on their own. Another factor that may contribute to a lack of coherence is that definitions of community policing may also contain elements of other police models. An example of this is the definition of community policing given by Bayley (1994), who speaks of CAMPS: consultation, adaptation, mobilization, and problem solving. However, the latter element can better be seen as belonging to the problem-oriented police model (see also below). If elements of other police models and organizational features are left aside, there remain five elements of community policing that are often referred to in the international literature. Despite all kinds of varieties in the organization and practice of community policing, there seems to be a general consensus about the

C

C

418

relevance of these elements: bridging the distance between the police and citizens; a focus of the police on a broad range of problems in the neighborhood (not only crime, but also social disorder and feelings of insecurity); a preventive approach in addition to a reactive one; cooperation with other agencies in the prevention and control of crime, disorder, and other neighborhood problems; and the promotion of citizen involvement. The question of whether or not these five elements are achieved in practice is a matter for empirical research.

Backgrounds Community policing started in the United States in the 1970s as a reaction to the failures of the dominant police styles of those days. Kelling and Moore (1988) divide the history of American policing into three different eras. First there is what they call the Political Era, which continued until the early 1900s. In those years, American police forces gained their authority from local political leaders and were adjuncts to local political machines. The police provided a wide range of services to citizens. The primary tactic of the police was the foot patrol. Although the police were organized at a local level, their organizations were highly centralized and quasi-military. This political strategy had significant weaknesses, such as the rise of police corruption, discrimination against strangers, and relative inefficiency. The second era is called the Reform Era, which started in the early decades of the twentieth century with the work of A. Volmer and later on O.W. Wilson, who wanted to reform the police organization and strategy. These reformers rejected politics as the basis of police legitimacy. Instead, the law and police professionalism were presented as the new basis for the legitimacy of the police. As a result, police departments became relatively autonomous organizations. The reform policy narrowed the main function of the police to law enforcement and control of crime. The police officers’ work became

Community Policing

standardized and uniform. The relation between the police and citizens came primarily to be defined in neutral and remote professional terms. Citizens were seen as passive recipients of professional crime control services. Preventive patrol by car and rapid response became the main strategies of this reformed police force. In the 1960s and 1970s, it became clear that this professional police model (as it was called, although it was not at all very professional, as Stone and Travis (2011) have argued), had serious shortcomings. Several factors contributed to this: the rising level of crime and fear of crime, complaints – especially from members of ethnic minorities – about the way they were treated by the police, and the new critical awareness as a result of civil rights and antiwar movements. Moreover, scientific research revealed that many of the central strategies of this professional police force (such as preventive patrol by car or standardized criminal investigation) proved to be highly ineffective. The third era, starting in the 1970s, may be seen as a reaction to the failures of the standard police model. In this Community Policing Era (Kelling and Moore call it a Problem-Solving Era), there is a renewed emphasis on close relations with the community and the citizens. The definition of the police function broadens into a community strategy. According to Kelling and Moore (1988), organizational decentralization is inherent to community policing. The involvement of police officers in the diagnosis of and response to community problems makes it necessary to locate the power of operational and tactical decision making at lower levels within the organization. Community policing requires intimate relations with citizens and instead of a claimed monopolistic responsibility for crime control, there should be strategic alliances with neighborhood and community groups. Community policing is based on a wide variety of strategies, such as foot patrol (rediscovered as an important instrument for regaining citizens’ trust and promoting feelings of security), information gathering, problem solving, victim services, and others.

Community Policing

Although this analysis by Kelling and Moore deals with developments in the United States, there are certainly similarities with the background to the rise of community policing elsewhere. The failures of the standard police model, a loss of police legitimacy and citizens’ trust, and higher levels of crime and feelings of insecurity contributed to the adoption of this new police model, in addition to the wish of police organizations elsewhere to imitate and follow the American police models, which enjoyed a high reputation. The historical backgrounds to community policing are even more complex than this standard view of the history of American policing reveals. The shift from a standard police force to community policing is also related to changes in the prevailing views on democracy and the relation between the state and society. Moreover, in some cases, especially in the United States, the model of community policing may be related to more or less romantic notions of community (Herbert 2006). This seems to be of less relevance in some European countries, for example, where community or neighborhood is more often viewed as relatively neutral scales for the organization of the police and police-citizen relations.

Related Police Models For the past few decades, community policing has become a dominant policing perspective in many countries. Other policing models were introduced, which are often seen as related or even integral to this community policing perspective. These models may also be viewed as responses to the failure of the (so-called professional) standard police model, but they diverge in their strategies and priorities. Nevertheless, such models are conceptually close to community policing and often they are assumed to be innovations in community policing. Three such models are briefly discussed here: problemoriented policing (Goldstein 1990), disorder policing (or broken windows policing)

419

C

(Wilson and Kelling 1982), and reassurance policing (Innes 2004). Problem-Oriented Policing The concept of problem-oriented policing was introduced by Goldstein (1990). In his view, much police work suffers from what he calls a means-over-ends-syndrome: police work is often defined more in terms of the “means” of police work (such as rule enforcement, surveillance, or rapid response) than by its goals. According to Goldstein, the work the police do requires them to deal with a wide range of behavioral and social problems that arise in the community. Often police work mainly consists of ad hoc responses to a seemingly endless stream of isolated and recurring incidents. A problemoriented approach is meant to make an end to this unproductive process. The police should conduct a detailed, systematic analysis of the causes and factors underlying a series of comparable events and incidents. Interventions by the police should be based on such detailed analyses and will differ for each problem (Goldstein 1990). This should contribute to a more preventive form of policing. Problem-solving strategies should not only be based on the traditional instruments and resources available to the police, but also on those of other agencies. In this respect, problem-oriented policing is a form of partnership or multiagency policing. Several reviews of studies that evaluate police work have concluded that community policing may only be effective if it is applied in combination with a problem-oriented approach (Eck 2004; Skogan and Frydl 2004). In spite of this, there are also important differences between problem-oriented and community policing. First, the relevance of a problem-oriented approach is not confined to community policing (Tilley 2003). Moreover, one of the main goals of community policing is to contribute to an improvement in or restoration of the relations between the police and citizens. Insofar as citizens are involved in a problem-oriented

C

C

420

approach, there is only a functional relation between the police and the citizens, who primarily are seen as the “eyes and ears” of the police (Tilley 2003). Disorder Policing According to the model of disorder policing, the police should focus on the restoration of physical and social order in neighborhoods. In this way, the police might reduce feelings of insecurity among citizens, enhance their trust in the police, and prevent and counter crime. This theory achieved recognition, thanks to a classical article by Wilson and Kelling published in 1982, entitled “Broken Windows.” The broken windows theory assumes that forms of physical and social disorder (such as incivility, physical deterioration, or disorderly social conduct), especially if they accumulate and persist for a relatively long time, will result in feelings of insecurity and the breakdown of informal social control in the community. Wilson and Kelling (1982) explain that a broken window acts as a signal to potential offenders that people living in this area are indifferent to what is happening in their neighborhood. As a consequence, one broken window (which is not speedily repaired) will result in more broken windows and more serious forms of crime in the neighborhood. Instead of reducing the police to playing the role of the criminal justice system’s “front end,” keeping the peace and maintaining social order should be restored as the force’s main tasks (Kelling and Coles 1996). The main goal for the police should be to promote informal social control. Many scholars view broken windows policing and community policing as opposites. Broken windows policing is identified as an aggressive approach to minor offenses and disturbances. In that case, community-oriented policing is associated with a more “gentle, friendly, concerned, or low-key method.” According to a second view, disorder policing is a specific instantiation of community policing. According to this line of thought, the broken windows theory is considered a source of

Community Policing

inspiration for the well-known form of community policing that was developed in Chicago in the 1990s. Reassurance Policing In the last decade, a new model of policing has come to the fore, called reassurance policing. A central assumption in this model is that the presence of visible, uniformed authorities will reassure citizens, reduce feelings of insecurity, and restore trust in the police. This model was presented in the UK in reaction to the fact that since the mid-1990s, the objective level of crime had improved, while many citizens still perceived public safety as a serious, even worsening problem (Tuffin et al. 2006). According to this model, the police should focus on problems that have the greatest impact on citizens’ feelings of security (“signal crimes”). According to Innes (2004), visible and troublesome incidents are occupying the “collective memory” of residents; they function as warning signs for future threats. The police should develop “control signals,” actions that communicate a sense of regained peace and order. These signals convince citizens that police officers are concerned about local security. Although this model places great emphasis on the importance of visibility, accessibility, and proximity of the police, these conditions, although perhaps necessary, are not seen as sufficient to promote feelings of security and trust in the police. Evaluations of reassurance policing in the UK show that, in addition to police visibility, other important factors are problem-solving strategies, citizen involvement, and the promotion of citizen initiatives (Tuffin et al. 2006). Extending the notion of reassurance with problem-solving strategies, cooperation with other agencies, and the focus on “signal crimes” makes this theory quite complex.

The Chicago Experiment: CAPS Disorder policing takes the “community needs” of residents as its starting point. In Wilson and

Community Policing

Kelling’s terms, the neighborhood indicates what the “appropriate level of public order” should be. The police should make use of the “preventive capital” of citizens. In Disorder and Decline (1990), Wesley Skogan reflected on the problematic nature of determining order. In heterogeneous neighborhoods, neither the police nor citizen-organizations can claim that their vision of order is authoritative and liable to impose compliance. In these conditions, Skogan says, it is a challenge to involve citizens and other relevant parties explicitly in determining and selecting disorder problems that should be tackled. Order is negotiated rather than imposed. In this way, policing becomes the object of an open, deliberative process. Actually this view formed the starting point of an ambitious, intensive police innovation program, the Chicago Alternative Police Strategy (CAPS). This program “reinvented” some of the core features of earlier community policing approaches (prevention, cooperation, problem solving, etc.), but added some “new” aspects: encouraging residents and other stakeholders to find solutions for local security problems and giving priority to problems of disorder. Since 1993, CAPS has been implemented in all of Chicago’s 279 police beats. All beats have monthly meetings attended by residents, police officers, and other professionals, the aim being to identify and prioritize local problems together, and to develop plans for dealing with them. The police explicitly accepted the broken windows philosophy, and consequently also the expansion of the police mandate. Nevertheless, CAPS was launched by the city administration. The most difficult task was to recruit the police force’s support for the program (Skogan 2006). The innovative element of CAPS lies not so much in its methods of citizen participation, but rather in the rigorous, long-term organization and political will to establish and maintain partnerships in every neighborhood. Evaluation studies (Skogan and Hartnett 1997; Skogan 2006) show that the monthly beat meetings are attended by an average of 20 citizens. CAPS attracts residents who had not participated earlier in other programs. Increased

421

C

participation of women, blacks, and poor citizens was noted. A remarkable finding is that participation in black neighborhoods is just as high as in white ones, even in high crime districts where the police were unpopular. Between 1994 and 2003, confidence in the police increased considerably. Citizens’ views of police effectiveness, responsiveness and demeanor within the “three Chicago’s” (white, black, and Latino) improved substantially (Skogan 2006). After studying Chicago beat meetings, Fung (2004) pointed out several problems. Some beats failed to address priority problems. Police resources were often allocated inequitably. The “natural” course of participation led to “conflictual paralysis” in one beat and domination by wealthy and well-educated residents in another. Often, energetic facilitators succeeded in breaking through this “laissez faire, first come, first served” style, managed to put problems of under-represented subgroups on the agenda, and called in intervention teams to take action against drug houses (for example). Thus, beat meetings need powerful facilitators. Fung argues that even in neighborhoods that lack resources or have serious internal conflicts, citizens gain more from the new deliberative arrangements than from the bureaucratic organizational forms that existed prior to the introduction of CAPS. In a case study of a neighborhood watch group in Beltway, a suburban district in Chicago, Carr (2005) shows that citizens received support from the police and other local public services only after CAPS had been introduced. As a result, small groups of active citizens were able to mobilize resources and initiate partnerships. Carr challenges the assumption that neighborhoods need strong social cohesion in order to get crime and disorder under control. Without CAPS, Carr stresses, Beltway activism would have been impossible. The main factor that drives citizen participation is trust in public professionals. Even in poor ethnic neighborhoods, consultation and partnerships may be successful. What is important is to overcome “legal cynicism” and the tendency of residents to tolerate antisocial behavior (Carr 2005).

C

C

422

Implementation A recurrent complaint about community policing concerns its implementation. In Mastrofski’s (2006) view, police work in the United States still does not meet the high ambitions of community policing, despite more than two decades of reform and the formal commitment of many police organizations to this model. In practice, there are many obstacles to full and effective implementation. According to Mastrofski, community policing in the United States did not transform the police in the way that was originally envisaged. Klockars (1988) was still more pessimistic, even in the early days of community policing. In his view, the rhetoric of community policing wraps the police in romantic, powerful, and unquestionably good aspirations, which the police, however, cannot realize in practice but only in what he calls “ersatz terms.” Although the past few decades have seen many studies on community policing, there is a shortage of detailed research on the practical implementation of this police model. Often its implementation is treated as a black box. Many studies focus on formal or organizational aspects of this form of police work, while seemingly disregarding the reality of community policing at street level. One of the main findings in the limited number of studies that have focused on the street level of community police work is that community police officers often continue to spend a great deal of time doing traditional police tasks. This may be the result of several factors, such as the prevailing police culture (Greene 2000) or limited expertise (Eck 2004). Police leaders may be rather cynical about community policing or may fear that the increased discretionary room allowed to community officers may have adverse implications for their own position. However, the most serious problems in the implementation of community policing may arise in improving relations between the police and citizens, especially in those urban areas that most need community involvement.

Community Policing

Studies in other countries reported comparable results. For example, in Denmark, Holmberg (2002) found that in practice, many of the community policing principles are often embraced only very partially or in unintended ways. The main impediments in this country included large working areas, the burden of a lot of paperwork, and the priority of other tasks. In the Netherlands, the five main ambitions of community policing (proximity, focus on a wide range of problems in the neighborhood, preventive strategies, cooperation with other agencies in the control of crime and disorder, and promotion of citizen involvement) were only carried through to a limited extent (Terpstra 2010). In contrast to the United States, these problems were not so much caused by a (supposed) police culture nor by romantic notions of community (Herbert 2006). More important was the fact that the concept of community policing still has a vague professional image, even among Dutch community police officers themselves. Moreover, community police work is often rather open and indefinite in nature. Although community policing usually has high ambitions, the officers involved often have only limited resources, skills, and expertise to achieve them. In addition, the support they receive from their management, superiors, and colleagues is often quite meager, and sometimes contradictory.

Governance and Accountability A common element of community policing initiatives is attempting to bridge the distance between the police and citizens. It is often assumed that community policing may encourage local governance and accountability of the police (Skogan and Hartnett 1997; Somerville 2009). A main argument underlying CAPS was that it would enable citizens to provide local knowledge and resources to the police and that they would be better able to monitor police officers and hold them accountable. Following this line of argument, community policing may even be presented as a form of democratization.

Community Policing

However, community policing is not always seen as beneficial to police accountability. Organizational decentralization, a growing intimacy between police officers and citizens, and the increased use of discretion by police officers may also be viewed as potential threats to police accountability (Brogden and Nijhar 2005; Herbert 2009). In fact, community police officers are confronted with very divergent forms of accountability, reflecting hierarchical, egalitarian, and more individualist cultures. This complexity results from the fact that a multiplicity of actors, each with their own views and interests, have a stake in the implementation of community policing. The (local) government, (internal) superiors, colleagues, external partners and (representatives of) the public try to “govern” the community officers and hold them accountable. As a result, these officers are not only held accountable within the traditional hierarchical relations, but also horizontally and vertically downward by partners or citizens (Somerville 2009). In countries where the police organization is not strictly local (which is generally the case outside the United States), it is useful to distinguish between nonlocal and local forms of governance and accountability in community policing. Nonlocally initiated forms are mainly bureaucratic and managerial, representing hierarchical and individualistic cultures, respectively. In addition, community police officers may not only be accountable to the local government and politicians, but also to other police professionals, partners, and (groups of) citizens. These latter local forms of governance and accountability are usually more egalitarian (although there may also be some hierarchical elements). The traditional culture of command-and-control is hard to align with the complexity of community policing, the need for a flexible, tailor-made approach, and the need for officer discretion. In the 1990s, new managerial forms of governance and accountability were introduced in many police organizations, such as systems of performance management with quantitative targets. These forms do not fit in with the

423

C

localized nature of community policing work. They often create unintended consequences that are difficult to combine with community policing, such as a standardization of the work or a fixation on those activities that can more easily be quantified and measured. Professional types of governance and accountability are often presented as a solution to the problems associated with hierarchical and managerial forms. Here the professional discretionary power of the community police officer is highly valued. Community police officers are not only asked to follow rules and procedures, but also to use their autonomy in a professional way and to show their moral involvement with local problem solving. However, these professional forms of governance and accountability are scarcely ever realized in practice, despite being what many community police officers would prefer in their relation with their superiors. For the past few decades, many countries have witnessed a growing interest in citizen participation in relation to policing and public safety. However, citizen participation scarcely contributes to local police governance/accountability. Control over the police by partners or citizens may be perceived as conflicting with the views, routines, and priorities of the police (Terpstra 2011). In the United States, Herbert (2006) interpreted this as the result both of a police culture that emphasizes police self-protection combined with lofty and unrealistic expectations about “community” and local democracy. Comparable problems were reported in the Netherlands, where more modest and pragmatic meanings of community and the relation between community policing and local democracy prevail. Here the minimization of citizen influence on the police was used by community police officers to reconcile conflicting demands and expectations about their work.

Effectiveness The issue of effectiveness of community policing raises several important questions. First, what

C

C

424

should be meant by the effects of community policing? Generally it is assumed that an evaluation of the impact of community policing should not only focus on the question of whether it results in a reduction of crime and social disorder, but also whether it contributes to feelings of security and to greater levels of citizens’ trust in the police. Secondly, how are the (potential) effects of community policing to be measured? Studies evaluating the impact of complex programs like community policing require a longitudinal and quasi-experimental research design. An adequate interpretation of the results of such an impact evaluation demands in addition that a detailed process or implementation evaluation should be used. If there is a shortage of reliable information about the implementation of the intervention or program, the effects or impact will be hard to interpret. The internationally best known evaluation of community policing is that of the Chicago Alternative Policing Strategy (CAPS) done by Skogan and his colleagues. This study shows that opinions about the police in Chicago steadily improved in the years that CAPS was implemented. An increasing number of citizens feel that the police in their area are dealing with problems that concern residents and that the police are prepared to cooperate with residents in solving these problems. The increasing visibility of the police (partly a result of more frequent foot patrols) improved citizens’ evaluation of the police. Moreover, many residents feel less insecure, partially because they see the police more often. Fear of crime declined substantially among the groups experiencing the greatest levels of fear. In Chicago, many forms of crime (both violent crime and property crime) declined sharply. However, in general, the fall was comparable to trends in other American cities. The evaluation of CAPS also showed that residents in certain areas reported sharp improvements in neighborhood conditions. African-American residents in particular reported improvements in relation to, for instance, abandoned buildings or graffiti, and different forms of social

Community Policing

disorder. In summary, community policing according to the CAPS model had a positive impact on citizens’ opinions about the police, feelings of security, and neighborhood problems and disorder (Skogan and Hartnett 1997; Skogan 2006). Since the late 1990s, especially in the United States, there have been several metaevaluations of the effectiveness of community policing, often comparing it with other police models and strategies. Sherman et al. (1997) showed that community policing which does not explicitly focus on specific risk factors, generally, has no impact or only a minor one on the level of crime. According to this review, the effectiveness of community policing may be improved if it is supplemented with a problem-solving approach or if residents are involved in the assessment of police priorities in their area. Skogan and Frydl (2004) concluded that in general, the effectiveness of community policing is hard to determine because this concept is used for a wide variety of police practices and strategies. They found no evidence that methods that are often associated with community policing, like neighborhood watch, general foot patrol, storefront offices, or community meetings, contribute to a reduction of crime. However, community policing may have a positive impact on neighborhood problems. Improved relations between the police and residents may also diminish fear of crime and may cause residents to perceive less social and physical disorder in their neighborhood. According to Skogan and Frydl (2004) community policing may have a greater impact on the reduction of crime if it is supplemented with elements of two other police models: hot spot policing and problem-oriented policing. Although available research gives important indications of the potential impact of community policing, there are still major questions left. Community policing is an internationally heterogeneous concept and practice. It is often unclear if conclusions from mainly American studies may be generalized to other jurisdictions. There are also indications that the effectiveness

Community Policing

of community policing may differ between different types of neighborhoods and for different categories of citizens.

Future Direction The conclusions of studies on the implementation and effects of community policing have raised serious questions about the value and practical feasibility of the community policing model. They have tempered naı¨ve and optimistic ambitions that were sometimes connected to the model (Herbert 2006). In some countries, such as Norway, community policing never gained a dominant position. The reform of policing in that country for the past two decades was more strongly motivated by a problem-oriented approach (Larsson 2010). In addition, in some European countries like the Netherlands, both an increasing emphasis on managerial discourse and a shift to a more repressive and punitive political climate meant that community policing lost much of its dominance (although it remains an important part of police work) (Terpstra 2011). Moreover, since the introduction of this model, police organizations were confronted with other fundamental changes that did not fit neatly within the framework of community policing (such as cybercrime). As a result, community policing lost some of its attraction as an overarching framework for the police (Stone and Travis 2011). Despite these developments, one might expect that community policing is here to stay, although perhaps in a more pragmatic, modest way. The fundamental notion was and remains that for the promotion of a legitimate, fair, and effective police, it is essential to close the distance between citizens and the police and to create different forms of cooperation between them at the neighborhood level. It may be expected that as a result, community policing will continue to be an important element in the future, although adapted to very different and changing circumstances.

425

C

Related Entries ▶ Broken Windows Thesis ▶ Communities and the Police ▶ Democratic Policing ▶ Impacts of Community-Oriented Policing ▶ Law of Community Policing and Public Order Policing ▶ National and Local Policing ▶ Order Maintenance Policing ▶ Problem-Oriented Policing ▶ Reassurance Policing and Signal Crimes ▶ Theories on Policing and Communities

References Bayley DH (1994) Police for the future. Oxford University Press, New York/Oxford Bayley DH, Shearing CD (1996) The future of policing. Law Soc Rev 30(3):585–606 Brogden M, Nijhar P (2005) Community policing. National and international models and approaches. Willan, Cullompton Carr PJ (2005) Clean streets: crime, disorder and social control in a Chicago neighborhood. NYU Press, New York Eck JE (2004) Why don’t problems get solved? In: Skogan WG (ed) Community policing. Can it work? Wadsworth, Belmont, pp 185–206 Fung A (2004) Empowered participation: reinventing urban democracy. Princeton University Press, Princeton Goldstein H (1990) Problem-oriented policing. McGraw-Hill, New York Greene JR (2000) Community policing in America: changing the nature, structure, and the function of the police. In Horney J (ed) Policies, processes, and decisions of the criminal justice system. Criminal justice, vol 3. U.S. Department of Justice, Washington, DC, pp 299–370 Herbert S (2006) Citizens, cops, and power. The University of Chicago Press, Chicago Herbert S (2009) Community policing and accountability. In: Grabosky P (ed) Community policing and peacekeeping. CRC Press, Boca Raton Holmberg L (2002) Personalized policing. Results from a series of experiments with proximity policing in Denmark. Policing: Int J Police Strat Manag 25(1):32–47 Innes M (2004) Signal crimes and signal disorders: notes on deviance as communicative action. Br J Sociol 55(3):335–355 Kelling GL, Coles CM (1996) Fixing broken windows, restoring order and reducing crime in our communities. The Free Press, New York

C

C

426

Kelling GL, Moore MH (1988) The evolving strategy of policing. National Institute of Justice, Washington, DC Klockars CB (1988) The rhetoric of community policing. In: Greene JR, Mastrofski SD (eds) Community policing: rhetoric or reality. Praeger, New York, pp 239–258 Larsson P (2010) Ideology as a cover up: community policing in Norway. J Pol Stud 3(16):233–241 Mastrofski SD (2006) Community policing; a sceptical view. In Weisburd D, Braga AA (eds) Police innovation. Contrasting perspectives. Cambridge University Press, Cambridge, pp 44–73 Ponsaers P (2001) Reading about “community (oriented) policing” and police models. Policing: Int J Police Strat Manag 24(4):470–496 Rosenbaum DP (1998) The changing role of the police: assessing the current transition to community policing. In Brodeur JP (ed) How to recognize good policing. Problems and issues. Sage, Thousand Oaks, pp 3–29 Sherman LW et al (1997) Preventing crime: what works, what doesn’t, what’s promising: a report to the United States Congress. National Institute of Justice, Washington, DC Skogan WG (1990) Disorder and decline: crime and the spiral of decay in American cities. The Free Press, New York Skogan WG (2006) Police and community in Chicago. A tale of three cities. Oxford University Press, Oxford/ New York Skogan WG, Frydl K (2004) Fairness and effectiveness in policing. The evidence. The National Academies Press, Washington, DC Skogan WG, Hartnett SM (1997) Community policing, Chicago style. Oxford University Press, New York/ Oxford Somerville P (2009) Understanding community policing. Policing: Int J Police Strat Manag 32(2):261–277 Stone C, Travis J (2011) Toward a new professionalism in policing. National Institute of Justice/Harvard Kennedy School, Boston/Washington, DC Terpstra J (2010) Community policing in practice: ambitions and realization. Pol: J Pol Pract 4(1):64–72 Terpstra J (2011) Governance and accountability in community policing. Crime, Law Social Change 55(1):87–104 Tilley N (2003) Community policing, problem-oriented policing and intelligence-led policing. In: Newburn T (ed) Handbook of policing. Willan, Cullompton, pp 311–339 Trojanowicz R, Bucqueroux B (1990) Community policing: a contemporary perspective. Anderson, Cincinnati Tuffin R, Morris J, Poole A (2006) An evaluation of the impact of the national Reassurance Policing Programme (No. 296). Home Office Research, Development and Statistics Directorate, London Wilson JQ, Kelling GL (1982) Broken windows: the police and neighborhood safety. Atlantic Monthly 243:29–38

Community Service in Europe

Community Service in Europe Frieder D€unkel1 and Tapio Lappi-Sepp€al€a2 1 Faculty of Law, University of Greifswald, Greifswald, Germany 2 National Research Institute of Legal Policy, Helsinki, Finland

Overview Community service orders have been introduced in England and Wales 1972 and since then have become one of the most important community sanctions in many European countries and outside Europe. Community service has gained importance, in particular, for offenders who are not able to pay a fine, but also to replace short-term imprisonment. It has an essential role as the principal community sanction in Spain, Hungary, England and Wales, Scotland, Latvia, and all Nordic countries. The scope of application varies considerably when looking at the range of hours that can be imposed. In Europe, one regularly finds a scope between 40 and 240 h, in the USA, between 100 and 500 h. Empirical studies have shown that the implementation of community service has been successful. In this regard, it is important that most offenders assigned to community service programs are completing their working hours. The available evidence suggests that community service is (at least) a promising alternative in terms of reducing recidivism. Several results point to significant differences between offenders who are sentenced to imprisonment and offenders who participate in community service. Recent meta-analyses demonstrate a stronger effect size (.07) than earlier ones, which may be explained by the better structured programs considering “what-works” criteria. The authors advocate to extend community service and to ensure that it comes to replace imprisonment (instead of replacing other noncustodial sanctions).


Historical Development Historically forced labor has a long tradition in human history. In penal history, it was one of the core elements of prison life. Forced prison labor is still widely accepted and exempted from the ban on forced labor as expressed in international human rights standards. First efforts to ban forced labor date back to 1930 when the International Labour Organisation (ILO) initiated the Convention No. 29 and in 1957 a second one (Convention No. 105) after ILO had been linked to the UN in 1946. Both Conventions aimed at fighting against slavery and any other form of involuntary (forced) labor (see de Jonge 1999, pp. 321 ff.). The ban on forced labor can also be found in art. 8 of the International Covenant on Civil and Political Rights (ICCPR) and in art. 4 of the European Convention for the Protection of Human Rights and Fundamental Freedoms (ECHR). Art. 8(3) (c) of the ICCPR, however, allows for forced or compulsory labor “normally required of a person who is under detention in consequence of a lawful order of a court, or of a person during conditional release from such detention.” The ECHR in the same wordings allows for forced labor only during detention and on conditional release. Strange enough to permit forced labor during the post-release stage, both Conventions do not clearly differentiate between remand and sentenced prisoners. Theoretically, one could think of forced labor in remand custody, but as a consequence of the presumption of innocence (see art. 6 ECHR), it is a common opinion and jurisprudence to exclude remand prisoners from compulsory work. The ILO Convention No. 29 had allowed forced labor only for sentenced prisoners, not for remand prisoners or those conditionally released (see de Jonge 1999, p. 327). The German Constitution as many other national laws allows forced labor only for sentenced prisoners (see art. 12(3) Federal Constitution, Grundgesetz). Therefore, from an international human rights view, it might be difficult to justify community service as an independent sanction as far as it is not strictly bound to replace a prison sentence. In that case, it can be justified with the argument

427

C

a maiore ad minus that forced labor in the community is a less intrusive sanction than imprisonment plus forced labor. If, however, community work does not replace a prison sentence, but other alternatives such as fines or conditional discharge without intervention, it can be questioned as being a violation of the ban on forced labor. The German Constitutional Court has found a way out of this dilemma by going back to the roots of the ban of forced labor in the German Constitution after World War II: The constitutional ban was implemented as a fundamental human right with regard to forced labor in the preceding Nazi-era, where thousands of people were killed by hard labor in concentration camps and elsewhere. The legislator wanted to outlaw such inhuman hard labor, which in no way is comparable with some few hours of labor, in particular if its purpose is the rehabilitation (resocialization, education) of the offender (see Bundesverfassungsgerichtsentscheidungen, BVerfGE 74, pp. 102 ff., 122 ff.). Germany did not introduce community service as an independent sanction in the Criminal Law for adults of at least 21 years of age, but only as a substitute sanction to prevent imprisonment for fine-defaulters (1983). This is different in the field of juvenile justice where the German Juvenile Justice Act (JGG) provides for community service as an independent sanction as well as a condition of diversion or of a suspended sentence (see D€unkel 2011a). The practice extensively uses community service: 44 % of all sentenced juvenile and young adult offenders (aged 14–21) in 2010 received a community service order (see D€unkel 2011a; Heinz 2012). First community service programs were introduced in the USA with female traffic offenders in Alameda County, California, in 1966, with local initiatives following in several counties throughout the USA (Wright 1991, p. 40). The aim of different initiatives was to create community service programs as a viable alternative to incarceration (van Ness 1986, p. 194). Community service as an alternative to prison in Europe first has been introduced in 1972 in England and Wales. Other European countries followed in the 1980s: Italy 1981, Denmark and

C

C

428

Portugal 1982, France 1983, the Netherlands 1985 (see Albrecht and Sch€adler 1986). With the exception of Denmark, the Scandinavian countries introduced community service only in the 1990s. Denmark started an experiment with community service – as mentioned – already in 1982. However, the sanction was not introduced in a wider scale in the Nordic sanctions systems until the 1990s. Community service appears in different forms. Finland and Norway treat community service as an independent sanction (although Norway renamed “community service” in 2001 as “community punishment”). In Denmark and Sweden, community service is attached either to conditional imprisonment or a probation order. In Finland, long (over 1 year) conditional prison sentences may be combined with a short (20–60 h) community service order. In Denmark, community service can be combined also with fines and unconditional imprisonment. In addition, community service may be attached with separate conditions concerning residence, school attendance, or work. Also Norway allows specific conditions regarding the offender’s dwelling, work, and treatment. The maximum number of community service hours varies from 200 (Finland) to 420 (Norway). The range of community service in Denmark was from 30 to 240 h, but the maximum was raised to 300 h in 2012. Countries differ also in the sense, how strictly community service has been defined as an alternative to imprisonment, and not to other sanctions. Finland has followed the strictest policy in this respect, by adopting a specific two-step procedure. First, the court is supposed to make its sentencing decision by applying the normal principles and criteria of sentencing without considering the possibility of community service. Second, if the result of this deliberation is unconditional imprisonment (and certain requirements are fulfilled), the court may transform the sentence into community service. In principle, community service may therefore be used only in cases in which the accused would otherwise have received a sentence to unconditional imprisonment. Other countries are less strict in this point. In connection of the 2001 reform in

Community Service in Europe Community Service in Europe, Table 1 Community service in Scandinavian countries 2008–2009 The Use of Community Service Sentences in Scandinavia in 2008–2009 Figures per 100,000 pop and % of court dispositions Fin % Swe % Den % Nor % Unconditional 130 158 160 219 prison (11 %) (19 %) (24 %) (45 %) Community 64 (5 %) 64 (8 %) 59 (7 %) 54 (13 %) service

Norway, the scope of community punishment was extended to be used, not only instead of imprisonment, but as a replacement penalty for juvenile offenders who had previously been sentenced to supervised conditional imprisonment. In Iceland, the decision over community service is taken by the prison administration, not the courts. This avoids effectively the risks of netwidening (but could raise questions from constitutional point of view in some other countries). In the turn of the 2000s, several Scandinavian countries completed law reforms in order to increase the use of community service. Sweden created a combination of community service and suspended sentence, thus increasing the number of annual cases from 2,000 to around 4,000. Denmark changed its policy in 2000 by allowing community service to be used also for drunken driving (which was previously forbidden). Within a period of two years, this increased the number of sentences from 1,000 to 4,000. Norway, in turn, tried to increase the credibility of community service by changing the title to community punishment by including also other elements in the sentence and by expanding the scope of application also to drunken driving. This resulted in an increase from around 500 cases to the present little over 2,500 cases (Table 1). Relative to population community service is used roughly on a same scale in all countries, with Finland and Sweden on the lead (64) and Denmark and Norway following (59 and 54). A longer trend from 1997 onwards shows that three out of four countries expanded the use of community service in the shift of the millennium, while Finland had used this alternative more extensively already from the mid-1990s (Fig. 1).

Community Service in Europe Community Service in Europe, Fig. 1 The number of court-imposed community service orders/ per 100,000 population

C

429 The number of community service order in four Nordic countries / 100 000 pop 1997–2008

90 80 70 60

C

50 40 DEN FIN NOR SWE

30 20 10

Non-European countries have introduced community service as well, in particular in the Anglo-Saxon world (Australia, New Zealand, South Africa, and USA), but also in South America, recently, community service is developed as an alternative sanction. In these countries (as in Europe already earlier), juvenile criminal law is a forerunner for reforms of the sanctions system (all juvenile justice jurisdictions dispose of community service orders, see Tiffer-Sotomayor 2000).

Legal Conditions and Basic Philosophies of Community Service Community service involves the performance of unpaid work, during leisure-time and within a given period, for the good of the community. The offender “shall pay back to the community via unpaid work” (Goldson 2008, p. 78; D€ unkel et al. 2011, p. 1673). Thus, the original philosophy of community service fits very well with the special preventive aim of punishment (resocialization or with regard to juvenile justice: education) as well as to the idea of restorative justice. In the context of restorative justice, another orientation is outlined by Wright (1991, p. 44): “The emphasis of community service is

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

0

not on punishment nor on rehabilitation; rather, it is on accountability.” It focuses “not on offenders’ needs but their strengths; not on their lack of insight but their capacity for responsibility; not on their vulnerability to social and psychological factors but their capacity to choose. These differentiate a rehabilitative response from a restorative/community service response to crime. And punitive elements of community service orders may attend its imposition, within a restorative system, only as by-products of the offender’s commitment of time and effort” (Wright 1991, p. 44). However, in the last 20 years, some countries have redefined community service and also renamed it as “punishment in the community” or “community punishment” and thus emphasizing the repressive goal of retribution. This was particularly the case in England and Wales (1998), but also in Norway (2001). Therefore, the nature and content of community service may be rather different from jurisdiction to jurisdiction. The status and contents of community service may vary. It may: • be imposed as an independent sanction or as an adjunct to another sanction • replace only prison sentences or also other penalties

C

430

Although community service is widely accepted in the field of juvenile justice, some countries explicitly exclude the very young age groups, e.g., under 16 (England and Wales) or 15 (South Africa). The range of hours (and implicitly the range of prison sentences that can be replaced) varies considerably: In Europe, the range mostly is from 40 to 240 h, often with a maximum of 120 h for juvenile offenders. Even in countries with normally great similarities in their penal policy such as the Scandinavian countries, large variations can be seen: In Finland, the maximum is 200; in Denmark, 300; in Norway, 420 h. In South Africa, the range is from 50 to 240 h (to be completed in 1 year). In New Zealand, the range with 40–400 h is considerably wider. Up to 200 h should be computed in one year, more than 200 h in two years. If offenders do not meet the requirements of the sentence, up to three months of imprisonment or a fine of up to 1,000 $ (NZ) can be imposed (see http://www.corrections.govt.nz/communityassistance/corrections-in-the-community/communitywork.html). In the USA, the range is regularly between 100 and 500 h (to be computed in one year).

International Human Rights Standards Concerning the Imposition and Execution of Community Service Human rights aspects are of major importance in the decision-making process as the principle of proportionality (Verh€ altnism€ aßigkeitsgrundsatz), derived from the principle of Rechtsstaat (rule of law, art. 20(3) of the Constitution), is to be considered for the imposition of community service, i.e., the number of working hours imposed must be related to the seriousness of the offense. In most countries, therefore, the maximum number of hours is fixed by law. In Germany, where community service as an independent sanction is available only in juvenile justice, no maximum is fixed by law. Therefore, the principle of proportionality is of special importance to avoid unjustified high number of several 100 h.


The Recommendation of the Council of Europe concerning Community Sanctions or Measures (CSM) of 1992 (Rec. (92)16) stipulates that “the nature, content and methods of implementation . . . shall not jeopardise the privacy or the dignity of the offenders or their families, nor lead to their harassment. . . . Safeguards shall be adopted to protect the offender from insult and improper curiosity or publicity” (Rule 23). The European Rules for Juvenile Offenders Subject to Sanctions or Measures (ERJOSSM) of 2008 (Rec. (2008)11) – as earlier Recommendations of the Council of Europe and the UN – assess that “sanctions or measures shall not humiliate or degrade the juveniles subject to them” (Rule 7). And: “Sanctions or measures shall not be implemented in a manner that aggravates their afflictive character or poses an undue risk of physical or mental harm” (Rule 8). In the commentary to these rules, reference is made to some forms of community work that can stigmatize offenders such as wearing special uniforms which identify them as offenders and which would not be consistent with the rules (Council of Europe 2009, p. 37). The practice of wearing orange uniforms as it is the case in England and Wales as well as in the Netherlands (and in the USA or South Africa) therefore is a violation of these human rights standards. From a European point of view, totally violating basic human rights standards has been the practice in some US states (and still is in Arizona) to make offenders work in so-called chain gangs, i.e., stigmatizing them by working in a kind of medieval outfit tethered together. Two more human rights issues have to be mentioned in this context: Rule 37 of the ERJOSSM states that the costs of implementation shall not be borne by the juveniles or their families. And community work shall not be undertaken for the sole purpose of making a profit (Rule 45 ERJOSSM). Therefore, the demand of the South Australian Corrections Service to make the communities pay 200 $ per a day of community service with good reasons has been rejected by the local authorities (see http://www.abc.net. au/news/2012-04-05/community-service-chargecouncil/3934500).


Statistical Developments: Europe and Scandinavia The use of community service as compared to other community sanctions and measures and imprisonment can be examined most effectively on the basis of enforcement statistics. The Council of Europe’s penal statistics Space I (survey year 2009) and Space II (survey year 2010) provide information both on stock rates and flow rates (/100,000 population) in Europe. Figure 2 displays the number of persons serving community service or other community sanctions under the supervision of probation services at a given day (31.12.2010). Figure 3 displays the same information on the base of flow statistics. Community service has a noteworthy role in over 20 European nations with data in the Council of Europe’s Space II survey. It has an essential role as the principal community sanction in Spain, Hungary, England and Wales, Scotland, Latvia, and all Nordic countries. Many of the countries with high rates of community sanctions are also countries with relative high incarceration rates (see D€ unkel/LappiSepp€al€a/Morgenstern/van Zyl Smit 2010, 1058 ff). Figure 4 compares the use of community service and imprisonment on the base of enforcement statistics. Stock- and flow-statistics give the same result. There is no correlation between the number of people in community service and in prison at any given day (stock). Neither is there a correlation between the number of people starting community service or entering in prisons during the year (flow). The latter finding is, however, heavily influenced by one outlier (Spain). Should this observation be excluded, one would find a weakly positive correlation. In other words, countries that use extensively community service use also imprisonment more. This finding is confirmed by a comparison between all community sanctions and measures and imprisonment (below) (Fig. 5). It seems that, in general, the use of imprisonment is not compensated by community sanctions. Countries with high incarceration rates are

431

C

often (but not always) countries with high community sanctions rates. And countries with low incarceration rates seem to manage also with lower rates of community sanctions. However, one needs to take into account that this comparison includes only community sanctions with supervision under the probation service. It excludes formal types of sanctions such as suspended and conditional sentences without supervision, as well as fines. Should these measures be taken into account, low incarceration countries like the Nordic countries and Germany would rank high in the scale of noncustodial sanctions, but low in incarceration (see LappiSepp€al€a in this volume in the chapter on fines).

Evaluation Research: Recidivism and Reintegration into Society Completion of Community Service Orders, Implementation of Programs, etc. Community service in most countries is organized by the probation service and/or private nonprofit organizations of offender rehabilitation. One major problem in implementing community service is to organize working places and to allocate offenders to adequate working facilities. In particular, in large rural countries or local municipalities, it is sometimes difficult to find working possibilities close to the offender’s residence. As community work has to be unpaid, the kind of work should be of a nature not to compete with regular paid work in the first labor market. Therefore, trade unions are keen to restrict community service to extra work which cannot be provided by normal enterprises. Empirical studies have shown that the implementation of community service has been successful in many countries (see above) and today is the most important alternative sanction apart from fines and suspended sentences (probation). In this regard, it is important that most offenders assigned to community service programs are completing their working hours. Only a small minority fails. Factors for noncompliance or failure to complete the sentence are a too high number of hours imposed

C

C

432


Community service

Other community sanctions

Spain Hungary Eng&Wales Netherlands Latvia Belgium Luxembourg Estonia Spain (Cat) Bulgaria North Ire France Denmark Ireland Portugal Norway Finland Sweden Moldova Iceland Switzerland Armenia Poland Lithuania Austria Croatia Greece Malta Azerbaijan Georgia Italy 0

200

400

600

800

Persons under the supervision of probation services. Stock 31.12.2010 / 100 000 pop.

Community Service in Europe, Fig. 2 Persons under the supervision of probation services. Community service and other community sanctions and measures. Stock

31.12.2010 (Space II 2010 and Hildebrandt 2012 for the Nordic countries)

and/or personal problems of the offender to get to work (problems with alcohol and/or drugs, personality problems, unstableness, lack of practical experience with work, refusal to work etc.). The experience of the probation services shows that offenders in these programs often show multiple problems of reintegration which demonstrates the necessity of further rehabilitative measures compared to the sole allocation to a working facility. One remarkable program in Germany should be mentioned. In Germany, community service is provided only for fine-defaulters as a substitute to imprisonment for nonpayment of a fine (Norway has in 2012 introduced a similar type of sanction

(“Bo¨tetjeneste”) and also Finland currently is considering one). The German programs go back to the year 1983, but until the mid-1990s, most federal states ran those programs on a rather limited scale. The federal state of MecklenburgWestern Pomerania had a specific problem as 22 % of prisoners in adult prisons were finedefaulters. Therefore, in 1998, a state-wide project was founded that systematically created working facilities and in particular a system of contracting offenders who did not pay and react to the imposition of fine-default imprisonment. The restructuring of probation and private offender rehabilitation services resulted in


433

Community service

C

Other community sanctions

Spain Hungary Netherlands Latvia Engl. & Wales Bulgaria Estonia Scotland Spain Luxembourg Belgium Czech Republic Denmark North. Ireland Sweden Finland Iceland Portugal Norway Moldova Austria France Ireland Cyprus Poland Switzerland Lithuania Croatia Azerbaijan Armenia Georgia Greece Malta Italy

C

0

200

400

600

800

1000

1200

Persons under the supervision of probation services. Flow 2010 / 100 000 pop.

Community Service in Europe, Fig. 3 Persons under the supervision of probation services. Community service and other community sanctions and measures. Flow

31.12.2010 (Space II 2010 and Hildebrandt 2012 for the Nordic countries)

a structure of systematic allocation to 1,600 working facilities all over the large federal state of Mecklenburg-Western Pomerania. One special element of the project was to deliver social work and care for specifically difficult offenders who otherwise would not have completed community work. About 70 facilities have established such intensive care programs for fine-defaulters. Another pillar of the project was that finedefaulters, even after they had been sent to prison, were allocated to community service programs and therefore often were released after a few days. The results during the 3-year pilot project starting in 1998 were remarkable: Daily

numbers of fine-defaulters decreased from around 120–100 to 50–70, i.e., were reduced by half (see D€unkel and Scheel 2006). The results were so convincing that the Ministry of Justice decided to permanently appoint six social workers who are occupied with electing and allocating offenders to working facilities. The project has achieved a sustained success (the last numbers are from 2010), keeping fine-defaulters in prison almost at a low level achieved in 2001. The placement in community service facilities was in about 75 % of the cases successful, i.e., the offender completed at least partly community service and thus the state saved money by

434


Aze Lit

Est

250 Pol

200 150 100 50

Aus Ice

0

Swz

Spa

Lux E&W

Malt Arm Bul Gre Ire Por

Hun

Bel

Prisoners 2009. Flow / pop (Space II)

Lat

300

R2 Linear = 0.027

Prisoners 2010 (adjusted)

350

1400

1000

Net

Fra Nor Den

Cyp

1200

R2 Linear = 0.001

C

Lat

800 Swz

Sco

600 Den Cro Fra Ita

400 200

Bul E&WNet

Lux Fin Aze Ire Bel

0

50 100 150 200 Community service 2010. Stock / pop. (Space II)

100

0

Spa

Hun

200

300

400

500

Community service 2010. Flow / pop. (Space II)

Lit

250

Est

Aze Pol

200 150

Lux Rom Cro

100 50

Malt Pop

Spa Hun

Bul

E&W Fra Bel

Ita Nor Ice

Ger Net Swz Swe Fin

0

200

Gre

1400 1200

600

CSM total 2010. Stock / pop. (Space II)

Lat

1000

Sco

800 Swz

600 400 200 0

400

Cyp

R2 Linear = 0.017

Lat

300

R2 Linear = 0.353

Prisoners 2010 (adjusted)

350

Prisoners 2009. Flow / pop (Space II)

Community Service in Europe, Fig. 4 Prisoners and persons under community service Stock 31.12.2010 and flow 2009 (Source: Council of Europe, Space I and II)

Cro Nor Den Lit Ita Malt Fra

Bul

E&W Por Bel Ire Rom Aze

0

400

Pol

Hun Spa

800

1200

CSM total 2010. Flow / pop. (Space II)

Community Service in Europe, Fig. 5 Prisoners and persons under supervision (all sanctions). Stock 31.12.2010 and flow 2009 (Source: Council of Europe, Space I and II)

avoiding imprisonment. In total, the state of Mecklenburg-Western Pomerania in 2010 saved about 1.6 million €, whereas the cost for the project (mainly the six social workers mentioned above) is about 450,000 € per year. Thus, the cost-benefit is more than one million € per year (see D€ unkel 2011b). Replacing Imprisonment or Net-Widening? The Example of Scandinavian Countries All Nordic countries adopted or increased the application of community service in the 1990s and early 2000s. The following diagrams

illustrate how these changes in the sanction structures were reflected in the application of other community sanctions and imprisonment in Denmark, Finland, Norway, and Sweden (see in more detail Lappi-Sepp€al€a 2008) (Figs. 6, 7). In Finland, community service came to replace prison sentences in 1993 without any notable net-widening. In Denmark, the extension of community service coincided with the reform that abolished the short-term (1–2 weeks) prison sentences. As it appears, the community service came to replace part of these prison sentences. But it also seems that part of previous


C

435

The adoption of community service in Finland in 1993 100 %

The rejection of short prison sentences and the expansion of community service in Denmark in 2000

90 %

100 %

80 % 70 %

80 %

C

60 % 60 %

50 % 40 %

40 %

30 % 20 %

20 %

Community service

Conditional

Prison

Community service

2007

2008

2006

2005

2003

2004

2002

2000

2001

1999

2008

2006

2004

2002

2000

1998

1996

1992

1994

1990 Prison

1998

0%

0%

1997

10 %

Conditional

Community Service in Europe, Fig. 6 The relative share of community sanctions and imprisonment (/100,000) – Finland and Sweden (Source: Lappi-Sepp€al€a 2012)

The change of community service to community punishment in Norway in 2003 100 % 90 % 80 % 70 % 60 % 50 % 40 % 30 % 10 %

Prison Supervision

Community service Conditional

2008

2007

2005

2006

2004

2003

2002

2001

2000

1999

1998

0% 1997

2008

20 % 2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

The expansion of community service in Sweden 1999 100 % 90 % 80 % 70 % 60 % 50 % 40 % 30 % 20 % 10 % 0%

Prison Community punishment conditional all

Community Service in Europe, Fig. 7 The relative share of community sanctions and imprisonment (/100,000) – Sweden and Norway (Source: Lappi-Sepp€al€a 2012)

short-term prison sentences was replaced also by conditional sentences (Figs. 6, 7). In Sweden, the expansion of community service in the year 2000 has replaced both conditional sentences and prison sentences. In 2002 Norway changed the name of “community

service” to “community punishment” in order to enhance the credibility of the sanction. This increased the use of the sanction. However, this increase does not seem to be reflected in the share of prison sentences, but rather in the decline in the use of conditional imprisonment.

C

436


Community Service in Europe, Fig. 8 Reconviction rates after prison and community service in Finland, 1992

Reconviction rates after prison and community service (Finland 1992) 80 70 Prison Reconvicted %

60 CSO 50 40 30 20 10 0 6 mths

1y

2y

3y

4y

5y

Follow up period

It looks like four out of five Nordic countries (Iceland included) have been successful in replacing prison sentences by community service. This success is associated with legislator’s aims and intentions. In Finland, it was clearly stated that community service should be used only in cases where the accused would otherwise have received an unconditional sentence of imprisonment. This was ensured with specific legislative arrangements. Results were also seen in the statistics. Along with the increase in the number of community service orders from zero to 3,500, the number of unconditional sentences of imprisonment decreased by a little over 3,000. In Finland, annually some 3,500 community service orders are imposed by the courts. This represents around 35–40 % of the sentences of imprisonment which could have been converted (sentences of imprisonment of at most 8 months). This corresponds to some 400–500 prisoners (10–15 % of the prison population) of the daily prison population. Assessing Reoffending Effects and Other Benefits A Finnish study used quasi-experimental design and compared two matched groups of offenders: one sentenced to community service in that part

of the country where community service was in use on experimental basis, and the other group of offenders with similar background and convicted for similar offenses (mainly drunken driving, which has been the major offense in Finland for community service). The follow-up period was 5 years. Only new sentences leading to conditional or unconditional imprisonment or community service were counted as recidivism. The study revealed a constant pattern showing that community service group had fewer reconvictions throughout the follow-up period. The differences in reconviction rates varied depending on where the counting was started. If started from the court’s decision the difference after 5 years was 60 % for community service and 66 % for prison group. If started from the completion of the sentence, the figures were 62 % and 72 %. And if counting of the follow-up period in the community service group starts from the court’s decision and in the prison group from the release to parole (which would be sensible), the difference in reconvictions would be 60 % (community service) and 72 % (prison, see Muiluvuori 2001) (Fig. 8). In Denmark, Susanne Clausen (2007) has examined the preventive effect of community service compared to imprisonment. The study is


limited to two offending categories: violent crime and traffic violations, including drunk-driving. In the case of traffic offenders, a change in the Danish penal code in 2000 made it possible to convert short prison sentences (up to 60 days) to CS. Capitalizing on this reform, Clausen compares recidivism levels between individuals sentenced to CS after the reform took place and those sentenced to unconditional imprisonment before the reform. As far as violent offenders, the quasiexperimental properties of the research design emanate from a change in the judicial policy that took place in the late 1990s. Within a relative short time period, the Danish courts increased the use of CS for violent offenders by a significant margin. In Clausen’s research, the treatment group consists of violent offenders sentenced to CS in 2000–2001 while the control group includes individuals sentenced to prison in 1997–1998. To further enhance the comparability between the offender groups, Clausen’s research controls for the impact of a large number of common risk factors, including age, gender, ethnicity, drug or alcohol misuse, mental problems, housing situation, education, employment, family type, and income. The subjects were tracked for 2 years for new crimes, starting from the date of conviction or release from prison (see Clausen 2007). Findings from multivariate models suggest that, in general, characteristics like age, income, and prior criminality are more important as predictors of recidivism than the nature of the criminal sanction. On the other hand, the results provide some evidence of reduced recidivism associated with CS. For example, young traffic offenders (under age 25) were 36 % less likely to reoffend than those sentenced to prison. Among violent offenders, the unemployed seemed to benefit the most from community service. To explain these findings, Clausen suggests that CS protects young traffic, but also violent offenders from the stigmatizing from the stigmatizing effects of a prison sentence. On the other hand, by providing mandatory employment, CS may restore a sense of responsibility and routine for the previously unemployed offenders, thus enhancing their ability to reintegrate in the

437

C

mainstream society. These benefits notwithstanding, Clausen’s study failed to confirm that CS helps to lower recidivism across most offender populations. The modest effects associated with CS may reflect the fact that it was compared to either conditional or very short prison sentences which, in most cases, are unlikely to create much damage. At the same time, there is little reason to suspect that CS will increase recidivism. Thus, to the extent it is less costly and considered more humane than prison sentence, it seems like an attractive alternative. In a methodologically advanced study, Killias et al. 2010a divided offenders randomly in community service and control group (prison). Recidivism was studied using four indicators: (1) whether offenders were convicted and (2) the number of convictions; whether (3) offenders were arrested and the (4) number of arrests. In addition, the authors compared how much the offenders had advanced, how many arrests they had before and after the sentence. By all measures, the community service group survived better. However, the small size of the sample kept the statistical significance rates low. In 2010, Killias and colleagues published evidence from two controlled, randomized experiments that assess the causal effect of community service and electronic monitoring on different post-sentencing outcomes (Killias et al. 2010a, b). While the first study (Killias et al. 2010a) finds no difference between community service and traditional custodial sentences, the finding of the second study (Killias et al. 2010b) suggests marginally significant differences between two types of noncustodial sanctions, electronic monitoring and community service, in favor of the former. However, the evidence is still relatively limited, particularly as most of the existing studies rely on rather small samples, which may explain the frequent result that the effects found are nonsignificant. Problems related to sample sizes have been effectively avoided in a largest study based on longitudinal official record data on adult offenders in the Netherlands (Wermink et al. 2010). The study compares recidivism after community service to that after imprisonment with a sample of 4,246 offenders. To account for

C

C

438

possible bias due to the selection of offenders into these types of sanctions, a large set of confounding variables was checked using a combined method of “matching by variable” and “propensity score matching.” The findings demonstrate that offenders recidivate significantly less after having performed community service compared to after having been imprisoned. In relative terms, community service leads to a reduction in recidivism of 46.8 % compared to recidivism after imprisonment. This finding holds both on the short term and long term, and for male and female offenders (Wermink et al. 2010) (Table 2). The available evidence suggests that community service is (at least) a promising alternative, in terms of reducing recidivism (using the Maryland University methodology scoring, see, for example, MacKenzie 2006). Several results point to significant differences between offenders who are sentenced to imprisonment and offenders who participate in community service. The recent meta-analysis of Bonta et al. (2006) demonstrates a stronger effect size (.07) than earlier meta-analyses, which the authors explain by the better structured programs since 1995 considering “what-works” criteria (Bonta et al. 2006, p. 115). In connection with community service, also other “non-reconviction benefits” need to be taken into account. These other beneficial features include positive contact to work life (and the resulting enhancement of offender’s economical situation), better self-control over substance abuse, and better preservation of family ties. “Importantly, results also indicate that community service participants fare better with regard to long-term social benefit dependency and income, especially when we disregard the possible effects of time trends” (Andersen 2012, p. 26). In addition, violent offenders and offenders convicted for misdemeanors showed lower recidivism rates than the imprisonment group. A problem, still deserving attention is how to deal with offenders, whose substance abuse prevents the use of community service. One answer is provided in the form of Swedish contract treatment.

Community Service in Europe Community Service in Europe, Table 2 Recidivism rates after community service compared to imprisonment Study Muiluvuori (2001) Finland

Quasi-experimental (during the experimental reform) Matched groups by age, sex, and prior convictions

Clausen (2007) Denmark

Quasi-experimental (before and after the reform)

Results Recidivism: New prison sentence in 5 years from the court decision Community service (60 %) Prison (66 %), nonsignificant Recidivism: New prison sentence in 5 years from the completion of the sentence Community service (62 %) Prison (72 %), significant Recidivism: New recorded offense 2 years from the court decision Traffic. Reoffending risk for young traffic offenders (drunk-driving) 36 % lower in CS group Violence. Largest benefit of CS for unemployed

Multivariate analyses (central socio-demographic factors) Killias et al. Recidivism: Convicted in 2 (2000) Switzerland years Randomized experiment Community service (21.4 %) Prison (25.6 %), nonsignificant Recidivism: Number on convictions in 2 years Community service (0.39) Prison (.64), nonsignificant Recidivism: Arrested in 2 years Community service (33.3 %) Prison (38.5 %), nonsignificant Recidivism: Number of arrests in 2 years Community service (0.76, but before 1.66) Prison (2.18, before 1.69), nonsignificant But see marked change in arrest rates before and after (above) Wermink et al. (2010). Community service leads to Netherlands a reduction in recidivism of 46.8 % compared to Matched comparisons recidivism after variable by variable imprisonment (continued)

Community Service in Europe Community Service in Europe, Table 2 (continued) Study Andersson and Alexandersson (1994) Sweden Bonta et al. (2002) Canada

Bonta et al. (2006) Canada

Results No significant difference in recidivism rates in regression analyses Meta-analysis (37 studies) Effect size – 0.03 (¼ 3 % less recidivism) Meta-analysis (57 studies) Effect size – 0.07 (¼ 7 % less recidivism)

Outlook: The Potential of Community Service to Reduce the Prison Population and Increase the Reintegration of Offenders European sanction policies have been characterized by two different trends during the 1990s and 2000s: an increasing use of prison and the adoption of new community sanctions. The first one reflects the growing punitive and populist trends in national crime policies; the latter seeks to counteract this development by offering more constructive, rational, and humane substitutes to incarceration. The overall effect of the expansion of community sanctions may appear disappointing: The number of alternatives has grown along with the increasing number of prisoners. Neither does it seem that countries that use extensively community alternatives have – in general – lower incarceration rates. Still, it would be premature to conclude that the efforts to enhance the use of community sanctions have been futile. Some jurisdictions have gained true success, giving thus also information of the models that do work. Failures, on the other hand, may offer important lessons on what not to do. In short, community sanctions do provide one functional tool against the overuse of imprisonment, once a number of lessons from different jurisdictions have been drawn upon. The key questions are: (1) how to ensure that these sanctions are applied in the first place, (2) how to ensure that they come to replace imprisonment (instead of replacing

439

C

other noncustodial sanctions), and (3) how to uphold and maintain the general credibility of these sanctions. Current problems and achievements have been summarized as follows (Lappi-Sepp€al€a 2003, see also van Kalmthout 2000): 1. Extra barriers should be constructed in order to ensure that the new alternatives are really used instead of imprisonment. In most countries, community service seems to substitute prison sentences only in roughly 50–60 % of cases. This rate can be improved by demanding directly – as is the case in Finland – that only prison sentences may be commuted to community service (leading to a “replacement rate” of over 90 % in Finland). Another way would be to define new alternatives as modes of enforcement of prison sentences, as has been done in Iceland. 2. Effective use of new alternatives and coherent sentencing practices require clear (statutory) implementation criteria. The courts should be given clear guidance as to when and for whom new sanctions are to be used. The role and position of new alternatives in the existing penal system (how they relate to other sanctions) should also be clarified. 3. The overall success of any community sanction requires resources and proper infrastructure. Community-based sanctions can only be applied within a community-orientated infrastructure geared to the specific requirements of these sanctions. Their implementation is dependent on the existence of an organization like the probation service. Often cooperation with private, semipublic, and public organizations or institutions is also required. The state and the local communities should provide the necessary resources and financial support. 4. Supervision, support, and swift reactions are needed in order to keep the failure rates down and to maintain the general credibility of new sanctions. The less control and supervision, the higher is also the dropout rate. There should also be a clear and consistent practice when the conditions of the sentence are violated. Varying and sloppy practices create

C

C

440

mistrust and resistance on the part of public prosecutors, the judiciary, and the public. 5. Arrangements should be made in order to enhance the motivation of the offender for cooperation and mutual trust. New alternatives usually require the offender’s consent and cooperation. Treating the offender, not as a passive object of compulsory measures, but as an autonomous person, capable of reasoned choices, is a value by itself, and as such, it should be encouraged whenever possible. In addition, experience indicates that explicit and well-informed consent is a highly motivating factor for the offender. Through his/her consent, the offender has also become committed to the required performance in a manner that gives hope for good success rates. 6. Issues of equality and justice must not be neglected. Community sanctions may often lead to discrimination, since they are easily used for socially privileged groups of offenders. Measures must be taken to avoid social discrimination. Clear and precise implementation rules and procedures are one important means to this end. Another way is to tailor the system of community sanctions to meet the demands of different offender groups with their different problems. Sweden, for example, has a specific sanction – “contract treatment” – for those who suffer from drug or alcoholic addiction as a substitute for shortterm prison sentences. Finland adopted in 2011 electronically monitored supervision order, supported by program work for offenders who are excluded from community service due to their addiction problems. 7. The idea has to be sold over and over again. Initial success does not guarantee that things will go well also in the future, just by themselves. Prosecutors and judges may lose their confidence, the enforcement agencies may lose their motivation, and the general public may withdraw its support. Maintaining the general credibility of the community sanctions and demonstrating their appropriateness is an ongoing process which does not end with the adoption of the requisite legislation and the arrangement of an initial training phase. The


key groups, responsible for the implementation of the sanctions, must be given constant training and general information of the general benefits of community sanctions and the drawbacks of the wide use of custodial sanctions. Taking care of community relations is also important: The community should be informed of the benefits and crime control potential of community sanctions. Also the value of volunteer work needs a clear recognition. Finally, the practices must be subject to impartial scientific evaluation in order to obtain necessary information for further development. 8. Do not turn community service into a mere punishment in the community. The original objective of community service was to provide a socially constructive and less damaging alternative to imprisonment. Community service has rehabilitative and reintegrative potential, which must not be “sold” during the efforts to widen its applications and finding political support. Policy planners should resist the temptation of presenting community service to the public only as a “credible” and “demanding,” if not outright humiliating, alternative – as has taken place in some jurisdictions. Community service should be oriented toward the restorative justice and the rehabilitation ideal and implemented that way (which would be in line with international human rights standards) and not be left to penal populism and repressive orientations of “punishment in the community.”

Recommended Reading and References Albrecht H-J, Sch€adler W (1986) Community service – a new option in punishing offenders in Europe. MaxPlanck-Institut f€ ur ausl€andisches und internationales Strafrecht, Freiburg i. Br Andersson T, Alexandersson, L (1994) Samh€allstj€anst som alternativ till f€angelse. Utv€ardering av 1990˚ -PM 1994:3. 1992 a˚rs fo¨rso¨ksverksamhet. BRA Andersen SH (2012) Serving time or serving the community? Exploiting a policy reform to assess the causal effects of community service on income, social benefit dependency and recidivism. University Press of Southern Denmark, Odense

Community Service in Europe Bonta J, Wallace-Capretta S, Rooney J, McAnoy K (2002) An outcome evaluation of a restorative justice alternative to incarceration. Contemorary Justice rev 5(4), pp 319–338 Bonta J, Jesseman R, Rugge T, Cormier R (2006) Restorative justice and recidivism: promises made, promises kept? In: Sullivan D (ed) The handbook of restorative justice: a global perspective. Routledge, New York, pp 108–120 Clausen S (2007) Samfundstjeneste – Virker det? Juristog Økonomforbundets Forlag, København Council of Europe (2009) European Rules for juvenile offenders subject to sanctions or measures. Strasbourg, Council of Europe Publishing. D€unkel F, Lappi-Sepp€al€a T, Morgenstern C, van Zyl Smit D (2010) (eds.) Kriminalit€at, Kriminalpolitik, strafrechtliche Sanktionspraxis und Gefangenenraten im europ€aischen Vergleich. Mo¨nchengladbach, Forum Verlag Godesberg. D€unkel F (2011a) Germany. In: D€ unkel F, Grzywa J, Horsfield P, Pruin I (eds) Juvenile justice systems in Europe. Current situation and reform developments, vol 2, 2nd edn. Forum Verlag Godesberg, Mo¨nchengladbach, pp 547–622 D€unkel F (2011b) Ersatzfreiheitsstrafen und ihre Vermeidung. Aktuelle statistische Entwicklung, gute € Praxismodelle und rechtspolitische Uberlegungen. Forum Strafvollzug – Zeitschrift f€ ur Strafvollzug und Straff€alligenhilfe 60, pp 143–153 D€unkel F, Pruin I, Grzywa J (2011) Sanctions systems and trends in the development of sentencing practices. In: D€unkel F, Grzywa J, Horsfield P, Pruin I (eds) Juvenile justice systems in europe. Current situation and reform developments, vol 4, 2nd edn. Forum Verlag Godesberg, Mo¨nchengladbach, pp 1,649–1,716 D€unkel F, Scheel J (2006) Vermeidung von Ersatzfreiheitsstrafen durch gemeinn€ utzige Arbeit: das Projekt “Ausweg” in Mecklenburg-Vorpommern. Mo¨nchengladbach, Forum Verlag Godesberg Goldson B (2008) Dictionary on youth justice. Willan, Cullompton Heinz W (2012) Das strafrechtliche Sanktionensystem und die Sanktionierungspraxis in Deutschland 1882–2010 (Stand: Berichtsjahr 2010) Version: 1/ 2012’, Internet-Publication http://www.uni-konstanz. de/rtf/kis/Sanktionierungspraxis-in-Deutschland-Stand2010.pdf Hildebrandt S (ed) (2012) Nordisk statistic for kriminalforsorgen I Danmark, Finland, Island, Norge og Sverige 2006–2010. Direktoratet for Kriminalforsorgen, Copenhagen de Jonge G (1999) Still ‘slaves of the state’: prison labour and international law. In: van Zyl Smit D, D€ unkel F (eds) Prison labour: salvation or slavery? Dartmouth, Aldershot, pp 313–334 Killias M, Gillie´ron G, Villard F, Poglia C (2010a) How damaging is imprisonment in the long-term? A controlled experiment comparing long-term effects of community service and short custodial

441

C

sentences on re-offending and social integration. J Exp Criminol 6, pp 115–130 Killias M, Gillie´ron G, Kissling I, Villettaz P (2010b) Community service versus electronic monitoring – what works better? Br J Criminol 50, pp. 1155–1170 Klement C (2010) Samfundstjeneste. En effektevaluering. Justitsministeriets Forskningskontor. http://www. justitsministeriet.dk/sites/default/files/media/Arbejd somraader/Forskning/Forskningsrapporter/2011/Sam fundstjeneste_-_En_effektevaluering.pdf. Accessed Aug 2011 Lappi-Sepp€al€a T (2003) Fines. Community sanctions as a means to restrict the use of imprisonment? In: European Conference for Probation (Ed) Probation in Europe. Bulletin of the confe´rence permanente Europeénne de la Probation, no 27. http://www. cepprobation.org/uploaded_files/juni%202003%20PDF %20format%2027%20-%20E.pdf Accessed June 2003 Lappi-Sepp€al€a T (2008) Crime prevention and community sanctions in Scandinavia. Tokyo: UNAFEI (Annual report for 2007 and resource material series no 61). http://www.unafei.or.jp/english/pdf/RS_No74/ No74_00All.pdf Lappi-Sepp€al€a T (2012) Penal policies in the Nordic Countries 1960–2010. J Scand Stud Criminol Crime Prev 13:85–111. MacKenzie DL (2006) What works in corrections: reducing the criminal activities of offenders and deliquents. Cambridge University Press, Cambridge Muiluvuori M (2001) Revidivism among people sentenced to community service in Finland. J Scand Stud Criminol Crime Prev 2(1), pp 72–82 Space I (2010) Council of Europe. Annual penal statistics by Aebi M F, Delgrande N. University of Lausanne, Switzerland. http://www3.unil.ch/wpmu/space/files/ 2011/02/SPACE-1_2010_English1.pdf Space II (2010) Council of Europe. Annual penal statistics. Persons serving non-custodial sanctions and measures in 2010. Survey 2010, by Aebi MF, Marguet Y, Delgrande N. University of Lausanne, Switzerland. http://www3.unil.ch/wpmu/space/files/ 2011/02/Council-of-Europe_SPACE-II-2010-E1.pdf Tiffer-Sotomayor C (2000) Jugendstrafrecht in Lateinamerika unter besonderer Ber€ ucksichtigung des Jugendstrafrechts in Costa Rica. Forum Verlag Godesberg, Mo¨nchengladbach van Kalmthout AM (2000) Community sanctions and measures in Europe: a promising challenge or a disappointing Utopia? In: Council of Europe (ed) Crime and criminal justice in Europe. Council of Europe Publishing, Strasbourg, pp 121–133 van Ness D (1986) Crime and its victims. Inter Varsity Press, Downers Grove van Zyl Smit D, D€ unkel F (eds) (1999) Prison labour: salvation or slavery? Dartmouth, Aldershot Villettaz P, Killias M, Zoder I (2006) The effects of custodial vs. non-custodial sentences on re-offending: a systematic review of the state of knowledge. Campbell Syst Rev 2006, p 13

C

C

442

Wermink H, Blokland A, Nieuwbeerta P, Nagin D, Tollenaar N (2010) Comparing the effects of community service and short-term imprisonment on recidivism: a matched samples approach. J Exp Criminol 6, pp 325–349 Wright M (1991) Justice for victims and offenders: a restorative response to crime. Open University Press, Bristol

Community Supervision ▶ Behavioral Management in Probation

Community-Oriented Policing

Community Supervision

United States represents an exception to this rule. This is because comparative research between federal states has become more common than earlier. Comparisons with other countries, be they European, South American, Asian, or African, remain less frequent. This is despite the efforts of a number of leading scholars who refer to experiences and sentencing models in other countries. This entry gives an account of the subject matter of comparing criminal sentencing, the purposes of comparing, and the typical approaches of comparing criminal sentencing. In so doing, the entry discusses the issues at stake when comparing and the limitations that flow from the different approaches to comparing criminal sentencing. The entry’s last section devotes special attention to the relevance of globalization to the act of comparing across borders.

▶ Community Policing

The Subject Matter of Comparing Sentencing

Comparative ▶ Comparative Incarceration

Comparative Case Analysis ▶ Linkage Analysis for Crime

Comparative Courts and Sentencing Rasmus H. Wandall Faculty of Law, University of Bergen, Bergen, Norway

Overview Comparing sentencing arrangements is a vital part when developing and studying criminal sentencing. Few reform programs move ahead and little research is done without consulting experiences within other jurisdictions. To some degree, the

For the present purposes, criminal sentencing covers actions through which it is determined what penal sanctions an offender must be subjected to following a criminal offence. Although this description, like any other, is ethnocentric, it does cover the process of police, prosecutorial, and court decision-making, the sanctioning that follows, just as it includes the organizational structures that sustains and develops the decision-making processes. Typically, three different subject matters come into play when comparing criminal sentencing. The most frequent is the outcome of sentencing: the sanction type (prison, alternative sanctions, fines, etc.) and the amount of sanction (length of prison term, size of fine, hours of community service). For example, comparing the rate of imprisonment per capita or the number of prison sentences in the United States and in England and Wales. Another subject matter of comparison targets the legal structures that govern the sentencing decision-making, for example, comparing the many different constructions of numerical sentencing guidelines that have been built in different jurisdictions in the United States since 1980 or

Comparative Courts and Sentencing

comparing the Swedish statutory sentencing law with any of these numerical guideline systems. A third subject matter of comparison takes seriously the social practices of the everyday handling of cases and the organizational framework of sentencing, in court as well as in prosecution. Inspired by the traditions of socio-legal scholarship, these studies rely neither on the law nor aggregate sentencing outcomes but compare the organization, operations, and views of those who handle criminal cases, for example, comparing judges’ views on sentencing in Finland and in England and Wales or comparing court systems and their political roles in industrialized nations. These different subject matters should not be mistaken for one another. They bring different consequences to the comparative analysis and most importantly provoke us to ask how things would look with a different subject matter of comparison.

Purposes of Comparing Criminal Sentencing As in comparative criminology in general (Nelken 2011), we can discern three typical purposes of comparing criminal sentencing decisionmaking: (1) the policy purpose, (2) the social scientific positivistic purpose, and (3) the contextual or interpretive purpose. The three purposes often overlap, yet are distinct. First, there is a long tradition for using comparative methodology in policy research. In most sentencing reforms policy-makers look to other jurisdictions’ experiences as a tool to find the best solution to their own challenges of sentencing reform. This practice of transferring sentencing experiences from one jurisdiction to another raises a number of comparative concerns. These are addressed later in the entry. Second, the social scientific positivistic purpose is about comparing sentencing decision-making and its organization to test theories about relevant causes and effects of criminal sentencing. For example, testing the explanatory value of income inequality to the level of prison rates in different countries or explaining the change from discretionary to numerical guideline informed sentencing in many

443

C

jurisdictions in the United States. The third and interpretivist purpose of comparing is to unravel the contextual embeddedness of criminal sentencing and studying how sentencing decision-making, its regulation, and its procedures and sanctions are differently constructed. One example would be to show that sentencing discretion and its boundaries are intrinsically linked to local cultures.

Approaches to Comparing Criminal Sentencing The approach to comparing is intertwined with both the purpose and the subject matter of comparing. Indeed, as the headings below confirm, there is a close connection between the different purposes of comparing sentencing and the different approaches to it. Nevertheless, it is both important and helpful to make the distinction when presenting three widespread and different approaches: 1. The legal categorical approach 2. The predictive and explanatory social scientific approach 3. The interpretivist approach Legal Categorical Approach A legal categorical approach is deeply connected to the legal structure as subject matter and has legal scholarship and dogmatics as its basis. The approach is used to categorize sentencing laws and systems, typically based on functional characteristics and their legal historical heritage. This approach helps to elucidate important structural properties of sentencing laws, but it also functions to cover the many relevant social, cultural, and political differences between jurisdictions. The best-known categorization in wider comparative studies is that by comparative legal scholars Ko¨tz and Zweigert (1998) in their introduction to comparative law. They systematically organize the different legal systems in legal families: the Anglo-American, the romanistic, the germanistic legal family, the Nordic legal family, then add the law in the Far East, and finally religious legal systems. In comparative sentencing research, the categorization is most often seen in the simple

C

C

444

distinctions between Anglo-American common law and continental models of civil law. The Nordic countries are sometimes placed in the continental and civil law category and sometimes recognized as also carrying elements recognizable in the Anglo tradition of law. Though it is disputed whether distinctions such as civil law and common law and adversarial and inquisitorial law can serve as units of analysis, it is indisputable that these distinctions are actually used and that they at least historically carry some legitimacy. Civil law and inquisitorial models arose from Roman law and were subsequently shaped by European continental legal developments. Common law and the adversarial model have their origin in English law which subsequently spread to the United States of America and other places where the English sought influence. Common law is characterized by the central role of the trial, the judicial decisions, and the recorded precedents, all of which are important vehicles for the development of law. Contrary in civil law models, through codification, every offence is ideally preconceived in written legal statutes and the judge has the role of reading, understanding, and applying the statute content correctly – and most importantly has the final word. In a sense, the key axis in this model is that between the legal subject and the written rule, whereas in the common law model, it is rather that between the legal subject and the trial. How does this help to compare criminal sentencing? First of all, a distinction is often made between the civil law tradition of structuring sentencing through legal statutes and the common law tradition of governing sentencing through trial experiences. This is a frequent way of comparing continental European and AngloAmerican sentencing law. Indeed, France and Germany do have complete criminal codes defining every offence and stipulating the sentencing options for each. Many Anglo-Saxon jurisdictions still have no general criminal code, and still place the courts and the trial in a comparatively more significant role in sentencing. The basic structures of sentencing law are different. One could point to the adversarial nature of plea bargaining in England and Wales


and in the United States and contrast it to the continental styles of legality in sentencing procedure. Both in Germany and in France, it is the judge who has the final decision, even when procedures involve consensual elements. Problems with these distinctions between common and civil law arise when they are used to describe entire countries and sentencing systems. Despite the historical heritages, sentencing laws and practices have evolved remarkably over time, and today it is problematic at best to describe entire systems as common or civil law and as inquisitorial or adversarial. As Pizzi (2008) illustrates, it is difficult to categorize the entire sentencing systems of the United States as common law systems as such. More than half of its jurisdictions have, since 1980, implemented legal sentencing rules that place the court under substantial legal control. The techniques employed include minimum sentencing laws and administrative sentencing guideline systems combined with various administrative governance institutions and enforced monitoring mechanisms. In England and Wales legislative structuring of sentencing has increased since the middle of the twentieth century. Today the sentencing approach towards many criminal offences is governed by binding sentencing guidelines, promulgated by a national sentencing council, and enforced by the upper courts. It is more descriptive then in these jurisdictions to emphasize the joint structuring of sentencing by the upper courts, the more recent administrative sentencing guidelines, and the legislature. In the same way it makes little sense to categorize the Italian system as a civil law system as the country has implemented plea bargaining in a wide area of criminal cases. Most other continental countries have implemented some version or other of consensual procedures, eating away at the traditional principle of legality and central role of the judge assuring the law to rule sentencing (Thaman 2010). This does not mean that these basic distinctions between civil and common law and between inquisitorial and adversarial systems serve no analytical purpose. First of all, the distinctions are a welcome tool to reduce the otherwise dominant use of jurisdictional borders as a key distinction (Nelken 2007). Second, if we handle


these concepts as representing legal traditions in the way that Glenn (2010) does in his comparative legal theory, we could use the distinctions as carriers of norms and legal forms, travelling from one country to another, constantly evolving with the local and different pressures of law, politics, and culture in each jurisdiction, and only expressing themselves in much more concrete ways as rules, practices, and arrangements that reflect more than a singular legal tradition (Field 2009). As such these distinctions do serve the purpose of sorting out general lines of traditional sentencing and procedure to further investigate their concrete implementations. There are attempts at breaking free from the dominant distinctions of civil law and common law. For example, Frase (2008) explores the use of the established comparative criminal procedure model developed by Damaska (1991). Justice systems are categorized on two diagonal dimensions. In one dimension, systems are tending to be hierarchical or to be coordinate systems of justice. In the other dimension, systems are tending to be policy implementing or conflict solving. This does allow more flexibility with respect to legal categorical comparisons. However, it does not change the problem that legal categorizations of entire systems using such simple categories or dimensions lack significant descriptive and analytical power. More fundamentally, comparative legal theory has evolved to challenge the legal categorical approach. Recognizing the increasing traffic between borders of legal ideas, constructions, and institutions, advanced comparative legal theory today seeks to accommodate for the mixed legal systems that have resulted. It is accordingly put forward that most legal systems are mixes of different legal constructions and solutions. They therefore represent different traditions and countries from which ideas and models for legal constructions and institutions have been borrowed, imposed, codeveloped, or other. The above described widening use of plea bargaining or similar consensual procedures are good examples, as are sentencing commissions and councils, sentencing guidelines systems, sentencing information systems, problem-solving courts,

445

C

alternative dispute resolution mechanisms, and prosecutorial diversion programs. The headlines that are used to make sense of these developments include “legal transplants” (Watson 1993), “legal irritants” (Teubner 1998), “law as ¨ r€uc€u 2002), and “legal borrowtransposition” (O ing” (Nolan 2009). Partly the concern here is to accommodate for the significant role that borrowing plays in comparative studies. Partly the concern is if such transport of legal institutions from one jurisdiction to another makes sense at all within the framework of comparative legal theory. The fundamental question is if a legal rule is bound to its social and cultural meanings, how can it move elsewhere without lousing this meaning? Yet, despite these significant developments, the distinctions between civil and common law and between adversarial and inquisitorial procedures continue to play a substantive role in scholarly writings on comparative criminal sentencing law. Predictive and Explanatory Social Scientific Approach As positivistic social science in general, this approach compares variation in sentencing arrangements between jurisdictions, such as countries, states, or regions, in order to identify the causes and consequences of sentencing arrangements. Different from the legal classificatory approach, all social phenomena can be used in the study of sentencing whether legal or not. The approach is used to compare various aspects of sentencing. For example, comparing the degrees of voluntariness in sentencing guidelines to see the effects on levels of punishment, comparing fine levels or the frequency and length of prison sentences, comparing different paths that reduce prison sentences, comparing the use of alternatives to imprisonment and their sentencing arrangements, comparing the effects of sentencing reforms on court practices. comparing to explain increases in prosecutorial authority in criminal sentencing, comparing the consequences of political culture on criminal sentencing and court practice, identifying the causes and effects of different sentencing policies and the structural reforms of sentencing, (Tonry and

C

C

446

Hatlestad 1997; Tonry 2007, 2009; Tonry and Frase 2001), and comparing in order to test arguments about correlations between penal levels and sentencing ideologies. Though these many approaches differ from one another, they can all be broken down to accounts of variation between jurisdictions in the correlation between dependent and independent variables. Four issues are particularly sensitive when using this approach. The first concerns the sample and population. Predictive and explanatory models make use of sample countries to generalize, but the extensive variation between countries makes it difficult to generalize both regionally and globally about sentencing. For example, Cavadino and Dignan (2007) argue that penal levels, measured by overall imprisonment rates, vary with the type of political economy. The four proposed typologies are the neoliberal (USA, UK, Australia, New Zealand, South Africa), the conservative corporatist (Germany, France, Italy, the Netherlands), the social democratic (Sweden, Finland), and the oriental corporatism (Japan). Their aim is to find similarities between groups of countries on significant variables that simultaneously distinguish those countries from others belonging to different typologies. Using sample countries from North America to east Asia, the authors can make the claim of sampling a global scale of modern industrialized countries. However, many countries pointing in different directions are not included. For example, what would happen if China is in the comparison? China has a very high rate of imprisonment, but does not adhere to a neoliberal societal model (Nelken 2011). Within the African scale of modern industrialized countries, it is only South Africa that is included in the comparison. However, Kenya is equally a comparatively large market economy in Africa, yet with much lower prison rates than South Africa. Similarly, Lappi-Sepp€al€a (2011) compares correlations between prison rates and a number of social and political structures in 30 different countries. He argues that among others, moderate penal policies are related to consensual and corporatist political cultures with high levels of social trust and political legitimacy. The sample


consists of 30 countries from continental Europe (hereunder Russia), North America, and it includes Australia and New Zealand. Yet, it does not include any Asian (apart from Russia) or African countries. For the present purposes we are reminded that explanatory models are limited to the countries chosen for comparison and that it is problematic to generalize both regionally and globally about correlates to sentencing and sanctioning. The second issue concerns the dependent variable. Whether it is prison rates, the use of alternative sanctions, fines, the legal framework of sentencing, or sentencing uniformity, it is central to a positivistic approach that the dependent variable is comparable across jurisdictions. When comparing prison rates, what counts as prison? Is weekend prison and open prison with daily outside working or educational activities outside included? Is electronic monitoring and home imprisonment included? Does it make sense to compare sentences served under the conditions in local prison facilities in the state of Virginia with those of Germany? Furthermore, increasingly important is the measuring of alternative sanctions. Its features, characteristics, and use have expanded rapidly over the last decades. But while many of the headlines are the same across borders, the content of alternative sanctions rarely is. Some are keenly aware of this challenge (Farrington et al. 2004). Taken wider, the issue here is that looking at criminal sentences across nations assumes sufficient non-variation both in sanctioning mechanisms and in competing social sanctioning mechanisms. For example, in the context of Japan in which disciplinary culture varies markedly from that of England and Wales and that of the United States, there is a range of other measures that are used for social control. South Africa, even though it does represent a punitive system, also represents local governance technologies that are applied to situations in which other countries, like England and Wales, France, or Norway, would use formal sanctions of imprisonment and community service. A social scientific predictive or explanatory approach must consider these factors.


The third issue that must be considered relates to the independent variables. When comparing two or more jurisdictions, everyone of the independent variables needs to be comparable across jurisdictions and relevant variations between them must be reflected. This includes both criminal offence variables, procedural variables, and variables related to the criminal offenders, and it includes the social, political, and cultural structures of the sentencing jurisdiction. Both legally and empirically, violent offences may look different, the offenders may belong to different groups of society and the form of trial may vary from place to place. In some countries less serious offences are not categorized as crimes but as administrative wrongs and are handled accordingly. Some countries rely on police and prosecution to finish many cases before they reach court. In some jurisdictions, like England and Wales, timely confessions are awarded measurable reductions in criminal sentencing, while in other countries confessions only affect the form of evidence and trial. Likewise, in some jurisdictions, there is little space for the inclusion of a personal investigation report, while in others it takes center stage. A further and important concern with independent variables is the covariance between them. For example, Lappi-Sepp€al€a (2011) relates low levels of inequality, democracy, high social trust, and welfare society with moderate penal levels. However true this may be, it is associated with considerable methodological concern that there is a high covariance between these particular variables. A related concern is to what extent we can meaningfully distinguish sentencing from any of the relevant independent variables. One important example is the ties between sentencing and procedural arrangements. Many criminal justice systems allow for a range of alternatives to trial: case diversions, conditioned charge withdrawals, and plea agreements, to mention but a few examples. Many such procedural dispositions are closely related to the expected sentence. This makes sentencing a procedural variable also and thus makes it difficult to distinguish between sentencing and procedure. Another example is

447

C

the connection between criminal sentencing, strategies of other justice institutions, and overall penal policies. The typical assumption is that penal policy – to some degree – determines the direction of sentencing practices and that sentencing is distinct from activities of other justice institutions. However, in most Western countries the individual branches of penal administration have increased their interdependencies in a number of different ways such as shared strategies, policies, goals, measures, daily interacting operations, and overall common infrastructures. The consequence is an increasingly larger difficulty of distinguishing the effects of one agency from the other. This naturally makes measuring such variables across borders difficult. The fourth issue with the positivistic approach is the interpretivist concern that comparing variable elements of sentencing regimes is problematic because every element is subject to the individual meaning structures of the compared systems, societies, and cultures. For example that judicial discretion is deeply imbedded in the cultural constructs of the society in which it is practiced, as are the meaning of “crime” and of “community” sanctions and of “youth” problems. See further below about the interpretivist approach. There are no easy answers to any of these concerns and questions. Yet, this should not discourage the use of this approach. Rather, important limitations of what can be inferred from such comparisons and the extraordinary care we must take when designing and interpreting comparisons should be borne in mind. Interpretivist Approach The interpretivist approach takes off where the explanatory and predictive approach stops. It lays bare the cultural and contextual embeddedness of penal arrangements. For example, on the basis of comparing the United States with France and Germany, arguing that the punitive level in the United States is conditioned by the particular American resistance to state power and drive towards egalitarian social status (Whitman 2003). Another example could be contrasting the judicial discretion in Islamic law courts in

C

C

448

Morocco with that of courts in the United States’ jurisdictions, showing how local (cultural) concepts and ideas inform the exercise of judicial discretion. Comparing judges’ accounts and views on sentencing in different countries can show that any understanding has to be sensitive to the judges’ different perceptions of harm, crime, sanction, and sentencing objectives. Characteristically, interpretivist approaches use the sentencing and penal arrangements as indicators of wider cultural and political arrangements. However, taking culture seriously is not reserved to the interpretivist approach. Also predictive and explanatory approaches make use of culture. It is oftentimes difficult to spot the difference between the approaches, and in many studies the two are mixed. The interpretivist approach firstly requires the observation and understanding of local cultural concepts and meanings and of how they manifest themselves in the different structural layers of criminal sentencing. This is a methodologically difficult task demanding linguistic, cultural, and historical sensitivity that only few have and only the fewest have the months and years to invest. Is one barred from taking this approach if one does not speak the language? How long must one stay in a given place? These are relative questions that limit the reliability of interpretive comparative studies, but that should not bar them. Some scholars use the collaboration between two or more colleagues, representing the legal cultures compared, as the central methodology. Another important issue of the interpretivist approach is that the boundaries of what counts as legal and penal culture around sentencing can at most be loosely defined. We may cautiously characterize it as conveying the sets of ideas, expectations, and values that in practice and in policy-making is held with regard to penal law, the criminal process, and the sentencing. A third important issue is that far from everything comes down to local cultures. Comparative studies today take seriously that techniques, ideas, concepts, and meanings travel and are sometimes shared by many jurisdictions, however, different concrete shape they acquire


locally. In relation to the specific transport of legal concepts and ideas, see also above about the legal categorical approach. For example, many scholars have described a Western development towards curtailing judicial discretion, increasing the level of standardization and formalization of criminal sentencing. Aas (2005) calls this a development towards “sentencing at a distance.” A significant body of research has looked at how such transnational trends and ideas merge with local legal and political cultures. Some have described how the policy implementation of three-strikes policies takes on local particularities in different countries. Others have described how transnational movements of new court types result in remarkably different forms of courts when institutionalized in different cultures of criminal justice. Yet another example is the different legal constructions of a principle of equality in sentencing that has emerged in different countries with the otherwise transnational development of standardization and formalization of criminal sentencing. The state of Minnesota provides one of the most prominent examples of numerical sentencing guidelines with a strong institutional tie to prison capacity constraints. Equality takes on the character of equality between formal categories, distancing sentencing from an individualized conception of equality. In Denmark (or Germany, Norway, or Sweden), the narrative guidelines and close ties between legal and appellate guidance of sentencing, institutionalized in the professionalization of judging and sentencing, maintain equality in sentencing in its individualized conception. It is a particular strength of the interpretivist approach that it is well adapted to grasp the reflexivity of transnational and local ideas, concepts, and meanings.

Globalization and Comparative Criminal Sentencing Disregarding purpose or approach, it is today difficult to carry out comparative studies of any area of criminal justice without regard to the transnational or globalized field that the


compared fields of practice are in. Globalization may be described as the way in which societies and communities over the last centuries are increasingly interconnected and in which processes of human and societal development are intertwined. Focusing on the processes that stem from this development, Savelsberg (2011) illustrates with three examples of globalization: well-documented conditions of late modernity, global-level scripts offering solutions to challenges across nations, and a new type of global law has emerged in vast areas of application, including penalty bringing with it both new doctrines and institutions (e.g., the international permanent and ad hoc criminal courts). Globalization first of all affects a traditional approach to comparative criminal sentencing by questioning the nation state and the jurisdictional borders as key units of comparative analysis. Significant processes involve transnational developments, cross border activities, and entire fields of norms and institutions that have moved to a level of international problem-solving. Does that mean that we cannot compare nation states and their practice of criminal sentencing? No. For sure, an argument can be made that it is significantly problematic not to include a transnational dimension in a comparative study of criminal sentencing. However, globalization does not necessarily move the field of study away from the local level of governance and practice. Rather, globalization has become one of the many forces and institutions that matter on a local level, suggesting that we should rather accept and include in our studies the subtle processes between the local and the global. In that way, it modifies the object of study and it forces us to expand on the methods of our comparison. Fourcade and Savelsberg (2006) have developed well-suited themes to such a process perspective in sentencing: first, increasing the focus on agency which locally constructs global domains through local actions; second, recognizing the mutually constitutive relationship between global and local practice; third, emphasizing that the process of globalization itself is tied to transnational power flows with the United States as the

449

C

power holder; and fourth, bearing in mind that every transnational and global convergence trend is locally manifested and accordingly is interpreted by local national institutions through the lenses of local historical and local political structures. Judicial courts and sentencing cannot be understood as one or the other but rather as dialectic between forces on the local and the global level.

Conclusion Comparative approaches are important to policy and research about criminal sentencing. Globalization processes challenge the way that we conceptualize national borders and compare countries. But at the same time, globalization underlines the deep necessity of comparing, however, reflexively. This entry has touched upon important aspects of comparing criminal sentencing, laying bare significant possibilities and issues at stake with the different purposes of comparing, and with the different approaches to comparing sentencing. Accordingly, a legal categorical approach serves valuable purposes of organizing legal traditions but also suffers from the blindness to political, social, and cultural differences. An explanatory or predictive social scientific approach serves to advance our understanding of general questions of sentencing arrangements and provides valid descriptions of causes and effects of national sentencing and sanctioning practices. Nevertheless, it comes with serious concerns about its scientific assumptions, ranging from the units of analysis, its samples, as well as its construction of variables. Finally, an interpretivist approach suffers from the inevitable inability to generalize, yet provides an unparalleled means of understanding the depth and technologies of the relationships between courts, sentencing, and the local cultural context. As such, the interpretivist approach serves to challenge and extend the understanding derived from positivistic approaches. And as with most other things, it is the combination and not the choice of approach that provides the best result.

C

C

450

Related Entries ▶ Comparative Legal Cultures ▶ Comparative Political Economy of Punishment ▶ Comparing Police Systems Across the World ▶ Drug Courts ▶ Effective Community Supervision Programming ▶ Historical and Comparative Perspectives on Incarceration ▶ International Sentencing ▶ Managerial Court Culture ▶ Mitigation and Aggravation at Sentencing ▶ Multisite Trials in Criminal Justice Settings ▶ Penal Philosophy and Sentencing Theory ▶ Plea Bargaining ▶ Problem-Solving Courts ▶ Prosecutorial Discretion ▶ Sentencing as a Cultural Practice ▶ Sentencing Guidelines in the United States ▶ Sentencing Research ▶ Structured Sentencing Outside the United States ▶ Theories of Punishment

Recommended Reading and References Aas KF (2005) Sentencing in the age of information. From Faust to Macintosh. Glasshouse, London Aas KF (2007) Globalization and crime. Sage, London Cavadino M, Dignan J (2007) Penal systems. A comparative approach. Sage, London Damaska MR (1991) The faces of justice and state authority. A comparative approach to the legal process. Yale University Press, New Haven Farrington D, Langan P, Tonry M (eds) (2004) Crossnational studies in crime and justice. U.S. department of justice. Office of justice programs. Bureau of justice statistics Field S (2009) Fair trials and procedural traditions in Europe. Oxf J Leg Stud 29(2):365–387 Fionda J (1995) Public prosecutors and discretion: a comparative study. Clarendon series. Oxford University Press, Oxford Fourcade M, Savelsberg J (2006) Introduction: global processes, national institutions, local bricolage: shaping law in an era of globalization. J Law Soc Inq 31(3):513–519 Frase R (2008) Sentencing and comparative law theory. In: Jackson J, Langer M, Tillers P (eds) Crime,

Comparative Courts and Sentencing procedure and evidence in a comparative and international context. Essays in honour of Professor Mirjam Damaska. Hart, Oxford, pp 65–79 Glenn P (2010) Legal traditions of the world. Sustainable diversity in law, 4th edn. Oxford University Press, Oxford Hardie-Bick J, Sheptycki J, Wardak A (2005) Introduction: transnational and comparative criminology in a global perspective. In: Sheptycki J, Wardak A (eds) Transnational and comparative criminology. Glasshouse, London, pp 1–17 Jacob H, Blankenburg E, Kritzer H, Provine DM (1996) Courts, law, and politics in comparative perspective. Yale University Press, New Haven Koichi H, Harris R, Hough M, Ville R, Zvekic U (1995) Probation round the world. A comparative study. Routledge, London Ko¨tz H, Zweigert K (1998) An introduction to comparative law, 3rd edn. Oxford University Press, Oxford Lappi-Sepp€al€a T (2011) Explaining imprisonment in Europe. Eur J Criminol 8:303–328 Nelken D (2007) Comparing criminal justice. In: Maguire M, Morgan R, Reiner R (eds) Oxford handbook of criminology, 3rd edn. Oxford University Press, Oxford, pp 139–157 Nelken D (2009) Contrasting criminal justice. Ashgate, Aldershot Nelken D (2010) Comparing criminal justice. Making sense of difference. Sage, London Nelken D (ed) (2011) Comparative criminal justice and globalization. Farnham, Surrey, England Nolan JL (2009) Legal accents. Legal borrowing. The international problem-solving court movement. Princeton University Press, Princeton ¨ r€ O uc€ u E (2002) Law as transposition. Int Comp Law Q 51(2):205–223 Pakes (2010) Comparative criminal justice, 2nd edn. Willan, Cullompton Pizzi WT (2008) Sentencing in the US: an inquisitorial soul in an adversarial body? In: Jackson J, Langer M, Tillers P (eds) Crime, procedure and evidence in a comparative and international context. Essays in honour of Professor Mirjam Damaska. Hart, Oxford, pp 65–79 Savelsberg J (2011) Globalization and states of punishment. In: Nelken D (ed) Comparative criminal justice and globalization. Farnham, Surrey, England, pp 69–86 Tata C, Field S (eds) (2010) Connecting legal and social justice in the neo-liberal world? The construction, interpretation, and use of pre-sentence reports. Punishment and society. Int J Penology 12(3) Teubner G (1998) Legal irritants: good faith in British law or how unifying law ends up in new divergences. Mod Law Rev 61(1):11–32 Thaman SC (ed) (2010) World plea bargaining. Consensual procedures and the avoidance of the full criminal trial. Carolina Academic, Durham Tonry M (ed) (2007) Crime and justice. A review of research. Volume 36. Crime, punishment,

Comparative Incarceration

451

and politics in a comparative perspective. Chicago University Press, Chicago Tonry M (2009) Explanations of American punishment policies. A national history. Punishment and society. Int J Penology 11:377–394 Tonry M, Frase R (eds) (2001) Sentencing and sanctions in western countries. Oxford University Press, New York Tonry M, Hatlestad K (eds) (1997) Sentencing reform in overcrowded times: a comparative and cross-national perspectives. Oxford University Press, New York Watson A (1993) Legal transplants, 2nd edn. University of Georgia Press, Athens, Georgia Whitman JQ (2003) Harsh justice. Criminal punishment and widening divide between America and Europe. Oxford University Press, New York

Comparative Criminal Justice ▶ Comparative Punishment

Political

Economy

of

Comparative Incarceration Douglas B. Weiss1 and Doris Layton MacKenzie2 1 University of Maryland, College Park, MD, USA 2 Justice Center for Research, The Pennsylvania State University, University Park, PA, USA

Synonyms Comparative; Cross-national; Incarceration rates

Incarceration;

Overview One of the most distinctive features of the criminal justice system in the United States is the large number of individuals incarcerated in its prisons and jails. Since the early 1970s, the United States has experienced approximately 40 years of nearly uninterrupted growth in prison and jail populations. There were approximately

C

1.6 million people incarcerated in state and federal prisons, and approximately 750,000 individuals were held in county jails in 2010 (Minton 2011). Only in recent years has the growth in incarceration slowed. Incarceration rates in the United States are several times greater in magnitude than those of other advanced Western democracies. A variety of explanations have been proposed for why incarceration is so much greater in the United States than in similarly developed nations. Some suggest that the United States has much greater levels of crime than in other Western democracies, while others argue that a “tough on crime” movement that has increased penalties is responsible for this growth. Research tends to indicate that the high rates of incarceration are mostly due to differences in sentencing policies and practices and not to exceptionally high levels of crime. Further evidence that policy changes can have a significant impact on incarceration rates can be found by examining the experience of other countries. In the early 1950s, Finland had an incarceration rate several times greater than that of its Nordic neighbors. A profound shift in the thinking about the use of incarceration led to a series of changes to sentencing practices in Finland. As a result of these changes, Finland has gradually reduced its incarceration rate over the past 50 years and now has one of the lowest incarceration rates in all of Europe. Germany is another nation that has controlled prison population growth through policy changes. The experience of Germany and Finland in controlling growth in incarceration suggests that changes to current sentencing practices in the United States can be effective at reducing prison populations.

Incarceration Rates in the United States and other Western Democracies The United States incarcerates more of its population than any other country in the world. Despite being home to just 5 % of the global population, the United States incarcerates about 25 % of the world’s prisoners (Walmsley 2009). At year-end 2010, there were more than

C

C

452

1.6 million prisoners in US federal and state prisons for a rate of 497 prison inmates per 100,000 US residents (Guerino et al. 2011). As of midyear 2010, there were nearly 750,000 inmates held in local jails for a rate of 248 jail inmates per 100,000 US residents (Minton 2011). These rates are slightly down from their peaks in 2007 when prison and jail incarceration rates were 506 and 259, respectively. When prison and jail inmates are combined, the incarceration rate in the United States at year-end 2010 was approximately 730 inmates per 100,000 US residents (International Centre for Prison Studies nd). The current level of incarceration in the United States is exceptional by both historical and comparative standards. Current levels of incarceration in the United States are the highest they have ever been in history. Throughout most of the twentieth century, incarceration rates in the United States remained relatively stable at a level of approximately 100 inmates per 100,000 US residents – a rate comparable to those of other Western democracies at the time. The stability of incarceration levels in the United States led Blumstein and Cohen (1973) to theorize that societies maintain stable levels of punishment. The theory did not gain much traction, however, as incarceration rates began nearly four decades of continuous growth almost simultaneous to the publication of the stability of punishment hypothesis. Incarceration rates in the United States are also exceptional in comparison with other nations. The United States has the highest prison population rate in the world according to the International Centre for Prison Studies. In 2010, the nations that came closest to the level of incarceration in the United States are Rwanda, Georgia, and Russia. Drawing comparisons between these nations and the United States, however, would be misguided as these nations differ along a number of different dimensions such as levels of economic and social development. When one compares the United States to nations that are more similar in terms of economic and social development, however, the difference in incarceration rates is even greater. Incarceration rates of advanced Western democracies such as


England, Germany, the Netherlands, Sweden, and Australia are considerably smaller. While the United States has an exceptionally high incarceration rate relative to other advanced Western democracies, it is not the only nation that has experienced increases in incarceration rates in recent history (International Centre for Prison Studies nd). Incarceration rates in the nations of England and Wales, Scotland, and Australia increased by approximately 50 % between 1992 and 2010, while Netherland’s incarceration rate has nearly doubled during that time. Only a few nations have managed to reduce or maintain relatively stable incarceration rates as Finland and Germany have done. Between 1992 and 2010, Finland’s prison population rate has dropped by more than 10 %. Germany did experience an increase in incarceration since 1992; however, there has been a recent downward trend since 2004.

Explaining Cross-National Differences in Incarceration Rates It is important to first acknowledge the difficulty inherent in making comparisons between nations in terms of incarceration use. Differences observed between nations in incarceration may be an artifact of different recording and reporting practices rather than a real substantive difference. One important difference between nations is the determination of who is included in the calculation of the incarcerated population. Some nations may include jail inmates in their calculation of their incarceration rates while other nations may exclude jail inmates. The determination of who should be counted as a prisoner can have a substantial impact depending on the size of the jail population. With this caveat in mind, there is a relatively small body of literature that has examined and sought to explain cross-national differences in incarceration rates. This research has investigated the relationship between incarceration rates and a variety of social, political, and economic factors. Findings regarding which factors affect incarceration rates have been somewhat


inconsistent due to methodological differences between studies. For instance, some studies draw comparisons between a large number of countries which differ from each other along a number of dimensions, while others focus on a smaller number of countries that are more comparable in terms of economic and social development. It is important to keep this in mind as making comparisons between vastly different nations may lead one to draw erroneous conclusions. The factor most often thought to affect incarceration rates are crime rates. Homicide rates in the United States are about as twice as high as those of other advanced Western democracies, such as Finland, which has one of the highest homicide rates in Western Europe (United Nations Office on Drugs and Crime 2011). However, homicide accounts for a very small proportion of total crime and should not have a very large impact on incarceration rates. Other than homicide, crime and victimization rates in the United States are not significantly different from those of other advanced Western democracies (Farrington et al. 2004; Van Dijk et al. 2007). Few studies have found support for the notion that differences in crime rates can account for disparities in incarceration rates (Lynch 1988; Neapolitan 2001). Neapolitan (2001) found a significant relationship between homicide rates and imprisonment rates in his study of 148 nations. Lynch (1988) also found differences in incarceration rates between the United States, England, Canada, and West Germany could be explained in large part by differences in the levels and types of crime in these nations. Most research has found, however, that crime rates are not directly associated with incarceration rates (Young and Brown 1993; Lappi-Seppala 2007). Incarceration rates and crime rates in the United States have moved independently since the early 1970s. Incarceration rates have been increasing in the United States since the early 1970s regardless of changes in crime rates. Despite similarities in crime trends since 1960, Canada has maintained a relatively stable incarceration rate while the United States experienced its incarceration boom (Doob and Webster 2006).

453

C

It is more probable that crime rates indirectly affect incarceration rates by affecting attitudes toward punishment among politicians and the general public. Rising crime rates may contribute to increased punitive attitudes and increasingly severe criminal penalties that would result in increased use of incarceration. This may explain the divergent trends in incarceration between the United States and Canada. Doob and Webster (2006) attribute Canada’s stable incarceration trend since the 1960s to the lack of changes in sentencing practices despite increasing crime rates. In contrast, the indeterminate sentencing scheme used in the United States came under fire from a number of different fronts in the 1970s. The Martinson (1972) report, which claimed nothing works in rehabilitation, and Judge Marvin Frankel’s (1973) criticism of indeterminate sentencing contributed to profound changes in sentencing practices in the United States including the adoption of determinate sentencing, mandatory minimums, habitual offender laws, and the reduction or elimination of parole (MacKenzie 2006). There is good evidence that suggests these changes to sentencing and correctional policies were largely responsible for incarceration growth in the United States. Blumstein and Beck (1999) attribute 88 % of the increase in imprisonment in the United States between 1980 and 1996 to changes in sentencing practices. During this time, there was an increase in both the number of offenses for which one may be imprisoned and in the length of prison time served. The type of legal system used in a country is another factor that can affect incarceration rates. Ruddell (2005) and DeMichele (2010) found the common-law legal system used in the United States and other Anglo nations to be associated with significantly higher incarceration rates than nations that operate under a system of RomanoGermanic or Nordic law. Incarceration rates tend to be low under Nordic law regimes (e.g., Finland) and moderate under Romano-Germanic law legal regimes (e.g., Germany). DeMichele (2010) attributes the difference in incarceration rates between legal regimes to several factors including the balance of power among the

C

C

454

courtroom workgroup, differences in courtroom practices (e.g., trial structure), and the involvement of lay participants in the judicial process. Certain aspects of the common-law legal regime, such as the jury trial, increase the involvement of lay participants in making decisions regarding punishment. Heinous crimes that receive heavy media coverage may result in public demands for increasing the severity of punishment. Jacobs and Kleban (2003) found higher incarceration rates occur in nations where the general public has the greatest influence over corrections decisions. The lowest incarceration rates were found in nations where decisions about corrections were made by unelected experts, e.g., judges. Another political factor that may affect incarceration rates is the level of conservative political strength. Studies done in the United States tend to find a relationship between conservative political strength and incarceration (Sutton 2000; Jacobs and Helms 1996). However, comparative crossnational research has not found this relationship to be significant (Jacobs and Kleban 2003). Jacobs and Kleban (2003) speculate that the failure to find this relationship in other advanced democracies may reflect the fact that decisions about corrections are insulated from popular opinion. Economic factors provide another potential explanation for differences in national incarceration rates. However, research has failed to find a clear association between economic factors and incarceration. Ruddell (2005) found no relationship between incarceration rates and income inequality, unemployment, or inflation in a sample composed of the richest 100 nations as determined by GDP. Jacobs and Kleban (2003) also fail to find a relationship between incarceration and unemployment. However, they find that the most economically developed nations with the highest per capita GDP are more likely to have higher proportions of their population in prison. Incarceration rates in these nations may simply be higher because they have more resources to build and stock prisons. Incarceration rates may also be affected by structural factors of the economic system. Sutton


(2004) found a relationship between incarceration rates and unemployment, inflation, and income inequality in a study of 15 affluent capitalist democracies. However, the relationship between incarceration rates and these factors disappeared when measures of institutional structure, such as union and left party strength, are considered. Other factors that may affect incarceration rates include out-of-wedlock birthrates (Jacobs and Kleban 2003) and the degree of population heterogeneity in a nation (Jacobs and Kleban 2003; Ruddell 2005). In sum, there are a variety of social, cultural, and economic factors that account for the disparity in incarceration rates between the United States and other advanced Western democracies. There is mixed evidence that suggests incarceration rates are related to crime rates or economic factors such as unemployment or inequality. Instead, there is a good body of research that suggests incarceration rates are affected by changes in criminal justice policies and the type of legal system used in a nation. Further evidence that changes to criminal justice policies can have a significant impact on incarceration rates will be presented in the next section.

Controlling Prison Population Growth It is clear that changes to criminal justice policies and practices can have a significant impact on incarceration rates. The operation of the criminal justice system in the United States underwent profound changes in the early 1970s that contributed to incarceration growth. These changes include the switch to a determinate sentencing system, the adoption of mandatory minimums and habitual offender laws, and truth-insentencing legislation. Combined, these policies have increased the number of offenses for which one could be imprisoned and increased the amount of time offenders serve in prison. If these policy changes were responsible for the growth in incarceration, then it is important to ask whether policy changes could reverse the past 40 years of incarceration growth in the United States. The answer to this question may be found


by examining the experience of other nations that have dealt with extraordinarily high incarceration rates. In particular, Finland’s experience with incarceration could serve as a guide for how incarcerated populations in the United States might be reduced. In the early 1950s, the incarceration rate in Finland was around 200 prisoners per 100,000 inhabitants. While this rate is a fraction of the current imprisonment rate in the United States, at the time it was nearly twice the rate in the United States and was three to four times greater than other Nordic countries. Lappi-Seppala (2001) identified three possible factors that explained Finland’s exceptionally high incarceration rate at that time. These factors include (1) a unique “cultural climate in which severity was not measured on the same scale used in the other Nordic countries” (106), (2) a rigid penal system characterized by high minimum penalties, and (3) severe sentencing practices for relatively common crimes such as drinking and driving. These three factors also seem characteristic of criminal justice practices in the United States today. Criminal sentences in the United States tend to be more severe than those of other advanced Western democracies, and the greatest disparities in punishment exist among less serious property crimes. In a comparison of punitiveness among eight advanced Western democracies, Blumstein, Tonry, and Van Ness (2005) find that offenders convicted in the United States between 1980 and 1999 served considerably longer periods of time in prison than did offenders who were convicted in other countries. Further, they find the disparity in time served between the United States and other nations is greatest for less serious crimes such as motor vehicle theft. Lynch (1993) had previously found that average sentence lengths in the United States in the 1980s were considerably longer than those in Australia, Canada, England and Wales, and West Germany. However, the amount of time served was relatively similar for violent crimes in these five nations. Prisoners serving sentences for property crimes in the United States spent more time in custody than those in Australia, Canada, and England and Wales.

455

C

Changes in these three factors were responsible for the gradual decarceration that took place in Finland since the mid-1950s (Lappi-Seppala 2001). At the time, there was a profound shift in the ideological thinking of criminal justice policy which had till then emphasized (coercive) rehabilitation of offenders. Unlike the movement against rehabilitation in the United States where the punishment philosophy was replaced by the goals of incapacitation and deterrence, the backlash against rehabilitation in Finland was replaced by a philosophy of restitution that emphasized (1) the minimization of costs of crime and crime control to society and (2) creating a fair distribution of costs among the offender, society, and the victim. Rather than serving as the primary instrument of criminal punishment, incarceration was seen to be one possible option among others. This change in philosophy was accompanied by the recognition that the most effective crime prevention may not occur through fear of punishment (deterrence) but rather through increasing the legitimacy of the law and subsequently enhancing the internalization of values and morality. This change in philosophy and the inability to justify the exceptionally high rates of incarceration in Finland relative to other Nordic countries prompted a series of legislative and policy reforms that would ultimately reduce incarceration levels in Finland. These reforms included reduced penalties for several common crimes, the expanded use of noncustodial alternatives, and the development of sentencing alternatives specifically designed to reduce the number of offenders sentenced to imprisonment. A complete discussion of these reforms can be found in Lappi-Seppala (2001) and Tornudd (1993). Reduction of penalties for theft and drunkdriving offenses were two reforms that had a significant impact in reducing the number of offenders in Finnish prisons (Lappi-Seppala 2001). These reforms were designed to both (1) reduce the proportion of convicted offenders who would receive a custodial sentence and (2) reduce the sentence lengths of those who were incarcerated. Through a legislative reform

C

C

456

enacted in 1972 that redefined serious forms of theft and introduced a new penalty range, the percentage of convicted offenders who received a custodial sentence for theft dropped from 38 % in 1971 to 11 % in 1991. In addition, the median sentence for offenders convicted of larceny dropped from 7.4 months in 1971 to 2.6 months in 1991. Reforms pertaining to drunk-driving offenses focused on replacing custodial sentences with noncustodial alternatives such as community service. Similar to the success of the reforms directed at theft offenses, drunk-driving-related reforms reduced the percentage of convicted offenders sentenced to unconditional imprisonment from 70 % in 1971 to 12 % in 1981. The expanded use of noncustodial alternatives was another major reform that contributed to reduced incarceration rates in Finland. Specifically, Finland has relied heavily on the use of community service as an alternative for short terms of imprisonment Lappi-Seppala 2004. This alternative was effective at reducing incarceration because it was specifically designed to ensure that the only offenders who would receive it would be those destined for incarceration. While sentencing alternatives have been used in the United States, they have not been very effective at reducing the use of incarceration as a punishment. Instead of diverting convicted offenders from custodial sentences, the use of sentencing alternatives in the United States has contributed to a “net-widening” effect since those offenders receiving sentencing alternatives would not be incarcerated otherwise. Finland adopted a two-step procedure during the sentencing phase to ensure that only those who were going to initially receive a custodial sentence were eligible for the community service alternative (Lappi-Seppala 2004). In the first step, the court determines a sentence for the offender using the normal criteria for sentencing. If the offender is sentenced to unconditional imprisonment, then the court can decide whether to commute the sentence into community service provided a number of criteria have been met. These criteria include (1) the consent of the convicted person, (2) the determination that the convicted offender is able to carry out


the requirements, and (3) that there is nothing in the prior criminal history of the convicted person that would preclude the use of community service (e.g., violent history). After several years of being implemented on an experimental basis, community service orders started being used nationwide in 1994. An examination of trends in Finnish court sentencing practices suggests that community service orders, used in this way, have been successful in reducing the flow of offenders into prison. Since 1994, approximately 30 % of convicted offenders sentenced to an unconditional prison sentence have received a community service order instead. Even as Finland took these steps to reduce its incarceration rate, crime trends in Finland were remarkably similar to those of its Nordic neighbors (Falck et al. 2003). Crime rates in Finland, Norway, Sweden, and Denmark increased from the 1960s through the early 1990s when they reached a more stable trend. While the increase in crime in Finland was unremarkable when compared to its neighbors during this time, the decline in incarceration rates during this same period was exceptional. In 1950, the incarceration rate in Finland was 187 inmates per 100,000 population. By the late 1990s, the incarceration rate in Finland dropped below 60 – a level comparable to those of its neighbors. Incarceration rates in Finland have since increased; however, all of the Nordic nations have experienced similar increases (National Research Institute of Legal Policy 2011). Germany is another nation that has taken measures to control the growth of incarceration at a time when crime rates were increasing. Although Germany has not experienced as high of an incarceration rate as has Finland, it has faced problems related to the heavy use of incarceration such as severe overcrowding (Graham 1990). Heinz (2006) traces the history of sentencing practices in Germany as far back as 1882, where nearly 80 % of convicted offenders were sentenced to unconditional prison sentences. Since then, Germany has gradually rolled back its use of custodial sentences in favor of noncustodial alternatives such as fines. Since the mid1980s, less than 10 % of convicted offenders in


German courts have received unconditional prison sentences, while approximately 70 % of offenders were ordered to pay fines. While the use of imprisonment has declined throughout the twentieth century, modern sentencing practices in Germany can be traced back to legislation enacted in 1969 (Weigend 2001). The criminal law reforms that took effect in 1969 were explicitly designed to restrict the use of imprisonment to only the most serious cases. This aim was to be achieved through several measures including (1) the abolishment of prison sentences of less than 1 month; (2) the decriminalization of many petty offenses, such as public order and traffic offenses, which were subsequently turned into administrative infractions punishable by fines only; and (3) the discouragement of sentences of less than 6 months. Noncustodial alternatives, primarily fines and probation, were used in lieu of short terms of imprisonment. The criminal law reforms of 1969 also began a trend in the development of sanctions other than those that have been traditionally used, i.e., imprisonment, probation, and fines (Weigend 2001). Some of these alternative sanctions have not been used very often, such as warning with suspension, while others are being increasingly used by the German courts, such as community service and restitution programs. Not all of these additional sanctions are directed at reducing the number of convicted offenders sentenced to imprisonment. For example, the sanction “warning with suspension of punishment” allows courts to suspend punishments consisting of dayfines and sentence the offender to probation instead. On the other hand, legislation enacted in 1994 allows the courts to mitigate the penalty or suspend sentences of less than 1 year of imprisonment if the convicted offender either participates in a victim-offender reconciliation program or makes some effort to provide financial restitution to the victim. Finland and Germany’s efforts to control and reduce prison populations may serve as an example for the United States to reduce its incarcerated population. While these two nations differ from the United States along a variety of dimensions, they both demonstrate that policy changes can be

457

C

effective at controlling and even reducing prison populations. Certain features of the United States, such as its common-law legal system, may make it more difficult to achieve the level of success that Finland has had in its history. However, it is unlikely that incarceration rates in the United States will drop to levels of other advanced Western democracies absent any changes to sentencing policies and practices.

Conclusion The incarceration rate in the United States far surpasses that of any other advanced Western democracy. This disparity in incarceration can be attributed to differences along a variety of factors including sentencing policies and the common-law legal system used in the United States. While some factors may be difficult to control because they are foundational, e.g., common-law legal system, there is evidence that changing sentencing and corrections policies can have a considerable impact on incarceration rates.

Recommended Reading and References Blumstein A, Beck AJ (1999) Population growth in U.S. prisons, 1980–1996. Crime & Justice 26:17–61. Blumstein A, Cohen J (1973) A theory of the stability of punishment. J Crim Law Criminol 64(2):198–207 Blumstein A, Tonry M, Van Ness A (2005) Cross-national measures of punitiveness. In: Tonry M, Farrington DP (eds) Crime and punishment in Western countries. University of Chicago Press, Chicago, pp 1980–1999 DeMichele M (2010) Three worlds of western punishment: a regime theory of cross-national incarceration rate variation, 1960–2002. Doctoral dissertation Doob AN, Webster CM (2006) Countering punitiveness: understanding stability in Canada’s imprisonment rate. Law Soc Rev 40(2):325–368 Falck S, von Hofer H, Storgaard A (2003) Nordic criminal statistics, 1950–2000. Department of Criminology, Stockholm University, Stockholm Farrington DP, Langan PA, Tonry M (2004) Crossnational studies in crime and justice. Bureau of Justice Statistics, Washington, DC Graham J (1990) Decarceration in the Federal Republic of Germany: how practitioners are succeeding where policy-makers have failed. Br J Criminol 30(2):150–170

C

C

458

Guerino P, Harrison PM, Sabol WJ (2011) Prisoners in 2010. Bureau of Justice Statistics, Washington, DC Heinz W (2006) Penal sanctions and sanctioning practice in the Federal Republic of Germany 1882–2004. The Konstanz repositories on crime and sanctioning. Retrieved 30 Oct 2009, from http://www.uni-konstanz.de/rtf/kis/sanks04_eng.htm International Centre for Prison Studies (nd) World prison brief. Retrieved 10 Feb 2012, from http://www. prisonstudies.org/info/worldbrief/ Jacobs D, Helms R (1996) Towards a political model of incarceration: a time-series examination of multiple explanations for prison admission rates. Am J Sociol 102:323–357 Jacobs D, Kleban R (2003) Political institutions, minorities, and punishment: a pooled cross-national analysis of imprisonment rates. Soc Forces 82(2):725–755 Lappi-Seppala T (2001) Sentencing and punishment in Finland: the decline of the repressive ideal. In: Tonry M, Frase RS (eds) Sentencing and sanctions in western countries. Oxford University Press, Oxford Lappi-Seppala T (2004) Penal policy and prison rates: Long-term experiences from Finland. The National Research Institute of Legal Policy, Helsinki. Lappi-Seppala T (2007) Trust, welfare, and political economy: cross-comparative perspectives in penal severity. In: Traskman PO (ed) Rationality and emotion in European penal policy: Nordic perspectives. DJOF, Copenhagen Lynch JP (1988) A comparison of prison use in England, Canada, West Germany, and the United States: a limited test of the punitive hypothesis. J Crim Law Criminol 79(1):180–217 Lynch JP (1993) A cross-national comparison of the length of custodial sentences for serious crimes. Justice Q 10(4):639–660 MacKenzie DL (2006) What works in corrections: reducing the criminal activities of offenders and delinquents. Cambridge University Press, Cambridge MacKenzie DL, Weiss DB (2009) Other countries have successfully reduced incarceration rates without increasing crime: we can do it! Vict Offenders 4(4):1–7 Minton TD (2011) Jail inmates at midyear 2010 – statistical tables. Bureau of Justice Statistics, Washington, DC National Research Institute of Legal Policy (2011) Crime and criminal justice in Finland 2010. National Research Institute of Legal Policy, Helsinki Neapolitan JL (2001) An examination of cross-national variation in punitiveness. Int J Offender Ther Comp Criminol 45(6):691–710 Ruddell R (2005) Social disruption, state priorities, and minority threat: a cross-national study of imprisonment. Punishm Soc 7(1):7–28 Sutton JR (2000) Imprisonment and social classification in five Anglo-American democracies. Am J Sociol 106:350–386 Sutton JR (2004) The political economy of imprisonment in affluent western democracies, 1960–1990. Am Sociol Rev 69(2):170–189

Comparative Legal Cultures Tornudd P (1993) Fifteen years of decreasing prisoner rates. National Research Institute of Legal Policy, Helsinki United Nations Office on Drugs and Crime (2011) 2011 Global study on homicide: trends, contexts, data. United Nations Office on Drugs and Crime, Vienna Van Dijk JJM, van Kesteren JN, Smit P (2007) Criminal victimization in international perspective: key findings from the 2004–2005 ICVS and EU ICS. Boom Legal Publishers, The Hague Walmsley R (2009) World prison population list. International Centre for Prison Studies, London Weigend T (2001) Sentencing and punishment in Germany. In: Tonry M, Frase RS (eds) Sentencing and sanctions in western countries. Oxford University Press, Oxford Weiss DB, MacKenzie DL (2010) A global perspective on incarceration: how an international focus can help the United States reconsider its incarceration rates. Vict Offenders 5(3):268–282 Young W, Brown M (1993) Cross-national comparisons of imprisonment. Crime Justice 17(1):1–49

Comparative Legal Cultures David Nelken Cardiff University Law School, Cardiff, UK

Overview This entry offers an overview of the main problems in defining and using the concept of legal culture, with special attention to how this may be relevant for scholars of criminal justice. Scholars of criminal justice may be interested in comparing legal cultures to make sense of similarities and differences in crime levels or punitiveness (Nelken 2000, 2010). Alternatively, they may want to understand how trends in criminal justice spread by means of legal transplants or efforts at harmonization or collaboration as well as how they are reinterpreted or resisted. This entry will first seek to clarify the meaning of the term legal culture and discuss how it may be used in doing research. It will then consider more specifically three of the key issues that arise in carrying out such investigations concerned with generating substantive findings about patterns of law or criminal justice in different

Comparative Legal Cultures

jurisdictions. These concern units and boundaries, elements and aggregates, and coherence and change.

What Is Legal Culture? Studies of law in relation to culture can cover a large range of topics including the role of culture in law (and the challenges of legal pluralism and multiculturalism) to the part played by law as culture, seen as a way of making meaning. Those interested in the relationship between law and culture might wish to study law as a cultural artifact, examine the way it becomes present in everyday life and experience, or through the media, or consider the role of law in accommodating cultural defenses or protecting cultural treasures. This entry, however, is mainly interested in seeing how the concept can be of use in research into cultural variation in how law is experienced in social life and, especially, differences in the role law plays and expected to play. There are obvious implications here for students of criminal justice seeking to tease out the relationship between formal and informal social controls. Research into comparative legal cultures can involve investigations into the extent to which law is party or state directed (bottom-up or top-down), the role and importance of the judiciary, or the nature of legal education and legal training. It may concern ideas of what “law” means (and what law is “for”). It is possible to discern legal culture in different approaches to regulation, administration, and dispute resolution. There may be important contrasts in the degree to which given controversies are subject to law, the role of other expertises, and the part played by “alternatives” to law, including not only arbitration and mediation but also the many “infrastructural” ways of discouraging or resolving disputes. With reference to criminal justice, the use of legal, non-state, or “informal” forms of social control deserves attention. But legal culture remains a highly debated term. Even Lawrence Friedman, who pioneered its use in social science, describes it as

459

C

“an abstraction and a slippery one” and that if he were to start over he might not use it again (Friedman 2006). There are several kinds of disagreements about what legal culture means or should mean (Nelken 1995, 2006, 2007, 2012). There is, for example, the question whether or not legal culture is a useful term as compared to alternative concepts such as legal tradition or legal ideology. Then there is discussion about what the term actually refers to – for example, to attitudes and/or behavior, or whether legal culture refers more appropriately at the popular or at the institutional level. Confusion can easily arise where what appears to be a debate about the correct interpretations of a given legal culture in fact trespasses into these other areas. Blankenburg, for example, takes Friedman to be arguing that it is “folk” culture – popular pressure for legal change – that shapes differences in legal development and response. He offers, by contrast, a comparison of the very different use of courts in Germany and the Netherlands, countries that he argues, have similar folk cultures and very different institutional outputs. For him, infrastructural arrangements represent the key to differences in legal culture, and in his view, “there is no legal culture outside of institutions” (Blankenburg 1997). Yet, what Blankenburg wants to call legal culture others would call institutions or structures, as seen in the related debate over whether the low level of use of law in Japan is to be attributed to deliberate cultural avoidance of litigation or is rather a result of structural arrangements that block access to the courts (Nelken 1997). It is also not always easy to tell where definitional questions end and empirical enquiries begin. How are we to distinguish legal culture from political, economic, or religious culture? Or take Friedman’s important distinction between “internal legal culture” which refers to the role in the law of legal professionals and “external legal culture” which refers to those individuals or groups who bring pressure to bear on the law to produce social change. Do lawyers belong to “internal legal culture” (as servants of the courts) or to “external legal culture” as agents of social groups and of individual litigants.

C

C

460

What about the relationship between external legal culture and culture in general? What of the way judges incorporate lay definitions of appropriate behavior into their activities? It can also be important to distinguish between legal culture and “the culture of legality” (stressing the “legal” before the word culture). The latter term – which corresponds very roughly to what in English is called “the rule of law” – is particularly common in those jurisdictions or parts of jurisdictions, for example, in the former Soviet Union, Latin America, or the south of Italy, where state rules are often avoided or evaded. These are places where – from a state perspective – there is a culture of illegality, and the point of talking of “legal culture” in such cases is to point to the normative goal of getting “legality” into the culture of everyday social and political life. This move seeks to reorient the behavior of such populations toward (state) law and/or encourage state law to respect certain limits of action. Whereas legal culture is a descriptive/explanatory term, the “culture of legality” is a normative and evaluative one. Keeping descriptive and normative meanings apart is necessary if we are to examine the sorts of legal culture that are more or less conducive to creating “the culture of legality”. But it can also be argued that talk about culture or legal culture is rarely merely descriptive and is more often an interested intervention in debates. For the observer, cultural ideas are contested and connected to relations of power. Cultural repertoires include polarities, such as values and practices, ideas and habits, and innovations along with commonsensical ways of doing things. Culture is the product of historical influences rather than evolutionary change. It is marked by hybridity and creolization rather than uniformity or consistency (Merry 2012). At the same time, however, for social actors themselves, “law” and “culture” are words whose interpretation and definition have illocutionary effects (“this is the law” “that behavior is inconsistent with our culture”). Thus, the term “legal culture” may itself be used by judges, politicians, or others, in the course of making claims about what is or is not consonant with a given body of


law, practices, or ideals. This use, as much prescriptive as descriptive or prescriptive through being descriptive, can “make” the facts it purports to describe or explain. Some scholars think the term must capture what these legal actors are trying to do (Webber 2004). As this suggests, culture is a term that easily lends itself to misuse, providing an easy alibi for resistance to change. For Patrick Glenn, the idea of culture is suspect (as a replacement for talk of race) because of both its origins and its consequences. Usage also includes “essentialist,” overdetermined, overbounded, and xenophobic applications. With cultural analyses, scholars are tempted to orientalize behavior as foreign and irrational and ignore or downplay the importance of economic and related political drivers of behavior (Glenn 2003, 2004). The German word kultur indeed emerged as a defensive term used in romantic opposition to the French universalizing idea of civilization (for which today’s discourses of democracy and human rights could be considered equivalents). Yet, there is nothing to stop the scholar from describing (and commenting on) such uses of the concept. And Friedman’s use of the word culture is in fact much closer to the nonessentializing French idea. He stresses convergence and the role of modernity as we move to a global legal culture based round individualism, equality, and human rights (Friedman 2006). Given that the goal of comparative work is to take us beyond ethnocentrism, we also need to ask whether there can be a universal concept of legal culture. How far is any given conception of legal culture inevitably itself a “folk concept” rather than an “observer’s” concept? For example, the Western idea of law as linked to the nation-state fits uneasily into places with overlapping plural legal systems. Instead of accepting Friedman’s own view of internal and external legal culture as neutral analytical categories, David Engel instead offers a cultural reading of them (Engel 2012). “Two aspects of Friedman’s imagery,” he tells us, “are particularly noteworthy. First, law is spatialized. It has an inside and an outside, and legal culture forms a kind of membrane between the two spaces.


Second, the internal space is dead, desiccated, and inert. It is a place of bones without flesh and words without life. By contrast, the external space is alive, vital, and active. Outside the law, one finds life forces that act like water in the desert and bring to life the otherwise barren world of law.” Referring by contrast to the approach taken by the noted anthropologist Clifford Geertz (Geertz 1973), Engel points to Friedman’s circumscribed view of law as a dusty, entombed skeleton versus Geertz’s broad and unbounded view of classical Indian law as sun and cattle! “Friedman’s colorful imagery of law and legal culture,” he argues, “presupposes a pluralism of legal and normative spaces, whereas Geertz’s description of Indian law presupposes the opposite – a unity of law, government, culture, and belief” (Engel 2012). For Engel, this concern with law having an “inside” and an “outside” is itself a reflection of the “project” of modernity, aimed inter alia at securing liberal legalism and secularism. Friedman’s conception of legal culture is linked to the aspiration to keep politics and economics subservient to law as well as the value of civil society playing its proper role as the source of the legal. Maintaining the metaphorical separation of inside and outside, and thereby affirming the “relative autonomy” of law, is central to the project of modernity. “The theory of modern law “presupposes that no one group in the society has a privileged access to religion and moral truth,” and the proper role of law is to establish a secular and neutral “process for conflict resolution” rather than to endorse one set of cultural practices or religious beliefs over another.” On the other hand, Friedman’s distinction may not carry all the “liberal legalist” implications that Engel attributes to it. It has been Friedman’s abiding concern to deny law’s autonomy so to show that it is what lies outside the law that actually animates and shapes it.

Working with the Idea of Legal Culture For this concept to be of value heuristically, it is essential to do more than treat legal culture as

461

C

a descriptive synonym for the legal system or some part of it. If our goal is explanation, it can be helpful to start from what appear to be puzzling features of the role and the rule of law within a given society (Nelken 2004). Why do the UK and Denmark complain most about the imposition of EU law but then turn out to be the countries which have the best records of obedience? Why does Italy, whose public opinion is most in favor of Europe, have such a high rate of noncompliance? Why does Holland, otherwise so similar, have such a low litigation rate compared to neighboring Germany? Why in the United States and the UK does it often take a sex scandal to generate official interest in doing something about corruption, whereas in Latin countries why does it take a major corruption scandal to excite interest in marital unfaithfulness? Such contrasts can lead us to discover new insights – into the importance of “enforcement” as an aspect of law, the reasons people use courts, or the way shame and guilt cultures condition the role of law. How we then put the term legal culture to use in such enquiries will depend on the disciplinary framework developed. To decide the appropriate strategies and methods to adopt, eyes must focus on disputes between (as well as within) continually evolving disciplines such as sociology, political science, or anthropology. For example, in line with competing approaches to social theory, legal culture will manifest itself through institutional behavior, as a factor shaping and shaped by divergences in individual legal consciousness, as a pattern of ideas that lie behind behavior, or as another name for politico-legal discourse itself. Studies of legal culture can also have different aims. Trying to make comparisons of legal systems more sociologically meaningful requires avoiding the risks of reifying culture as determinant and constraining of individual choices. A concern, by contrast, with illustrating the ongoing process of meaning – making in and through law overlaps with an interest in studying legal consciousness (Silbey and Ewick 1998, Silbey 2001). For many authors, legal culture is measurable by asking people questions about their attitudes to

C

C

462

the law. But we may also need to make sense of patterns of behavior of which the participants themselves are unaware. Even well-informed people living in India mistakenly think that the reason courts are slow is because the country has such a (relatively) high rate of litigation (Galanter and Krishnan 2003). Americans are convinced that their tort system regularly produces excessive and undeserved awards. But it turns out that, in large part, this impression is manufactured by the media (Holmes and McCaan 2004). Those societies where legal professionals show least concern for what Anglo-American writers describe as the “gap” or gulf between the “law in books” and “law in action” may not be those where the gap is least problematic but those where the gap is in fact overwhelming. A problem that in regularly leveled against explanations that rely on the idea of culture is that of tautology. Why do Japanese do what they do – because they are Japanese! It adds little to say that Italians suffer extensive court delays because of their legal culture if legal delays are part of what we are referring to when we speak of Italian legal culture. The danger of tautology needs to be overcome if we hope to develop policy-relevant social science explanations showing how variables produce outcomes. If the concept of legal culture accounts for why a transplant takes or not, can the success of transplants provide clues to legal culture? (Kurchiyan 2009). Interpretative approaches, however, may be less embarrassed by using culture as “cause” and consequence. If social actors take “culture” to represent or require certain behavior or values, this may provide a reason for their actions. In the 1980’s, the appearance of league tables of relative levels of incarceration induced Finland (interested in being seen as similar to other Scandinavian countries) to move toward the average European imprisonment rate by reducing its prison population; Holland for its part felt enabled to do the opposite. Likewise, how far East European legal cultures come to resemble Western legal cultures, assuming this is desirable, in part depends on how far politicians, policy makers, and legal actors there believe they can escape the patterns inherited from the past.


Another way of avoiding tautology is to see legal culture as something that needs to be explained rather than as itself the explanation. Take again the question why Italy suffers such long court delays. Treated as part of legal culture, the issue is to discover the legal, economic, political, religious, and factors that help reproduce this feature of Italian legal life. In the first place, there are relevant laws, especially those that have to do with civil and criminal procedure. There is also the management and organization (or lack of organization) of the courts and legal profession and claims about the supply of law not keeping pace with the demand, economic interests, political priorities, etc. (Nelken 2004). The problem of course is that it can then prove surprisingly difficult to decide what factors are the crucial ones, especially as the relevant facts can often be elusive and are open to very different interpretations. Large companies can make use of judges as (paid) arbitrators outside the normal trial system. But small businessmen (the backbone of economic life in Italy) would seem to gain little from the current situation. If so, why do they not put more pressure on the politicians to do something? Lawyers and banks profit from this situation. But do they have veto power on reform? Hence, rather than the idea of culture providing pat solutions to our question, it leads (as perhaps is appropriate) to an infinite regress of causal puzzles regarding what shapes it.

Controversies: Three Issues in Comparing Legal Cultures Research on comparative legal cultures necessarily mobilizes some assumptions about what it is that holds a culture together. But it may also help to revise these. We shall consider here questions connected to units and boundaries, elements and aggregates, and coherence and change. Units and Boundaries What is the unit we want to compare? In particular, is it (still) safe to identify the boundaries of legal culture with those of national jurisdictions? The search to understand and explain legal


culture at the level of the nation-state certainly continues to be an important ambition of comparative law and comparative sociology of law, as seen in titles such as the “Japanese approach to law,” “Dutch legal culture,” and “French criminal justice”. But the student of legal culture will often want to focus at levels below and above that of the nation-state. On the one hand, the focus could be on the culture of the local courthouse, the working norms of different social and interest groups and professional associations, and the networks of individuals involved in pursuing, avoiding, or mediating disputes. On the other, we could study international institutions and regulators and the so-called third cultures of international trade, communication networks, and other transnational processes. It can also be rash to assume any necessary “fit” between law and its environing national society or culture. Legal systems always have been modified by a variety of processes of borrowing, imitation, and imposition. Nation-states have never been the exclusive or even predominant source of norms. Their increasing insertion in larger bilateral or multilateral structures and networks means that there is now a wide gap between the (global) sites where issues arise and the places where they are managed (the nation-state). Different kinds of units emerge as objects and as agents of control. Instead of governments, the talk now is increasingly of “governance,” of how power is exercised at a series of other levels and by other institutions, in collaboration or otherwise with state bodies. The “denationalization” of rule making means that transnational public and semipublic networks substitute, to an increasing extent, for national governments. Rule formulation and settlement increasingly take place within new agencies of transnational governance, such as NAFTA, the OECD, and the WTO, but also in many lesser-known public-private forums. Naturally, this process varies by different areas of legal and social regulation. A contrast is often made between, on the one hand, those areas of law that are relatively internationalized, such as international business contracts, antitrust and competition policy, Internet and

463

C

new technology, labor law, social law, and environment law, and, on the other hand, family law and property law. But experts in these latter fields frequently report evidence of international trends and cross-cultural influences even there. As the world increasingly is tied together by trade and communication, people have the sense of living in an interdependent global system marked by borrowing and lending across porous cultural boundaries that are saturated with inequality, power, and domination. All this means that the purported uniformity, coherence, or stability of given national cultures will often be no more than an ideological projection or rhetorical device used by some of those within or outside a given society or other context (Coombe 1998). So we need to avoid reifying national or other stereotypes and recognize that these may often be no more (or less) than “imagined communities” or “invented traditions”. We may also need to be cautious about using terms that suggest boundaries at a time when many argue that it would be more appropriate to speak of “flows” (Appudurai 1995). But claims about the decline of the nationstate should not be taken too far. Many new states have been created recently, and “failing states” are more often the cause of concern rather than seen as harbingers of a new order. Differences between legal cultures may mobilize or reflect wider social and cultural patterns that roughly coincide with national political boundaries. Such boundaries often coincide with language and cultural differences and represent the source of common statistics. The imposition of a common legal code and the common training of legal officials form part of attempts to achieve and consolidate national identity. And “borders” continue to play important instrumental and symbolic roles, not least in responding to immigration. There is even an empirical basis for psychological differences in national traits in the way people relate to each other (Hofstede 1980). Such different historically conditioned sensibilities may persist over quite long periods (though careful research is needed to avoid confusing short-term and longterm trends).

C

C

464

Even if we accept that the state is – in many respects – losing its centrality, it does not follow that what happens in the field of criminal justice necessarily follows this general logic. Criminal law continues to be a powerful icon of sovereignty, and the nation-state persists as a key site where the insecurities and uncertainties brought about by globalization are expected to be “resolved.” It is even claimed that the state may “act out” in responding to some crime problems precisely because it has lost power elsewhere (Garland 1996). More functionally, some argue that states are obliged to enforce a new harsher type of order required because of the dismantling of welfare protections mandated by neoliberalism (Wacquant 2009a, b). Each country may also have its own reasons for increasing punitiveness. If the United States has seen “governing through crime” in a range of domestic settings (Simon 2007), in many European societies state power has been used mainly to criminalize noncitizen flows. In places such as South Africa, the state has to underline its ability to provide public safety in order to convince the “global economy” that it is a “safe place” in which to do business. One of the most important tasks of the student of legal culture is to try to capture how far globalization represents the attempted imposition of one particular legal culture on other societies. Importing countries are offered both by the Anglo-American model whose prestige is spread by trade and the media and by national versions of the more intellectually impressive continental legal systems embodied in ready packaged codes. The Anglo-American model is characterized by its emphasis on the link between law and the economy (rather than law and the state) and its reliance on legal procedures that prioritize orality, party initiative, and negotiation inside law. More than any particular feature of legal procedure, however, what does seem to be spreading is the common law ideology of “pragmatic legal instrumentalism,” the very idea that law is something which does or should “work,” together with the claim that this is something which can or should be assessed in ways which are separable from wider political considerations.


Elements and Aggregates A problem linked to the task of identifying the unit(s) of legal culture is that of distinguishing the elements of (legal) culture as opposed to treating it as an aggregate. Cotterrell famously criticized Friedman for treating legal culture both as an element – discussing why people turn to law – and as an aggregate-as when speaking of American legal culture or modern or global legal culture (Cotterrell 1997). Glenn too tells us that what he finds particularly problematical is the employment of culture as a “holistic signifier” and as a “variable” (Glenn 2004). On the other hand, the distinction between aggregates and elements is not a hard and fast one. All wholes can be incorporated into yet larger ones, just as all elements can be broken into yet smaller ones. For example, the group “attitudes” toward the use of law that are at the center of Friedman’s use of legal culture can also be broken down into smaller elements. Indeed, Friedman thinks it is plausible to speak of each individual’s legal culture. And these individual attitudes or opinions are in turn themselves composed of measurable responses to a range of particular questions. Whether it is appropriate to go down or up in levels of abstraction will depend on the purpose of an enquiry, for example, whether are, comparing whole societies, or are seeking to explain interconnections within them. Recently, Merry has argued that we need to break up the idea of legal culture into what she calls four “social dimensions.” The first is the practices and ideologies within the legal system, everyday way of getting things done, shared assumptions about good and bad clients, and other internal rules and practices, some of which are based on legal doctrine and others on categorizations shared by the wider society, such as ideas of race and gender. (As she says, this corresponds to Friedman term “internal legal culture.”) Then there is the public’s attitude toward the law. Is the legal system a source of corruption and ethnic preference, for example, or viewed as an institution that offers the rule of law for all people equally, regardless of their background? This, she suggests, is somewhat


similar to what Friedman calls external legal culture. Thirdly, there is the question of legal mobilization, which refers to how readily people define their problems in legal terms, when they turn to the law for help. A fourth dimension is legal consciousness, the extent to which an individual sees him or herself as embedded in the law and entitled to its protections. Experience with the law, both good and bad, can change legal consciousness. It may encourage further use or may drive the litigant to avoid the law next time. Merry argues that these last two aspects offer the best way to understand the cultural dimensions of law and its relationship to a social context as well as providing a more satisfactory analysis of the processes of translation across legal fields and the hybridism of these fields (Merry 2012). Nevertheless, others insist on the need to grasp the aggregate as a whole. Stewart Field, for example, argues that “actors do not live in a world of differentiated elements of institutions, formations, structures of feeling and traditions, however useful the distinctions may be as a heuristic device. The ultimate and impossible challenge that legal anthropologists and comparative lawyers must set themselves is to try to get something of the subjective feel of the normative pressures operating on the legal ‘other’ while making those pressures explicit in a way that native legal actors would not and perhaps could not do.” The goal for him is to show how “the interpretation of a wide range of operative legal concepts is shaped by distinct social connotations, because these practices only ‘make sense’ within particular sets of legal and broader social contexts and relationships. No doubt, these meanings are fragile, contested, and subject to change. But the argument is that there are distinctive cultural ‘logics’ at work, distinct ways of seeing into which the researcher must struggle to enter” (Field 2012). Coherence and Change The last way of thinking about what gives culture its hold on us is to ask what makes for the (alleged) coherence of legal cultures and the way this relates to the possibilities of change.

465

C

Do aspects of law in society come in “packages”? How do elements turn into a whole? What gives a unit its “unity”? For many writers, coherence (others might call it path dependency) is the key to “explaining” continuities in patterns of ideas and practices over time. It predicts how it is likely to respond to attempts at legal transfers. But for those critical of the concept of legal culture, the presumption of coherence is the problem not the solution. Comparative lawyers and philosophers of law tend to define the term with reference to the activities of the various legal professionals and jurists who bear the responsibility of (re)producing purported coherence in legal materials. But they say little about what type or degree of coherence is required in actual practice, and their way of using the term is somewhat narrow for sociolegal enquiry. A variety of possible phenomena may be involved – opinions and attitudes, behaviors, texts, and institutional and organizational norms, ideas, and ideals. What are at stake may be psychological pressures to consistency, pressures for group conformity, institutional and organizational controls or routines, or even loyalty to legal or religious texts. For some scholars, there can be interdependence between the elements that make up a given legal culture (as between the number of lawyers and the number of trials). Alternatively, coherence may be imposed by commentators, for example, when they categorize types of legal culture or legal “communities” in terms of ideal types. But others think that any connection exists (only) insofar as participants talk about it “as if” it is real. Merry, as we have seen, sees legal culture as no more than a framework with different overlapping elements with no necessary overall coherence – little more than a series of topics to investigate concerning institutions, attitudes, mobilization, and consciousness. For her, “Legal practices tend to be hybrid and creolized, formed of borrowings, transplants and translations of other legal practices in other places and times. Hence, it is unlikely that it will be coherent. Indeed, as with culture more generally, lack of coherence is what gives actors room for manoeuvre and innovation” (Merry 2012).

C

C

466

Legrand takes “the notion of “culture” to mean the framework of intangibles within which a community operates, which has normative force for this community (even though not completely and coherently instantiated), and which determines the identity of a community as community” (1997). But if culture is, to a large extent, a matter of struggle and disagreement, the purported uniformity, coherence, or stability of given national or other cultures will often be no more than a rhetorical claim manipulated by members of the culture concerned or projected by outside observers. In particular, as Roger Cotterrell tells us, it is judges and lawyers who attempt to sell the ideology of law as a “gapless system,” whatever they may know from their everyday practice in their offices or in the courts (1997). Much depends on scale and perspective. Under scrutiny, even defined areas of law, such as those governing family relations, are far from coherent. But, seen cross-nationally, even apparently unconnected branches of law may in fact manifest remarkable levels of cultural similarity within a given society. As Whitman has claimed recently, in replying to criticisms of his “culturalist” approach to penal law, “the pattern that we see in comparative punishment is also the pattern we see in many other areas of the law. Indeed, I would claim it as a virtue of my book that it shows that punishment law cannot be understood in isolation from the rest of the legal culture. For example, American workplace harassment law differs from German and French workplace harassment law in very much the same way. The same is true of comparative privacy law . . .just as it is true of the law of hate speech and everyday civility” (Whitman 2005). But even if legal culture possesses coherence, this does not stop it from changing. It can be salutary to recall the rapid transformation in attitudes toward “law and order” in the short period that elapsed from Weimar to Hitlerian Germany. What is clear is that any serious work on legal culture must make sense of both continuity and change. The strain toward coherence – whatever causes it – explains relative lack of change, the difficulty of change, and even the


direction of change. But it does not stop all change. To go back to the previous examples of country differences, in legal culture, England and Wales have recently witnessed some important corruption scandals to do with parliamentary expenses that were not connected with sex. And Berlusconi, Italy’s last prime minister, is now being pursued for sex scandals apart from any corruption implications. Nonetheless, it is still fair to say that, in the UK, scandals concerning business-politics links are still much less salient than in Italy and Berlusconi’s fall from power depended more on the financial crash than his private life. In an ever more interconnected world, even resistance to outside influence is an active process rather than a result of cultural inertia.

Related Entries ▶ Cross-National Performance in Policing ▶ Cultural Criminology ▶ Effective Community Supervision Programming ▶ International Policy Transfer ▶ International Responses to Victims in Criminal Justice ▶ Sentencing as a Cultural Practice ▶ Transnational Exchange of Forensic Evidence

Recommended Reading and References Appudurai A (1995) Modernity at large: cultural dimensions of globalization. University of Minnesota Press, Minneapolis Blankenburg E (1997) Civil litigation rates as indicators for legal culture. In: Nelken D (ed) Comparing legal cultures. Dartmouth, Aldershot, pp 41–68 Coombe RJ (1998) Contingent articulations: a critical cultural studies of law. In: Sarat A, Kearns P (eds) Law in the domains of culture. University of Michigan Press, Ann Arbor, pp 21–65 Cotterrell R (1997) The concept of legal culture. In: Nelken D (ed) Comparing legal cultures. Dartmouth, Aldershot, pp 13–32 Engel DM (2012) The uses of legal culture in contemporary socio-legal studies: a response to Sally Engle Merry. In: Nelken D (ed) Using legal culture. Wildy, London

Comparative Political Economy of Punishment Field S (2012) Comparative youth justice and legal cultures. In: Nelken D (ed) Using legal culture. Wildy, London Friedman L (1994) Is there a modern legal culture? Ratio Juris 7:117 Friedman L (2006) The place of legal culture and its position in the sociology of law. In: Freeman M (ed) Law and sociology. Oxford University Press, Oxford Galanter M, Krishnan JK (2003) Debased informalism: Lok adalats and legal rights in modern India. In: Erik GJ, Thomas CH (eds) Beyond common knowledge: empirical approaches to the rule of law. Stanford University Press, Stanford, p 96 Garland D (1996) The limits of the sovereign state: strategies of crime control in contemporary society. Br J Criminol 36:445–471 Geertz C (1973) The interpretation of cultures. Basic Books, New York Glenn HP (2003) The nationalist heritage. In: Legrand P, Munday R (eds) Comparative legal studies: traditions and transitions. Cambridge University Press, Cambridge, p 4 Glenn HP (2004) Legal cultures and legal traditions. In: van Hoeck M (ed) Epistemology and methodology of comparative law. Oxford University Press, Oxford, p 32 Hofstede G (1980) Culture’s consequences: international differences in work related values. Sage, Beverly Hills Holmes W, McCaan M (2004) Distorting the law. Chicago University Press, Chicago Kurchiyan M (2009) Russian legal culture: an analysis of adaptive response to an institutional transplant. Law Soc Inquiry 34:337 Legrand P (1997) Review of David Nelken’s comparing legal cultures. Camb Law J 56:646 Merry SE (2012) What is legal culture? An anthropological perspective. In: Nelken D (ed) Using legal culture. Wildy, London Nelken D (1995) Understanding/invoking legal culture. In: Nelken D (ed) Special issue on legal culture, diversity and globalization. Social Legal Stud 4(4):435–52 Nelken D (1997) Puzzling out legal culture. In: Nelken D (ed) Comparing legal cultures. Dartmouth, Aldershot Nelken D (ed) (2000) Contrasting criminal justice. Ashgate, Aldershot Nelken D (2004) Using the concept of legal culture. Austl J Leg Phil 29:1–28 Nelken D (2006) Rethinking legal culture. In: Freeman M (ed) Law and sociology. Oxford University Press, Oxford, pp 200–224 Nelken D (2007) Defining and using the concept of legal culture. In: Orucu E, Nelken D (eds) Comparative law: a handbook. Hart, Oxford, pp 109–132 Nelken D (2010) Comparative criminal justice: making sense of difference. Sage, London Nelken D (2012) Using legal culture: purposes and problems. In: Nelken D (ed) Using legal culture. Wildy, London

467

C

Silbey S (2001) Legal culture and legal consciousness. In: Encyclopedia of the social sciences. Elsevier/ Pergamon Press, New York, pp 8623–8629 Silbey S (2005) After legal consciousness. Annu Rev Law Soc Sci 1:323 Silbey S, Ewick P (1998) The common place of the law: stories from everyday life. University of Chicago Press, Chicago Wacquant L (2009a) Prisons of poverty. University of Minnesota Press, Minneapolis Wacquant L (2009b) Punishing the poor: the neoliberal government of social insecurity. Duke University Press, Durham Webber J (2004) Culture, legal culture, and legal reasoning: a comment on Nelken. Austl J Legal Philos 29:25–36 Whitman JQ (2005) Response to Garland. Punish Soc 7:389

Comparative Police Systems ▶ Democratic Policing

Comparative Political Economy of Punishment Nicola Lacey All Souls College Oxford, University of Oxford, Oxford, UK

Synonyms Comparative criminal justice

Overview Recent decades have seen a revival of interest in the broad social, political, and economic forces which are shaping criminal justice policy in Western democracies. The global economic changes which began in the 1970s – recession, the contraction or collapse of manufacturing industries, the growth of unemployment, and the creation of a large sector of people either unemployed or employed in insecure forms of work – it is argued, have eroded the consensus which

C

C

468

sustained postwar penal welfarism. As rising crime across Western countries gradually produced a situation in which the experience of criminal victimization and of managing the risk and fear of crime became normal features of everyday life, crime became increasingly politicized, leading to exclusionary criminal justice policies (Pratt 2007). In making this argument, scholars like Garland (2001), De Giorgi (2006), Reiner (2007), Wacquant (2009), and Young (1999) have refined the Marxist account deriving from Rusche and Kirchheimer (1969) with an analysis of cultural and political factors, thus avoiding the pitfalls of economic reductionism. Yet in each case, an emphasis on structural forces tends to direct attention away from variations in the institutional framework through which those forces are translated into concrete social policies in different countries. While each of these accounts, therefore, suggests hypotheses which are ripe for comparative investigation, a comparative focus does not feature prominently in their analyses (though see Garland 2007). Not all “late modern” democracies, however, have opted for a neoliberal politics (Lacey 2012). And many have managed to sustain relatively moderate, inclusionary criminal justice systems, while the British and American systems have, albeit to different degrees, been moving toward ever-greater penal severity. Even as between Britain and the USA, the differences in terms of the overall scale and quality of punishment are striking. Countries like Denmark, Germany, or Sweden fulfill the prophecy of “penal populism” yet less accurately (Lacey 2008; Cavadino and Dignan 2006). But how do these resemblances across types of political economy hold together over time, and why do they produce systematically different patterns of punishment? Comparative institutional approaches seek to answer these questions. They argue that progress can be made toward a genuinely explanatory account by drawing on the political-economic analysis of comparative institutional advantage and of the capacities for strategic coordination inherent in differently ordered systems. Structural forces are seen as

Comparative Political Economy of Punishment

mediated not only by cultural filters but also by economic, political, and social institutions: it is the institutional mediation of cultural and structural forces, and its impact on the interests and identities of relevant social actors, which is argued to produce the persistent variety in criminal justice which, notwithstanding globalization (Lacey 2011), we see across systems at similar stages of capitalist development (Sutton 2004; Tonry 2007).

Institutional Approaches Recent comparative institutional analysis (Lacey 2008) builds on the distinction between “liberal” and “coordinated” market economies developed by political scientists Peter Hall and David Soskice (2001). A “coordinated market economy” (CME) functions in terms primarily of long-term relationships and stable structures of investment, not least in education and training oriented to company- or sector-specific skills, and incorporates a wide range of social groups and institutions into a highly coordinated governmental structure. Such a system is argued to be more likely to generate incentives for the relevant decision-makers to opt for relatively inclusionary criminal justice for it is a system which is premised on incorporation and hence on the need to reintegrate offenders into society and economy. Typically, moreover, the interlocking institutions of coordination of the CMEs conduce to an environment of extensive informal social controls; this in turn supports the cultural attitudes which underpin and help to stabilize a moderated approach to formal punishment. A “liberal market economy” (LME) – of which the extreme case, significantly for any argument about criminal justice, is the USA – is typically more individualistic in structure, is less interventionist in regulatory stance, and depends far less strongly on the sorts of coordinating institutions which are needed to sustain long-term economic and social relations. In these economies, flexibility and innovation, rather than stability and investment, form the backbone of comparative institutional advantage. It follows


C

469

Imprisonment per 100,000

Homicide per 100,000

2002/3

2006

2009/10

2011

2008

701 155 141 115 128

737 186 148 125 140

743å 203 154 133 153

199 155 155

5.2 1.3 1.2 1.2 2.2*

100 98 73 58 70 58

128 94 82 77 75 60

94 88 78 71 61 72

87 74 59 73

1* 0.8 0.9 1.4* 2.5 0.6*

Liberal Market Economies USA New Zealand England & Wales Australia Scotlanda Co-ordinated Market Economies Netherlands Germany Sweden Denmark Finland Norway å

Prison rate for 2009 * Homicide rates for 2007 a Prison rates for Scotland are estimated on the basis of average daily prison population for 2002/3, 2006/7, 2009/10 Adapted from: Hall and Soskice (2001), Barclay and Tavares (2003), Cavadino and Dignan (2006), Hall and Gingerich (2004), International Centre for Prison Studies (accessed 2011). European Sourcebook of Crime and Criminal Justice (2010); UNODC (accessed 2010); General Register for Scotland (accessed 2011); Prison Statistics Scotland (accessed 2011)

Comparative Political Economy of Punishment, Fig. 1 Punitive measures in cross-national comparison

that, particularly under conditions of surplus unskilled labor – conditions which LMEs are also more likely to produce – the costs of a harsh, exclusionary criminal justice system are less than they would be in a CME (Fig. 1). Political Economy, Imprisonment, and Homicide This institutional approach sees the distinction between different varieties of advanced capitalist political economy as a powerful tool in building an understanding of the inclusionary and exclusionary dynamics of criminal justice systems. For, particularly as developed in the most recent political economy literature, which explores the relationship between the regime of economic production and labor market institutions and a cluster of social and political institutions, the distinction has an analytic reach into a wide range of political and economic variables. The various factors which characterize coordinated and liberal systems are closely intertwined: the number of variables further implies that there may be

paradigm and less central cases of each type of system. The features of the LME/CME distinction of greatest relevance to criminal justice are as follows: 1. The structure of the economy: production regimes, labor markets, education and training, and disparities of wealth The departure point for an analysis of two main varieties of capitalism – an individualistic liberal model whose comparative advantage lies in flexibility and a coordinated model in which comparative advantage depends on long-term investment in workers and their skills – lies in the changing nature of economic relations. In Britain and the USA, for example, the dynamics of an LME have accelerated markedly over the last 30 years, as many of the attitudes and values which sustained, respectively, the postwar welfare state settlement and the Great Society programs have come to be eroded by a more aggressively market-oriented culture (Reiner 2007). This culture is itself premised

C

C

470

in part on the imperative of high performance amid increasing global economic competition, with the collapse of Fordist production regimes and the availability of cheap manufactured goods from countries like South Korea, China, and India. The inevitable upshot is structural economic insecurity for low-skilled workers. In a short-term economic culture, the bottom third of the workforce risks becoming a socially as well as economically excluded group. In the CMEs, by contrast, a longer-term economic culture appears to have survived increased international competition and the collapse of Fordism. Within the political economy of comparative advantage, this is seen as a function of several interlocking factors: the nature of the economic activities in which these countries have concentrated their efforts, the close incorporation of employers as well as unions in the management of the economy, and the implications of each of these factors for the structure of education and training. Unlike the increasingly flexibilized LMEs, many CMEs excel in producing high-quality goods which depend on industry-specific, nontransferable skills. In this context, employers have strong reason to invest in education, training, and apprenticeship systems. They also have strong reason to use their considerable bargaining power with government to press for generous welfare provision for workers who are temporarily unemployed but whose skills remain necessary to the economy. With the higher levels of investment in education and training typical of these economies, which also demonstrate lower disparities of wealth and higher literacy rates (Iversen and Soskice 2009), the costs of pursuing socially exclusionary policies in areas such as criminal justice are relatively high. This implies that many of the general theories of increasing penal severity are based on an account primarily applicable to LMEs as opposed to CMEs, whose high-skill production regimes were less strongly affected by the collapse of Fordism. In the LMEs, increasing relative deprivation consequent on flexibilization of labor markets


and growing disparities of both income and skills pose a huge challenge for inclusionary criminal justice policies (Reiner 2007, Chap. 3), particularly in a world in which mass communications and increased levels of education imply the cultural inclusion of the relatively deprived within the individualistic values of a consumer society from which they are economically excluded. What is more, this relative deprivation has increased along a number of dimensions, accentuating both differences between the richest and the poorest and the difficulty of moving from a position in the bottom one third or so into the relatively advantaged majority (Young 1999). In this context, prisons have become a mechanism for disciplining those excluded from the legitimate economy. During this period there has been a large increase in the absolute and relative size of the harsher end of the American and, to a lesser extent, British criminal justice systems. In England and Wales, the imprisonment rate virtually doubled – from 79 to 152 per 100,000 – between 1970 and 2011 (http://www.prisonstudies.org/info/worldbrief/ wpb_country.php?country¼169 accessed August 25th 2011), while in the USA, it quintupled – from 153 to 743 per 100,000 – between 1974 and 2009. This is reflected not only in the scale of imprisonment and policies such as mandatory minimum sentences but also in a weakening of political sensibilities in favor of human rights and decent conditions for prisoners. 2. Political systems: electoral arrangements and the bureaucracy The analysis of two systematically different variants of contemporary capitalist economy deploys not only economic variables but also a variety of other institutional variables. Among the most important of these for criminal justice is electoral politics. Given the increasing salience in many countries of criminal justice to politics and the force of electoral discipline on democratic governments, the relationship between popular attitudes to crime and punishment and the political system constitutes an important


variable in any attempt to understand the differences between contemporary penal systems in relatively similar societies. This is the case even if popular attitudes conducing to fear of crime and to penal severity are on occasion stimulated by government rhetoric and policy (Beckett 1997; Roberts and Hough 2002). Whatever the causation here – it seems likely that it moves in both directions – once certain popular attitudes and expectations are created, they in turn create significant electoral constraints. How directly they are registered in the electoral system (Barker 2009; Miller 2008), and hence exert discipline on governing parties, is therefore likely to be an important factor in explaining the institutional capacity of different systems to sustain moderate criminal justice policies. In this context, it is significant that there is, empirically, an association between CMEs and proportionally representative (PR) electoral systems, whereas LMEs tend to have first-past-the-post, winner-takes-all systems (the key exceptions being New Zealand and Scotland: Lacey 2012). The policy-making autonomy of CMEs is constrained by the need to negotiate with groups incorporated in the governmental process. CMEs with PR systems are more oriented to effective participation in and contribution to policy making – at least for groups integrated within subsisting sociopolitical structures – than are LMEs whose electorate gets a one-shot say in policy making at election time. But this consensusbuilding dynamic may make the CMEs less heteronymous in the light of swings of popular opinion. While decisive winners of first-pastthe-post elections in LMEs may feel relatively unconstrained by popular opinion early on in their terms, their unmediated accountability at the ballot box makes them highly sensitive to public opinion as elections loom. What is more, as party affiliations among the electorate weaken, governments’ increasing dependence on the approval of a large number of “floating” or reluctant voters, who regard crime as a threat to their well-being, may feed into the political salience of criminal

471

C

justice. Under the conditions attendant on the collapse of Fordism since the 1970s and in the light of the salience of increasing relative deprivation to the scale and the perceived seriousness of crime problems, there may thus be a stronger association between the politicization of criminal justice and the impact of penal populism in majoritarian, two-party LMEs such as the USA, with decisive implications for the harshness of punishment. In this context, the further empirical fact that PR-based CMEs are more likely to elect left of center governments and to display lower disparities between the best and worst off is significant (see Iversen and Soskice 2006). Electoral structure, in other words, has implications for both partisanship and the substance of political, social, and economic outcomes. There is further reason to think that this difference in the electoral structure of LMEs and CMEs may have some important implications for upward pressure on punishment. As many commentators have observed, one of the developments which has fed the trend toward penal populism in several LMEs is the emergence of well-organized single-issue pressure groups, notably those representing victims of crime (Pratt 2007, Chap. 3; Gottschalk 2007, Chaps. 5–7). On the face of it, one would expect that single-issue political groups would find it harder to get their voices heard in a majoritarian, two-party system than in a PR system which incorporates a number of smaller parties. This expectation would not apply, however, where a particular single issue appeals widely to floating or reluctant voters. Such has been the case of crime in the USA over the last 30 years. As Simon (2007, pp. 26ff, 159ff) has shown, both the popularity of harsh criminal policy among median voters and the relative simplicity of enacting such policy – the lack of need, for example, to develop complex new bureaucracies to administer or implement increased criminalization – have proved a potent temptation to US politicians and other elected officials. Furthermore, while concerns about crime reflected in victims’ movements may well find

C

C

472

a footing in the PR environment, particularly among smaller parties who may hold the balance of power, their adoption and implementation will ultimately have to be negotiated out in the complex bargaining process typical of PR systems and will hence tend to be more insulated from the dynamics of emotive campaigns than is typical in the majoritarian systems. What is seen in the latter is a cycle of mutual reinforcement, grounded in a set of incentives conducing to politicians’ focus on single issues such as criminal justice which are, superficially, easy to demonstrate that they have acted upon, such action in turn leading to heightened public identification of the salience of crime problems and to heightened expectations of governmental capacity to resolve them through tough criminal policy. This dynamic is especially strong where both major parties adopt a tough stance on law and order (Newburn 2007). The scale and impact of this sort of cycle is vividly illustrated by the fact that, in addition to its unplanned expansion of the prison system, the British Labor government has been estimated to have enacted no fewer than 3,000 additional criminal offences in the 8 years between its election and the end of 2005 (Morris 2006). It is also worth noting a further difference between the political systems to be found in LMEs and CMEs, itself correlated with the PR/majoritarian distinction. In most CMEs, deference to the expertise of the professional bureaucracy – that is, the civil service, often including not only policy advisers, penal system officials, and prosecutors but also judges – tends to be higher than in LMEs. This is in part because the coalition politics typical of PR systems imply a less polarized political environment in which governments feel less need to retain total control of policy making. By contrast, particularly in recent years, the tendency in majoritarian systems has been for governments to prefer to work with their own, politically appointed advisers, and to ignore the advice of technically neutral civil servants wherever this interferes with political expediency. Particularly in the UK, the


increasing domination of parties by their leaders has fed this dynamic. This not infrequently leads to a situation in which pressure groups such as those representing victims of crime are constructed as the relevant “experts” for the purposes of consultation in the development of policy, weakening both an important constraint on ad hoc policy making and coordination with criminal justice professionals (Pratt 2007, Chap. 3; Simon 2007, Chap. 3). In LMEs, this feeds into a dynamic in which politicians’ decisions become ever less insulated from the flow of perceived public opinion – a factor which has been a crucial driver of penal harshness in several countries. In CMEs, by contrast, the strength of a professional bureaucracy, along with deference to expertise, have been identified by a number of scholars as conditions key to the maintenance of moderate criminal justice policies (Cavadino and Dignan 2006, pp. 35–36; Savelsberg 1994, 1999; Tonry 2007, pp. 31–32). 3. The welfare state Another key difference between contemporary capitalist democracies lies in the institutions associated with the welfare state (Esping-Andersen 1990; 1996). Here again, political economies at relatively similar levels of development, characterized by broadly liberal-democratic political structures, have taken markedly different paths. In terms of Esping-Andersen’s famous typology, while countries with liberal welfare regimes like the USA and Britain have adopted neoliberal policies committed to “rolling back the state” and curtailing public expenditure (Hillyard and Tombs 2005), the Nordic, socialdemocratic countries have maintained their welfare states more or less intact (Pratt 2008a; Pratt 2008b), with Northern European countries such as Germany adopting a pattern closer, in terms of generosity of provision and scope of coverage, to their Nordic than to their British neighbors. Among variables in political-economic structure, the welfare state has received most sustained attention from comparative criminal


justice scholars. It is obviously plausible that the impact of relatively generous welfare provision on the reduction of poverty has a knockon effect on crime. Less obviously, there is evidence that it is also associated with levels of punishment. Downes and Hansen (2006) have shown, in a study covering 18 countries, that those spending a higher proportion of their GDP on welfare have lower imprisonment rates – a relationship which has grown stronger over the last 15 years. Similarly, Beckett and Western (2001, pp. 44, 48, 55; Western 2006) have demonstrated systematic differences among states within the USA, in which those with relatively low social welfare spending are also those with higher prison populations. But the precise causal mechanisms here are unclear. Sometimes the argument takes a cultural form: the inclusionary instincts represented in generous welfare state policies are likely to be reflected in criminal justice policy. Beckett and Western (2001), for example, argue that welfare regimes vary according to their commitment to including or excluding marginal groups: the more inclusive systems exhibit both higher welfare spending and lower imprisonment rates. Albeit making some observations about the relationship between the welfare state and the structure of the economic and political systems, Pratt (2008a), in his recent analysis of “Scandinavian exceptionalism,” attributes the generosity of the Scandinavian welfare systems primarily to a “culture of equality” with long historical roots. But what explains this varying commitment to egalitarianism and inclusion and underpins the political support necessary to sustain it? Clearly, long-standing institutional arrangements are typically articulated with, and stabilized by, distinctive cultural attitudes such as the strong Scandinavian commitment to social solidarity and equality (Bondeson 2005, p. 189). But there is also evidence that the distinctive structures of welfare states are articulated with the political and economic dynamics already discussed. In other words,

473

C

it may be not just possible, but good economic sense for some countries to maintain generous welfare provision even in the face of increasing competition from countries who are not investing public resources in this way. Certain economic arrangements, in short, themselves foster a culture of solidarity or support for the welfare state – a culture which is in its turn important in sustaining the political support needed to sustain generous welfare institutions. A range of explanations has focused on precisely such an articulation of the welfare state to the structure of the economy. Within a liberal market system in a flexible economy, governments have chosen to maximize incentives to rejoin the labor market – a strategy that has had sufficient plausibility with a critical mass of the electorate because of the high degree of transferable skills within the workforce. Within the labor markets of countries with less flexibility, where long-term investment in less transferable skills (as in Germany), or an extensive public sector providing employment for women and services for dual-career families (as in the Nordic countries), is still key to comparative advantage, it makes sense to give relatively generous support to workers who experience periods of unemployment, rather than encouraging them to retrain or to find work in new sectors of the economy. Generosity of welfare provision and relatively secure employment relations are, under certain conditions, just as good a basis for economic success and stability as flexibilization and welfare cuts. And though there has been some recent reduction of welfare benefits in several of the European and, to a lesser extent, Nordic countries, with generosity of provision in the countries like Germany depending heavily on status as an insider to the high-skill economy, the remaining differences between European welfare regimes remain significant and seem unlikely to be eroded in the near future (Hall 2007, p. 39; Iversen 2007, p. 278). 4. Constitutional structure: decision-makers, veto points, and constraints on criminalization

C

C

474

Beyond political and economic institutions, the constitutional structure which provides parameters for the institutional environment already considered, and for the legal system through which much criminal policy is implemented, may also be important in explaining national differences in criminal justice policy. Three such variables might be thought to entail systematic differences in countries’ capacities to develop and sustain moderate penal practices. These are the distribution of decision-making power among different actors; the structure of legal institutions, and in particular of the tenure and selection of the judiciary and prosecutors; and the impact of constitutional framework on the definition of crime. The discussion of political systems above indicated that one important variable in shaping the reception of popular concern about crime into criminal justice policy is the degree to which political decision-making is insulated from the flow of electoral opinion. But a further factor seems likely to be important here. This is the distribution of veto points across the system, with consequent potential for checks and balances to be invoked, and for reactive policies to be blocked or delayed, thus leading to a more reflective and negotiated style of policy making. On the face of it, such veto points might be taken to be correlated with federal as opposed to unitary systems and with bi- or multicameral as opposed to unicameral legislatures. But this is not inevitable. For example, in the USA, most criminal justice policy is developed at state level or even at city or county level, so that the federal structure cannot exert the sort of inhibiting power which it has in Germany. And in the bicameral legislature of the UK, the constitutionally privileged place of the House of Commons restricts the degree to which the upper House can act as a block on policy formation. This is not to say that, for example, US federalism has been unimportant to the way in which criminal policy has developed (Miller 2008; Stuntz 2001). The need to address multiple constituencies can, as in the


case of campaigns to abolish capital punishment (Garland 2010), place barriers in the path of criminal justice reform, while the highly decentralized form of US government allows local politicians to emphasize popular issues the costs of which their own constituents will not have to bear (Soskice 2009; Lacey 2011; Stuntz 2011). And the creation of federal structures can itself feed upward trends in punishment: this appears to have been the case in the USA through both direct and indirect (via state-imitation) effects of the federal Uniform Determinate Sentencing Act and, in Europe, in recent EU initiatives arguing for cross-union statutory minima for certain serious crimes (Pratt 2007, p. 169). There is scope here for a careful empirical analysis of the ways in which systems with more and less diffused structures of criminal justice decision-making have responded to external pressures toward penal severity. A priori, the distribution of veto points and the need to coordinate decision-making points in federal systems where key aspects of criminal justice policy have to be centrally or constitutionally determined would appear to be important variables. The size or scale of the system is also likely to be relevant. In relatively small jurisdictions like the Netherlands or Scotland, both the capacities for central negotiation and the intensity and influence of elite policy networks may also be important factors. They are differences which may favor either the stabilization of moderate policies via coordination between elite networks, as in the 1970s in the Netherlands, or rapid shifts of policy through precisely the same mechanism, as in that country in the late 1980s (Cavadino and Dignan 2006, p. 123). Leaving the question of scale aside, however, the key feature of the long-established PR systems, whose parties tend to represent sectoral interests, is their resulting orientation to negotiation across groups. This is likely to conduce to a wider distribution of veto or delaying points, and hence underpins an association between CMEs and checks on the politicization of criminal justice.


Second, it is plausible that the selection, training, and tenure of judges and other key criminal justice officials will have distinctive implications for the environment in which penal policy is developed and implemented. To take only the most obvious example, a system like that of the USA, in which many judges are elected, is one in which a key barrier between popular demands for punishment is, if not removed, seriously weakened. This becomes important under conditions in which criminal justice is highly politicized. We might draw an analogy here between the election of many US judges and the fact that the vast bulk of criminal cases in Britain are heard by lay magistrates. This arrangement would be unthinkable in the highly professionalized systems of Northern Europe or Scandinavia. While English magistrates are not answerable to popular opinion in the style of elected officials, they are less buffered by a professional expertise and culture. Their place in the administration of criminal justice suggests that relatively low importance attached to expertise may find its roots deep in the history of British sociopolitical arrangements rather than merely being a product of the recent dynamics of criminal policy. In many Northern European and Nordic CMEs, the judiciary continues to be regarded as a key partner in the development as well as the implementation of criminal justice policy. In LMEs like the USA and the UK, by contrast, the rise of penal populism has seen an increasingly hostile and unstable relationship between government and judiciary. The judiciary conceives their independence as inconsistent with any overt incorporation in governmental negotiations, and the government is accordingly inclined to regard the judiciary as an irksome and even irresponsible thorn in the flesh of its criminal policy. Though constitutional or human rights structures such as the European Convention or the US Constitution may provide judges with some tools to resist certain government excesses of punishment or criminalization,

475

C

these have tended to be relatively weak in the face of a determined executive with a clear legislative majority (Lacey 2009). In this context it is interesting that judiciaries in the Anglo-Saxon, common law, liberal market countries with strong traditions of judicial independence appear to have suffered a decline in public status and authority in recent years (see Simon (2007, Chap. 4); while their European, civilian cousins, traditionally of lower status and more intimately linked to the state bureaucracy, appear so far to have escaped a similar fate. Third, systems exhibit markedly variable constitutional constraints on the content of criminal law and punishment. These constraints have deep historical roots. Whereas in most European countries, the process of constitutional modernization entailed a clear differentiation between criminal justice and administrative “police power,” in common law systems such as Britain and the USA, no such differentiation has ever been fully institutionalized (Dubber 2005; Whitman 2003). Legal and constitutional constraints on the substance (as opposed to the procedure) of both criminalization and punishment are, accordingly, weaker in these systems. Again, there is an association between this variable and the LME/CME distinction. 5. Institutional capacity to integrate “outsiders” Another key feature of contemporary societies is the increasing mobility of the social world from the late 1960s on. This mobility has a number of dimensions. In a wealthvaluing culture and flexible economy, with relatively high levels of education, there may be more mobility between social classes; in a globalizing economy characterized by transnational political structures like the EU, marked by relatively cheap international travel and mass communications, there is more geographical mobility. These developments have added new layers of complexity to one of the central challenges for any democratic system of criminal justice: that of “reintegrating” offenders into society and economy.

C

C

476

This is a complex issue. But it is an important corrective to what might the temptation to think that the highly coordinated systems of Europe and Scandinavia today are necessarily better placed to sustain democratically acceptable levels of penal moderation than their liberal market Anglo-Saxon counterparts. For the structure of this problem is significantly different in the two sorts of system. While the laisser-faire and individualistic culture typical of LMEs may well make it relatively easy to integrate geographical or “cultural” “outsiders” like recent immigrants wherever they find access to the labor market, the more intensively group- and skills-based system of the CMEs may well pose significant challenges in terms of integrating newcomers into the representative and decision-making structures which have helped to sustain a relatively moderate criminal justice policy with relatively high institutional capacity for reintegration. CMEs are, in short, good places to be incorporated insiders, but hard systems to enter from the outside (Lacey 2008, Chap. 3).

Future Directions The comparative political-economic analysis of crime and punishment, using the tools of contemporary political science, is at a relatively early stage of its development. But it has opened up an extensive research agenda which promises to refine the understanding of how the dynamics of crime and punishment are shaped by forces external to the criminal process and of how those forces themselves relate to patterns of institutional structure and incentives which vary systematically across national borders. At a time at which much criminal justice scholarship is preoccupied with globalization, institutional approaches grounded in comparative political economy are a telling reminder that vast differences between criminal justice systems of the developed world continue to obtain, as well as a crucial resource for understanding how they are produced and, accordingly, for assessing whether they are likely to persist.


Related Entries ▶ Comparative Courts and Sentencing ▶ Comparative Legal Cultures ▶ Comparing Police Systems Across the World ▶ Historical and Comparative Perspectives on Incarceration ▶ History of Corrections ▶ Informal Social Control ▶ Public Opinion about Sentencing ▶ Sentencing Commissions ▶ Sex Offenders and Criminal Policy ▶ Social Control Theory of Sexual Homicide Offending

Recommended Reading and References Barker V (2009) The politics of punishment: how the democratic process shapes the way America punishes offenders. Oxford University Press, New York Beckett K (1997) Making crime pay. Oxford University Press, New York Beckett K, Western B (2001) Governing social marginality. In: Garland D (ed) Mass imprisonment: social causes and consequences. Sage, London, pp 35–50 Bondeson U (2005) Levels of punitiveness in Scandinavia: descriptions and explanations. In: Pratt J, Brown D, Brown M, Hallsworth S, Morrison W (eds) The new punitiveness. Willan Publications, Cullompton Cavadino M, Dignan J (2006) Penal systems: a comparative approach. Sage, London De Giorgi A (2006) Rethinking the political economy of punishment. Ashgate, Aldershot Downes D, Hansen K (2006) Welfare and punishment in comparative perspective. In: Armstrong S, McAra L (eds) Perspectives on punishment. Oxford University Press, Oxford Dubber MD (2005) Police power. Columbia University Press, New York Esping-Andersen G (1990) The three worlds of welfare capitalism. Polity Press, Cambridge Esping-Andersen G (1996) Welfare states in transition. Sage, London European Sourcebook of Crime and Criminal Justice Statistics – 2010, fourth edition (4th ed. Den Haag: Boom Juridische uitgevers. Onderzoek en beleid series, no. 285, Ministry of Justice, Research and Documentation Centre [WODC]). www.europeansourcebook.org/ ob285_full.pdf Garland D (2001) The culture of control. Oxford University Press, Oxford Garland D (2007) High crime societies and cultures of control. In: Ostermeier L, Paul B (eds) Special Issue, Kriminologisches Journal

Comparative Political Economy of Punishment Garland D (2010) Peculiar institution: America’s death penalty in an age of abolition. Oxford University Press, New York General Register Office for Scotland. Revised Mid-year Population Estimates 1982–2000. www.gro-scotland. gov.uk/statistics/theme/population/estimates/mid-year/ revised1982-2000.html Gottschalk M (2007) The prison and the gallows. Cambridge University Press, Cambridge Hall PA (2007) The evolution of varieties of capitalism in Europe. In: Hancke˙ B, Rhodes M, Thatcher M (eds) Beyond varieties of capitalism. Oxford University Press, Oxford Hall PA, Soskice D (2001) An introduction to the varieties of capitalism. In: Hall PA, Soskice D (eds) Varieties of capitalism. Oxford University Press, Oxford, pp 1–68 Hillyard P, Tombs S (2005) Towards a political economy of harm: states, corporations and the production of inequality. In: Hillyard P, Pantazis C, Tombs S, Gordon D (eds) Beyond criminology: taking harm seriously. Pluto Press, London, pp 30–54 International Centre for Prison Studies. World prison brief. http://www.prisonstudies.org/info/worldbrief/ Iversen T (2007) Economic shocks and varieties of government responses. In: Hancke˙ B, Rhodes M, Thatcher M (eds) Beyond varieties of capitalism. Oxford University Press, Oxford Iversen T, Soskice D (2006) Electoral Institutions and the politics of coalitions: why some democracies redistribute more than others. Am Polit Sci Rev 100:165–181 Iversen T, Soskice D (2009) Distribution and redistribution: the shadow of the nineteenth century. World Polit 61(3):438–486 Lacey N (2008) The prisoners’ dilemma: political economy and punishment in contemporary democracies. Cambridge University Press, Cambridge Lacey N (2009) Historicising criminalisation: conceptual and empirical issues. Mod Law Rev 72(6):936–960 Lacey N (2011) Why globalisation doesn’t spell convergence: models of institutional variation and the comparative political economy of punishment. In: Crawford A (ed) International and comparative criminal justice and urban governance. Cambridge University Press, Cambridge Lacey N (2012) Political systems and criminal justice: the prisoners’ dilemma after the coalition. Current Legal Problems 2012. doi:10.1093/clp/cus002 Lacey N (forthcoming 2012) Punishment, (Neo-) liberalism and social democracy. In: Simon J, Sparks R (eds) The sage handbook of punishment and society. Sage Publications Miller LL (2008) The perils of federalism: race, poverty, and the politics of crime control. Oxford University Press, New York Morris N (2006) Blair’s “Frenzied Law-Making”. The Independent 16 August

477

C

Newburn T (2007) “Tough on Crime”: penal policy in England and Wales. In: Tonry M (ed) Crime, punishment and politics in comparative perspective: crime and justice: a review of research, vol 36. University of Chicago Press, Chicago, pp 425–470 Pratt J (2007) Penal populism. Routledge, London Pratt J (2008a) Scandinavian exceptionalism in an era of penal excess: part I the nature and roots of Scandinavian exceptionalism. Brit J Criminol 48: 119–137 Pratt J (2008b) Scandinavian exceptionalism in an era of penal excess: part II does Scandinavian exceptionalism have a future? Brit J Criminol 48:275–292 Reiner R (2007) Law and order: an honest citizen’s guide to crime and control. Polity Press, Oxford Roberts J, Hough M (eds) (2002) Changing attitudes to punishment: public opinion, crime and justice. Willan Publishing, Cullompton Rusche G, Kirchheimer O (1969) Punishment and social structure, (first published, in German, 1939). Russell Sage, New York Savelsberg JJ (1994) Knowledge, domination, and criminal punishment. Am J Sociol 99:911–943 Savelsberg JJ (1999) Knowledge, domination and criminal punishment revisited. Punishm Soc 1:45–70 Simon J (2007) Governing through crime: how the war on crime transformed American democracy and created a culture of fear. Oxford University Press, New York Soskice D (2009) American exceptionalism and comparative political economy. In: Brown C, Eichengreen B, Reich M (eds) Labor in the era of globalization. Cambridge University Press, New York, pp 51–93 Stuntz WJ (2001) The pathological politics of criminal law. Mich Law Rev 100:505–600 Stuntz WJ (2011) The collapse of American criminal justice. Harvard University Press, Cambridge Sutton J (2004) The political economy of imprisonment in affluent western democracies, 1960–1990. Am Sociol Rev 69:170–189 The Scottish Executive. Statistical Bulletin Criminal Justice Series (August 2007) CJr/2007/7. www.scotland. gov.uk/Resource/Doc/196743/0052707.pdf The Scottish Government. Statistical Bulletin Criminal Justice Series (20 December 2010). www.scotland. gov.uk/Publications/2010/12/15154652/24 Tonry M (2007) Determinants of penal policies. In: Tonry M (ed) Crime and justice: a review of United Nations Office on Drugs and Crime. UNODC Homicide Statistics. www.unodc.org/unodc/en/dataand-analysis/homicide.html Wacquant L (2009) Punishing the poor: the neoliberal government of social insecurity. Duke University Press, Durham Western B (2006) Punishment and inequality in America. Russell Sage, New York Whitman JQ (2003) Harsh justice. Oxford University Press, Oxford Young J (1999) The exclusive society. Sage, London

C

C

478

Comparative Punishment ▶ Institutional Theories of Punishment

Comparing Police Systems Across the World Rob I. Mawby University of South Wales, Newport, South Wales, UK Department of Natural Sciences, University of Gloucestershire, Cheltenham, UK

Overview Until comparatively recently, most of the writing – and indeed research – on the police had focused on the situation in the United States and England and Wales. Indeed, it was only with David Bayley’s (1985) seminal text that any systematic crossnational approach was adopted. This lack of an international comparative dimension was surprising, given the centrality placed on comparative analysis by sociologists and political scientists. It was also a notable omission given that senior police administrators across the world, such as Raymond Fosdick in New York and Sir Charles Jeffries, a British colonial official, had long been aware of differences in the nature of police systems between societies and used their awareness to adopt new ways of policing (Mawby 1990). Practitioners are still involved in international comparisons. However, more of the recent contributions have been made by academics. International comparative research has become an essential ingredient of the established academic agenda of police studies. One method of so doing, of which the following entries are excellent examples, is to compare specific policing issues cross-nationally. Another is to compare different police systems. However, ironically, the very fact that police systems and policing methods are discussed so regularly though the international – academic and practitioner – police families lends

Comparative Punishment

weight to the argument that in postmodern societies, convergence has led to a reduction in the contrasts between different police systems. Having previously compared Anglo-American police systems with four other models, the continental, communist, colonial, and Far Eastern systems (Mawby 1990, 1999a, 2011a), the issue is revisited here. Specifically, this entry considers how far the continental, communist, and colonial systems, all examples of control-dominated systems, are still relevant to a contemporary critique of policing across the world.

Models of Alternative Police Systems As other entries in this volume testify, there is a distinction between policing – as a process – and the police as an institution that might be responsible for many other services that are only tenuously related to maintaining order or preventing crime. Focusing on the police as a state-based organization, the nature of “the police” varies markedly between countries and over time (Bayley 1985; Findlay and Zvekic 1993; Mawby 1990, 1999a, 2011a). Distinctive models can be distinguished in terms of police legitimacy, structure, and function. Legitimacy implies that the police are granted special authority by those in power, whether this is an elite within the society, an occupying force, or the community as a whole. Structure implies that the police is organized, with some degree of specialization and with a code of practice within which, for example, the extent to which use of force is legitimate is specified. However, the extent of organization or specialization, and the types of force considered appropriate, will vary. Finally, function implies that the role of the police is concentrated on the maintenance of law and order and the prevention and detection of offenses, but there might be considerable differences in the balance between these and in the extent to which other duties are assigned to the police. On these three criteria, the public police model that emerged on mainland Britain and the USA in the nineteenth century differed from the

Comparing Police Systems Across the World

centralized, autocratic arm of state authority that preceded it on continental European. Equally, a control-dominated system can be identified with traditional policing in the colonies established by Britain and its European neighbors and communist Europe. However, there are, arguably, as many differences between countries within a model as there are between alternative ideal types, a point made forcefully by Anderson and Killingray (1991, 1992) in the context of a colonial police system. Equally, there are often variations within a country. For example, in Canada marked variations exist between the centralized RCMP and local urban and provincial police. The Canadian example is a useful one, because it challenges the assumption that the control-oriented models cited above are categorically different from Anglo-American police systems. While the latter might be identified with a democratic, community-oriented police system, the task of assigning any specific police system to a community-oriented model is contentious. Although this may be the type of democratic policing that many aspire to, and Western democracies have been keen to influence such police developments elsewhere (Bayley 2006; Marenin 1998; Pino and Wiatrowski 2006), it is difficult to nominate any one country as even approaching achieving it (Brogden 1999; Mawby 1990). This is evident if we disaggregate the core components of a community-oriented system. Such a police system is one where the main function of the police is to provide a public service that addresses the wider needs of the community. Maintaining order is important, but the emphasis is more on crime as symptomatic of community problems than as an affront to authority. Such a model assumes that the police are accorded considerable legitimacy by local communities. The police are consequently generally organized and managed locally, and barriers between police and public are minimized. Community policing and problem-oriented policing typify this approach. Anglo-American police clearly fall short of these ideals. In England and Wales, for example, often eulogized as the home of community

479

C

policing, the modern police system emerged at least in part as a means of maintaining order in the midst of working class protest and was brutally deployed following the first World War to break the national miners’ strike, a strategy revisited when Margaret Thatcher’s government used the police to break the miners’ strike in the early 1980s (Fine and Millar 1985). And in many cases, police were recruited from rural areas to work in the cities, undermining the claim that they were local citizens in uniform. In the USA, where police systems have been traditionally locally based, personnel have been recruited locally, and officers have engaged in a wide range of “noncrime” responsibilities; the image of the police as a militaristic body charged with fighting the “war” against crime is equally pervasive. Increased militarization of the police to “fight” the “war” against terrorism has reconfirmed this, as Belur notes in this volume. Elsewhere, Bayley’s (1991) early presentation of the Japanese police as community based and welfare oriented has been questioned (Aldous 1997; Leishman 1999; Miyazawa 1992). It may, therefore, be that the key strength of specifying a community-oriented model is as an ideal type, in the Weberian sense, so as to better evaluate police systems and changes within them. In stark contrast to a community-oriented system, a control-dominated system is one where the main function of the police is to maintain order and where the population generally fails to recognize the legitimacy of the state and its agents, the police. In such societies, the police may carry out a range of administrative tasks on behalf of the state but rarely provide a public service that addresses the welfare needs of the community. The police are, consequently, generally organized and managed centrally and have many paramilitary qualities. In some cases, the distinction between police and military is negligible. Nevertheless, it can be argued that the democratization of postcommunist and postcolonial societies and the cross-national interchange of ideas within the international police family have resulted in police structures and methods from one country being imported to others, leading to convergence. It is this argument that is more fully

C

C

480

discussed in this entry. The focus is on three police models that have traditionally been considered control oriented: the European continental system, communist police, and the colonial system. In each case, the original model is described before moving on to consider how specific police systems have changed and how far democratic, community-oriented policing is more evident.

The Continental Model: The Greek Police Example Discussions of an alleged continental European policing system have a long history, and Fosdick’s (1969) account of continental police at the beginning of the twentieth century is the first of many attempts to identify key characteristics of the police systems of continental Europe (Mawby 1990). In terms of function, the role of the police in continental societies has traditionally tended to be wide-ranging, with a particular emphasis upon political control, termed “high policing” in the French context, in addition to crime control. Continental systems have also been associated with a range of administrative responsibilities, with relatively less emphasis on welfare or service functions. However, there is a marked difference between the French, Italian, and Spanish police on the one hand, as classic examples of this ideal type, and their counterparts in countries such as the Netherlands or Scandinavia. The need for strong policing might imply that continental police systems would also be characterized as centralized and paramilitary, but this was not always the case. For example, Iceland and Switzerland have, respectively, district and canton-based systems, and the Netherlands reorganized in 1993 into 25 regional forces (Jones 1995). More characteristic of the traditional continental model is a structure where one centralized, militaristic force is counterbalanced by either a second or by local city forces. The French are traditionally identified with this model, where the national Gendarmerie-covered rural areas and urban


areas were policed initially by local forces and latterly by the Police Nationale; the maintenance of at least two police forces allowing governments to ensure that no one institution achieved too much power. While the police in most continental countries carry firearms, it is also the case that in many countries, there is at least one centralized force that evidences significantly greater militaristic qualities. In France, for example, the Police Nationale traditionally came under the Ministry of the Interior, whereas the Gendarmerie has been a military force under the Ministry of Defence, with a two-tier entry system, barrack accommodation, and impressive armaments. In the past, continental police systems were also distinguished in terms of their lack of public accountability, being directly responsible to the head of state. While this is less easily reconciled with the liberal democracies of postwar Europe, it is still the case that public accountability is more restricted in countries where the police are more centralized and militaristic. Moreover, the move toward democracy does not inevitably bring with it a more democratic police system. The Greek situation well illustrates this (Mawby 2011b). An assessment of the Greek police as it developed after gaining independence from the Ottoman Empire in 1830 suggests that the system was steeped in the continental European tradition. This was partly the result of external influence (especially from Britain, France, and Russia), partly the law and order problems prioritized internally. The Chorafylake, formed in 1833, was modeled on the French Gendarmerie and charged with the task of supporting the army in protecting the fledgling state from insurrection. It remained the major policing agency for the next 150 years, adapting and being reformed to meet new challenges perceived to threaten public order. For example, it was also used to counter the emerging “threat” of communism and to assert Greek sovereignty in newly “acquired” territories (Rigakos and Papanicolaou 2003). However, the prioritization of high policing meant that the Chorafylake seemed unwilling or unable to deal with conventional crime problems, most notably


in the expanding cities. As a result, municipal police forces were created in the late nineteenth century. They were disbanded in 1906 but then in 1920, a national force was reconstituted to cover the main cities and the island of Corfu. Known as the Astynomia Poleon, it was conceived as a civilian force, modeled on the London Metropolitan Police, with its own training system and a pay structure designed to present it as a more professional alternative to the Chorafylake. To a certain extent, the establishment of separate forces, a paramilitary Gendarmerie to police rural Greece and a civilian urban equivalent, paralleled developments in many other European countries. Both forces were armed and explicitly conservative and anticommunist, the latter tendency being reinforced in the immediate period after World War 2 as the police were positioned as a force against communism, both internally and externally, and specialist units, the Hellenic National Intelligence Service (KYP) and the LOK Special Forces, were created with alleged supported from the US. LOK was actively involved in the 1967 coup that led to the military dictatorship of the colonels (1967–1974) and a further shift toward paramilitary policing with the creation of the Greek Military Police. The return to a democratic system in the 1980s and the creation of a civilian police system thus emerged from the foundations of a centralized and militaristic police, mandated to uphold right-wing governments, with little or no broader accountability. The structure of the new police was encapsulated in the 1984 Act with the Chorafylake and Astynomia Poleon merged into the new Hellenic Police. Unlike in postcommunist societies (see below) though, there was no dramatic change to police personnel. However, as Rigakos and Papanicolaou (2003: 286) note, this new police system was intended to become a “genuine social service.” But the ensuing structure contained a hybrid mix of its forbearers, with innovations constrained by political and social expediency. Thus, while it was answerable to a new Ministry of Public Order, the Hellenic Police continued the centralized, militaristic tradition of the Chorafylake, being justified with reference

481

C

to the universal mantra of a “war on crime” and the only slightly less common assumption that the military model was a safeguard against corruption. In one sense, any justification for such a model appears weak. Thus, Greece had, and continues to have, a relatively low crime rate. The International Crime Victim Survey (ICVS), which included Greece for the first time in 2003/2004, ranked its rate for most crimes as well below average, albeit levels of fear were higher than in most other countries (van Dijk et al. 2008). It was, rather, the political threat, both communist and Turkish inspired (and more recently riots fuelled by the financial crisis), and concerns over crossborder contagion (illegal immigration and drug trafficking) that have been used to justify the need for strong policing. Reflecting such concerns, a number of developments in the past 20 years have been aimed at toughening security responses, including: • The reformulation of the Special Suppressive Anti-Terrorist Unit (EKAM), originally introduced in 1978, when the Hellenic Police was created • The creation of the Border Guard Police Service (BGPS) in 1998 to tackle illegal immigration • The formation of the Special Guard Service (SGS) in 1999 to protect sensitive sites from terrorist attacks • The introduction of Criminality Prevention and Repression Squads (CPRS), with a particular focus on Romas and illegal immigrants • The creation of the Department of Police Special Controls (DPSC) in 1995 with Special Controls Squads (SCS), mainly active in areas of social deprivation There have, of course, been countertrends. Nevertheless, the Greece police has maintained its tradition as a powerful, centralized, and militaristic organization. It is against this backcloth that we need to contextualize the response to incidents such as the fatal shooting of Alexis Grigoropoulos in December 2008 that provoked riots both at the time and a year later and the confrontations on the streets of Greece’s major

C

C

482

cities following the financial crisis in the Spring of 2010 and subsequently. More broadly, it is illustrated in the feelings of the Greek population, expressed in the ICVS, where the Greek police rate poorly (van Dijk et al. 2008).

Moving Out of Communism: The USSR and the New Europe The claim that there is, or was, a distinctive communist police system has also been the subject of debate. For example, it is plausible to argue that it was in many ways conceived out of the police models of continental Europe. There are also clear differences between the police systems of Russia, the “colonial power” of Central/Eastern Europe, and the People’s Republic of China (PRC) (Mawby 1990). As elsewhere, then, the extent to which countries’ police systems can be categorized into an ideal type is influenced by a host of variables such as social structure, prevailing culture, the influence of other countries and, in the case of communist countries, the nature of the previous regime and the distinctive form taken by the revolution. The emergence of communism took starkly different paths in the USSR and PRC (Kowalewski 1981). While in China, decades of civil war culminated in the creation in 1949 of a rural-based popular government that was dependent upon peasant support and saw the Mass Line as a mechanism for forging conformity; in the USSR the revolution of 1917 was all but completed by 1920, leading to minority Bolshevik control of a country where grassroots support for the new regime was minimal. These conditions therefore paved the way for a highly centralized economy and state, with control exerted through a party elite and where the former Tsarist Secret Police provided a model for the new secret police, the cheka, which operated above the centralized militia. Clearly a major role of the police was to deal with crime. However, the extent to which crime control was the main role of the police is questionable, with the police playing a key political role in maintaining the regime. The cheka’s


mandate was to control the border and to prevent internal counterrevolutionary activity, a mandate it accepted with such ruthlessness that it was disbanded in 1922, although its replacements – the GPU, MGB, and, later, the KGB – were also counterintelligence organizations with considerably more power than the militia. Meanwhile the militia played a major part in political control through the regulation of passports and identity cards, through which it had a direct influence on restricting population movement. The political role of the police was not the only way in which the functions of the police were wide-ranging. However, whereas in China, these included welfare and social service roles; in the USSR the emphasis was on administrative responsibilities. The centralized nature of policing was also evident. Any degree of local autonomy that existed in theory appears to have been purely fictional. Furthermore, the USSR’s militia was essentially militaristic, being routinely armed with a rank structure equivalent to the military and recruits drawn from either the party or the armed forces. This acknowledgement of party membership was crucial to an understanding of the legitimacy of the police in communist societies. The interrelationship of party and police was perhaps the most significant feature of the system. In both the USSR and PRC, the police included a disproportionate number of party members, but in addition the party independently exerted considerable influence on police practices. In the USSR, with an early emphasis on a written constitution, the militia was in theory accountable to the law, although the security forces have traditionally operated above the law. However, the influence of the party on the entire criminal justice system meant that the concept of separation of powers was entirely absent. Similarly, attempts to involve the public in the policing process seemed to result in either additional party control through the creation of police aides (at best) or a complex informer system (at worst). Clearly the emphasis changed as the balance of power shifted and new priorities gained ascendancy. For example, Lenin’s preference for apportioning some power to factory committees


illustrated an example of locally based crime control that was dismantled in Stalin’s ruthless pursuit of a totalitarian central regime, but reemerged under Khrushchev, with the introduction of comrades’ courts, the campaign against Parasites and the People’s Guard. The police systems of Warsaw pact countries were closely modeled on that of the USSR and thus unsurprisingly incorporated political and administrative responsibilities, a strong centralized secret police and a centralized, militaristic uniformed police. Being accountable to the party, there was little popular mandate. In postcommunist societies they were commonly acknowledged to have been “repressive” systems (Mawby 1999b). It is also notable that during the mid-1980s, as these regimes faced more overt public opposition, the repressive political role of the police became more pronounced. In Czechoslovakia, the police were associated with brutality in the period leading up to 1989. In the GDR the police played a key role in protecting the Honiker government in the early 1980s, while in Poland the police attempted to repress Solidarity. Just as political reform in the former USSR and Eastern Europe has been varied, so changes to the police have differed in scale. On the one hand, it appears that change in Russia and many of its former states has been minimal (Pustintsev 2000; Beck and Chistyakova 2002; Galeotti 2003). On the other hand, police reform in some, but by no means all, of its Central/Eastern European “colonies” was accorded priority in the embryonic periods of the new regimes. However, the move from a communist to a democratic police system has not been plain sailing in these societies in transition (Zvekic 1996, 1998). Given the importance of the police in sustaining the “old” system, it is scarcely surprising that the new democratic regimes of Central and Eastern Europe should have prioritized changing them, and in the early days of the new regimes, changes to police systems were considered crucial (Mawby 1999b). However, while the most radical changes occurred in East Germany, where its police were absorbed into West Germany’s, elsewhere changes were less radical. Changes to police personnel appear to have been widespread in the

483

C

early phase of development. But the function, structure, and legitimacy of the police have been affected rather less. Some changes have taken place, but they have perhaps not been as radical as was envisaged at the time of the collapse of communism in Central and Eastern Europe. This reflects both the difficulty of radically changing established institutions and the changing priorities in postcommunist societies, where governments have identified an escalating crime problem. Social, economic, and cultural changes added to the pressure on criminal justice systems in general and the police in particular. Economic inefficiency, collapse of living standards, and social dislocation bred a criminogenic environment, and rising levels of crime were well illustrated in official statistics. Victim survey data, while being relatively recent, also suggest that from comparatively low rates under communism crime rates in societies in transition rapidly caught up with those in the West. Certainly, public anxiety increased (Zvekic 1996, 1998; Mawby 1999b), albeit the most recent data suggest that both crime rates and public anxiety have been reduced (van Dijk et al. 2008). The ICVS also suggests that in some former Warsaw pact countries public attitudes toward the police have improved, although they still remain relatively negative. While these factors tended to put a break on police reform, the influence of Western Europe, especially as societies in transition sought EU membership, and the USA is also important. Political alliances with the West, through Interpol, EU membership, and joint training operations have drawn postcommunist societies closer to the West, and especially closer to Western Europe and the USA (Marenin 1998). In the latter case, however, US concern to emphasize the “war” against crime, and especially terrorism and the international drug problem, has taken priority over any emphasis upon democratizing and decentralizing the police.

Colonial Policing: UK and US Colonialism A third police system that has been consistently recognized in the literature is the colonial model.

C

C

484

In many respects it corresponds to the continental model – not surprising given that much of the administrative and legal structure of European states was based on earlier Roman institutions, where the Romans were themselves colonists. This also serves to remind us that the British were not the only colonists (Cole 1999). However, there has been relatively little research on the police of colonial Spain, Portugal, and France or indeed of the colonial model adopted by the USA in its “unincorporated territories,” such as Puerto Rico, which came under US control in 1898 following the Spanish-American War. Focusing on the British Empire, it is arguable that the Westminster government created a different type of police system for its colonies, one that was more appropriate for the control of a subjugated population. The model it used was the one first established for Ireland, where the police could not rely on public consent, which was then introduced, with modifications, throughout Britishcontrolled Africa, Asia, and the Caribbean. Colonial police may be characterized as, in terms of structure, relatively centralized and militaristic (e.g., armed and living as units in barracks), in functional terms, giving more priority to public order tasks, but also having a number of administrative responsibilities, and deriving their legitimacy from their colonial masters rather than the indigenous population. There has, however, been considerable debate over the distinctiveness of a colonial model. On the one hand, Brogden (1987) has argued that the differences between British and British colonial policing have been exaggerated. On the other hand, Anderson and Killingray (1991, 1992) claimed that the differences between colonial systems exceeded the similarities. What is clear is that the British government saw the establishment of a strong police apparatus as central to the establishment of control and legitimacy across its empire. To enforce control, the police were formed according to a militaristic model, although not necessarily armed, nor was the police system necessarily centralized: in India the provinces or states had their own police forces although the Indian Police Service was responsible for the recruitment, training, and deployment


of senior ranks; in Nigeria, at least two forces operated in different parts of the country. However, clearly the police acted on behalf of the British government and had little local mandate, frequently operating with draconian powers. A further illustration of the lack of local influence was the common practice of recruiting staff from either the British military or from elsewhere in the empire, ensuring that the police did not establish close relationships with the indigenous population. The police were part of the administrative structure, and their roles reflected this. They were involved in ensuring that local government ticked over quietly. While crime control might have been important, especially where it involved British nationals as victims, maintaining order and eliminating dissent were pivotal (Arnold 1986). The police were engaged in putting down political protest, including labor disputes, while in Hong Kong, guarding against the threat of communism was central to their mandate (Anderson and Killingray 1991; Travers and Vagg 1993). While crime and disorder issues within the indigenous community received little priority, this meant – ironically – that in some countries, such as Hong Kong, communitarian forms of selfpolicing were tolerated. The fact that the British government experienced similar problems throughout its empire provided an important push toward conformity. This was strengthened, though, with centralized administration and control through the Colonial Police Service, based in London, central training for officers, the formation of the Inspector General of Colonial Police in 1948, and the practice of transferring senior officers between different countries. However, just as there were differences in emphasis between different parts of the empire, so the model shifted at different points in time. In particular, in the conflict building up to independence, which in many cases culminated in armed insurrection, the response of the British government was to accentuate the key features of the model. So, for example, police numbers were increased, central control was strengthened, police arms were improved, and links between police and military were enhanced (Anderson and


Killingray 1992). That this should happen at a time when the legitimacy of the British government and its police was being challenged is scarcely surprising. What is equally important to stress, though, is that it makes problematic the transition from a colonial to a democratic, community-based police system. Moreover, given the inevitability of conflict and disorder after independence, the social control functions of the colonial police came to be valued by the new regimes. In consequence, the strategy, for example, in India (Arnold 1986), seems to have been to replace the (British) officer class but preserve core features of the system. Thus, even with significant political change, police systems often continued as before, an issue illustrated in Northern Ireland (Independent Commission on Policing in Northern Ireland 1999). This is also evident in the Caribbean. In Jamaica, for example, which gained independence from Britain in 1962, the Jamaica Defence Force (JDF) continued the colonial police tradition. In the context of political unrest and corruption, a significant illegal drug trade, high homicide rates, and the establishment of the rigidly defined geographical zones known as garrisons, safe havens controlled by organized crime, paramilitary policing has been evidenced through joint police-military operations and special units such as the Mobile Reserve and the Crime Management Unit (CMU) (Harriot 2000, 2003). In Puerto Rico, despite significant political changes, high crime rates, especially homicide, and the existence of public housing projects that are virtually no-go areas, have been used to justify a continued military-style police system, symbolized through the Mano Dura Contra el Crimen (Hard Hand against Crime) (Dinzey-Flores 2011) and the brutal shooting by a police officer of community leader, Miguel A. Caćeres Cruz in 2007. External intervention, particularly by the USA, has been significant in many parts of the Caribbean, with funding and expertise provided in an attempt to move toward more effective, and possibly more democratic, forms of policing (Bayley 2006; Bowling 2010; Pino and Wiatrowski 2006). However, while Wilson et al. (2011) report some evidence of improvement in

485

C

Trinidad and Tobago, perhaps the key message here is the difficulty of introducing significant changes, particularly in the context of internal and cross-border crime and disorder issues.

Resistance to Change: The Continuation of Control-Dominated Police Systems The evidence discussed here suggests that control-dominated police systems are still evident across the world and that it is still relevant to talk about the continental, communist and colonial models, albeit the key features of these have been modified to varying degrees. The question then arises as to why there has, apparently, been resistance to change. Two broad generators of change can be identified: internal pressure and external influence. Internal changes may be predicated by regime change. This is particularly well illustrated in the case of former Eastern Bloc countries and postcolonial societies but also applies to the Greek example. In each case, the transformation of the traditionally control-dominated police into a community-oriented police was commonly advocated by aspiring leaders. However, a perceived threat to law and order led to a dismantling of police reform agendas, for example, in Eastern Europe (Mawby 1999b), with subsequent reforms less radical than had been anticipated (Beck et al. 2006), while in the case of postcolonial societies, new governments also sometimes retained old police systems in order to establish and assert their authority (Anderson and Killingray 1992). The Greek and Caribbean examples also illustrate the extent to which governments’ “identification” of major crime problems and political threats has been used to justify the continuation of a militaristic police. The broader processes leading up to regime change are illustrated in Fig. 1. During the death throes of colonial and communist societies, regimes under threat acted to preserve the status quo by enhancing the militaristic features of the police, for example, by police-military cooperative ventures, the creation of elite, specialist public order police squads, and increases to police

C

C

486

Comparing Police Systems Across the World Continental

Communist

Colonial

Community policing

Function

Control

Control

Control

Service

Structure

Militaristic Centre/Local

Militaristic Centralised

Militaristic Varied

Civilian Local

Legitimacy

Central government

Party

Occupying power

Local Community

Comparing Police Systems Across the World, Fig. 1 Police systems in transition: the problems of changing the police

hardware. The result was that such systems contrasted even more with a community-oriented system. The possibility of a steady transition from control to community oriented thus faced even more obstacles. In this situation, the importance of external influences becomes crucial. External influence is particularly important in postmodern societies where similar influences are prevalent across national boundaries and where examples of innovative developments in one society are readily available as examples of best practice elsewhere. Formal pacts add a further impetus to change. For example, the emergence and expansion of the European Union has involved greater crossborder cooperation and, consequently, increased pressure toward the harmonization of policy. Allied to this, Europol was ratified in 1999 as the EU organization for cross-border coordination between national law enforcement agencies, providing collation, analysis, and dissemination of information, and a European Police College (CEPOL) was established in 2001. A further source of external pressure involves Western democracies, with their alleged democratic police systems, offering a supportive role in the transformation. As noted above, the USA has been influential in the Caribbean, while it also founded a police college in Budapest, providing training for key staff from societies in transition. The UK has fulfilled a similar role, establishing the Knowhow Foundation to provide support in its former colonies and later extending the scheme to Central and Eastern Europe. However, it is easy to identify tensions here without acceding to Brogden’s (1987) dismissals of the differences between Anglo-American and

control-dominated systems. Western democracies have a long history of sacrificing democratic principles when other interests are perceived as paramount, and this equally applies to police reform (Bayley 2006; Pino and Wiatrowski 2006). One aspect of this relates to organized cross-border crime, including drug trafficking, people smuggling, and money laundering, where the “war” against crime may be advocated in advance of democratic policing ideals. Examples of this are evident in the content of US training priorities at its Budapest college and in US policy in the Caribbean. Of even more significance is the political threat. As Aldous (1997) demonstrated for postwar Japan, US policy oscillated between prioritizing the democratization of the police through decentralization and encouraging a strong police as barrier against the communist threat, the latter point being reiterated in South Korea (Lee 1990). Similar fears concerning the spread of communism within Europe can be identified vis a vis US support for hard-line policing in Greece. In the light of such examples, the influence of US policy on the democratization of Islamic states in the aftermath of the Arab Spring should be greeted with caution. While internal pressures toward democratic policing are by no means inevitable, external influences are often ambiguous. The assumption that police systems across the world are inevitably converging toward a democratic, communityoriented ideal is thus highly questionable. Variations between the police systems of different countries are still evident, with the continental, communist, and colonial models still relevant for any analysis of variations between the police systems of different countries.


Related Entries ▶ British Police ▶ Causes of Police Legitimacy ▶ Communities and the Police ▶ Community Policing ▶ Cross-National Performance in Policing ▶ Democratic Policing ▶ Dutch Colonial Police ▶ French Colonial Police ▶ Gendarmerie Policing ▶ German Police Until 1918 ▶ National and Local Policing ▶ Order Maintenance Policing ▶ Police Use of Firearms ▶ Policing of Peacekeeping ▶ Police in the Police State ▶ Policing in Developing Democracies ▶ Privatization of Policing in an International Context ▶ Role and Function of the Police ▶ Strategies of Policing Terrorism ▶ Women in Policing

Recommended Reading and References Aldous C (1997) The police in occupation Japan. Routledge, London Anderson DM, Killingray D (eds) (1991) Policing the empire. Manchester University Press, Manchester Anderson DM, Killingray D (eds) (1992) Policing and decolonisation. Manchester University Press, Manchester Arnold D (1986) Police power and colonial rule: Madras 1859–1947. Oxford University Press, Oxford Bayley DH (1985) Patterns of policing. Princetown University Press, New Brunswick Bayley DH (1991) Forces of order: policing modern Japan. University of California Press, Berkeley Bayley DH (2006) Changing the guard: developing democratic police abroad. Oxford University Press, New York Beck A, Chistyakova Y (2002) Crime and policing in postSoviet societies: bridging the police/public divide. Policing Soc 12:123–137 Beck A, Chistyakova Y, Robertson A (2006) Police reform in post-soviet societies. Abingdon: Routledge Bowling B (2010) Policing the Caribbean: transnational security cooperation in practice. Oxford University Press, Oxford Brogden M (1987) The emergence of the police: the colonial dimension. Brit J Criminol 27:4–14

487

C

Brogden M (1999) Community policing as cherry pie. In: Mawby RI (ed) Policing across the world: issues for the twenty-first century. UCL Press, London, pp 167–186 Cole B (1999) Post-colonial systems. In: Mawby RI (ed) Policing across the world: issues for the twenty-first century. UCL Press, London, pp 88–108 van Dijk J, van Kesteren J, Smit P (2008) Criminal victimisation in international perspective: key findings from the 2004–2005 ICVS and EU ICS. The Hague: Boom Legal Publishers (accessed at http://rechten.uvt. nl/icvs/pdffiles/ICVS2004_05.pdf) Dinzey-Flores ZZ (2011) Criminalizing communities of poor, dark women in the Caribbean: the fight against crime through Puerto Rico’s public housing. Crime Prev Community Saf 13:53–73 Findlay M, Zvekic U (eds) (1993) Alternative policing styles: cross-cultural perspectives. Kluwer, Boston Fine B, Millar R (eds) (1985) Policing the miners’ strike. Lawrence & Wishart, London Fosdick RB (1969) European police systems, Revisedth edn. Patterson Smith, Montclair Galeotti M (2003) Russian police reform: centralisation, paramilitarisation and modernisation. Crime Justice Int 19(70):17–19 Harriot A (2000) Police and crime control in Jamaica: problems of reforming ex-colonial constabularies. The University of the West Indies Press, Kingston Harriot A (ed) (2003) Understanding crime in Jamaica: new challenges for public policy. The University of the West Indies Press, Kingston Independent Commission on Policing in Northern Ireland (1999) A new beginning– policing in Northern Ireland (Patten Report). Independent Commission on Policing in Northern Ireland, Belfast (see also www.belfast.org. uk/ report.htm) Jones T (1995) Policing and democracy in the Netherlands. Policy Studies Institute, London Kowalewski D (1981) China and the Soviet Union: a comparative model for analysis. Stud Comp Communism 14:279–306 Lee SY (1990) Morning calm, rising sun: national character and policing in South Korea and in Japan. Police Stud 13:91–110 Leishman F (1999) Policing in Japan: east Asian archetype? In: Mawby RI (ed) Policing across the world: issues for the twenty-first century. UCL Press, London, pp 109–125 Marenin O (1998) United States police assistance to emerging democracies. Policing Soc 8:153–167 Mawby RI (1990) Comparative policing issues: the British and American experience in international perspective. Routledge, London Mawby RI (ed) (1999a) Policing across the world: issues for the twenty-first century. UCL Press, London Mawby RI (1999b) The changing face of policing in central and eastern Europe. Int J Police Sci Manag 2:199–216

C

C

488

Mawby RI (2011a) World policing models. In: Natarajan M (ed) International crime and justice. Cambridge University Press, New York, pp 409–415 Mawby RI (2011b) The Greek police in context. In: Cheliotis LK, Xenakis S (eds) Crime and punishment in contemporary Greece: international comparative perspectives. Peter Lang, Oxford, pp 491–503 Miyazawa S (1992) Policing in Japan: a study on making crime. State University of New York Press, New York Pino N, Wiatrowski MD (eds) (2006) Democratic policing in transitional and developing countries. Ashgate, Burlington Pustintsev B (2000) Police reform in Russia: obstacles and opportunities. Policing Soc 10:79–90 Rigakos GS, Papanicolaou G (2003) The political economy of Greek policing: between Neo-Liberalism and the Sovereign state. Policing Soc 13:271–304 Travers H, Vagg J (1993) Crime and justice in Hong Kong. Oxford University Press, New York Wilson DB, Parks RB, Mastrofski SD (2011) The impact of police reform on communities of Trinidad and Tobago. J Exp Criminol 7:375–405 Zvekic U (1996) Policing and attitudes toward police in countries in transition. In: Pagon M (ed) Policing in Central and Eastern Europe: comparing firsthand knowledge with experience from the west. College of Police and Security Studies, Ljubljana, pp 45–59 Zvekic U (1998) Criminal victimisation in countries in transition. UNICRI, Rome

Compliance and Corporate Crime Control Wim Huisman School of Criminology, VU University Amsterdam, Amsterdam, The Netherlands

Synonyms Corporate crime control; Regulatory compliance; White-collar crime

Overview Legal definitions of economic crime have long been in dispute. Formal prosecution of a crime needs an (individual) accused and a legally defined deviant act. Common parlance and media, however, speak of “economic crime”

Compliance and Corporate Crime Control

when they refer to corruption, financial manipulation, or fraudulent enrichment. As a reflection of typical status of the offenders, their crimes are dubbed “white-collar crime.” Moreover, the media often blames corporations and public agencies rather than individuals. Braithwaite (1984: 6) describes corporate crime as “the conduct of a corporation, or of employees acting on behalf of a corporation, which is proscribed and punishable by law.” Corporate crime, therefore, encompasses a wide array of illegal activities that are criminally, civilly, and administratively proscribed and which may be committed by individual managers as well as by the firm they work for. Thus, modern regulations have introduced fines and conditional exclusion (e.g., from competition) for well-defined organizational misconduct. Additionally, enterprises may try to control misconduct by internal self-regulation and may penalize deviant employees and managers, using internal sanctions. Their regulatory definition of “misconduct” need not be restricted to criminal deviance but may include all sorts of agency (particularly company) activities which can lead to social damage and to public shaming (such as for health risks, environmental pollution, etc.). Consequently, corporate misconduct can be considered to something of a sliding scale of wrongdoing: crime, punishable by state sanction to organizational misconduct inspected by internal compliance management of corporations and other agencies. It is obvious that this scale ranges from “hard” criminal law with strict procedural rules (such as formal burden of proof and the right to defense) to “soft” regulation with managerial techniques of control. These concepts of “hard” and “soft,” however, do not indicate the severity of imposed sanctions but rather the legal definition of the rules and the procedures applied to control them. As is well known, the threat of hard sanctions, particularly if the misconduct is hard to prove, regularly leads to plea-bargaining or an out of court for a settlement (Cullen et al. 2006). It follows from these definitions that there is quite some overlap of criminological research on corporate crime with business studies on


compliance with regulation in corporations and other organizations. Corporate crime theories combine insights from criminology and organizational sciences. The leading paradigm blames organizations’ environments and structural characteristics for inducing managers and employees to commit crimes in a business context while otherwise behaving as law-abiding citizens. Prevention measures as well as repressive intervention focus on corporate strategy, structure, and culture to identify and change crime-fostering organizational conditions. Recently, however, there is an academic trend toward analyzing personality traits and criminal careers of corporate actors. Research on the frequency and causes of regulatory (non) compliance and corporate crime has to consider criminal justice interventions as well as corporate compliance management.

Research on Corporate Compliance The dependent variable in studying compliance with corporate regulation may be seen on a gliding scale from the “hard” side of criminology which labels rule violations as “corporate crime” or “white-collar crime” to the “soft” side of political science and management which look at obedience with increasing mass of internal and external rules. As a result, the study of regulatory (non)compliance is rich and diverse, applying various methodological and theoretical approaches. Parker and Nielsen (2011) distinguish two methodological approaches. The first is straightforward testing of theories which identify organizational characteristics and external factors that are associated with (non)compliance. This approach is concerned with measuring the prevalence and levels of (non)compliance as dependent variable. The explaining variables are typically individual motivation, organizational characteristics, and environmental factors such as industry culture and the regulation itself. The outcomes of such rigorous empirical studies are often used to design normative models for better regulation, which are expected to produce higher

489

C

levels of compliance. Such “evidence-based” regulatory policies rely on explanatory theory building about causes of compliance and noncompliance. The second approach aims at the social construction of the very notions of compliance and noncompliance. It tries to uncover multiple interpretations and meanings of compliance together with the power relations that lead to more or less social acceptance or denial of legitimacy. They use discourse analyses and empirical observation of interactions which define, maintain, or change notions of compliance. For instance, in her study on the enforcement of occupational safety and health regulation in the United Kingdom, Hutter concluded that defining, achieving, and maintaining compliance are an interactive process between regulator and regulated (Hutter 1997). Both approaches complement each other. Internally, corporations use the term “compliance” for activities by which they try to ensure internal regulations, designing procedures and employ compliance officers. Externally, “compliance” stands for a style of regulation and enforcement, as opposed to a “deterrent” enforcement style.

Prevalence and Causes of (Non) Compliance and Corporate Crime The Prevalence of Compliance and Noncompliance In criminology, studies of new forms crime usually start with assessments of the scale in terms of the prevalence of offenses and the level of damage. Edwin Sutherland (1949/1983) is seen as the academic founding father of crimes committed in the context of otherwise legitimate business. He coined the term “white-collar crime” when he conducted a first study on regulatory noncompliance by America’s 70 largest corporations. He found sanctions for violations of government regulation in all of them, 97 % were recidivist and 60 % he even labeled as “habitual” criminals with more than four violations. Clinard and Yeager (1980) replicated this study by

C

C

490

measuring noncompliance by America’s 500 largest corporations. They found widespread violation within several branches of industry with a concentration on a small number of habituals: 13 % of the offenders were responsible for about half of all violations. It is extremely difficult to estimate the prevalence of corporate crime or corporate noncompliance. Since organizational crime is generally less visible and harder to define, the “dark numbers” are even higher than with ordinary crimes. Official statistics direct much more attention to conventional crime than to whitecollar crime. Victims of manipulation, fraud, and embezzlement often do not detect their losses, and even if they become aware, may be reluctant to file reports due to fear of negative publicity. The police, customs, and other agencies have to handle variable or ambiguous legal definitions so that corporate crime is not uniformly defined, reported or recorded. Specific forms of business regulation are dispersed over several agencies which not always coordinate their enforcement (e.g., environmental regulation, occupational safety regulation, trade regulation, consumer protection regulation, etc.) Each of the numerous agencies may have its own forms of record keeping reflecting the expertise and priorities of the regulated field, but not reflecting the actual prevalence of noncompliance. Nevertheless, registrations of these enforcement agencies show that corporate rule breaking is widespread in many areas of regulation and branches of industry. Motivation for Compliance and Noncompliance Criminological theories on the causes of criminal behavior look at two types of explanatory variables: motivation – why people want to commit crimes, and opportunity – why people are able to commit crimes. The same variables can be found at the positive end of the compliance spectrum. Business regulation requires corporations to take certain action; corporations also have to be willing and able to comply. Thus, corporate deviance is often due to omission: Corporations violate rules as a result of failure to act in circumstances


where that is necessary, e.g., financial reporting, installing prescribed equipment, or avoiding environmental pollution. There is some debate as to whether to look for motives at the level of the organization or individual managers and employees. Managers’ social programming tends to the same action in same situations (Coleman 1987), and thus, their motives are identified with the goals of the organization as a whole. Some scholars even see the motivation of individuals as “irrelevant” in explaining corporate crime (Braithwaite 1984; Coleman 1987). At the firm level, three types of motives for compliance and noncompliance are distinguished: economic, social, and normative motives. Economic motives refer to the extent to which the firm is committed to maximizing its own economic or material utility, such as increasing turnover and profit. Social motives refer to the extent to which a firm is committed to earning the approval and respect of significant people with whom an actor interacts including other businesses, trading partners, employees, customers, local communities, and the wider public. Normative motives refer to the extent to which the firm is committed to obeying the regulation for its own sake because of a sense of moral agreement with the specific regulation or a generalized sense of moral duty to comply. Empirical research has made it clear that there are a wide range of relevant motives and they interact with each other in reciprocal ways: Personal motives of managers are extrapolated to the motivations of the organization as a whole, and on the other hand, these shape the motivations of individual managers (Parker and Nielsen 2011). The financial crisis and large accounting fraud cases have renewed the attention for personality traits that influence individuals’ motives for corporate crime. Large financial accounting frauds such as ENRON have been related to the personality traits of corporate executives. Public, political as well as academic debate on the causes of the financial crisis of 2008 have indicated the roots of the current credit crunch to be the result of excessive risk taking, moralistically referring to “greed” (Huisman 2011). Research has even


491

C

linked masculinity to risk-taking financial markets (Coates and Herbert 2010). To prevent excessive risk taking and to change “masculine” corporate cultures, new laws prescribe certain quota of the members of corporate executive boards to be female. In recent studies, several personality traits are related to white-collar crime, such as the need for control, bullying, charisma, fear of falling or failing, company ambition, lack of integrity, narcism, and a lack of social conscience (Blickle et al. 2006; Bucy et al. 2008; Piquero et al. 2010; Ragatz and Fremouw 2010).

autonomy of departments. However, lack of internal control shows up in diverse organizational structures: In highly bureaucratic structures, managerial responsibility for shop-floor level misconduct tends to be diffused over several hierarchical layers, while in decentralized – loosely coupled – organizational structures the, more autonomous, subsidiaries might escape the monitoring of the parent company (Keane 1995). Organizational cultures may transmit ruletransgression by “normalizing” lawbreaking by neutralization and rationalization (Shover and Hochstetler 2002).

Opportunities and Organizational Capabilities Opportunities for crime can be identified as means to commit crime as well as the lack of mechanism preventing it. While the general perception of organizational deviance sees it in the light of intention and economic motives, empirical studies have identified less serious causes such as widespread routine, and other types of organizational shortcomings (Van de Bunt and Huisman 2007; Vaughan 2011). Motivation to comply is of secondary importance if a firm does not possess the capacity to comply (Parker and Nielsen 2011). In this view, corporate rule breaking is the outcome of bounded rationality. Managers do not possess all the relevant information on the cost and benefits of options for action and organizational procedures lead to suboptimal outcomes. Landmark cases of organizational deviance – such as the Ford Pinto and the NASA Challenger – have falsified the initial explanation of these cases as being the result of rational and amoral calculation on the basis of cost-benefit assessments. Instead, the accidents were framed as being the result of organizational failure. Most studies focus on the meso-level and try to identify criminogenic organizational strategies, structures, and cultures. These studies show that corporations set their ambitions high while tolerating poor practice regarding the means by which these goals should be attained (Wang and Holtfreter 2012). This dichotomy of ends and means varies with the amount of

Multilevel Framework Several authors brought the explanations together in an integrated framework of three categories of explanatory variables: motivation, opportunity, and (lack of) control (Coleman 1987; Shover and Bryant 1993). Integrative theories on the causes of corporate crime stress the interaction of variables on the institutional macro-level, the organizational meso-level, and the individual micro-level. Kramer and Michalowski (2006) place the three variables on three levels in a matrix with nine cells. To give some examples, on the macro-level, motivation is operationalized by the culture of competition characteristic for business, and that on the meso-level poses economic pressure on corporations to set ambitious targets which on the micro-level can provide individual managers with neutralizations to break rules to attain these goals. On the macro-level, the structure of the market determines the distribution of legitimate and illegitimate means to accomplish goals, while at the meso-level, the opportunities to attain goals are determined by the organizational structure. On micro-level, this can result in individual employees experiencing illegal means as the most attractive option for achieving goals. At the macro-level, the element of control points to the lack of administrative, political, and social control on the conduct of. If at the meso-level, the internal control structure is ineffective, it will result in an organization where on the micro-level, managers and employees rationalize illegal means for attaining organizational goals.

C

C

492

Corporate Crime Control Academic models have suggested various strategies for monitoring compliance and controlling corporate crime. Some of these are used in practice. Generally, they distinguish two ideal-types of corporate crime control: a deterrence strategy and a compliance strategy. Alternative models develop integrative strategies. These models are based on assumption about the causes of compliance and noncompliance, as discussed in the previous section. Deterrence Strategies The deterrence style is referred to as “command and control”: Regulations are drafted by government agencies and imposed on business, and compliance is strictly monitored. Regulatory enforcement is concentrated on detecting and sanctioning violations. It is based on the assumption that compliance and violation are led by economic motives. Companies are seen as rational actors calculating costs and benefits of compliance and noncompliance. A deterrence strategy assumes that offenders will refrain from future offending, if detection and punishment occur with sufficient frequency and severity. However, the strategy is counterproductive when it takes away the voluntary motivation to comply. Simpson (2002), who spent much of her academic career testing the corporate deterrence theory, concluded that we know very little about whether, when, and why deterrence might work as a strategy for compliance enforcement. Compliance decisions are not determined by objective cost and benefits of compliance and noncompliance, but rather by the subjective perceptions of corporate managers. However, there is hardly any empirical evidence that deterrent enforcement styles succeed in influencing this perception. Instead, industry and organizational characteristics have stronger effects on compliance (Simpson 2002). The application of civil or administrative law for regulatory enforcement can have punitive or even incapacitating business effects, for instance, issuing administrative fines or revoking licenses. Nevertheless, the deterrence style is


mainly associated with criminal law. In practice, however, criminal prosecution of corporate offenders is scarce. Most cases of corporate crime are not prosecuted or at best, they lead to a financial settlement. The reasons for lack of priority compared to street crime are multilayered: Specialized expertise is necessary due to the technical complexity of cases of white-collar crime; there are coordination problems due to the many agencies involved in the enforcement of economic laws; white-collar offenders and large corporations might be able to hire better lawyers capable of finding weaknesses in the prosecution’s case; and finally because of legal difficulties, such as establishing who is liable in corporate entities (proving mens rea) and as well as the legal ambiguity of many regulations aimed at businesses. Nevertheless, heightened public “blaming” has increasingly led to prosecution and sanction in spectacular cases such as accounting fraud at ENRON, the Ponzi scheme of Madoff, and the large-scale bribery at Siemens. The deterrence strategy also assumes that potential violators will be deterred by the punishment of offenders. In a landmark study, Thornton et al. (2005) found limited support for this assumption. In a survey of 233 American firms, it was examined whether companies learn about cases of severe penalties, in the same industry, and whether this knowledge changes the firms’ compliance-related behavior. They found that less than half of the companies could identify the “signal cases” of severe legal penalties. They concluded that “explicit general deterrence” does not enhance the perceived threat of legal punishment, but it may serve as a reminder to check on reliability of internal compliance routines and as a reassurance that compliance is not foolish. Due to the difficulties of directly influencing compliance-related behavior by legal penalties, contemporary regulators have set their hopes on reputation damage as a result of the “naming and shaming” effect of publicly sanctioning companies. Substantial reputation damage might result from publicly shaming corporate misconduct, even though there is not much knowledge about


the impact in terms of financial and social reputation damage for business (Van Erp 2013). Compliance Strategies While official rhetoric is tough on tackling corporate crime and noncompliance, empirical studies show that most agencies regulating business apply enforcement styles that rely on cooperation rather than confrontation; they use advice and negotiation rather than coercion and formal legal action. “Compliance assistance” is the preferred official enforcement strategy of several regulatory agencies (Van de Bunt and Huisman 2007). Where violation results from organizational shortcomings, negotiation and consultancy as enforcement style can be used to teach a corporation how to comply with regulations. Cooperative regulation is used to stimulate companies to maintain internal compliance management systems. Compliance management systems are viewed as important tools for companies to organize and monitor regulatory compliance. This might possibly lead to the introduction of “meta-regulation” at a distance: Instead of a “rule-based” approach with detailed prescriptions for business operations, the government creates “principle-based” regulation, containing broad duty-of-care provisions and the obligation for businesses to maintain compliance management systems. This is the case when regulation no longer dictates with what kind of equipment a certain level of emission should be reduced but prescribes a compliance system that ensures a sustainable business operation. Enforcement is then based on reports provided by the companies themselves, and it is aimed at ascertaining whether compliance mechanisms are working effectively. However, such an approach may prove counterproductive in the case of intentional violation (Gunningham 2011). From the rationality paradigm, lenient enforcement and negotiation were initially rejected. Academic studies on the causes of noncompliance have given legitimacy to addressing causes of noncompliance: organizational incompetence and corporate cultures failing to reinforce the value of compliance.

493

C

Whether compliance management systems have had a positive impact on noncompliance has to date received little attention in empirical research. Compliance programs create awareness and give instructions on how to comply. Kaptein (2011) found a positive effect of ethical codes of conduct on compliance. It was observed, however, that the effectiveness of corporate compliance officers could be compromised when they report to the business line management, rather than the corporate board. Compliance programs could even be considered to be more cosmetic than real, largely serving a PR function. In some cases, they may actually increase or facilitate corporate offenses, by regulatory capture or keeping up appearances (Friedrichs 2010). Integrative Models The criticism of different styles of regulation and enforcement is as old as the propagation of “new,” “smarter,” and more “responsive” models. Academics who suggest these models try to find an optimal mix of compliance and deterrence. A currently adopted model is that of “responsive regulation” (Ayres and Braithwaite 1992). The central notion of “responsive regulation” is that enforcement styles should be tuned to the motives and competences of individual companies laid down in an “enforcement pyramid.” Regulators should start at the bottom of the pyramid, assuming that a company is willing to comply voluntarily. When this is shown not to work, regulators should escalate up to deterrenceoriented strategies. The regulatory pyramid has become the paradigm for a “sequential” combination of compliance and deterrence styles of enforcement. It has been adopted by regulatory and enforcement agencies worldwide. In academic debate, the model has had a mixed reception. Empirical studies on its use show that it is very difficult to apply the entire pyramid in practice. It is hard to assess the actual motives of companies, it is difficult to shift between enforcement styles, and it ignores other environmental influences on compliance than state-regulation (Gunningham 2011). As a result of the above insights, new and more sophisticated models have evolved such as

C

C

494

“smart regulation”: adopting the principles of responsive regulation, but expanding regulation to third parties as surrogate regulators. As all modern regulatory theories, smart regulation advocates government to facilitate and mobilize other network members rather than direct intervention. Smart regulation has to organize public enforcement so that it can reinforce third parties to help with regulating (Gunningham and Grabosky 1998). Situational Crime Prevention Responsive models seem to be primarily oriented at influencing the motivation behind compliance and noncompliance. The bottom of the enforcement pyramid aims at normative motives for compliance (the general attitude that one should comply with regulation), the next step aims at social motives (the realization that others also comply), and the top is aimed at economic motives (increasing the costs of violation). As motivation has proven difficult to measure, criminology has shifted the focus on studying and changing opportunities. An opportunity perspective focuses on the specific characteristics of a situation, and studies the process through which an offense is committed. In other words, it studies the “how,” rather than the “why.” Situational crime prevention addresses the environments which shape crime opportunities, and the modifications which may diminish criminal opportunities. It identifies five characteristics of criminal opportunity: effort required to carry out the offense; perceived risk of detection; reward from committing the offense situational conditions that may encourage criminal action; and excuses and neutralizations of the offense. Preventive measures are best found in the immediate environment of the crime. Crime prevention, then, becomes a matter of five principles: increasing the effort and risk of detection, reducing circumstantial provocation, and reward and removing excuses. Situational crime prevention theory departs from the assumption that potential offenders decide whether or not to engage in crime, based on the attractiveness of criminal opportunities, in terms of costs and benefits of crime. White-collar


crimes may be committed without clear criminal intent, since regulation is often ambiguous. Nevertheless, offenders are assumed to make more or less rational choices to commit crime, or at least, are to some degree sensitive to the changes in their environment which influence costs and benefits. Benson and Madensen (2007) propose situational crime prevention theory, which has been widely used to analyze traditional forms of street crime, to white-collar crime. They argue that the analysis of the opportunity structure of whitecollar crime through situational crime prevention theory can provide useful starting points for prevention. In fact, they argue that an opportunity perspective is more fruitful than an approach based on offender characteristics or motive: “we suggest that focusing on how is likely to be more productive than focusing on why” (Benson et al. 2009, p. 176). They suggest that crime prevention by means of the alteration of opportunity structures represents “a more fundamental way of thinking about the problem of white-collar crime control” than those of competing schools of thought, in the sense that it “implicitly underlies” other approaches (Benson and Madensen 2007, p. 623). Furthermore, this analysis leads to newer and more effective methods of prevention and control, because potential points of intervention are illuminated. Consequently, they urge whitecollar crime scholars to apply the situational prevention approach, and have made the first step with a brief analysis of opportunity structures in health care fraud. Others have tried to apply situational crime prevention to environmental crimes (Dorn et al. 2007). Considering the strong claims by Benson et al., more empirical research is needed to test the usefulness of situational crime prevention theory for prevention of particular types of white-collar crime.

Future Developments and Directions for Research Most of the relevant variables in explaining corporate compliance and noncompliance have been


identified in different strings of research, often in critical response to one another. The main and current research question is how all these relevant variables interact in producing compliance and noncompliance. New and more advanced research designs are needed to unravel the existing process of interaction. It has to integrate measurement by objective indicators which try to establish the causal relations with independent variables focusing on hypotheses about what produces compliance on the one hand, with interpretative approaches which appreciate that compliance and noncompliance are defined in processes of social construction on the other hand. The ambiguities in the definition of compliance could be one of the variables in explaining noncompliance. Research on the regulation of business and regulatory enforcement has long moved beyond the dichotomy of compliance and deterrence. Scholars are still searching for the optimal mix of enforcement styles. The two-dimensional problem has been expanded to a threedimensional challenge, surpassing another dichotomy – the dyadic relationship between regulator and the regulated – and by introducing third party regulators. Harnessing actors and resources outside the public sector also fits in situational crime prevention. Yet, although these more sophisticated models are based on empirical insights as to compliance and noncompliance, the models themselves still have to be tested in empirical research.

Related Entries ▶ Corporate Crime Decision-Making ▶ Corporate Crimes and the Business Cycle ▶ Corporate Liability ▶ Crimes of Globalization ▶ Deterrence of Tax Evasion ▶ Environmental Regulation and Law Enforcement ▶ Managerial Court Culture ▶ Naming and Shaming of Corporate Offenders

495

C

▶ State-Corporate Crime ▶ Victims of Corporate Crime ▶ Women and White-Collar Crime

Recommended Reading and References Ayres I, Braithwaite J (1992) Responsive regulation: transcending the deregulation debate. Oxford University Press, Oxford Benson ML, Madensen TD (2007) Situational crime prevention and white-collar crime. In: Geis G, Pontell H (eds) International handbook of white collar crime. Springer, New York, pp 309–326 Benson ML, Madensen TD, Eck JE (2009) White-collar crime from an opportunity perspective. In: Simpson SS, Weisburd D (eds) The criminology of white-collar crime. Springer, New York, pp 175–193 Blickle G, Schlegel A, Fassbender P, Klein U (2006) Some personality correlates of business white-collar crime. Appl Psychol Int Rev 55(2):220–233 Braithwaite J (1984) Corporate crime in the pharmaceutical industry. Routledge and Kegan Paul, London Bucy PH, Formby EP, Raspanti MS, Rooney KE (2008) “Why do they do it?” The motives, mores, and character of white-collar criminal. St John’s Law Rev 82:401–571 Clinard MB, Yeager PC (1980) Corporate crime. Free Press, New York Coates J, Herbert J (2010) Endogenous steroids and financial risk taking on a London trading floor. Proc Natl Acad Sci 105(16):6167–6172 Coleman JW (1987) Toward an integrated theory of white-collar crime. Am J Sociol 93(2):406–439 Cullen F, Cavender G, Maakestad W, Benson M (2006) Corporate crime under attack: the fight to criminalize business violence, 2nd edn. LexisNexis Matthew Bender/Anderson Publishing Company, Newark Dorn N, Van Daele S, Vander Beken T (2007) Reducing vulnerabilities to crime of the European waste management industry: the research base and the prospects for policy. Eur J Crime Crim Law Crim Just 15(1):23–36 Friedrichs DO (2010) Trusted criminals: white collar crime in contemporary society. Wadsworth Cengage Learning, Belmont Gunningham N (2011) Strategizing compliance and enforcement: responsive regulation and beyond. In: Parker C, Nielsen V (eds) Explaining compliance: business responses to regulation. Edward Elgar, Cheltenham, pp 199–221 Gunningham N, Grabosky P (1998) Smart regulation: designing environmental policy. Oxford University Press, Oxford Huisman W (2011) Corporate crime and crisis: causation scenarios. In: Sociology of crime, law and deviance, Economic crisis and crime. Emerald/The JAI Press, Bingley, UK, pp 107–125

C

C

496

Hutter BM (1997) Compliance: regulation and environment. Clarendon, Oxford Kaptein SP (2011) Towards effective codes: testing the relationship with unethical behavior. J Bus Eth 99(2):233–251 Keane C (1995) Loosely coupled systems and unlawful behaviour: organization theory and corporate crime. In: Pearce F, Snider L (eds) Corporate crime; contemporary debates. University of Toronto Press, Toronto, pp 168–181 Kramer RC, Michalowski RJ (2006) The original formulation. In: Michalowski RJ, Kramer RC (eds) Statecorporate crime. Wrongdoing at the intersection of business & government. Rutgers University Press, New Brunswick, pp 18–26 Parker C, Nielsen V (2011) Explaining compliance: business responses to regulation. Edward Elgar, Cheltenham Piquero NL, Schoepfer A, Langton L (2010) Completely out of control or the desire to be in complete control? An examination of how low self-control and the desire for-control relate to corporate offending. Crime Delinq 56(4):627–647 Ragatz L, Fremouw W (2010) A critical examination of research on the psychological profiles of white-collar criminals. J Forensic Psychol Pract 10:373–402 Shover N, Bryant KM (1993) Theoretical explanations of corporate crime. In: Blankenship MB (ed) Understanding corporate criminality. Garland Publishing, New York, pp 141–176 Shover N, Hochstetler A (2002) Cultural explanation and organizational crime. Crime Law Soc Change 37:1–18 Simpson SS (2002) Corporate crime, law and social control. Cambridge University Press, Cambridge Sutherland EH (1949) White collar crime. Dryden Press, New York Sutherland EH (1983) White collar crime. The uncut version. Yale University Press, New Haven Thornton D, Gunningham N, Kagan R (2005) General deterrence and corporate environmental behavior. Law & Policy 27:262–288 Van de Bunt HG, Huisman W (2007) Organizational crime in the Netherlands. Crime Just Rev Res 35:217–260 Van Erp J (2013) Naming and shaming. In: Encyclopedia of criminology and criminal justice. Springer, New York Vaughan D (2002) Criminology and the sociology of organizations. Analogy, comparative social organization, and general theory. Crime Law Soc Change 37:117–136 Vaughan D (2011) Theorizing the dark side. In: Linstead S, Marechal G, Griffin R (eds) Organization studies, special issue on the dark side of organizations. SAGE publications, Thousand Oaks Wang X, Holtfreter K (2012) The Effects of corporationand industry-level strain and opportunity on corporate crime. J Res Crime Delinq 49(2):151–185

Compstat

Compstat James J. Willis Department of Criminology, Law and Society, George Mason University, Fairfax, VA, USA

Overview Since its implementation in 1994 in the United States under former Commissioner William Bratton and Deputy Commissioner, Jack Maple, of the New York City Police Department, Compstat has become widely recognized as a major innovation in US policing and abroad. Compstat is a strategic management system whose reform elements are designed to overcome some of the traditional constraints and limitations of bureaucratic administration and make police organizations more responsive to changes in their crime environment. According to its doctrine, Compstat decentralizes decisionmaking to middle managers operating out of districts, holds these managers strictly accountable for their performance, and increases a police organization’s capacity to identify, understand, and respond to crime problems as they emerge. The evidence that Compstat reduces crime is not clear and consistent, but its originators and proponents credit Compstat with impressive crime drops and improvements in neighborhood quality of life in New York City. Based on the publicity it has received, including a 1996 Innovations in American Policing Award sponsored by the Ford Foundation and Harvard University’s John F. Kennedy School of Government, other police leaders, delegates, and politicians have flocked to New York City to observe Compstat in action. Impressed by what they have seen, they have implemented Compstat in their own agencies. Consequently Compstat has become a global commodity and many countries, including England, Australia, and Canada, have since adapted methods of assessing performance from the NYPD Compstat model. Research on Compstat’s relationship to another popular

Compstat

reform, community policing, indicates that they may work independently when implemented in the same police organization: each having little effect on the other. This suggests opportunities for their integration, but what form such a model would take is uncertain. What is clearer is that Compstat may be evolving in a new and potentially different direction. According to its supporters, “predictive policing” is an extension of Compstat’s focus on using analyses of timely crime data to drive police strategies. In this case, crime and noncrime data are made readily available and combined with forecasting, modeling, and sophisticated statistics to help make predictions about where crime is likely to occur in the future. What specific role Compstat plays in predictive policing has not been studied systematically, but this new development suggests interesting avenues for future research.

Compstat’s Doctrinal Elements and Its Diffusion When Commissioner Bratton assumed leadership of the NYPD, his assessment of the organization was that it suffered from bureaucratic dysfunction. Functional specialization impeded coordination, an inflexible hierarchy hindered the rapid flow of information, timely crime data were unavailable, and the department had lost sight of its overarching mission, namely, to reduce serious crime (Bratton and Knobler 1998: 209; Silverman 1999). To streamline operations, Bratton borrowed state-of-the-art management doctrines espoused by organizational development experts in the private sector. These principles included a commitment to establishing priorities, finding creative approaches to new challenges, and using information scientifically to drive decision-making (Willis et al. 2007). According to this “reengineering” approach (Hammer and Champy 1993), successful organizations respond to uncertainties in their environments by reprioritizing their goals and revamping core structures for their successful accomplishment. Compstat, whose name comes from

497

C

a computer file name, was the centerpiece of Bratton’s reform efforts (Silverman 2006: 221). At the core of NYPD’s Compstat model are four crime reduction elements that are designed to make police organizations rational and more responsive to management direction: (1) accurate, timely information made available at all levels in the organization; (2) selection of the most effective tactics for specific problems; (3) rapid focused deployment of people and resources to implement those tactics; and (4) relentless follow-up and assessment to learn what happened and make subsequent tactical assessments if necessary (Bratton and Knobler 1998). Fundamental to this Compstat approach is the delegation of decision-making authority to middle managers (or precinct/district commanders) with territorial responsibility. These commanders are then held directly accountable for reducing crime in their precincts and given the necessary resources (specialist units, detectives, etc.) for accomplishing this goal. Alongside these elements, the NYPD also developed eight crime control and quality-oflife strategies that were disseminated throughout the department. Their primary purpose was to provide commanders with guidance on how to craft specific approaches to reduce crime and disorder problems in their precincts, including youth violence, drug and gun crimes, and public nuisances such as “squeegee window washers” (Safir, n.d.: 6). However, case studies on Compstat in other departments suggest these features of the NYPD model have not been as widely implemented as its frequent “crime control strategy meetings” that are the centerpiece of the Compstat process (Willis et al. 2007). During these meetings precinct commanders appear before the department’s top echelon to report on crime problems in their districts and what they are doing about them. This occurs in a datasaturated environment where crime analysts collect, analyze, and map crime statistics to spot crime trends and help precinct commanders identify patterns among crime incidents. Top administrators then use this information to quiz district commanders on the crime in their beats and to hold them responsible for solving them. Failure

C

C

498

to provide satisfactory responses to their inquiries may lead to stern criticism or possibly removal from command. Based on what those who developed Compstat have written, as well as what those who have studied Compstat have observed, researchers have identified six core elements that have emerged as central to the development of Compstat (Weisburd et al. 2003): • Mission clarification. Top management is responsible for clarifying and promoting the core features of the department’s mission. Mission clarification includes a demonstration of management’s commitment, such as stating those goals in specific terms for which the organization and its leaders can be held accountable – for example, reducing crime by 10 % in a year. • Internal accountability. Operational commanders are held accountable for knowing their commands, being well acquainted with the problems in the command, and accomplishing measurable results in reducing those problems – or at least demonstrating a diligent effort to learn from that experience. Those who fail to do so can suffer adverse career consequences such as removal from command. • Geographic organization of operational command. Operational command is focused on the policing of territories, so central decisionmaking authority about police operations is delegated to commanders with territorial responsibility (e.g., districts). Functionally differentiated units and specialists (e.g., patrol, community police officers, detectives, vice) are either placed under the command of the district commander or arrangements are made to facilitate their responsiveness to the commander’s needs. • Organizational flexibility. The organization develops the capacity and the habit of changing established routines to mobilize resources when and where they are needed for strategic application. • Data-driven problem identification and assessment. Crime data are made available to identify and analyze problems and to track and assess the effectiveness of the department’s response.

Compstat

• Innovative problem-solving tactics. Police responses are selected because they offer the best prospects of success, not because they are “what we have always done.” In this context, police are expected to look beyond their own experiences by drawing upon knowledge gained in other departments and from innovations in theory and research about crime prevention. On occasion, this list has been expanded to include a seventh Compstat element or “external accountability.” This refers to Compstat’s capacity for making police decision-making more transparent to the public. Under Compstat, departments are externally accountable to the degree that they provide stakeholders with accurate and timely information about how well they are accomplishing their official crime control mission. Evidence suggests that Compstat has diffused rapidly. In a national survey of large (>100 sworn) police departments in the USA administered by the Police Foundation in 2000, a third of agencies reported they had implemented a Compstat-like program with a quarter claiming they were intending to do so. This study confirmed that Compstat had “literally burst onto the American policing scene” and was following a diffusion process as rapid as “innovations in other social and technological areas” (Weisburd et al. 2004: 15). Nor does the spread of Compstat appear to be slowing, according to another survey conducted in 2006 by scholars at George Mason University and funded by the Department of Justice’s Office of CommunityOriented Policing Services, 60 % of large police departments in the USA had implemented Compstat or a Compstat-like program (Willis et al. 2010a).

Compstat in Practice It is virtually a truism in organizational theory that organizational change, including police reform, rarely works in intended ways. Accounts of the New York City Police Department suggest that Compstat’s elements function like a well-oiled

Compstat

machine (Henry 2002; McDonald et al. 2002), but others characterize these evaluations as “advocacy” studies that lack rigorous scientific analysis by disinterested observers (Jang et al. 2010). A small but rich body of systematic research in other police departments that have sought to replicate the NYPD model has revealed some unintended effects and internal paradoxes in how it operates (Willis et al. 2007; Dabney 2010). Using data from surveys sent to a stratified sample of 615 police agencies nationwide, Weisburd et al. (2003) concluded that Compstat reinforces the paramilitary model of police organizations. Most notably, Compstat appeared to strengthen the command hierarchy by making middle managers more responsive to top leadership direction through fear of punishment for poor performance. This finding was significant because it showed that Compstat appeared to preserve, rather than transform, the “bureaucratic” or “paramilitary” model of police organization criticized by reformers as too rigid and punitive. Follow-up fieldwork at three police agencies of different size and geographic location confirmed the survey finding that Compstat’s accountability mechanism was generally the element most strongly implemented, along with a clear crime control mission and the display of crime statistics and maps at regular Compstat meetings (Willis et al. 2007). To some this could be interpreted as clear evidence of significant police reform. Compared to past police practices, Compstat reinforced the police crime-fighting mission, held middle managers more accountable for performance, and advanced the timely use of crime data in decision-making. These changes notwithstanding, the fact that departments had placed less emphasis on other key Compstat elements (geographic organization of operational command, organizational flexibility, and innovative problemsolving strategies) tempered any claims of radical change under Compstat. One possible explanation for Compstat’s uneven implementation in these three agencies was that those elements of Compstat that were most likely to be adopted were also those that were most likely to help the agency to appear progressive in the eyes of powerful stakeholders in the agency’s environment (e.g., local

499

C

politicians, community leaders, citizens’ groups). Like many public organizations, police agencies that appeal to powerful values and beliefs in their environment about what police structures and practices should look like stand to gain legitimacy and increase their chances of survival and of securing valuable resources. Increasing the agency’s attention to crime control, the spectacle of weekly performance evaluations, and glitzy displays of crime statistics help give the powerful impression that these three departments were doing something to reduce crime even though other structures designed to strengthen this focus remained fundamentally unaltered. Those elements which were least likely to be implemented were also those that represented the greatest departure from taken-for-granted beliefs about how police organizations should operate. So, for example, favoring experimentation with different problem-solving approaches over traditional law enforcement responses to crime could expose the organization to risk (as innovations rarely succeed the first time) and challenge widely held expectations that the primary role of the police is to engage in preventive and reactive patrol and make arrests. An in-depth case study of one police department applying Max Weber’s conceptualization of the modern bureaucratic organization to Compstat also revealed a paradox in how Compstat operated: the power of Compstat’s accountability mechanism undermined several of its other elements including collaboration (by fostering competition) and innovation (by shrinking tolerance for failure). This case study also revealed how Compstat’s strategic approach to adapting quickly to emerging crime problems clashed with classic bureaucratic structures for maintaining efficient routines (such as clear lines of authority and stable job assignments) (Willis et al. 2004).

Compstat’s Effect on Crime Compstat’s implementation in the NYPD corresponded to a greater drop in New York City’s crime rate than the national average. According to Silverman, “During 1994, the

C

C

500

overall New York City crime decline was 12 %, compared to 2 % nationally, and 17 % for 1995” (Silverman 1996: 1). Its supporters claimed that this was clear evidence of Compstat’s success as a mechanism for reducing crime and improving neighborhood quality of life (Bratton and Knobler 1998), but the association between Compstat and declining crime levels does not necessarily imply causation. Subsequent research seeking to assess Compstat’s effect on crime suggests a more complicated picture. This is owed in no small measure to the fact that Compstat often combines agency reorganization with the implementation of a wide range of crime control strategies (e.g., order maintenance policing, zero tolerance policing, hot spot policing) making it difficult to identify the specific Compstat mechanism responsible for any crime reduction. Applying the same descriptive analyses used by its supporters, Eck and Maguire challenge the claim that Compstat played a key role in reducing serious crime in New York City (2000). They examine data from the United States’ Uniform Crime Reports between 1986 and 1998 and show that the decline in New York’s homicide rate began long before Compstat was implemented. They also note that other large cities experienced significant declines in their homicide rates during the early to mid-1990s but had not implemented a Compstat program. For example, San Diego, San Antonio, and Los Angeles all experienced sharp drops in their homicide and violent crime rates during this period (Harcourt 2001). Eck and Maguire conclude, “On balance, the data do not support a strong argument for Compstat causing, contributing to, or accelerating the decline in homicides in New York City or elsewhere” (2000: 233). Similarly, Willis et al. using a simple pre-/posttest noted that crime was already in decline at the agencies where they conducted their on-site fieldwork and the rate of decline was the same or less steep following Compstat’s implementation (2007). More rigorous attempts to assess Compstat’s effectiveness as a crime control strategy do not provide clear and consistent results. Using piecewise linear growth models, Rosenfeld et al. (2005) examined the impact of three law

Compstat

enforcement initiatives, including Compstat, on homicide trends in three cities while making systematic comparisons to crime trends in other cities and controlling for other factors influencing crime trends (e.g., measures of social and economic disadvantage and police density). Based on their analysis, they concluded that there was insufficient evidence in support of the NYPD Compstat model’s impact on homicide trends. Others are also skeptical of Compstat’s capacity to reduce crime (Dixon 1998; Chilvers and Weatherburn 2004). Recent studies in Queensland, Australia, and Fort Worth, Texas, are more sanguine. They use time-series analyses to assess the impact of Compstat on crime. In Queensland, Compstat’s management accountability features were combined with a problem-oriented policing approach, but this was not the case in Fort Worth. Using crime data spanning a 10-year period (1995–2004 inclusive), the Queensland study concluded that Compstat generally reduced reported crime and was cost effective, but it was most effective at reducing property-related offenses. Similarly in Fort Worth, Compstat significantly decreased property and total index crime rates but had little effect on violent crime rates. What still remains unclear are the independent effects of the different elements of the multidimensional Compstat process (e.g., accountability, crime analysis, order maintenance strategies) on crime. In summary, when combining these findings on Compstat with a growing body of evidence on police effectiveness, it appears that when Compstat’s accountability and performance measurement structures support the implementation of police practices targeted and tailored toward specific offenses, offenders, or places, Compstat can lead to reductions in crime and disorder (Braga and Bond 2008: 599).

Compstat’s Relationship to Community Policing Compstat has emerged during the same period as community policing, another popular engine of police reform in the USA and around the globe.

Compstat

Community policing can be characterized as a philosophy and an organizational strategy designed to reduce crime and disorder through community partnerships, problem solving, and the delegation of greater decision-making authority to patrol officers and their sergeants at the beat level. It varies more than Compstat from place to place in response to local problems and community resources. Given the visibility of these reforms, it is natural to consider the relationship between the two when they are implemented in the same police organization: do they operate together, that is, one reinforcing the other, or are there points of conflict, where pursuing one makes it harder to pursue the other successfully? Alternatively, do they work separately, that is, each having little consequence for the other? Some have claimed that Compstat complements and supports community policing and even improves it (McDonald et al. 2002). According to this perspective, both reforms share the same concern with crime and quality-of-life offenses and seek to decentralize operations geographically. Where there are discrepancies, such as community policing’s weakly developed accountability mechanism, the other reform is simply able to compensate. A second group of researchers is less convinced that these reforms are compatible. Wesley Skogan, a well-known researcher on community policing, has commented on the tensions between Compstat’s data-driven accountability elements and the different currents that push community policing in the opposite direction (2006). Community policing’s focus on local neighborhood problems, developing police-community partnerships, and creative problem solving at the grassroots level are not key features of Compstat as it has evolved. It is also possible that these reforms operate independently from one another in order to minimize conflicts to existing organizational structures and routines. The body of empirical research on this issue is very small. Willis et al. identify seven core elements that the full implementation of Compstat and community policing demands and assess where their respective reform doctrines stand on these elements (see Table 1). Briefly, Compstat

501

C

focuses the police mission narrowly on serious crime, while community policing broadens it to include a wider variety of objectives including community problems, reductions in social and physical disorder, and citizens’ fear of crime. Furthermore, in contrast to Compstat, community policing requires that community members play a key role in defining what the police should be trying to accomplish. When it comes to making people feel responsible for their performance, Compstat is first and foremost a method to hold middle managers accountable for knowing what is going on in their areas and for devising timely, effective solutions to the most pressing problems. By contrast, there is less concern in the community policing literature with holding police feet to the fire of accountability. Both community policing and Compstat promote decentralizing decision-making authority, but Compstat concentrates on the delegation of authority to middle managers, while community policing has been far more interested in decentralizing decision-making to those responsible for doing the lion’s share of the police organization’s work: the rank and file. Organization flexibility refers to a department’s capacity to determine where a problem is and to change or disrupt department routines to do this. Community policing requires flexibility to meet the varying demands of different constituencies, while Compstat requires flexibility to put the key resources in the hands of the right people and to alter procedural routines to do what must be done to be effective in controlling crime. Both reforms emphasize the use of timely data to drive decision-making, but there are some important differences. Compstat relies on a department’s existing data systems to focus attention on serious crime, while community policing solicits input from community residents to identify a broader range of minor crimes and social disorder deserving of police attention. Moreover, community policing places higher value on sergeants and patrol officers participating in crime analysis, unlike Compstat which empowers district commanders to identify and analyze problems.

C

C

502

Compstat

Compstat, Table 1 A comparison of the doctrines of Compstat and community policing (Adapted from Willis et al. 2010a) Reform element Community policing Mission clarification Broadening of police mission to include wide variety of objectives Key role of community in defining police priorities Internal Not well developed accountability Decentralization of To lowest level in organization, especially patrol decision-making officers, in order to customize solutions to local problems Organizational Capacity to accommodate innovation and flexibility differing needs within communities resulting in strong neighborhood-level focus Data-driven problem Systematic identification of crime and disorder identification and problems by local residents and follow-up assessment assessments of success High value placed on rank-and-file officers participating in crime analysis Innovative problem Innovation at lowest levels of organization solving expected and valued Significant role of community in problemsolving process External Police consult with community on objectives accountability and progress toward them

Under Compstat and community policing, crime data are supposed to provide a basis for searching and implementing creative solutions to crime and disorder problems. Where they differ is that Compstat assigns most of this responsibility to middle managers using traditional crime and calls-for-service data, while community policing emphasizes the role of rank-and-file officers and the public in using customized sources of data and information to tailor specific solutions to local neighborhood problems. Finally, Compstat and community policing’s attempts to make police operations more transparent work in different ways. Under Compstat, departments are externally accountable to the degree that they provide stakeholders with accurate and timely information about how well they are accomplishing their crime control mission. According to community policing doctrine, external accountability goes far beyond merely providing citizens with standard crime measures. Under this model, police are accountable to the

Compstat Specific and highly visible crime reduction goal

Middle managers held strictly accountable for crime performance by leadership To middle managers

Capacity to reallocate resources for effective accomplishment of crime control objectives (putting resources in the hands of key decision makers) Systematic identification of serious crime problems and follow-up assessments of success High value placed on middle managers doing crime analysis Innovation by middle managers expected and valued

Police publicize traditional crime statistics as measures of agency performance

degree that they create a collaborative environment with local residents that is directly responsive to their concerns and to the degree that they foster an open dialog on what the police are doing and how well they do it. Based primarily on observations conducted from 2006 to 2007 at seven police agencies in the USA, Willis et al. describe how Compstat and community policing operated in relation to each of these core elements and assessed their level of integration (not at all integrated, low, moderate, or high). They concluded that Compstat and community policing operated largely independently from each other. Their simultaneous operation helped departments respond to a broader set of goals and wider variety of tasks than had they implemented just one reform. So, for example, community policing provided opportunities to meet with community groups and identify their concerns, while Compstat helped focus the police organization’s energies on fighting crime. However, those problems identified as most pressing

Compstat

by communities were rarely subject to the kind of intense and systematic scrutiny at regular Compstat meetings as serious crime. These findings provided the basis for recommendations on how Compstat and community policing might be integrated in order to be mutually reinforcing (e.g., constructing measures of community policing that can be prioritized, measured, and reported on at regular Compstat meetings) (Willis et al. 2010b). However, to date, these recommendations have not been subjected to empirical testing and thus remain speculative.

Predictive Policing and Future Research At the time of this writing (2012), Compstat’s emphasis on using timely crime data to address current crime problems may be evolving into a policing approach that attempts to forecast trends and then prevent crime from occurring in the future (Bratton and Malinowski 2008: 264). Currently in the stages of its early development, this model known as “predictive policing” uses “advanced analytics” in an attempt to anticipate or predict crime and act as a guide to “risk-based deployment” (Beck and McCue 2009: 19–20). Its elements include integrating crime data with other information sources inside and outside of the police agency (e.g., economic data on housing foreclosures); analyzing these data to identify and anticipate crime patterns related to people, places, or events; mobilizing resources; implementing strategies in response to these predictions; and linking accountability to performance targets rather than just past outcomes (Bratton et al. 2009). Similar to Compstat, this model draws on market practices commonly used in the private sector and specifically attempts to understand and predict consumer behavior. Companies like Walmart use business analytics to forecast demand and then make adjustments to their supply lines so that any increase in demand can be met successfully (Beck and McCue 2009). So, for example, in the event of a large weather event like a hurricane, Walmart stores stock up in advance on water, batteries, and other items to ensure that they have adequate supplies available.

503

C

According to Charlie Beck, Chief of the Los Angeles Police Department where predictive policing originated, “the predictive vision moves law enforcement from focusing on what happened to focusing on what will happen and how to effectively deploy resources in front of crime, thereby changing outcomes” (Pearsall 2010). The emergence of predictive policing opens up a rich vein of research opportunities including questions about the nature of its implementation in different police agencies and its effectiveness in reducing crime. Early anecdotal evidence regarding the latter is encouraging. The Santa Cruz Police Department in California reported an 11 % drop in burglaries for the 5-month period following the implementation of its predictive policing model compared to the same period the previous year. This model runs historical data on the location and time of burglaries through an algorithm. This then generates a list of locations and times with the highest probability of burglaries and auto thefts that is distributed to officers at roll call (Baxter 2011). Any evaluation of predictive policing should also include an assessment of the consequences of its adoption on the legitimacy of the police in the eyes of citizens and other community stakeholders. Of particular concern here should be the degree to which predictive policing heralds a more pronounced move toward the profiling of certain types of people than has been observed under its predecessor. Given predictive policing’s potential to link police databases with a variety of other information systems, such as the census, schools, or social services, the likelihood increases that specific groups will be targeted for additional police attention. Not only will this likely exacerbate existing racial disparities in the American criminal justice system, it also advances a troubling conception of justice and constitutional rights based on the risk of future offending rather than past practice (Harcourt 2007: 3). Some have suggested that predictive policing might unintentionally undermine the Fourth Amendment protections for some individuals (Ferguson 2011). Under current law in the USA, police officers can stop someone based on

C

C

504

“reasonable suspicion” that a crime is being, has been, or is about to be committed. According to the Supreme Court, an important contextual factor for determining reasonable suspicion is whether or not the stop occurs in a high-crime area. Thus predictive policing opens up the possibility that individuals living, visiting, or working in such areas will “have a lesser expectation of privacy than those in other non-high crime neighborhoods” (Ferguson 2011: 1). It might be that the promise of crime analysts applying advanced statistical methods to vast amounts of data, thereby improving the accuracy of results, convinces police chiefs that this is a viable crime prevention strategy. An alternative possibility is that ethical concerns about predictive policing’s analytic methods being used “to target individuals inappropriately for future crimes, or bad acts that they may commit but have not” could impede its progress (Beck and McCue 2009:23). Either way, the emergence of predictive policing presents an intriguing case for an assessment of Compstat as it appears to be developing, one that may lead to important empirical and theoretical insights about the changing nature of crime control policy and practice in the contemporary USA and elsewhere.

Related Entries ▶ Information Technology and Police Work

Recommended Reading and References Beck C, McCue C (2009) Predictive policing: what can we learn from Walmart and Amazon about fighting crime in a recession? Police Chief 76:18–24 Braga A, Bond B (2008) Policing crime and disorder hot spots: a randomized controlled trial. Criminology 46:577–607 Bratton WJ, Knobler P (1998) Turnaround: how America’s top cop reversed the crime epidemic. Random House, New York Bratton WJ, Malinowski SW (2008) Police performance management in practice: taking Compstat to the next level. Policing: A Journal of Policy and Practice 2:259–265 Bratton WJ, Morgan J, Malinowski S (2009) Fighting Crime in the Information Age: The Promise of

Compstat Predictive Policing. Discussion draft, November 18. Available online: http://publicintelligence.net/lapdresearch-paper-fighting-crime-in-the-information-agethe-promise-of-predictive-policing/ Baxter S (2011) Santa Cruz Police Have Success With Predictive Policing. Santa Cruz Sentinel, July 18. Available online: http://www.santacruzsentinel.com/ localnews/ci_18502786. Chilvers M, Weatherburn D (2004) The New South Wales Compstat process: its impact on crime. Aust N Z J Criminol 37:22–28 Dabney D (2010) Observations regarding key operational realities in a Compstat model of policing. Justice Quarterly 27:28–51 Dixon D (1998) Broken windows, zero tolerance, and the New York miracle. Current Issues in Criminal Justice 10:96–106 Eck J, Maguire ER (2000) Have changes in policing reduced violent crime? An assessment of the evidence. In: Blumstein A, Wallman T (eds) The crime drop in America. Cambridge University Press, Cambridge, pp 207–265 Ferguson AG (2011) Predictive policing and the fourth amendment. Am Crim L Rev. http://www.americancriminallawreview.com/Drupal/blogs/blog-entry/% E2%80%9Cpredictive-policing%E2%80%9D-andfourth-amendment-11-28-2011. Accessed 3 Jan 2012 Hammer M, Champy J (1993) Reengineering the corporation: a manifesto for business revolution. Harper Business, New York Harcourt BE (2001) Illusion of order: the false promise of broken windows policing. Harvard University Press, Cambridge, MA Harcourt BE (2007) Against prediction: profiling, policing, and punishing in an actuarial age. University of Chicago Press, Chicago Henry VE (2002) The Compstat paradigm: management accountability in policing, business, and the public sector. Looseleaf Publications, New York Jang H, Hoover LT, Joo H-J (2010) An evaluation of Compstat’s effect on crime: the forth worth experience. Police Quarterly 13:387–412 Mazerolle L, Rombouts S (2007) The impact of COMPSTAT on reported crime in Queensland. Policing: An International Journal of Police Strategies and Management 30:237–256 McDonald P, Greenberg SF, Bratton WJ (2002) Managing police operations: implementing the New York crime control model, Belmont, CA Pearsall B (2010) Predictive policing: the future of law enforcement? NIJ Journal 266:16–19 Rosenfeld R, Fornango R, Baumer E (2005) Did Ceasefire, COMPSTAT and Exile Reduce Homicide? Criminol Publ Pol 4:419–450 Safir H (n.d.) The Compstat process. Office of Management and Planning, NYPD Silverman EB (1996) Mapping change: how the New York City police department re-engineered itself to drive down crime. Law Enforcement News 15:1

Computer Forensics Silverman EB (1999) NYPD battles crime: innovative strategies in policing. Northeastern University Press, Boston Silverman EB (2006) Compstat’s innovation. In: Weisburd D, Braga A (eds) Police innovation: contrasting perspectives. Cambridge University Press, Cambridge, pp 267–283 Skogan WG (2006) Police and community in Chicago: a tale of three cities. University of Chicago Press, Chicago Weisburd D, Mastrofski SD, McNally AM, Greenspan R, Willis JJ (2003) Reforming to preserve: COMPSTAT and strategic problem solving in American policing. Criminol Publ Pol 2:421–456 Weisburd D, Mastrofski SD, Greenspan R, Willis JJ (2004) The growth of Compstat in American policing. Police Foundation, Washington, DC Willis JJ, Mastrofski SD, Weisburd D (2004) COMPSTAT and bureaucracy: a case study of challenges and opportunities for change. Justice Quarterly 21:463–496 Willis JJ, Mastrofski SD, Weisburd D (2007) Making sense of COMPSTAT: a theory-based analysis of organizational change in three police departments. Law and Society Review 41:147–188 Willis JJ, Kochel TR, Mastrofski SD (2010a) The co-implementation of Compstat and community policing: a national assessment. Department of Justice, Office of Community-Oriented Policing Services, Washington, DC Willis JJ, Mastrofski SD, Kochel TR (2010b) Recommendations for integrating Compstat and Community Policing. Policing: A Journal of Policy and Practice 4:182–193

Computational Criminology ▶ Agent-Based Assessments of Criminological Theory

505

C

Overview The 1s and 0s that make up digital data are incomprehensible to most people. Even though we have heard about them many times, we often have no idea how the 1s and 0s turn into useful data. Forensic analysts must make sense this data and present it to persuade others. This entry explains how the binary data (1s and 0s) are exactly equal to a slightly easier way of showing the data: hexadecimal. Hexadecimal is commonly seen as the most fundamental representation of the data. Once expressed, the organization of the data becomes easier to understand. Common structures and methods of discovering and explaining other structures are explained and then shown in two examples: carving a lost or deleted file and finding hidden data in a common JPEG photograph. In the first example, a step-by-step set of commands is used to find and recover a deleted picture. The commands demonstrate the same steps used by automated forensic packages. The second example examines data embedded in a JPEG photograph byte by byte. Using publicly available resources, the structures of embedded data are explained. With this information, an analyst’s ability to present data exceeds the ability of commonly used forensic packages. Along with the examples is a discussion of how these operations are handled by automated forensic tools. The whole is brought together in a discussion of how these techniques can improve testimony and the investigative effectiveness of a forensic analyst.

Introduction

Computer Forensics D. Kall Loper Department of Computer Science and Engineering, Lyle School of Engineering, Southern Methodist University, Dallas, TX, USA

Synonyms Low-level data forensics

Digital information in its raw form is expressed as ones and zeros, or more correctly as on’s and off’s. Each single piece of information: one or zero is called a bit. This understanding is part of popular culture, but is essentially useless. Almost all meaningful information is gathered into bytes or “words.” With Intel processors, the standard word is 2 bytes or 16 bits. Even though the fundamentals of chip architecture and low-level file structure may also seem useless to the day-to-day

C

C

506

business of an investigator or forensic analyst, the structure of data is critical to understanding digital evidence. While the majority of useful evidence comes from active user files, the majority of disk space is unallocated. This means that either direct interpretation of unallocated space forensic recovery or file carving is necessary to access that space. All three of these activities benefit from some knowledge of reading a hexadecimal presentation of the data. The term “hexadecimal presentation” is specifically chosen because it is merely another way to look at the data on the disk and in no way changes the content of the file.

Fundamentals of Data: An Explanation of Hexadecimal Presentation Data “presentation” refers to the way data is displayed to a user. A native presentation is the way the file was intended to be displayed to a user. Hexadecimal presentation shows the data in a raw form but summarized into couplets of hexadecimal digits. Hexadecimal digits include the familiar 0–9 of decimal but also include A–F. Where the more familiar decimal repeats its digits every 10 counts, hexadecimal repeats every 16. Since these counting systems use the same characters, it is common to specify which system is being used. Hexadecimal numbers are often followed by “xh” to denote hexadecimal (e.g., 00xh); similarly, decimal notation can be noted with “xd” if it is intermixed with hexadecimal. There is no mathematical difference between counting in hexadecimal (base 16), decimal (base 10), or even binary (base 2). Each notation can be converted to the other with absolutely no loss of information. For example, 255 (decimal) is exactly equal to FF (hexadecimal) which is exactly equal to 1111 1111 (binary). Internet addresses also demonstrate this fact. 192.168.0.1, a valid IP address, is expressed in decimal-dot notation. It can also be expressed in binary as 1100 0000. 1010 1000.0000 0000.0000 0001. However, it can most efficiently be expressed as C0.A8.0.1 in hexadecimal.

Computer Forensics

Hexadecimal notation is especially convenient for information based on 8-bit bytes. As can be seen in the IP address, two hexadecimal digits – a hexadecimal couplet – can completely express 8 bits. Rather than presenting eight digits in a row, the system uses two digits, making the information easier to interpret. The underlying structure of a file is often based on a byte structure. Either an 8-bit word (byte) or some multiple of them is used to hold data. In Windows programming, a 16-bit word is called a WORD, a 32-bit word is called a DWORD, and a 64-bit word is called a QWORD. When reading outside resources about file structure, these terms are often used freely with the assumption that the reader understands them. Resources that do not deal specifically with the Windows Application Programming Interface (WinAPI) may use “word” in a different way. Even when it only requires one or two digits to account for all possible values, a full byte or even a WORD or DWORD may be used to maintain byte alignment. This means that new fields within the data tend to start on the 0th, 4th, 8th, or 12th byte of a row of 16 bytes. Even when only single bytes are used, a full WORD or DWORD may be filled with padding to maintain byte alignment. Although any value can be used for padding, it is most common to find nulls (00xh). When working with tools that allow a user to change the number of columns displayed in hexadecimal, it can be useful to maintain the standard 16 byte width to allow easy recognition of byte aligned file structures.

Basic File Structures File types are structured according to the needs of their creators; thus, there are no universal rules about file structures. However, most file types use a header and body structure. Headers contain information about the file. A file signature value or “magic number” is often used to identify the file type. There is no guarantee that a signature value is unique to a given file type. The Web site wotsit.org maintains lists of file types by

Computer Forensics

extension and gives signatures values when available. Other commonly found fields in the header include file version number, count of records contained in the file, logical size of the file, checksums of file contents, and many others. Header information usually contains only information necessary to interpret the file structure and identify the file type. The body of the file contains the data used by the application interpreting the file. All of the possible data types cannot be listed here, but there are a few general types. Text-based file types tend to be presented in Unicode or ASCII equivalent. Most hex tools provide an ASCII interpretation to the right of the hex itself. Unicode corresponds to ASCII with Latin characters (the letters used in English). The only difference noticeable is that Unicode has columns of null (00xh) between the letters. Unicode needs the extra bits to encode accented Latin characters and non-Latin characters. However, in English they are not used so the bits are empty, thus null. The body of a file may also contain blended text and non-text regularly interspersed. This is the hallmark of a file with discrete records. Each record can either follow a precisely established format or contain its own header information. The contents of the record may need to be interpreted or may be readable as text. Repeating record structures make it much more likely that a file fragment will provide useful information. For example, partial Web browser history files can be recovered and read from the hex without recovering an intact file. In many cases, the human eye can spot repeating patterns or legible fragments of files that cannot be identified by automated file carving software. A few regular expression searches (RegEx) and a little knowledge of file structure can produce accurate and precise results.

Encapsulated File Structures: Encryption, Compression, and Encoding Sometimes it is not immediately apparent that data is part of a known file type. This may be because the data is encapsulated or layered in

507

C

other forms of processing. It is common practice in computer science to break a complex task into layers of activity. Each layer is responsible for a discrete task and either passes up or down to the next layer without interacting with the contents. In this way, multiple layers can be imposed on the data found on the disk. Processes that encrypt, compress, or merely convert data can make it impossible to directly analyze the data. The data on disk in no way resembles the original file data when viewed directly, but the original file data can be easily recalled by reversing the encapsulation or layering process. For useful analysis, the data must be converted to a readable format. Automated forensic packages can process this data back through the layers that were imposed on it or it can be done by isolating the file and processing it with a stand-alone tool.

Accessing the Data There are many tools available that allow an analyst to access data with a hexadecimal presentation. Rather than try to exhaustively list them, an example of each type can serve to illustrate them. XXD is a utility found on Linux systems that will present a file in hexadecimal. It can also be used in Windows with the Cygwin utility. XXD is called from the command shell (e.g., Bash). Unless the file to be examined is small, it is best to send the output to “less” or a GREP search as follows: XXD/evidence/image.dd | grep ffd8ffe1. This line will search a forensic image for a JPEG signature characteristic of a JPEG picture with EXIF data embedded in it. WinHex is a full featured hex editor that is commercially available. Care should be taken when using hex editors to not write to the file being examined. The interface is familiar to Windows users. Familiar menus and dialog boxes allow the user to open a file in hex presentation. Additional dialogs allow a user to open a partition or physical disk the same way. FTK Imager is a commercial-grade imaging and data access tool provided at no cost by AccessData. Imager can open physical devices (disks, CDs, flash drives, etc.), partitions, or files. A file system

C

C

508

hierarchy view similar to Windows Explorer allows a user to traverse a file structure and display data in hexadecimal or in its native presentation. FTK Imager is intentionally incapable of performing searches in the data.

Automated Forensic Packages and Hexadecimal-Level Analysis Automated forensic packages allow analysts to rapidly conduct searches across multiple data sets and access archive files (layered data) without intermediate steps. These tools are not rivals or replacements for lower-level tools; quite the contrary, they include low-level tools with other functions that add greater convenience for the analyst. Automated tools can contextualize the data within the file system or within search results. Automated packages can also prevent human error found in complex, repetitive tasks; however, they can mask unique aspects of the data through consistent presentation. For example, some forensic software takes a file that was never meant to be viewed by a user and presents it as a table of values or an html presentation (shows it as a Web page). While there is no “native” view of the file, hexadecimal presentation is the most accurate. If the software cannot interpret a field of data, it may simply be skipped in the “user-friendly” presentation. If the user relies on the tool, then there may never be a chance to see the skipped data until an opposing expert brings it forward. Most forensic experienced users are aware of the warnings to “know” their tools and the tool’s limitations, but without a firm basis in interpreting the hexadecimal data, they are left thinking that comparing the output of two forensic tools works as validation; it does not. Automated packages can also introduce a bias for intact data that can be cleanly presented by the package. There is no denying that automated packages are necessary to process the massive amounts of data found on even a moderately sized corporate system (very roughly 40 GB). It is self-evident that automated packages are needed for home systems that can run into the

Computer Forensics

multiple terabytes. Automated packages tend toward a clean presentation of native views. With huge amounts of data available, there may be justification in searching visible user files first. As this practice is taught to new examiners and becomes ingrained in experienced examiners or, worst of all, becomes part of policy in a police department or forensics practice, it leads to a bias toward intact data of known types. It is easy to automate searches in EnCase™ with EnScripts™. If policy is in place to exclude unknown file types or search only intact data, large amounts of relevant data can be completely missed. It is easy to avoid this with the full text indexing and searches found in FTK: the Forensic Tool Kit™ by AccessData. However, users soon learn to ignore or even explicitly filter out results from unallocated space or files fragments in favor of those that can be cleanly presented. Users of either package can overcome these limitations, but there is a tendency to become complacent with the tool. By getting in the habit of examining “corrupt” files, unknown file types, and fragmentary data, a forensic examiner builds a mastery of forensic examination that goes beyond document review or tool dependence.

Interpreting File Content The two lines in Fig. 1 are output from a hex editor. The first line identifies the column (in hexadecimal, there are 16xd columns). The second is the first 16 bytes of a JPEG photograph with EXIF data embedded. The JPEG signature is FF D8 FF E1xh (bytes 0–3). The JPEG type identifier is 45 78 69 66xh (bytes 6–9). This is followed by two null bytes: 00 00xh (bytes A–B). The next two bytes identify the order of following information 4D 4Dxh (bytes C–D). The ASCII interpretation of the bytes (4D 4Dxh) as MM indicates that values should be read left to right as read by Motorola chips (called big endian byte order). The other possible value is “II,” indicating that values should be read right to left as read by Intel chips (called little endian byte order). Much more information can be found in the following lines like the make and model of the camera used

Computer Forensics

C

509

0

1

2

3

4

5

6

7

8

9

A

B

C

D

E

F

ASCII

FF

D8

FF

E1

2F

F7

45

78

69

66

00

00

4D

4D

00

2A

ÿØÿá/÷Exif

MM *

Computer Forensics, Fig. 1 The first sixteen bytes of a JPG photograph with the hexadecimal and ASCII presentations of the data

C 0

1

2

3

4

5

6

37

B0

3D

65

B9

FF

D9

7

8

9

A

B

C

D

E

F

ASCII 7°=e¹ÿÙ·······

Computer Forensics, Fig. 2 The last seven bytes of a JPG photograph with hexadecimal and ASCII presentations of the data

to create this photograph. The GPS coordinates are also recorded in this particular photograph (not shown). Without an EXIF aware viewer, an analyst would not know this data was present without viewing it in hexadecimal; the userfriendly presentation shows only the photograph. There are thousands of file types in common use, it would be impossible to catalog them all in this entry, but technical details can be found on the Web in a few common resources. Microsoft file formats can be found in Microsoft TechNet or in the Microsoft Developer Network (MSDN). To identify an unknown file signature or extension, wotsit.org maintains voluminous catalogs of file types. Details of open-source file types can often be found through their SourceForge projects. Some file types are maintained by work groups comprised of representatives from many companies, academic institutions, governmental bodies, etc. Web searching these resources or simply looking for them in the results of a Web search for the first 4–8 bytes, in hexadecimal, from an unknown file will help find authoritative and reliable sources for file formats.

Recovering and Interpreting Data Files that have been deleted or corrupted can be recovered by using the techniques described in this entry. File carving is the process of recovering a file or fragments of a file from unallocated space (where the file system has marked the space as unused). The easiest way to do this is to search

for a header signature (FF D8 FF E1xh for a JPEG picture) and copy all the data between it and the file footer signature (FF D9xh for a JPEG) as seen below in the last line of a JPEG file (Fig. 2).

Recovering a File from Start to Finish In this example, a JPEG photograph was copied to 128 MB flash drive formatted with the FAT32 file system. An image of the physical drive was created with the Linux/Unix utility “dd” using a SMART boot disk on a host system. SMART is a commercial forensic package created by ASR data, but any forensically sound Linux boot disk will work. The raw, bit-stream copy of the flash memory is analyzed using hexadecimal tools mentioned previously. The purpose of this example is to comment on various file structures along the way to recovering the file. A few pointers on the commands will help in understanding what is happening. All commands are issued in a terminal window in the Ubuntu operating system. Multiple commands can be strung together with a “pipe” character (|). Commands can also have options that change the way they act. Each of the following commands defines an operation in the recovery of a file. These same operations are used by automated packages. The automated packages do not actually implement the Linux/ Ubuntu commands, but they have the same functions. The first step is to find the file to be recovered. As seen previously, the signature of a JPEG file is

C

510

FFD8FFxh. Choosing to include the fourth couplet, either E0xh or E1xh, limits the number of files found. Removing it has other interesting effects as will be seen later. “xxd” converts a file from a binary stream to a hexadecimal stream. By refining it with command options, it can be used to extract specific information. “grep” is a Unix environment search tool that has been ported out to many other environments. By sending the output of an entire file in a hexadecimal stream to grep, an analyst can search that stream very efficiently. The image file is named “image_dd.001.” UBUNTU Command: xxd -g 0 image_dd.001 | grep ffd8ff Output: 0400400: ffd8ffe000104a464946000101010048...... JFIF.....H By default xxd lists 16-byte lines with a hexadecimal line number. In this case, the line number is 0400400xh. It is not obvious, but that line number is a good sign for carving a complete file. Each sector is 512xd bytes. 512xd equals 200xh. Thus, any line number evenly divisible by 200xh is the first line of a sector. Most files start on sector boundaries, not in the middle of a sector. The first four bytes are ffd8ffe0xh. This is one of the possible JPEG signatures. These bytes are followed by the remaining 12 bytes in hexadecimal followed by the ASCII conversion of the data. One other consideration is that xxd gives output in a conveniently grouped byte pattern by placing a space between each hexadecimal couplet. “xxd –g 0” removes the grouping and makes the output searchable by grep. It is necessary to convert this line number to decimal. In Windows, open calculator and choose scientific view. Enter 400400 in hexadecimal and convert to decimal. In a Linux command line, also use a calculator: bc. Echo directs output to stdout (the terminal or a pipe). “bc” is the binary calculator. The following command tells echo to repeat the information “ibase ¼ 16; 400400” to bc. bc interprets the input base (ibase) as hexadecimal (16) and the number as 400400 with no other operations. The default base for bc is 10, so the output base stays 10. The output is the

Computer Forensics

400400xh converted to decimal. When using bc, remember the default input in hexadecimal is uppercase, but the default output of xxd is lowercase. This will be shown in the next example. UBUNTU Command: echo "ibase ¼ 16; 400400" | bc Output: 4195328 Once the header location is known, the footer signature location must be found to locate the end of the file. Line 400400xh was the first to contain a JPEG signature, so all previous lines can be skipped. Once again, xxd and grep can locate this value. UBUNTU Command: xxd -s 4195328 -g image_dd.001 | grep ffd9 Output: 0406 fc0: 20 86 73 13 7a ee 7f ff d9 00 00 00 00 00 00 00.s.z........... The line number is 0406fc0xh. The first nine bytes are part of the JPEG file. Bytes number eight and nine are the footer signature value (FF D9xh), marking the end of the file. The trailing nulls (00xh) are file slack space. It cannot be seen from the output above, but there are actually 55 bytes of file slack space before the end of the sector and in this case, the beginning of the next file. Those bytes could contain fragments of previous file(s) stored in that sector. In this case, they contain nulls because the media was securely deleted before use in this example. Again, it is necessary to convert the line number to decimal, but also to add the 9 bytes found in that line. The direct output of the line number needs to be converted to uppercase or produced from xxd using the “-u” switch to make the hexadecimal digits uppercase. The default output of xxd is lowercase and bc requires uppercase. Since decimal and hexadecimal use the same characters for 0–9, it may be hard to tell that the additional nine bytes are added in hexadecimal. UBUNTU Command: echo "ibase ¼ 16; 406FC0 + 9” | bc Output: 4222921 A final calculation will yield the file length. This can also be accomplished with a simple calculator.

Computer Forensics

511

C

0

1

2

3

4

5

6

7

8

9

A

B

C

D

E

F

ASCII

FF

D8

FF

E0

00

10

4A

46

49

46

00

01

01

01

00

48

ÿØÿá··JFIF·····H

Computer Forensics, Fig. 3 A JPG photograph header showing the JPG signature in hexadecimal and a file type identifier intended to be interpreted in ASCII

C UBUNTU Command: echo "4222921 4195328" | bc Output: 27593 With all the file location information available, the file itself can be carved out of the image and exported to a file system. In this case, it is simply exported to the working directory containing the image, but the output can be directed to any mounted file system. UBUNTU Command: dd if ¼ image_dd.001 of ¼ carved.jpg skip ¼ 4195328 bs ¼ 1 count ¼ 27593 Output: 27593 + 0 records in 27593 + 0 records out 27593 bytes (28 kB) copied, 0.515814 s, 53.5 kB/s There is also a new file called “carved.jpg.” It can be viewed with the xview command or with the tools of the desktop environment. To be clear, there is no difference between the carved file and the original. This can be seen by viewing the JPEG file and seeing the picture. It can also be mathematically confirmed by running an MD5 checksum of the original file and the carved file. In this case, the MD5s match, confirming that not one bit is different.

Interpreting Information in a JPEG File Aside from the picture itself, there is often other data embedded in a JPEG file. Two of the many types of JPEG file are discussed here. The first is the standard “JFIF” file. This is the type of file seen in the previous example. To comply with the JFIF standard, the first two bytes must be the image marker FF D8xh. The second two bytes are the application marker FF E0xh. These bytes are the header signature value. A five-byte

identifier follows these: 4A 46 49 46 00xh; the identifier has the ASCII values JFIF followed by a null terminator. This can be seen in (Fig. 3). The second type of JPEG file contains additional information embedded within the file. EXIF data is often included with Tagged Image File Format (TIFF) files and JPEG/TIFF files. With photographs, it is most often used to capture camera settings, manufacturer, model, and other details relevant to photographers. As cellular phone cameras have become more common, the data has begun to include other elements like GPS coordinates. EXIF data also includes metadata describing the JPEG file. It is possible to get creation date for the image from the camera rather than from the file system containing the image. Much of this information is typically available through media management software, but automated forensic packages have been slow to update the types of EXIF data offered. This means that without exporting each JPEG file and viewing them in media management software (not forensically sound software), a forensic examiner may miss critical data like the GPS coordinates of where the picture was taken or embedded data possibly used on social media sites. As an alternative, the analyst can learn how to decipher the contents of that data directly. In the following example, the first 48 bytes of a JPEG picture are used to illustrate how EXIF information is stored in TIFF and JPEG files (Fig. 4) (Table 1). The designers of the JPEG file, the Joint Photographic Experts Group, created the JPEG file to be versatile. A number of options were left open for future development or customization. Because of these fields could not be anticipated and listed in the original file specification, some fields were reserved specifically to describe subsequent field’s structures and locations. Image File Directories (IFDs) may either contain brief

C

512

Computer Forensics

0

1

2

3

4

5

6

7

8

9

A

B

C

D

E

F

FF

D8

FF

E1

2F

F7

45

78

69

66

00

00

4D

4D

00

2A

ÿØÿá/÷EXIF··MM·*

ASCII

00

00

00

08

00

09

01

0F

00

02

00

00

00

08

00

00

················

00

9E

01

10

00

02

00

00

00

08

00

00

00

A6

01

12

·············|··

Computer Forensics, Fig. 4 These are the first 48 bytes of a JPG photograph. The structure of these bytes is explained in Table 1, non-repeating elements of the header, and Table 2, repeated structures in Image File Directories (IFD’s)

Computer Forensics, Table 1 JPEG file header and partial application data Field SOI marker APPO marker APPO length Identifier Byte order Version Offset to IFD

Bytes 2 2 2 5 2 2 4

Sample FF D8 FF E1 2FF7 45 78 69 66 00 4D 4D 00 2A 00 00 00 08

Notes Start Of Image marker, also part of the file signature Application use marker, indicates the type of data in the file Length of application field: 12,279xd (varies with each file) “EXIF” and null terminator (00xh) “MM” indicates “big endian” byte order Version is always 42xd Image File Directories (IFD) contain EXIF data

Computer Forensics, Table 2 Image File Directories (IFD) Field tag (2 bytes) 01 0Fxh 271xd 01 10xh 272xd 01 12xh 274xd 01 1Axh 282xd Other IFDs omitted 87 69xh 34665xd

Field type (2 bytes) 00 02xh 00 02xh 00 03xh9 00 05 00 04xh

ASCII ASCII WORD Ratio

Length of field (4 bytes) 00 00 00 08xh 00 00 00 08xh 00 00 00 01xh 00 00 00 01xh

Data or pointer (4 bytes) 00 00 00 9Exh 00 00 00 A6xh 00 01 00 00xh 00 00 00 AExh

158xd 166xd 1xd 174xd

DWORD

00 00 00 01xh

00 00 00 0Cxh

204xd

data (IFD resident data) or indicate where more extensive data stores may be found by following a pointer to another location in the file. The IFD tag is shown in bold in the example above. Each IDF is 12 bytes. The table below explains the first two IFDs from the example above. The first field of the IFD identifies it as tag number 271 (10 F in hexadecimal equals 271 in decimal). This value can be found in the TIFF standard on page 35 and in a table on page 117. The TIFF standard document indicates that tag 271 is the make of the camera or scanner. The second field in the first IFD is the field type. According the TIFF standard on page 15, type 2 is 7-bit ASCII text. That means that the camera or scanner make will be reported in ASCII text. The third field of the first

IFD is the length. The make of the camera will be reported in 8 ASCII characters. The fourth and final field of the first IFD is either the data indicated (make of the camera) or a pointer to the make of the camera. The last field has four bytes; if the data length is greater than four, the value of this field is a pointer to where the data can be found. Counting 158 bytes from the APPO section (not the beginning of the file), there are 8 ASCII bytes that identify the camera’s make: “HTC-8900.” This cannot be seen the sample; the sample data ends before the third IFD’s field type is shown (Table 2). The second IFD indicates the model of the camera or scanner. With the camera phone that created this photograph, the make and model are

Computer Forensics

identical. The one difference that can be seen in the IFD is the pointer. The previous pointer was 158 bytes and the data occupied 8 bytes; the new pointer is 166 bytes (158 + 8 ¼ 166). The third IFD is the orientation of the picture. The field type indicates that this is a WORD (2 bytes). The length of field indicates that only one unit is used to store the data. Since the field type is WORD and only one unit is used, the data for this IFD occupies one WORD or 2 bytes of data. The data or pointer field will contain data if it is less than four bytes. This IFD has resident data. Only the first two bytes are used, so the remaining two bytes are padded with a null (00xh). The final IFD in the first group will indicate that there are no more IFDs with a pointer of 00 00 00 00xh or will give a pointer to the next IFD table. The tag type 34665 is reserved to indicate that the IFD does not contain displayable content but either a final entry or a pointer to the next set of entries. Tags higher than 32,768xd are considered private tags. They are typically used by organizations, hardware makers, or software makers for special data to be stored. There are no firm standards on where in the file an IFD can be stored. The presence of unique IFDs at unpredictable locations means that an analyst who is completely reliant on a given tool will likely miss data.

Both Finding and Interpreting Data in a JPEG Another lesson presented by exploring the EXIF data is that pointers within a file indicate the location of data anywhere within the file. EXIF data can be placed anywhere within the Application Use area, as defined by the APPO marker and length. The JPEG file standards document, used to decipher the bytes found in the JPEG header, indicates that a JPEG is a wrapper for content data. Part of that data is one or more pictures, but part of it can be text or some other form of data. The file examined for EXIF data actually contains two images. A search for the image marker: FF D8xh shows that there can be three markers in a JPEG file. There is one marker at the beginning of the file, another at the beginning of

513

C

a thumbnail, and a third at the beginning of the full-sized image. A fragment with a complete thumbnail may appear to be only a few lines of the greater image because the thumbnail does not display. The thumbnail may provide confirmation that the image was present on the system or it may show enough detail to prove useful. Without some knowledge of the underlying file structure, it is easy to ignore the image as being corrupt or incomplete.

Forensic Examination and Testimony It is rare that a forensic examiner making conclusions supported by the data is challenged. The most common scenario in law enforcement forensics is the simple production of documents or images that speak for themselves. The challenge often lies in finding the data and backing the results with properly documented procedure, but not always. The most common scenario for civil forensic examiners is simple preservation of data in a forensically sound manner and production of documents for review by either an attorney or a content expert. Again, the challenge lies in producing a result in a consistent and searchable format with well-documented procedure, but not always. For this type of case, an automated forensic package with consistent results from a process that can be replicated is the tool of choice. They allow large volumes of data to be processed reliably. For a subset of investigations in both civil and criminal forensics, the analyst must go deeper to understand what has happened to a particular file or in the system itself system. In the example above, the EXIF data contained in the picture includes GPS coordinates of where the picture was taken. The EXIF data in that picture also contains date and time stamps. Such times might differ from the file times that were created when the picture was downloaded from the camera. The EXIF data contains the make and model of the camera which can help match the camera to one seized at a crime scene. All of this information can help make a case, if the investigator is aware of it. In civil forensics, an “expert” may

C

C

514

testify that all of a given type of file was produced to the opposing side during discovery. Without the ability to interpret and understand file types that are not known to the automated package used, the “expert” may miss an entire file type with crucial data. Almost worse, the “expert” may not warn the attorney of the presence of such information. An entire legal strategy may fail based on a key fact that was absent to the uninformed side. Using the data from the JFIF and EXIF examples, it has been shown that files in unallocated space can be identified and recovered in a process known of file carving. By first locating the header signature of a file, then locating the footer signature, and copying all the data in between, an analyst can carve a file from unallocated space. This allows the analyst to demonstrate an understanding of the process used by automated tools that carve files. If segments of the file are stored in more than one location, a state called “fragmented,” an analyst can still derive useful information from EXIF data found in the initial fragment. Garfinkle (2007) published data that estimated 92 % of files found on an NTFS volume were intact or fragmented once. This means that most files can be found and carved intact. The remaining, fragmented, files can still yield useful results. If fragmentation is minimal, then file structure may indicate how to carve and combine the pieces.

Conclusion Understanding the structures of digital files is more than an academic exercise. Analysts who limit themselves to the output of the tool do not bring any special skill to their analysis. Automated tools play an important role in rapidly processing large amounts of information, but they should not be allowed to set boundaries on the evidence. Restricting the data considered to easily displayed files places boundaries on the data. The inability to interpret file fragments places boundaries on the data. Even self-evident results can be successfully challenged by calling into question the process that produced them or

Concept of the Police

the knowledge of the person who chose the process. Sometimes, an opposing expert can even pray on the ignorance of an analyst. The analyst either questions his or her own results or argues beyond his or her knowledge. The best defense to such challenges is a deep and thorough understanding of how the tools work and the structure of the data found as it sits on the disk.

Related Entries ▶ Automated and Manual Forensic Examinations

Recommended Reading and References Adobe Systems (1992) TIFF: Revision 6.0, http://exif.org/ TIFF6.pdf Aware Systems (2008) Tiff tag reference, GPS tags, http:// www.awaresystems.be/imaging/tiff/tifftags/ privateifd/gps.html Becker R, Payne S (2011) International data forensic solution center: computer forensics investigation training. International Data Forensic Solution Center, Phillipsburg Garfinkle S (2007) Carving contiguous and fragmented files with fast object validation. Digital Invest 4(S 1):2–12 Maischein M (2011) The TIFF file format: I-TIFF, http:// www.fileformat.info/format/jpeg/corion.htm. Murray J, Van Ryper W (1996) Encyclopedia of graphics file formats, 2nd edn. O’Reilly Media, Sebastopol, CA National Digital Information Infrastructure & Preservation Program (2011) Still images: tags for tiff and related specifications, http://www.digitalpreservation. gov/formats/content/tiff_tags.shtml Nugent T, Moolenaar B (1996) XXD manpage. http:// www.unix.com/man-page/OpenSolaris/1/xxd/ Pal A, Memon N (2006) Automated reassembly of file fragmented images using greedy algorithms. IEEE Trans Image Process 15(2):385–393 Pal A, Sencar H, Memon N (2008) Detecting file fragmentation point using sequential hypothesis testing. Digital Invest 5(S 1):2–13

Concept of the Police ▶ Democratic Policing

Conceptualizing of Police

Conceptualizing of Police Fabien Jobard Centre de Recherches Sociologiques sur le Droit et les Institutions Peńales (CESDIP), CNRS, Guyancourt, France

Overview Police as an institution cannot be said to have inspired in-depth conceptualization efforts from criminology and criminal justice so far. Scholars in this area show a clear preference for empirically exploring what individual police officers do or think, and the policing concept they most readily converge towards tends to hinge on the use of force, a notion developed by ethnomethodologist, Egon Bittner. This concept holds that what constitutes policing as such is “the distribution of nonnegotiably coercive force.” In another tradition, stemming from European political philosophy, police are heir to a dual dimension, i.e., its relationship to both knowledge and dogma. However, numerous affinities between these two schools of thought – the Anglo-American sociological school, with its interactionist inclinations, and the European school of political philosophy – suggest that the concept of police should in fact be understood through a dualist epistemological approach. Both in legal and knowledge terms – two fundamental dimensions of policing – police as an institution is highly idiosyncratic in that it concomitantly harbors both (a) the rule of law and violence and (b) knowledge and ignorance.

Fundamentals of the Police Concept The concept of police currently used in social science has been derived mainly from the legacy of American ethnomethodologist Egon Bittner, who considered the use of physical force as the “core of the police role.”

515

C

Egon Bittner and Nonnegotiable Coercive Force From his field observations of policing patrols, Egon Bittner was able to characterize the police by their distinctive function: “The role of the police is best understood as a mechanism for the distribution of non-negotiable coercive force employed in accordance with the dictates of an intuitive grasp of situational exigencies” (Bittner 1974/1990: 131). Situational exigencies are not defined in and of themselves but are left to the appreciation of the police the field: “somethingthat-ought-not-to-be-happening-and-about-whichsomeone-had-better-do-something-now” (Bittner 1974/1990: 249). Virtually all policing scholars today rely on this definition as their starting point. With it, the concept of police follows in the footsteps of early twentieth century German social scientist Max Weber, who had defined the state as an organization that retained a “monopoly on the legitimate use of physical force.” This concept is currently known as the Police Use of Force Paradigm (PUFP) and has several implications. First, whatever force is employed ought to be minimal, i.e., the exact amount of force required by the situational exigencies that are being dealt with, which should not be exceeded. Such a force is neither destructive nor annihilative, as opposed to military force. Besides all this, force is the natural, mechanical outcome of common-sense behavior in a critical situation. Bittner here still appeals to policemen’s intuition: to deal with troubles, policemen, ordinary individuals in a pacified society, draw upon their intuitive perception of situations in order to anticipate what the community is expecting from them. Force, with PUFP, is thus located on the demand, instead of the supply, side: “citizen demand is a factor of extraordinary importance for the distribution of police service” (Bittner 1974/1990: 252). It follows that PUFP, as a concept, pertains to police self-restraint, which gives sociological resonance to European legal theories on selfrestraining state power. In this respect too, it is linked to the concept of legitimate violence

C

C

516

introduced by Max Weber. The basic assumption, however, is that the police are the institution that intervenes in situations where the ultimate capacity of the police to dominate is not in doubt, as opposed to circumstances of armed rebellion and warfare, where military force is necessary. To Bittner, policing rests on social consent, which is why he considers that the police should not be resorted to in civil war situations, for instance, or even in the context of crowd control – a function that to him is devoted to military corps such as the gendarmeries in Europe or the National Guard in the USA. Controversies on the PUFP Model Canadian scholar Jean-Paul Brodeur has produced a powerful critique of the PUFP (Brodeur 2010: 103–128). He criticized Bittner for not discriminating between competence and performance. Indeed, empirical research has consistently shown that force is actually quite rarely resorted to by even the uniformed police patrolling the skid-row areas of large metropolises. Brodeur argued that a concept of police cannot be grounded in what the actual police rarely or never do: the concept should target what policing is, not what it may, or might or might not be (assuming a situation where the police, for want of public consent, would be unable to resort to force. . .). To Brodeur, the key and actual feature of any police was “legal lawlessness” (Brodeur 2010: 6) “Policing agents are part of several connected organizations authorized to use in more or less controlled ways diverse means, generally prohibited by statute or regulation to the rest of the population, in order to enforce various types of rules and customs that promote a defined order in society, considered in its whole or in some of its parts” (Brodeur 2010: 130). This concept is by no means incompatible with the PUFP; in fact, it proceeds from it: force, considered by Bittner to be the “core of the police role,” here becomes one of the many means of expression of the “legal lawlessness” granted to the police. Other authors, such as English historian Robert Reiner, have voiced suspicion towards more normative aspects of the PUFP. Reiner considers such concepts as policing by consent and demand-side


force (Reiner 1992: 139–146) as delusive. Reiner’s critique targets an entire trend of thought in AngloAmerican social science which, like Michael Banton for instance, concludes that: “The policeman on patrol is primarily a ‘peace officer’ rather than a ‘law officer’. Relatively little of his time if spent enforcing the law in the sense of arresting offenders; far more is spent ‘keeping the peace’ by supervising the beat and responding to requests for assistance” (Banton 1964: 127). Reiner, on the contrary, holds that if policing were done out of consent, there would be no need for resorting to force. Contemporary policing, however, is increasingly focusing on force and coercion. While most cases of domestic violence used to elicit mere verbal warnings, they are legally repressed today, and the police shall use coercion. Hence, to Reiner, the core of the police concept is not the potential capacity to use force, but “order maintenance,” i.e., the capacity to prevail over individuals as well as crowds. In both the USA (Kraska and Kappeler 1997) and Europe (Porta and Fillieule 2004), policing organizations appear to be increasingly militarized, especially for crowd control operations, as opposed to having their role restricted to social work and peacekeeping. Neither peace nor consent are given; actually, they might be might be further removed today than ever before. New Perspectives: Consent or Conflict? In a series of interviews with Brodeur, Egon Bittner has elaborated upon his concept of police (Brodeur 2007). His approach was an etymological one, strongly associating the concept of “police” to the Greek etymology “polis,” meaning “city.” In this perspective, what the police are dealing with is “a ‘polis’ that already redefines the minimal state, where it literally consists of the organization of conditions of urban life.” In this setting, what then the state provides is “creating conditions for the orderly coexistence of strangers” (Bittner in Brodeur 2007: 111). With such a definition, Bittner knows he is getting exposed to a conceptualization of the police that is not consensual but agonistic, i.e., a conception of policing that no longer rests on trust as a necessary preamble to any society but


on conflict understood as a key element of said society. In this perspective, policing could be said to be defined no longer by its means but by its ends, and its primary end or finality is to maintain order, i.e., ensure the coexistence of social groups in the city, in other words to perpetuate the (social or economic) domination of a given group over minorities. This perspective is very much in line with minority threat theories, which are quite popular in the USA and have used econometric analysis to show a strong correlation between police strength and the presence of black minorities (Kent and Jacobs 2005). The notes published by African-American writer James Baldwin, in which the police patrolling in urban ghettos are compared to “an occupying soldier in a bitterly hostile country,” are one in many testimonies of this (Baldwin, More Notes of Native Son, 1962, quoted by Sklansky 2005: 1772). In Europe, historical research usually focuses on the riotcontrol origins of policing organizations, whose primary goal was to control the urban proletariat on behalf of the “peace-loving propertied classes” mentioned by Silver (1967). In this perspective, the concept of police is correlated to supply, i.e., the state as the organization that perpetuates the domination of a given group over the others. In Bittner’s words, “the thorny question is whether police force can be justified on the supply side” – in Brodeur 2007: 118. It is in that spirit that Peter K. Manning, while still claiming to get his inspiration from a Bittnerinspired concept of police, offered the following conceptualization: “The police as an organization in Anglo-American societies, constituted of many diverse agencies, are authoritatively coordinated, legitimate organizations (. . .). As such, they require compliance to command from lower personnel and citizens and the ability to proceed by exception.” (Manning 2010: 68). Clearly, the order that the police are supposed to maintain is not just any order, or even the status quo, but a substantially defined one. In a nutshell, the police embody one order (the political order of the state) and defend another (the social order as is). Twice anointed, the police are “sacred”: to Manning, “the police engage in a form of magic” (2010: 59), and do so by embodying a transcendental entity –

517

C

public authority, the state, the collective – and deploying an imagery of self-sacrificial symbolism (whose most visible expression is the ritualized death of a policeman – see Manning in this volume). To Loader and Mulcahy (2007), this political disposition translates into local cultureabiding behavior: social order is not created by the police, but simply consolidated by them. To perform these political functions, the police may resort to exception. This notion originates in the work of German jurist Carl Schmitt, a dissident disciple of Weber and a controversial as well as influential figure, given his full allegiance to the Nazi regime. The notion of exception is useful to get a proper grasp of the specific nature of the police – both inside and outside the law – and to explore the concept of police in greater depth (Jobard 2012). Because it is unequal, social order allows the deployment of unequal policing. In relegated social spaces, where demand for policing is strongest, not only is policing supply scarcest, it is also the least democratic of all: ruthless, biased, undisclosed, and unfair. Most importantly, given the lack of resources and credibility of residents in these areas, police action is seldom likely to be challenged in court. In that sense – and barring exorbitant displays of police violence or irrefutable evidence such as video footage (Skolnick and Fyfe 1993) – police in these areas may act without risk of being disciplined by public justice, i.e., in an exceptional legal space. As summed up by Philip Stenning, “while public police are theoretically accountable for excessive uses of force through the processes of the criminal and civil law, it is well known that in practice it is very difficult to obtain either criminal convictions or civil judgments against public police officers for such wrongdoings” (Stenning 2000: 337). According to Schmitt, however, any entity that may behave this way in an exceptional situation is nothing short of the “sovereign,” i.e., the owner of the political, who designs the political borders of a society, defines who the enemy is in this society and who the friend is (Schmitt 1932). Hence, exceptional actions – i.e., decisions made with no risk of punishment in exceptional circumstances or spaces – are the actions that, according

C

C

518

to Schmitt, assert the political in our societies. In this perspective, assuming and recognizing that the police has a capacity to act in the context of exception, then the police can only be said to be the institution that brings to modern, rational, bureaucratic orders an irrational, passionate, untamed dimension, which is that of the political. The Many Expressions of the Concept of Police The way of the concept of police in social science has been a meandering one. Some contend that policing has more to do with social work than force nowadays. To others, force is just one element in the bigger picture, certainly not the whole picture, and policing action is characterized by its unpunished infringements on that element. What characterizes the police, then, is not coercive force but legal lawlessness. To others still, the police have to do with a disposition to order maintenance. In this perspective, police as an institution is rooted in adversity against the dangerous classes and is in charge of protecting the dominant groups. In another acceptation, more grounded in political theory, the concept of police would not be distinct from that of political order: the police institution embodies and defends the political order and can even be said, more specifically, to embody the political and its own particular passion in modern, rational, disenchanted societies. The police, from this point the view, is an institution that, despite being laden with modernity, rationality, and technology, carries along a political essence that escapes the Weberian modernization and disenchantment of the world (Weber’s “Entzauberung der Welt”). These various contemporary understandings are rooted in older, more European conceptions, those associated with policing sciences that started to develop in mainland Europe as early as the Renaissance.

The Police: History of a Polysemous Concept We have been concerned so far with how social sciences may conceptualize policing as it occurs


in contemporary societies. Academic usage would rather have had us describe the history and genesis of a concept before examining its most current developments. Contemporary conceptualization of the police, however, is peculiar in that it has been developing as if detached from historical knowledge. There are two reasons for this. First, the field has been dominated by North American social scientists, whose policing institutions, as we will see, have cultivated some distance with their mainland Europe counterparts during their development. As a matter of fact, however, the latter are what the police concept has been drawing upon ever since the seventeenth to eighteenth century. The second reason is the specific fields of said North American social scientists: being sociologists, ethnomethodologists, ethnologists, and sometimes jurists, their concern has never been to anchor their conceptualizations in the history of political ideas. What was being favored was the conceptualization of whatever was immediately observable. Yet the concept of police does have a history, and this history helps to understand certain aspects of policing today, especially its relationship to the political. Police Concept and Police Science Among the seminal texts on the history of the concept of police is Michel Foucault’s 1979 University of Stanford lecture, in which the French philosopher intended to describe “the rationality of state power (which) was formulated especially in two sets of doctrine: the reason of state and the theory of police” (Foucault 1981). For theoreticians of seventeenth- and eighteenth-century European police, “the police includes everything (. . .), sees to the living”: “‘Polizei’ is a positive task: it has to foster both citizens’ lives and the state’s strength (. . .) Life is the object of the police: the indispensable, the useful, and the superfluous. That people survive, live, and even do better than just that, is what the police has to ensure” (Foucault 1981). Theoreticized a century later by Hegel, “policing” is this government competency that is used to improve the living conditions of the people, perpetuate the population, and give it strength.


Everything falls within the realm of the police: the trade of grain, ideas, and beauty; the movements of men, capital, and cattle; the increase of well-being; the decrease of child mortality, illness, or intemperance; security in the widest possible sense; and the “abolition of disorder” (Knemeyer 1980). In such a context, policing knows no border. The police institution, as defined in “police science” textbooks, is totalitarian in that it aspires to take charge of the entire life of individuals as well as their interpersonal relationships. The policing project, the project of police science craftsmen, is to consider the population as an asset and its growth as a goal. Therefore, in European absolutist states, “police” encompasses repression and prevention, the mores, the culture, and the well-being; “police” is anything related to what is referred to as “social policy” or “social control” (Pankoke 1986). This is where the connection with the reason of state lies. The police are comprised of all the government institutions that contribute to state power towards the well-being of the population. However, the latter may quickly be relegated to the backdrop of the former. The object of policing is knowledge on things and people, the police is “the eye” that public authorities keep on society (L’Heuillet 2007). This is precisely where policing meets the reason of state: since the police contribute to the efficiency and profitability of governmental action, they are a reason of the state, supported by new knowledge such as statistics or accounting; since it conceives of no obstacle to the growth of the happiness of constituents, it stops at no means and may resort to force at any time. From this huge but scattered pool of knowledge that fed “police science” handbooks, one area was particularly successful in its development, i.e., the branch dealing with the forecasting, management, or repression of disorders and irregularities. The mad, the sick, the vagrant, and the young – unmarried and unattached – were increasingly targeted by scientists in these fields (Pasquino 1991). This branch of knowledge bridged the gap with “social policy” as a whole and, all along the nineteenth century, conquered its autonomy within policing knowledge as it is

519

C

known today: knowledge on crime and its various forms; bolstered surveillance of specific social environments; focus on juvenile crime or prostitution and later on foreigners; development of specific techniques for the tracking of individuals, from passports to physical descriptions through the spoken portrait, followed at the turn of the century, under the influence of Italian Lombroso and Frenchman Bertillon, by the development of various anthropometric techniques aiming to spot and arrest criminals and even “born criminals” (Piazza 2011). The Shapes of Police Science in the Classical Age The notion of police was born with the absolutist state. Starting from there, however, various versions of police science started to branch out, even leaving the Anglo-Saxon world aside. The most significant version, in terms of both the sheer size of its knowledge legacy and its influence on subsequent political philosophy (including Hegel’s, already mentioned), is that of German Polizeiwissenschaften, sometimes also referred to as “Kameralwissenschaften” or cameralism (named from the prince’s “chamber,” his cabinet, which points at the secrecy of consultations with or by his advisors). Around the late seventeenth century, in German states, “cameralists” started to shape the body of knowledge that was to lead to modern-day governmentality – as described by Michel Foucault – as well as police science and its focus on maximizing the productivity of both the social body and the state (Wakefield 2009). German cameral science has bequeathed us a huge corpus: more than 3,000 German language books on policing are known to have been published from 1600 to 1800 (Pasquino 1991; Stolleis 1996). The French kingdom has also produced an impressive number of reports, which however were established by public agents who had more to do with the policing function in its modern sense. Indeed, as early as 1667, the French monarchy had established a police authority in Paris, followed by a nationwide body of police commissioners and inspectors: the reports and other written documents that constitute French policing

C

C

520

science are in fact their correspondence. Nicolas Delamare, often mentioned by Michel Foucault, was one of them. “Policing science” may be an overstatement here, since – as opposed to what was going on in Prussia – this knowledge was neither formalized nor academic. It was practical knowledge on specific urban areas, focusing less on the general economy of the nation than on public order in a given city. Already a motley network of informants was playing a substantial part in the manufacturing of this knowledge on the city and the street. Strangely enough, policing science never reached the shores of Albion. While a few thinkers have used the term, thus testifying to the existence of a form of police science, they were quick to downplay its scope. Patrick Colquhoun, in his Treatise on the Police of the Metropolis (1795), for example, stated that “Police in this country may be considered as a new Science; the properties of which consist. . . in the Prevention and Detection of Crimes; and in those other Functions which relate to Internal Regulations for the well ordering and comfort of Civil Society.” Similarly, here is how Edwin Chadwick introduced the police in the report he produced for Sir Robert Peel in 1929 on “preventive police”: “the first great object of a Police, that to which every practical adoption should conduce, is to prevent the commission of crime. The second is, when crime has been committed, to detect and to bring to Conviction the perpetrators of it” (both quoted in Neocleous 1998: 440). In England, this rejection of policing as a state science is but the corollary of the rejection of continental mercantilism and more broadly the notion that the development of society should be mediated by the inquisitive knowledge and unlimited power of the state. In the spirit of the economic “laissez-faire” doctrine, the AngloSaxon notion of policing was quickly reduced to institutionalized force. Still, the linkage between police as an instituted force and police science – i.e., between ancient and modern conceptions of the police – is more than incidental. In our view, it is actually central.


Police Science and Modern Police Agencies What is the actual kinship between this police of the whole, as conceptualized by classical age police science, and contemporary policemen – patrolling our streets and collecting samples from bloodstains in our TV shows (as they also do in real life, only less frequently)? Both a conceptual link and an institutional historical process exist. First, let us explore the institutional process. The entire body of knowledge collected by police science, as well as knowledge tools such as accounting or statistics, all contributed to the birth of bureaucracy, this multifaceted government body overseeing whatever pertains to the knowledge required to ensure the prosperity of the state and its constituents. Alongside bureaucracy, the police pursued its development, but did so as an institution that had specialized in one function that of preventing: preventing disorder, preventing crime, and helping justice elucidate crimes. It follows that the police is not a branch of the bureaucracy: while bureaucracy belongs to the sphere of dogma, policing is circumstantial; and while bureaucracy is a vector of Weberian modernization, policing retains an ancien regime flavor, grounded in secrecy, dissimulation, and the general techniques of absolute power. France is a case in point. The French Revolution had no choice but to retain the police force established in 1667 by the absolute monarchy, with its urban brand of policing based on the surveillance of the territory and its dwellers, hinging on secrecy and malice, shielded from the institutions of control that the bureaucracy was starting to implement on its side. The point is that the police, in the minds of the 1789 revolutionaries (the decree on municipal police was promulgated on 14 December 1789), is a preamble to liberty: only enforced public tranquility makes it possible for a constitution (understood as a solemn break from absolutism) to be respected. Tranquility conditions liberty, and proper policing is a practical requirement for implementing a future constitution. Policing is thus circumstantial, paving the way for the rule of law and therefore preceding it. This is a decisive point, as it establishes the link between


Policey and modern police, between classical age concepts and contemporary conceptualizations. The Police and the Legal Order The reader may remember Bittner quoted as saying that the police realm of competence begins when something crops up “that-ought-not-to-behappening-and-about-which-someone-had-betterdo-something-now.” A few centuries earlier, French philosopher Montesquieu contended that “the business of the Police consists in affairs which arise every instant, and are commonly of a trifling nature: there is then but little need of formalities” (The Spirit of Laws, 1748: }25). Certainly, the classical age understanding of policing and the function of the police as it is conceived of in contemporary sociology are related. Continuity here is best expressed in urgency and the imperiousness of action (in }25, Book 26, Montesquieu gets more specific: “The actions of the Police are quick; they are exercised over things which return every day”). By so doing, with regard to the legal order, policing gets characterized by a singularity: police action cannot be entirely prescribed by the rule of law. Obviously, in contemporary democratic regimes, policing increasingly tends to be framed by legal standards; however, its fundamental characteristic is that it always eludes the law – otherwise the police would be powerless. Prudence is the name of the game (and prudence is one of the most constant traits theoreticians have been associating the police with, ever since the seventeenth century). With prudence, certainly, but also, as dictated by the circumstances, without restraint. As French MP Limodin put it while discussing some bill of law on policing in 1795: “policing implies, if I may say so, bordering on the arbitrary” (quoted in Napoli 2003: 220). Policing power thus appears to have been born in a legal interstice, namely, the “measure” (Napoli 2003, 2009). The measure is a legal text that does not really belong in the pyramid of standards; it is the form chosen by the government to regulate the minutest details in the lives of citizens. The measure aligns, regulates, and administrates first and foremost what can be

521

C

exchanged; in particular, it governs public roads and squares, as well as marketplaces. It will thwart whatever prevents exchanges and trade (crime as well as fraud or swindling – e.g., bad scale calibration or inaccurate bills) and, then, once the predators have been neutralized, will attempt to foster trade, increasing produced and consumed wealth. This precedence of policing over the rule of law and – to a stronger degree even – the subordination of the very existence of liberty to a policed order is what has prompted German philosopher Walter Benjamin to formulate an “ignominious” concept of police – which, although it has not been very popular with social scientists, has reached fame in philosophy: “(. . .) two forms of violence are present in an institution of the modern state, the police. True, this is violence for legal ends (it includes the right of disposition), but with the simultaneous authority to decide these ends itself within wide limits (it includes the right of decree). The ignominy of such an authority (. . .) lies in the fact that in this authority the separation of lawmaking and lawpreserving violence is suspended (. . .). It is lawmaking, because its characteristic function is not the promulgation of laws but the assertion of legal claims for any decree, and law-preserving, because it is at the disposal of these ends. The assertion that the ends of police violence are always identical of even connected to those of general law is entirely untrue. Rather, the ‘law’ of the police really marks the point at which the state, whether from impotence or because of the immanent connections within any legal system, can no longer guarantee through the legal system the empirical ends that it desires at any price to attain (. . .). Its power is formless, like its nowhere-tangible, all-pervasive, ghostly presence in the life of civilized states” (Benjamin 1921/1996: 242–243). Such a bifid and paradoxical disposition, which consists in making use of violence within the rule of law but also in making possible the very existence of the rule of law by resorting to violence, has always proved terribly challenging to reformers in their attempts to establish democratic forms of policing. Though the police, by

C

C

522

necessity, resist the imposition of external rules, they still have to yield to the rule of law; otherwise, policing becomes pure violence, governed only by the arbitrary, and tips the political regime scale to the side of the police state. In this tension between emancipation and domestication, the relationship between policing and knowledge becomes a key issue. Police and Knowledge: Knowledge and Culture The police are a dual institution, harboring both the rule of law and violence. Its relationship with knowledge is of a similar nature. Knowledge and the rule of law – as well as their opposites, ignorance, and violence – are constitutive elements of the concept of police. First, it should be noted that the police institution is also heir to classical age policy in that one of its tasks is to have knowledge of the society so as to anticipate whatever may threaten it. As early as the seventeenth and eighteenth centuries, surveillance activities have led policing apparatuses to produce considerable amounts of knowledge. Surveillance does not pertain exclusively to “high policing,” i.e., the political police. It trickles down to all levels of policing action, such as that of the patrolman engaged in a community policing program, who does his best to get a grasp of the life and people in the area. Knowledge on others is thus a prominent policing skill, so much so that some contemporary sociologists actually consider the police to be “knowledge workers” (Ericson and Haggerty 1997), because of their specific relationship to prevention, and thus to the forecasting, detection, and knowledge of the various risks that threaten society. As can be seen, the rule of law/violence pair is not the only one at stake in the conception of policing: knowledge also plays a significant part. If the notion of knowledge is to be expanded to a wider dimension – e.g., the cognitive dimension of policing – then the concept of police should include the notion of policing culture and give it prominence. This includes not only surveillance-produced knowledge but also the cognitive environment surrounding policing institutions.


The creation of a judicial police body was debated in Parliament during the French Revolution. The discussions partly hinged on whether this force should be autonomous or, on the contrary, subordinate to judicial power. One of the MPs, although a conservative, once declared that police action should not be restrained by the law and saw no other conceivable boundaries to policing power than the civic mindedness of the police themselves: “Be wise in picking up your police officers, and then let them enjoy the latitude without which their functions are void.” (Bon-Albert Briois de Baumetz, quoted in Napoli 2009). Powerlessness in codifying policing with external regulations is answered by the necessary “wisdom” of the agent. This resignation of policymakers when confronted with the police question exactly echoes the following remark by Anglo-Saxons police sociologists, reporting the very words of the police they had been observing: “you can’t police by book” or, as stated by a veteran police officer from the city observed in John Van Maanen’s seminal study on learning the trade: “There’s only two things you gotta know around here. First, forget everything you’ve learned in the academy ‘cause the street’s where you’ll learn to be a cop; and second, being first don’t mean shit around here” (Van Maanen 1978: 306). The French MP from 1791 considered the policeman’s civic mindedness as the only conceivable guarantee that civil rights would be respected by the police institution; the Union City veteran police observed by Van Maanen in the 1970s established that a good policeman must remain impervious to external ideas and in particular those taught at the academy. This is the birth of a central notion, namely, police culture. More than institutional arrangements or legal prescriptions, what defines the police and their institution, in a system where “police discretion” is both necessary and feared, is the culture (Wilson 1968). As a consequence, one of the most insightful concepts for explaining the use of deadly force by US police has been that of “administrative climate,” the local culture that actually determines which brand of policing is at work (Sherman 1980). Contrary to what happens with


other, more tightly regulated institutions, the occupational environment is not a tempering element of policing: it is a key element of policing. Thus, after having attempted to make urban police forces more democratic by playing either on their technical equipment, the relationship between their top management and elected officials or their organizational systems, US reformers have thought it more relevant to consider, following in the footsteps of the French revolutionary MP, that the “police officer is the agent of change” and to design police reform “in order to embrace police discretion and to find ways to make its exercise more informed, methodical, and collectively self-reflective, rather than to control discretion from above or from outside (the idea behind judicial oversight and civilian review boards)” (Sklansky 2005: 1776–1777). The fact that these attempts have all failed – beyond demonstrating the power of police unions, whose explicit purposes include protecting the immunity of policing agents – clearly shows that the question of knowledge aporetically converges towards the question of the rule of law. In the words of Napoli: “Ignoring legal texts is (. . .) the ultimate resource of the policing measure. At the end of the day, police action is based on an instance of non-knowledge” (Napoli 2009: 166). To put this into perspective and relate it to the wider picture, policing could be said to entertain a two-pronged relationship to knowledge. Knowledge is among its key modes of action, one that, ever since the conceptualizations of the seventeenth and eighteenth centuries, has turned the police into a formidable surveillance institution. However, policing has also been known to rely on its occupational or even vocational culture to counter all external regulatory attempts. The policemen’s intuition, flair, or common sense are, in their own and their leadership’s eyes, their core knowledge asset.

Conclusion : A Dual Concept of Policing Policing certainly ranks as one of the most challenging social objects to conceptualize. This

523

C

resistance, of course, owes a lot the secrecy opposed by the police institution to the external gaze. It owes even more, however, to its dual nature. Dualism, a philosophical movement whose early proponents included Descartes and Leibniz, has established that certain phenomena consist in the reunion or coexistence of two irreducible and irreconcilable components. In this perspective, policing is the reunion of two pairs of opposites: in it coexist the rule of law and violence and knowledge and ignorance. What most consistently shapes the concept of police is precisely this irreducible duality.

Recommended Reading and References Banton M (1964) The policeman in the community. Basic Books, New York Benjamin W (1996) Selected writings. Harvard University Press, Cambridge, MA Bittner E (1990) Aspects of police work. Northeastern University Press, Boston Brodeur J-P (2007) An encounter with Egon Bittner. Crime Law Soc Change 48:105–132 Brodeur J-P (2010) The policing web. Oxford University Press, Oxford, UK Ericson R, Haggerty K (1997) Policing the risk society. Clarendon, London Foucault M (1981) ‘Omnes et singulatim’: towards a criticism of political reason. In: McMurrin S (ed) The tanner lectures on human values, vol 2. University of Utah Press, Salt Lake City Jobard F (2012) Propositions on the theory of policing. Penal Fields (online publication) Champ peńal/Penal field, Vol. 9, 2012. http://champpenal. revues.org/8286 Kent SL, Jacobs D (2005) Minority threat and police strength from 1980 to 2000. Criminology 43(3): 731–760 Knemeyer F-L (1980) Polizei. Econ Soc 9:2 Kraska P, Kappeler V (1997) Militarizing American police: the rise and normalization of paramilitary units. Soc Probl 44:1–18 L’Heuillet H (2007) Baja polı´tica, alta policıá: un enfoque histo´rico y filoso´fico de la policıá. Prometeo, Buenos Aires Loader I, Mulcahy A (2007) Civilizing security. Cambridge University Press, Cambridge Manning P (2010) Democratic policing in a changing world. Paradigm, Boulder Napoli P (2003) Naissance de la police moderne: pouvoir, normes, socie´te´. La Dećouverte, Paris Napoli P (2009) Misura di polizia. Un approccio storicoconcettuale in era` moderna. Quad Stor 44:523–547

C

C

524

Neocleous M (1998) Policing and pin-making: Adam Smith, police and the state of prosperity. Polic Soc 8(4):425–449 Pankoke E (1986) Von ‘guter policey’ zu ‘sozialer politik’. ‘Wohlfahrt’, ‘gl€ uckseligkeit’ und ‘freiheit’ als wertbildung aktiver sozialstaatlichkeit. In: Sachßse C, Tennstedt F (eds) Soziale sicherheit und soziale disziplinierung. Beitr€age zu einer historischen theorie der sozialpolitik. Suhrkamp, Frankfurt a.M Pasquino P (1991) Theatrum politicum. The genealogy of capital. Police and the state of prosperity. In: Burchell G, Gordon C, Miller P (eds) The Foucault effect: studies in governmentality. Harvester Wheatsheaf, Hemel Hempstead Piazza P (2011) Bertillonnage: savoirs, technologies, pratiques et diffusion internationale de l’identification judiciaire. Criminocorpus, revue hyperme´dia (online, last consultation 20 July 2012). http://criminocorpus. revues.org/347 Porta D, Fillieule O (2004) Policing social protest. In: Snow D, Soule S, Kriesi H-P (eds) The Blackwell companion to social movements. Blackwell, Oxford, pp 217–241 Reiner R (1992) The politics of the police, 2nd edn. Toronto University Press, Toronto Schmitt C (1932/2007) The concept of the political. Chicago University Press, Chicago Sherman L (1980) Perspectives on police and violence. Ann Am Acad Polit Soc Sci 452:1–12 Silver A (1967) The demand for order in civil society: a review of some themes in the history of urban crime, police, and riot. In: Bordua D (ed) The police: six sociological essays. Wiley, New York Sklansky D (2005) Police and democracy. Mich Law Rev 103:1699–1830 Skolnick J, Fyfe J (1993) Above the law: police and the excessive use of force. The Free Press, New York Stenning P (2000) Powers and accountability of private police. Eur J Crim Policy Res 8:325–352 Stolleis M (ed) (1996) Policey im Europa der fr€ uhen neuzeit. Vittorio Klosterman, Frankfurt-am-Main Van Maanen J (1978) Observation son the making of policemen. In: Manning P, Van Maanen J (eds) Policing. A view from the streets. Newbery Award Records, New York Wakefield A (2009) The disordered police state: German cameralism as science and practice. Chicago University Press, Chicago Wilson J (1968) Varieties of police behavior: the management of law and order in eight communities. Harvard University Press, Cambridge

Concurrent Validity

Conditional Conviction ▶ Probation and Community Sanctions

Conditional Release ▶ Early Release from Prison

Confessions ▶ Interview and Interrogation Methods Effects on Confession Accuracy

Confirmation Bias ▶ Cognitive Forensics: Human Cognition, Contextual Information, and Bias

Constabularization ▶ Police and the Military Nexus

Construction Site Theft and Burglary ▶ Theft at Construction Sites

Concurrent Validity Content Analysis ▶ Self-Reported Offending: Reliability and Validity

▶ Quantitative Studies on Media and Crime

Control of Police Misconduct

Contextual Influences ▶ Cognitive Forensics: Human Cognition, Contextual Information, and Bias

Contingency Management ▶ Behavioral Management in Probation

Continuity and Change ▶ Individual Characteristics and General Strain Theory

525

C

functions into internal control mechanisms and external control mechanisms. The key internal control mechanisms (in which control functions are performed by the police agency itself) include the establishment of official rules, the roles performed by the police chief and the police administration, supervisors, and the internal system of control. The key external control mechanisms (in which control functions are performed by the organizations, institutions, and individuals outside of the police agency) include the legislature, prosecutors, and criminal courts, civil courts, the Supreme Court, and independent commissions. Lastly, the key mixed mechanisms (in which the functions are performed by an outside institution or agency with police officers constituting at least a part of the membership) include citizen reviews and the Commission on Accreditation for Law Enforcement Agencies.


Introduction

Sanja Kutnjak Ivkovic´ School of Criminal Justice, Michigan State University, East Lansing, MI, USA

Control of police misconduct includes a number of heterogeneous functions, from setting official agency policies and enforcing them, establishing supervision and accountability, detecting and investigating misconduct, and cultivating police culture intolerant of misconduct, to providing resources for control, controlling police agency’s own efforts to control misconduct, and limiting opportunities for misconduct. Some of these functions are reactive (e.g., disciplining police officers), while others are preventive (e.g., setting official policies, cultivating police culture intolerant of misconduct). The functions are performed by a network of diverse institutions such as the police agency itself, courts, mayors, media, and independent commissions. The traditional approach to the study of control mechanism has been to classify the institutions performing the control functions into internal control mechanisms (housed within the police agency) and external control mechanisms (housed outside of the police agency). In addition, there are a few mechanisms of control that have elements of both internal and external mechanisms (those

Synonyms Police accountability; Policing the police

Overview Control of police misconduct is a complex undertaking that includes a series of heterogeneous functions such as establishing supervision and accountability, detecting and investigating misconduct, and cultivating police culture intolerant of misconduct. These functions of control and accountability could be performed by a network of diverse institutions such as the police agency itself, courts, mayors, media, and independent commissions.The traditional approach toward the study of control mechanism has been to classify the institutions performing the control

C

C

526

housed outside of the police agency, but potentially having police officers as members).

Internal Mechanisms of Control and Accountability One of the most important roles in controlling misconduct and achieving accountability should be performed by the police agency itself. Yet, police agencies vary dramatically with respect to the intensity and thoroughness with which they control their employees’ conduct, as various commission reports have indicated (see, e.g., Christopher Commission 1991; Knapp Commission 1972; Mollen Commission 1994; Pennsylvania Crime Commission 1974). Police agencies have many diverse and interconnected ways in which they can strive to control police misconduct and enhance accountability. The control system “starts with the recruitment and selection process and continues with training and supervision, incorporating various aspects of rule establishment, communication, and enforcement that stimulate, allow, or prevent police officers from turning their propensity toward corruption into actual corrupt behavior” (Kutnjak Ivkovic´ 2005, p. 68). 1. Administrative Rules and Policies. Administrative rules and policies, usually compiled in the police agency’s standard operating procedure manual (SOP), are typically made by the police chief. The policies relevant for the control of police misconduct traditionally channel the use of discretion, describe appropriate conduct of police officers, prohibit inappropriate conduct, instruct officers to complete written reports after critical incidents, and require supervisory oversight (see, e.g., National Research Council 2004; Walker and Katz 2008). Over the last two decades, there seems to have been an increase not only in the number of police agencies that have official policies, but also in the extent of the official policies. A study by Barker and Wells in 1982 reveals that about one-quarter of the police agencies at the time had no written policies at all, and having a set of written policies seems to have


been related to the agency’s size. A more recent 2007 Bureau of Justice Statistics survey indicates that, regardless of the size of the population they serve, more than 95 % of the local police agencies had written policies regulating the code of conduct and appearance, and the use of lethal and nonlethal force by their employees. In itself, promulgation of rules does not suffice to control misconduct. Indeed, both the ambiguity of the rules and the presence of unofficial rules in conflict with the official ones could lead to misconduct. Various independent commissions (e.g., Knapp Commission 1972; Mollen Commission 1994; Pennsylvania Crime Commission 1974) recorded examples of police administrators creating unofficial rules which trumped the official ones (Kutnjak Ivkovic´ 2005). Empirical research suggests that police officers in the police agencies characterized as corrupt were more likely to say that the rules were not clear than the police officers employed in the police agencies relatively free of corruption were. Empirical studies (e.g., Fyfe 1979) report the reduction in the use-of-deadly-force incidents in the aftermath of the introduction of the more restrictive use-of-deadly-force rules. The administrative regulation of the highspeed pursuits seems to be directly related to the frequency with which officers engage in high-speed pursuits. For example, the introduction of a more restrictive policy of highspeed pursuits in the Metro-Dade Police Department substantially reduced its number of pursuits. By contrast, the introduction of a more relaxed policy of high-speed pursuits in the Omaha Police Department resulted in a dramatic increase of the number of pursuits the next year. The effect of administrative rulemaking on misconduct has been explored in several other areas (e.g., use of force, domestic violence); the evidence therein is more limited and shows less success (National Research Council 2004, p. 285). 2. Police Chief and Administration. The police chief and other highly ranked police administrators are the key to the police agency’s successful misconduct control. The nature of their


contributions can span across all aspects of traditional managerial functions such as planning, organizing, coordinating, and controlling. Although what police chiefs and highly ranked administrators may do could be limited by the mayor and other politicians, the public, media, and police unions as well as by the legal statutes, laws, court cases, and civil service rules, “[the police chiefs] may exert a substantial influence on the recruitment standards, training in ethics, leadership and management style, supervisory accountability and standards, internal control mechanisms, discipline, and rewards” (Kutnjak Ivkovic´ 2005, p. 70). Independent commissions (e.g., Christopher Commission 1991; Knapp Commission 1972; Mollen Commission 1994; Pennsylvania Crime Commission 1974) provide examples of police administrators’ failure to perform these functions and adhere to the expected standards by turning a blind eye to police misconduct. If in contradiction with the official rules, the police chief’s actual behavior sends a clear message to the police officers (see, e.g., Knapp Commission 1972, pp. 170–171). The majority of the police officers participating in a 2,000 nationwide survey agreed that “a chief’s strong position against the abuse of authority can make a big difference in deterring officers from abusing their authority” (Weisburd et al. 2000, p. 6). 3. Supervisors. While police chief and the highly ranked police administrators deal with the general issues potentially affecting the whole police agency, first-line supervisors are required to oversee their subordinates. Specifically, they are expected to monitor police officers under their command, review their reports, advise police officers when their performance is less than satisfactory, and file a report when they are aware that police officers had violated the rules (Walker and Katz 2008, p 482). As the overwhelming majority of the respondents (90 %) participating in a nationwide survey of police officers agreed, supervisors’ role in preventing misconduct is crucial (Weisburd et al. 2000, p. 6).

527

C

Independent commissions that investigated police misconduct in the 1960s/1970s documented cases of supervisors actively taking part in police misconduct (e.g., Knapp Commission 1972; Pennsylvania Crime Commission 1974). More recent independent commissions (e.g., Christopher Commission 1991; Mollen Commission 1994) tended to be more likely to find supervisors turning “a blind eye” or failing to take a stance on police misconduct. The Mollen Commission documented that Michael Dowd (a former NYPD police officer who stole money and drugs, participated in larger drug rings, became a drug dealer, and was eventually caught, tried, and sentenced to 14 years in prison), received glowing performance evaluations from his supervisors who even recommended him as “a role model” for other police officers (Mollen Commission 1994, p. 118). The Mollen Commission further found that supervisors in the NYPD regularly failed to review the search and seizure forms, arrest forms (p. 29), and overtime payment forms (p. 39) and question line officers under their supervision who were falsifying these forms. Similarly, the Christopher Commission (1991) concluded that the LAPD supervisors omitted to monitor the subordinates’ racist discussion on the MDT. In the aftermath of the Rampart Division scandal, the LAPD’s Board of Inquiry documented failures in the supervisors and wrote that “the practice of officers printing or signing a sergeant’s name to booking approvals and arrest reports was a particularly glaring illustration of poor CRASH supervision” (Los Angeles Police Department 2000, p. 61). Independent commissions (e.g., Christopher Commission 1991; Los Angeles Police Department 2000; Knapp Commission 1972; Mollen Commission 1994) reported that supervisors in these police agencies not only were overlooking line officer misconduct, but also were not held accountable by the police chief and the highly ranked police administrators for their failure to supervise. In fact, the Mollen Commission (1994) concluded that the

C

C

528

supervisors who were proactive and actually reported misconduct of their subordinates suffered informal punishment for their proper conduct. Since 1994, CompStat (Computer Statistics) has been used not only as a mechanism to reduce crime rates, but also as a mechanism to increase supervisor accountability. Developed by Bill Bratton and his top administrators in the NYPD, CompStat centers around regular meetings in which the top administrators require of middle managers to discuss crime problems in their areas and hold them accountable for its success or failure in dealing with the problems. CompStat became an instant hit and by 1999, one-quarter of the sample of police agencies with 100 or more employees have already implemented CompStat, and an additional one-third planned to do so in the future (Weisburd et al. 2003). However, the concept of accountability, as envisioned by CompStat, relates to the changes in the crime levels and does not extend to the control of rule-violating behavior of the subordinates. Moreover, whereas CompStat could operate as an accountability mechanism for the middle managers who participate in the CompStat meetings (e.g., Silverman 1999), it does not operate as an accountability mechanism for first-line supervisors who do not participate in the CompStat meetings (National Research Council 2004, p. 188). 4. Internal System of Control. Each police agency has a separate system set up within the agency to investigate allegations of police misconduct. Depending on the police agency size, overall public service demands, available resources, and number of complaints, the internal system of control can vary from a single police investigator assigned on the case-by-case basis to an elaborate network of police investigators housed in the main office and additional field offices. According to the 2,000 LEMAS, the majority of local police agencies (79 %) and state police agencies (84 %) had separate, permanent internal affairs offices. These internal affairs offices are traditionally directly accountable to the police chief.


The primary purpose of the internal affairs unit is fact-finding, while some other unit or individual (e.g., immediate supervisors, a board, citizen review) makes the decision about the case. The work of the internal affairs units could be classified as either reactive (e.g., investigation initiated by a complaint) or proactive (e.g., integrity tests). In reality, proactive investigations are more of an exception than the rule (Kutnjak Ivkovic´ 2005). The official procedure typically starts with the filing of a complaint by a citizen or a police officer, or with the submission of a report by a supervisor. During the investigation, the police investigators collect and examine physical evidence, interview witnesses, analyze records, and interview the accused police officer. As a consequence of the Garrity ruling (Garrity v. New Jersey, 385 U.S. 483 (1967)), police agencies separate investigations into a criminal investigation and an administrative investigation. In the administrative investigation, the accused police officer is not allowed to claim the Fifth Amendment privileges and is ordered to answer the questions truthfully. At the end of the investigation, the police investigator completes the report and turns over the evidence to the decision-making body. The disposition of the complaint is made either by the police officer’s chain of command review or by an administrative board through a disciplinary hearing. The nature of the hearing (e.g., adversarial) and the rules regulating it vary greatly across agencies. When the complaint is found sustained (i.e., there is sufficient evidence to prove that the police officer engaged in the rule-violating behavior), the decision-making body will mete out discipline. The severity of the discipline likely depends on the seriousness of violation, aggravating and mitigating circumstances, and the police officers’ prior history. Complaint rates vary greatly across the cities; Pate and Hamilton compared the complaint rates per 100 sworn officers across the six largest police agencies (1991, p. 144) and


reported variation in the complaint rates from 5.5 in Philadelphia to 36.9 in Houston. A 2002 survey of large local and state police agencies shows that the rates for the use of force complaints varied from 1.3 for the state agencies to 9.5 for municipal agencies (Hickman 2006, p. 2). Typically, police agencies sustain between 0 % and 25 % of all complaints (Dugan and Breda 1991; Hickman 2006, p. 4; Pate and Fridell 1993, p. 42; Perez 1994), with 8–10 % being typical or average (e.g., Pate and Fridell 1993; Hickman 2006). It seems that complaint rates are influenced by so many factors other than the actual rate of misconduct that these rates likely can tell more about the agency’s openness to complaints, the ease with which citizens could file complaints, and the level of legitimacy of the police, than about the actual level of misconduct. Wide differentials of complaint rates across the cities prompted scholars to recommend extreme caution in cross-agency comparisons (e.g., Hickman 2006; Klockars et al. 2006; Pate and Hamilton 1991; Perez 1994; Walker 2001). Having the system of internal control put in place does not guarantee that it will operate properly. Research studies and reports by independent commissions (e.g., Christopher Commission 1991; Knapp Commission 1972; Mollen Commission 1994; Pennsylvania Crime Commission 1974) documented many failures, from not establishing written guidelines and providing resources and manpower to the internal affairs units, to failing to investigate complaints, ignoring information, and openly hiding complaints (see, e.g., Christopher Commission 1991; Knapp Commission 1972; Mollen Commission 1994; Pennsylvania Crime Commission). Early warning systems have recently been added to the internal system of control in a number of police agencies. Early warning systems are proactive in nature. Police agencies collect information about individual police officers, such as use of force reports, accident reports, complaints, financial records, and other information (see,

529

C

e.g., Walker et al. 2000; Walker and Katz 2008). Once a police officer has a certain number of these items (e.g., sustained complaints), red flags are raised. The police agency engages in an intervention (typically informal counseling by the supervisor or retraining) and postintervention (monitoring the police officer for a certain period of time; Walker et al. 2000). The idea behind the early warning systems rests on the premise that if potential problems are spotted and addressed early, they would not become serious problems later. The limited empirical research (e.g., Walker et al. 2000) implies that the early warning systems might be effective in terms of reducing the number of use of force reports and citizen complaints by the officers subject to early warning systems.

External Mechanisms of Control and Accountability A number of institutions an organizations housed outside of the police agency take part in the control of police misconduct. 1. Legislature. The role of the legislature, both at the federal and state level, is to enact the laws that criminalize police misconduct and/ or establish civil liability for police misconduct. First, the rules of various federal and state criminal statutes (e.g., criminalization of theft, assault, murder), enacted by the legislature, could be applied to every citizen, including police officers, unless the police officers can successfully claim that they performed the activity as part of their legitimate police work. Second, federal and state codes also establish certain crimes which only public officials can commit (e.g., bribery of public officials and witnesses, Title 18 of the U.S. Code, Section 201, 1999; extortion by public officials, Title 18 of the U.S. Code, Section 872, 1999; deprivation of civil rights, Title 18 of the U.S. Code, Section 242, 1999). The legislation includes such acts as the Hobbs Act (Title 18 of the U.S. Code, Section 1951, 1999) and the

C

C

530

Racketeer Influenced and Corrupt Organizations Act (RICO; Title 18 of the U.S. Code, Sections 1961–1968, 1999). The advantage of RICO relative to other penal codes is the possibility to prosecute for a “pattern of racketeering activity” (Title 18 of the U.S. Code, Section 1961, 1999) and, therefore, to present the picture of systematic exploitation of the official position, instead of being forced to prosecute each and every act separately. Third, federal and state statutes also establish the grounds for civil liability for deprivation of civil rights (e.g., deprivation of civil rights, Title 42 of the U.S. Code, Section 1983; conspiracy to interfere with civil rights, Title 42 of the U.S. Code, Section 1985). Fourth, the Violent Crime Control and Law Enforcement Act of 1994 (Title 42 of the U.S. Code, 1994) authorizes the Attorney General to collect the data about the use of excessive force by the police officers and publish the summary. Consequently, the Bureau of Justice Statistics and National Institute of Justice started sponsoring the national police use of force database, collected by the International Association of Chiefs of Police and the police public contact survey, as well as expanded the LEMAS survey to collect the data on formal citizen complaints about the use of force (Hickman 2006). One of the problems with the legal codes is that the rules are vague and open for ambiguous interpretations. The Knapp Commission reported that in the 1970s, the rules governing the operation of bars were sound in principle, but were “so vague and ill-defined that they lend themselves to abuses in practice” (Knapp Commission 1972, pp. 147–148). Furthermore, the rules could be difficult to enforce. In the words of Patrick Murphy, a former Commissioner of the NYPD (Pennsylvania Crime Commission 1974, p. 420), “[b]y charging our police with the responsibility to enforce the unenforceable, we subject them to disrespect and corruptive influences, and we provide the organized criminal syndicate with illicit industries upon which they thrive.”


However, once the legal codes are enacted, the crucial question is whether they are enforced (see Criminal Courts; Civil Courts). 2. Prosecutors and Criminal Courts. The criminal justice system enforces the laws enacted by the legislature and provides the boundaries of police officers’ acceptable/unacceptable behavior by holding them criminally responsible for police misconduct. Like other citizens, police officers could be tried for “traditional crimes” like robbery, assault, or theft. In addition, unlike other citizens, police officers are public servants and could also be prosecuted, tried, and convicted for crimes that require that the person who committed the crime is a public employee (e.g., extortion, 18 U.S.C. 1951; criminal liability for deprivation of civil rights, 18 U.S.C. 242). There is no systematic database, either at the federal or the state level, that contains the data about all prosecutions and convictions for police misconduct. At the federal level, there were fewer than 50 convictions each year for law enforcement corruption in the period from 1992 to 1998 (Kutnjak Ivkovic´ 2005, p. 59), At the state level, the convictions are almost equally sparse and seemingly suggest that corruption is not a problem. However, the findings by the independent commissions contradict these reports. The Knapp Commission (1972) reported that corruption in the NYPD was widespread, ranging all the way to the rank of a captain. Yet, the prosecutors initiated the prosecution in only about 30 cases per year. Furthermore, severe sentences are unlikely; only one out of five police officers prosecuted, tried, and convicted received a prison sentence of more than a year (Knapp Commission 1972, p. 252). The Pennsylvania Crime Commission (1974), which investigated the extent and nature of corruption in the Philadelphia Police Department and found widespread corruption in the agency, reported that there were on average only seven arrests per year for corruption (Pennsylvania Crime Commission 1974, p. 446). The data also suggest that


criminal prosecutions for the use of excessive force, either federal or state, are rare (e.g., Cheh 1995, p. 241; Human Rights Watch 1998) and that the conviction rates are low. Out of about 8,000 police misconduct complaints that the Department of Justice receives annually (Cheh 1995, p. 241), about 3,000 are investigated and only 50 presented to the grand jury. 3. Civil Courts. Civil courts enforce the laws enacted by the legislature and provide the boundaries of acceptable/unacceptable behavior by holding police officers, police departments, and city administrators civilly liable for police misconduct. Compared to the number of contacts between the police and citizens, civil lawsuits are rare. Yet, the amounts paid to the successful plaintiffs vary across the country, from an average of 1.6 million annually in Cincinnati to an average of 35.8 million annually in Los Angeles (Kappeler 2006, p. 10). The legal basis for the civil lawsuit against the police could be found in both the federal codes and the state codes. Section 1983 lawsuit (Title 42 U.S. Code 1983) is considered a crucial tool in the fight against police misconduct. In Monroe v. Pape (365 U.S. 167 [1961]), the Supreme Court determined that police officers could be held liable for deprivation of the 4th Amendments rights under the civil rights statute. In Monell v. Department of Social Services (436 U.S. 658 [1978]), the Supreme Court established that municipalities could be held liable for police misconduct if it was pursuant to the agency’s policy or custom. Thus, these two decisions allowed citizens to sue both individual police officers and police departments, and according to the estimates, the number of the Section 1983 lawsuits increased dramatically since the 1960s (Cheh 1995, p. 250). As a consequence of the Supreme Court decision in Lyons (City of Los Angeles v. Lyons, 461 U.S. 95 [1983]), citizens can obtain only compensatory and punitive damages (Cheh, 1995, p. 255). The effect of civil

531

C

lawsuits on individual police officers is limited; the compensatory damages (and sometimes punitive damages) resulting from the lawsuit are typically covered by the city government. The effect of civil lawsuits on the police agencies is limited as well. Empirical research covering several jurisdictions shows minimal effect on police agencies. The Los Angeles County Sheriff’s Department seems to be an exception from this rule; in 1993, the Office of the Special Counsel – a form of citizen review – was established to investigate problems, recommend reforms, and reduce the costs of litigation. The reports imply positive changes. The situation has changed with the enactment of the 1994 Violent Crime Control Act (42 U.S.C. 14141, 1994), which has authorized the Department of Justice to act as a plaintiff and sue a police department when there is “. . . a pattern or practice of conduct by law enforcement officers . . . that deprives persons of rights, privileges, or immunities secured or protected by the Constitution.” The most recent data available by the Department of Justice (January 31, 2003) indicate that the investigation of 14 police agencies is ongoing, that, out of the 10 lawsuits, 4 ended with consent decrees and 6 with out-of-court settlements (Department of Justice 2010). The settlements or consent decrees require of the police agencies to engage in systematic and widespread reforms of the police agency, such as revising the use of force reporting system, establishing the early warning system, revising the complaint procedures, and improving training. In addition, each police agency is assigned a court-appointed monitor to ensure the implementation of these changes. Existing research did not examine the overall effect of these “pattern or practice” lawsuits on police agencies. The Vera Institute, serving as a monitor for the Pittsburgh Police Department, generally reports that the police are on track with the required changes (Vera Institute of Justice 2002). On the other hand, the reports concerning police agencies in Los Angeles

C

C

532

and Washington D.C. indicate that they failed to meet some of the deadlines (Walker and Katz 2008, p. 503). 4. The Supreme Court. The chances that the Supreme Court will hear any particular case are miniscule (the Supreme Court grants certiorari in less than 5 % of the cases filed). Compared to the lower-ranked trial courts, the Supreme Court, as the highest court of the country, has a wider role to play. The effect of its decisions is far reaching, well beyond the specific case under consideration. Thus, by establishing a nationwide precedent for future court cases, the Supreme Court performs not only a reactive role, but also a potentially even more important, preventive role. The relevant Supreme Court cases regulate proper conduct (e.g., the Fifth Amendment warnings given to the arrestees, as was decided in Miranda v. Arizona (372 U.S. 436 [1966])), as well as police misconduct (e.g., exclusion of illegally obtained evidence, as was decided in Mapp v. Ohio (367 U.S. 643 [1961]). The cases may focus on substantive law (e.g., sentencing for the violation of King’s constitutional rights under color of law, as was decided in Koon v. United States (518 U.S. 81 [1996])) or procedural law (e.g., prohibition of a denial of the Fifth Amendment rights to a police officer during a criminal case, as was determined in Garrity v. New Jersey, 385 U.S. 483 (1967)). Empirical research typically explores the influence of a specific landmark decision on police officer behavior. One such decision is Miranda v. Arizona (372 U.S. 436 [1966]), in which the Supreme Court held that a confession obtained during custodial police interrogation constitutes a violation of the 5th Amendment right against self-incrimination, unless the police provided specific warnings to the persons that they have the right to remain silent, that anything they say could be used against them, and that they have the right to counsel. The Miranda warnings have been routinely publicized in TV shows and motion pictures and have become a part of the American popular culture (see, e.g., Dickerson v.


United States, 530 U.S. 428 [2000]). The early research, already in progress when the Miranda decision was made, suggested that police officers rarely gave Miranda warnings. Later studies (Leo 1998) showed that the warning was issued routinely, but the style in which the warning was issued tended to be superficial. The most recent study (Leo 1998), relying on direct observation, confirmed that the police issued the Miranda warnings in about 96 % of the cases and thus showed that the Miranda decision has had a long-term effect on police behavior. 5. Independent Commissions. Independent commissions are established in the aftermath of a critical incident which resulted in a scandal (e.g., the Christopher Commission was established in the aftermath of the worldwide airing of the Rodney King beating tape; the Mollen Commission was established in the aftermath of the arrest of six police officers). The political pressure created pushes the mayor or the city manager to create the commission, composed of prominent community members and potentially policing scholars, with the purposes of investigating the nature and extent of police misconduct and proposing recommendations for the changes in the police agency. Prominent examples of independent commissions include the President’s Crime Commission (1967), Knapp Commission (1972), and Pennsylvania Crime Commission (1974). The work of independent commissions has the potential to reach beyond the individual case; they are in a position to establish the standards that could affect the way policing is done across the country (see, e.g., National Research Council 2004; Walker and Katz 2008). Yet, independent commissions are temporary and have no power to enforce their own recommendations (e.g., Kutnjak Ivkovic´ 2005; National Research Council 2004; Walker and Katz 2008). For example, although one of the recommendations by the Christopher Commission (1991) was the establishment of an early warning system in the LAPD, more than a decade later, the


LAPD did not have the early warning system put in place (Walker 2005, p. 179). Furthermore, the work of independent commissions could be severely affected by the lack of political independence (e.g., Pennsylvania Crime Commission 1974), insufficient legal authority (e.g., Knapp Commission 1972, p. 44), and inadequate resources (e.g., Pennsylvania Crime Commission 1974, p. 762).

Mixed Mechanisms of Control and Accountability The mixed mechanisms of control and accountability share some characteristics of external mechanisms (e.g., housed outside of the police agency) and some characteristics of internal mechanisms (e.g., have police officers as members). 1. Citizen Reviews. Citizen reviews are established with the purpose of providing an independent review of citizen complaints. Yet, empirical research suggests that about 23 % (15 out of 65) of citizen reviews have police officers as members (Walker and Kreisel 2001). This idea is not new; although it was first developed in the 1960s, the majority of the existing citizen reviews were established in the last two decades (e.g., Walker 2005). Walker (2001, p. 6) claims that by 2,000, “[o]ver 100 different agencies exist, covering law enforcement agencies that serve nearly one-third of the American population, and they are found in about 80 % of the big cities of this country.” Walker has classified citizen reviews based on functions they perform (Walker 2001). Only about 34 % of citizen reviews conduct the initial fact-finding completely independently from the police (Class I citizen review; Walker 2001). An additional 46 % provide input in the police investigation (Class II citizen input; Walker 2001; Walker and Kreisel 2001), and 17 % perform the function of the appellate review once the police investigation is completed (Class III citizen monitors). Lastly, Class IV citizen reviews (“citizen

533

C

auditors”), which constitute only 3 % of the citizen reviews, “do not investigate individual complaints, but are authorized to review, monitor, or audit the police department’s complaint process” (Walker 2001, p. 62). The focus of Class I to Class III citizen reviews is on individual cases. Even when they go beyond the single case and engage in the policy review as well, their work is bound by the issues raised in the cases. On the other hand, Class IV citizen reviews, which are authorized to review and monitor the police agency’s own complaint system, have a greater potential to provide meaningful feedback to the police agency. The San Jose Police Auditor and the Special Counsel to the Los Angeles County Sheriff’s Department are evaluated in the literature as more successful auditors, while the Seattle Police Auditor and the Albuquerque Independent Counsel are evaluated as the less successful (Walker 2005, pp. 165–166). Empirical research exploring the effectiveness of citizen reviews is limited. Whereas Kerstetter and Rasinski (1994) report that the level of public confidence in the complaint process should increase with the establishment of a citizen review, Sviridoff and McElroy (1989) find that both citizens and the police officers were unhappy with the New York City Civilian Review Board and thought that the board was biased against them. Another avenue of research tried to compare the citizen review outcomes in individual cases with the outcomes of individual cases decided by the police agency itself. A serious challenge with this type of empirical research is finding matching cases. Hudson (1972) compared the work of a citizen review (“Police Advisory Board,” which has subsequently been abolished) with the police investigation in Philadelphia and concluded that the citizen review sustained a lower percentage of complaints than the internal affairs unit did. However, the types of the cases handled by two institutions were not similar. 2. CALEA. The Commission on Accreditation for Law Enforcement Agencies (CALEA),

C

C

534

established in 1979, is an independent agency housed outside of the police. It is composed of 11 police officers and 10 community members (e.g., judges, professors, politicians). Its primary task is to serve as a source of selfregulation for the police. CALEA publishes standards and model official rules. Some of the rules and standards are required for police agencies seeking CALEA accreditation, while others are only recommended. For example, CALEA now requires the early warning systems for accreditation (CALEA 2006). As of September 2009, CALEA has 463 standards (CALEA 2010). The estimates are that about 500 police agencies have been accredited by 2008 (Walker and Katz 2008, p. 492). Participation in the CALEA accreditation program is voluntary. The drawback of this approach is that the police agencies in greatest need to change the official rules (e.g., characterized by widespread corruption) are the least likely to do so following CALEA standards. Furthermore, the CALEA standards provide just the minimum standards and do not even try to assess the optimal or ideal standards (Walker and Katz 2008, p. 494). Empirical research on the influence on CALEA accreditation on misconduct and accountability in the police agencies is very limited. Walker and Katz (Walker and Katz 2008, p. 493) provide examples of how accreditation reduced insurance costs, enhanced the use of force reporting, and improved procedures for juveniles in several agencies.

Related Entries ▶ History of Police Unions ▶ Law of Police Searches ▶ Law of Police Seizures and the Exercise of Discretion ▶ Law of Police Use of Force ▶ Police and the Excessive Use of Force ▶ Police Corruption


▶ Police Culture ▶ Police Discretion and Its Control ▶ Police Lying and Deception

Recommended Reading and References Barker T, Wells RO (1982) Police administrator0 s attitudes toward definition and control of police deviance. FBI Law Enforc Bull 51:8–16 CALEA Commission on Accreditation for Law Enforcement Agencies (2006). Standards for Law Enforcement Agencies. Fairfax, VA: CALEA CALEA Commission on Accreditation for Law Enforcement Agencies (2010). Standards for Law Enforcement Agencies. Fairfax, VA: CALEA Cheh MM (1995) Are law suits an answer to police brutality? In: Geller WA, Hans T (eds) And justice for all. Police Executive Research Forum, Washington, DC, pp 233–259 [Christopher Commission]. Independent commission on the Los Angeles Police Department (1991). Report of the independent commission on the Los Angeles Police Department, Los Angeles, Author Department of Justice (2010). FAQ. http://www.justice. gov/crt/split/faq.php#primsource Dugan JR, Breda DR (1991) Complaints about police officers: a comparison among types and agencies. J Crim Justice 19(2):165–172 Fyfe JJ (1979) Administrative interventions on police shooting discretion: an empirical examination. J Crim Justice 7:303–323 Hickman, M. J. (2006). Citizen complaints about police use of force, Washington, D.C, Bureau of Justice Statistics. Available at http://bjs.ojp.usdoj.gov/content/ pub/pdf/ccpuf.pdf Hudson J (1972) Organizational aspects of internal and external review of the police. J Crim Law, Criminol Police Sci 63:427–432 Human Rights Watch (1998) Shielded from justice: police brutality and accountability in the United States. Human Rights Watch, New York Kappeler VE (2006) Police civil liability, 2nd edn. Waveland Press, Long Grove Kerstetter WA, Rasinski KA (1994) Opening a window into police internal affairs: impact of procedural justice reform on third-party attitudes. Soc Justice Res 7(2):107–127 Klockars CB, Kutnjak Ivkovic´ S, Haberfeld MR (2006) Enhancing police integrity. Springer, Dordrecht [Knapp Commission] Commission to investigate allegations of police corruption and the city’s anti-corruption procedures (1972). Report on Police Corruption. New York, G. Braziller Kutnjak Ivkovic´ S (2005) Fallen blue knights: controlling police corruption. Oxford University Press, New York Leo RA (1998) The impact of miranda revisited. In: Leo RA, Thomas GC (eds) The miranda debate: law, justice, and policing. Northeastern University Press, Boston

Control Theory Los Angeles Police Department (2000). Board of inquiry into the rampart area corruption incident: executive summary. http://www.lapdonline.org [Mollen Commission] New York city commission to investigate allegations of police corruption and the anti-corruption procedures of the police department (1994). Commission Report. Author, New York National Research Council (2004). Fairness and effectiveness in policing: the evidence. committee to review research on police policy and practices. Wesley Skogan and Kathleen Frydl, editors. Committee on law and justice, division of behavioral and social sciences and education. The National Academies Press, Washington, DC Pate AM, Hamilton HE (1991) The big six: policing America’s largest cities. Police Foundation, Washington, DC Pate A, Fridell L (1993) Police use of force: official reports, citizen complaints, and legal consequences. Police Foundation, Washington, DC Pennsylvania Crime Commission (1974) Report on police corruption and the quality of law enforcement in Philadelphia. Author, Saint Davids: PA Perez DW (1994) Common sense about police review. Temple University Press, Philadelphia Silverman EB (1999) NYPD battles crime: innovative strategies in policing. Northeastern University Press, Boston Sviridoff M, McElro JE (1989) Processing complaints against police in New York city. Vera Institute of Justice, New York Vera Institute of Justice (2002). “Pittsburgh’s experience with police monitoring.” http://www.vera.org/project/ project1_1asp?section_id¼2&proejct_id¼13 Walker S (2001) Police accountability: the role of citizen oversight. Wadsworth Publishing Company, Belmont Walker S (2005) The new world of police accountability. Sage Publications, Newbury Park Walker S, Alpert G, Kenney D (2000) Responding to the problem officer: a national evaluation of early warning systems. National Institute of Justice, Washington, DC Walker S, Katz CM (2008) The police in America: an introduction, 6th edn. McGraw-Hill, Boston Walker Samuel, Betsy Wright Kreisel (2001). Varieties of citizen review. In: Roger G Dunham, Alpert G (ed) Critical issues in policing,4th edn. Waveland Press, Prospect Heights, IL, pp 338–355 Weisburd David, Rosann Greenspan, Edwin E. Hamilton, Hubert Williams, Kellie A. Bryant (2000) Police attitudes toward abuse of authority: findings from a national survey. Washington, DC: United States Department of Justice, National Institute of Justice, Research in Brief. http://www.ncjrs.gov/pdffiles1/nij/ 181312.pdf Weisburd D, Mastrofski SD, McNally AM, Greenspan R, Willis JJ (2003) Reforming to preserve: compstat and strategic problem-solving in American policing. Criminol Public Policy 2:421–456

535

C

Control Theory Marc Le Blanc School of Criminology and School of Psychoeducation, Universite´ de Montreál and Boscoville 2000, Montreal, QC, Canada

Synonyms Community control: criminality, community social organization, cultural organization, deviant opportunities, directs controls; Criminal phenomenon: crime, criminal, criminality; Event control: criminal event, routines activities, presentness, occasions; Personal control: criminal behavior, social control, self-control, models, constraints

Overview The central notion of our theory is control. Control as the exercise of restraining and directing influences over the criminal phenomenon. The criminal phenomenon involves three embedded levels: the crime, the criminal, and the criminality. In appropriate contexts, control triggers conformity in harmony with social and moral expectations. The restraining and directing influences for each levels of the criminal phenomenon are the result of four mechanisms: bonding, unfolding, modeling, and constraining. Our theoretical statement is an expansion of existing psychological and sociological theories and a mixture of various control constructs. As a consequence, our control theory is integrative. It is also isomorphic because the four mechanisms are expressed in particular ways for each levels of the criminal phenomenon. Finally, our theory is developmental because it uses the laws of thermodynamics to address the question of the explanation continuity and change in the criminal phenomenon.

C

C

536

Introduction Over the last 20 years, criminology has not witnessed any major theoretical innovations. Numerous theoretical perspectives were available: social disorganization, strain, control, cultural deviance, differential association, social learning, criminal personality, labeling, deterrence, and so on (Shoemaker 2005). These perspectives were elaborations of ideas of early twentieth-century theorists: Quetelet, Lombroso, Durkheim, Freud, Marx, Tarde, and others. Most of the theoretical perspectives explained why individuals became criminals. Durkheim, Freud, Reiss, Nye, and Hirschi have presented control theories. They proposed specific constructs and they accepted the same basic assumptions concerning human nature. Over the last four decades, control theory became and remains the most prominent empirically based criminological theory. Control theories share four assumptions about human nature and social order. The first assumption is that humans are unsocialized at birth, and during the life course their socialization is never perfect. These conditions favor the emergence and maintenance of a criminal propensity in individuals and communities. The second assumption is that a social order always implies some consensus on values and informal and formal mechanisms of interactions. However, social order is always transient. The third assumption is that communities and individuals are self-serving and have mutual influences on each other. The fourth assumption is that the situation, the individual, and the community modulate each other as much as there are forces internal to each of them. Our theory is multilayered. It targets three levels of definition and explanation of the criminal phenomenon: the crime, the criminal, and the criminality. Our theory is integrative. It combines constructs from various disciplines and theoretical perspectives. We constructed our theory assuming that different words and terms are different for various theorists but that their theoretical meanings and operational definitions are synonymous. Our theory is isomorphic.

Control Theory

The similarity of the explanatory structure from one level of definition of the criminal phenomenon to the other corresponds to the view of the fractal nature of the universe. We propose six generic constructs: two categories of exogenous factors, environment and background, and four control mechanisms, bonding, unfolding, modeling, and constraining. Finally, our control theory is developmental. Most criminological theoretical perspectives do not address the question of the explanation of continuity and change in the criminal phenomenon. We define a contextual perspective that allows for developmental principles like orthogenesis, sensitivity to the original state, and epigenetic probability. In this entry, we define levels of explanation of the criminal phenomenon: the crime, the criminal, and the criminality. Starting from these layers, we elaborate an integrative and developmental control theory. We define the components of the theory and their structural organization. This paper proposes static and a dynamic formulation of our theory. An extended version of our integrative and developmental multilayered control theory of the criminal phenomenon is available (Le Blanc 1997a, b, 2006, 2009). The Social Sciences and Humanities Research Council of Canada supported the theoretical and empirical research on our integrative and developmental multilayered control theory of the criminal phenomenon over four decades.

The Levels of Definition and Explanation of the Criminal Phenomenon The French criminologist Jean Pinatel, in a masterly effort to define the bases of criminology, proposed, in 1963, that criminologists should distinguish among three levels of the criminal phenomenon, namely, the criminality, the criminal, and the crime. Each level of the criminal phenomenon has its own perspectives, its own rationales, and its own methods. American criminologists rediscovered this fundamental principle (Hirschi 1979; Short 1985). However, the emphasis has been mainly on two levels, the criminality and the criminal.

Control Theory

The micro-level interpretation of the criminal phenomenon, the crime, refers to a small part of the criminal career of a person or of criminality. The criminal event has a beginning, a development, and an end, and the task of criminology is to ascertain the control mechanisms that sustain its appearance in a particular situation. The meso-level of the criminal phenomenon, the criminal, includes the study of his personal biological, psychological, and social characteristics in relation to offending. The dependent variable is any descriptive index of the criminal activity, participation, frequency, onset, and duration or any developmental measures, activation, aggravation, desistance, and trajectories (Le Blanc and Frećhette 1989). Independent control variables are numerous and vary during the life course. The macro-level of the criminal phenomenon, the criminality, is the sum of all offenses committed and criminals at a given time in a particular place, a school, a community, and a society. The rate of criminality is influenced by demographic, economic, political, and other macro control factors. The independent variables are indicators of the state of a society, a community, and a school. Accepting the need to specify the level of the criminal phenomenon when elaborating theoretical propositions implies that they are distinct and their explanatory factors matched to them. Psychologists distinguish three levels: the milieu, the person, and the situation. Sociologists refer to society, community, and institutions. In consequence, we posit that each level of explanation is only pertinent to a particular level of definition: a crime in a situation, a criminal with his offending, and a rate of criminality of a society. To respect the logic of the ecological fallacy and the fallacy of reductionism, explanatory variables of one level do not explain the dependent variable of another level. For example, societal variables may affect the person; however, only person variables can influence directly individual offending, and person variables cannot affect the rate of criminality. Each level of explanation is also autonomous. However, explanatory

537

C

variables are constantly interacting, and it is also implicit from a dynamic point of view that levels of definition reciprocally influence each other. At a specific point in time, the characteristics of a person and of his community, associated with previous behavior, will partially determine subsequent behavior that, in turn, will modify the characteristics of the person and of his community.

The Integrative Multilayered Control Theory of the Criminal Phenomenon Lenski (1988, p. 168) defines a multilayered theory as “one in which a broadly inclusive general theory establishes a covering principle from which a series of more limited special theories can be derived.” We propose to apply this definition to the levels of definition and explanation of the criminal phenomenon. A broadly inclusive general theory is what Wagner and Berger (1985) call an orienting strategy; it discusses guidelines for understanding the criminal phenomenon, what notions to include and how to relate them to each other. In criminology, there are many such strategies according to Shoemaker (2005). These limited theories present a plausible body of theoretical statements offered to explain a particular layer of the criminal phenomenon. In this entry, we develop a general control theory and limited theories for the explanation of the crime, the criminal, and the criminality. The Structural Statement of an Integrative Multilayered Control Theory The Structure of the Generic Control Theory Much of the recent literature forgets formulations of control theory that preceded Hirschi’s statement (1969) that limits the notion of control to Durkheim’s definition of the bond to society. All control theorists would agree with Empey’s statement that the core of control theories is (1978, p. 207) “. . .their emphasis upon the idea that delinquent and conformist behavior is a function of the ability of the child to control his antisocial impulses. They start from the assumption that children require training if they are to behave socially. Delinquent behavior will result

C

C

538

either if a child lacks the ability for effective training or because he has been trained badly.” This statement fits particularly well Durkheim’s definition of control in his 1934 book “L’e´ducation morale” and the last version of control theory that adds the notion of low self-control to the notion of bonds (Gottfredson and Hirschi 1990). As formulated, control theory applies to the meso-level of definition of the criminal phenomenon, the criminal. Some authors developed a control theory that applies to communities, cities, or society: Sutherland, Trasher, Shaw, McKay, and Kornhauser. Our general control theory of the criminal phenomenon states that, in a favorable context, control mechanisms are operand and will change in harmony with social and moral expectations, as a consequence conformity results and is maintained. Conversely, in an unfavorable context, control mechanisms are insufficient and inappropriate, and the criminal phenomenon emerges and persists. We use the term “control” according to its third literal definition in Webster’s dictionary (p. 245, 3b): “a mechanism used to regulate and guide the operation of a system.” The term “control” then refers to the definition of the central notion that Gibbs proposes (1989, p. 23): “. . .control is overt behavior by human in belief that (1) the behavior increases the probability of some subsequent condition and (2) the increase or decrease is desirable.” For Gibbs, the commission of an act or its omission is overt behavior. Overt behaviors manifest themselves in the form of an inanimate thing, human and nonhuman organisms, self-controls and external controls, proximal, sequential, or social. Gibbs argues that this notion of control is central for all the behavioral and social sciences. Gibbs’ definition of control is compatible with its literal definition: the exercise of restraining and directing influences. This definition has two advantages. First, it keeps us away from the strictly individual level formulation of the definition of control in the Durkheimian tradition, that is, the bond to society and internal and external constraints used during socialization. Second, its level of abstraction facilitates the formulation of our control theory for the three levels of

Control Theory

definition of the criminal phenomenon. Individuals, communities, and events can produce behaviors, acts, circumstances, or conditions that are purposive and desirable. Bonds and constraints are only one type of such overt behaviors. Gibb’s notion of control is also central to psychology. Lytton (1990) uses the umbrella of control system theory to review the literature on child development. Horowitz (1987) proposes a structural/behavioral control model of development. The notion of control is also dominant in criminology. Attachment and supervision for bonding theorists are forms of such regulating mechanisms. We could also argue that arrests, the perceived certainty of a sanction or opportunities, are forms of restraining influences for subsequent offending. They are constructs proposed by labeling, deterrence, and strain theorists. Favorable and unfavorable definitions for differential association theorists and reinforcements for learning theorists are also controls because they are directing influences on criminal behavior. In our generic control theory, there are four mechanisms of control: bonding, unfolding, modeling, and constraining. Two types of context modulate the action of these mechanisms, the environment and the background. Each category of control mechanism and each type of contexts represent numerous factors that have a potential impact on one particular level of the criminal phenomenon. The definitions of the control mechanisms are the following. Their name is written in capital letters in Fig. 1. Bonding refers to the various ways by which individuals are held together in a community, in a group, or interpersonally. Unfolding is the natural growth and development toward a desirable state of greater quality, the growth of a community or the development of the person according to social expectations. Modeling is the existence of patterns that can shape conformity, opportunities, models, or occasions that are available in a community or to individuals. Constraining is the regulation of conformity through various direct and indirect restrains; they are limits defined by a community or imposed by the social network of the person. These mechanisms are simultaneously and causally interacting

Control Theory

539

C

MODELLING

ENVIRONMENT

Opportunities Models Occasions

BONDING Social structure Social status Community control

Criminality Criminal Crime

Social organization Bonds Routine activities

CRIMINAL PHENOMENON

BACKGROUND

Direct controls Constraints Guardianship

UNFOLDING Setting Biological capacity Personal control

Legend:

Cultural organization Self-control Presentness

CONSTRAINING

In each box, the explanatory constructs for each level of the criminal phenomenon are enumerated in order: criminality, criminal and crime. Three types of arrows are theoretically active between all constructs and each type of arrows is illustrated with two constructs : Interactions at a specific time, age or moment : T1 Causal effects across times, ages or moments : T2 Time, age or moments Feedback from a time, age or moment : T3

Control Theory, Fig. 1 The structure of the integrative multilayered control theory of the criminal phenomenon

to produce conformity (all the arrows in Fig. 1). They also have their own life or ontogeny (the superimposed boxes in Fig. 1). This theory is systemic in the sense that it defines a structure, a sequence between its components, as well as reciprocal and directional relationships and feedbacks. It is also a dynamic theory because over time there is continuity and change within the control mechanisms as well as because of their mutual influences. In Fig. 1, the structure of the control theory is illustrated by the proximity of the mechanisms from the criminal phenomenon. This structure depends on the principle of prerequisites, on the distinction between continuity and change, and on existing empirical knowledge. The theory states that the exogenous factors do not have a direct impact on the criminal phenomenon. They are the environment (the complex of climatic, edaphic, and biotic factors) and the background (circumstances that modulate the control mechanisms). Two of the mechanisms of control, bonding and unfolding, are prerequisites; their impact is indirect on the criminal

phenomenon. They are the foundations of the general control mechanism. Without bonds, models cannot be significant and constraints cannot be operand. In consequence, an unbounded community or individual cannot be sensitive to direct controls or influenced by the models. In addition, since the unfolding mechanism refers to a desirable state, this precedes the influence of available models and constraints. The bonding and the unfolding mechanisms modulate the criminal phenomenon through the mechanisms of modeling and constraining. These mechanisms are proximal causes of the criminal phenomenon because they are more specific to the space-time dimension. Their nature changes with time. The bonding and the unfolding mechanisms are in a situation of reciprocal causation at a specific moment. The modeling and the constraining mechanisms are in the same situation. In sum, the bonding and unfolding mechanisms are the foundations and the continuity component of control, while the modeling and constraining mechanisms are catalysts of conformity. We postulate that the four mechanisms of control

C

C

540

are in a synergetic relation. They interact to produce an overall level of control of the criminal phenomenon. This synergy, as illustrated in Fig. 1, emerges also from the causal effects, the reciprocal relations, and the feedback effects. The superposed boxes containing the constructs represents the changes on the time, age, and moment dimensions of the evolution of controls. This structure of the theory applies to the three levels of the definition of the criminal phenomenon. In a multilayered theory, the components are present at all the levels of explanation of the criminal phenomenon. There is isomorphism in the content of the mechanisms of control at the layers of the crime, the criminal, and the criminality. The Content of the Control Theory of the Criminal Behavior

Reiss’ (1951) statement of control theory proposed the distinction between social and personal controls. Hirschi’s formulation of the bond (1969) did not include psychological constructs. This deficiency is overcome with the notion of self-control (Gottfredson and Hirschi 1990). Since the 1970s, the vast majority of etiological publications in criminology develops and tests these notions. At the level of the criminal, conformity to conventional standards of behavior occurs and persists, on one hand, if self-control is robust and the bond to society is firm and, on the other hand, if constraints are appropriate and models are prosocial. This self- and social regulation of conformity is conditional to the biological capacities of the person and his position in the social structure. Alternatively, criminal behavior emerges and continues when low self-control persists, the social bond is tenuous, the constraints are insufficient, and the deviant model is abundant. These causes of offending will be more efficient when the individual has some biological deficiencies and when he comes from a lower social class. The definitions of these six constructs are presented and operationalized in Le Blanc (1997a, b, 2006) and in numerous studies in criminology. The structure of the individual level control theory can be deduced from many empirical results. In causal modeling term, bonds (attachment

Control Theory

and commitment) and self-control are in a situation of reciprocal causation. They are developmental prerequisites of conformity. They are also continuous primary needs during the life course. This is not the case for the modeling (involvement and peers and adults models) and constraining (rules, supervision, punishment) mechanisms. During the life course, the models change – attitudinal and behavioral demands of society for children, adolescents, and adults – and constraints are transformed – from external to internal controls – with age and with the evolution of society. In consequence, modeling and constraining are consequences of the bonding and self-control mechanisms; they are proximal causal factors of conformity. There are also retroactive effects illustrated in Fig. 1. A retroactive effect runs in the opposite direction of the sequence defined by the structure of the theory, for example, from actual individual offending on subsequent state of bonding, self-control, modeling, and constraining. The Content of the Control Theory of the Criminality

We now move on to the level of criminality. We elaborate our community control theory with the content and the logic used for the personal control theory. Kornhauser (1978) argues that the social disorganization perspective is basically a control theory, even if it is primarily about communities. She indicates that this theoretical perspective has an overwhelming emphasis on the community context. She demonstrates that social disorganization theory includes the major components of a control theory: the strength of social bonds in the community as the foundation of control, the importance of direct controls, the weakness of culture, and the defective socialization to cultural values. Her model is based on three constructs: exogenous variables (economic status, mobility, heterogeneity), cultural disorganization, and social disorganization. She specifies the relationships between the exogenous variables and the structural and cultural community organization constructs without stating clearly the connection between these two forms of community organization. Our version of community control elaborates on this base.

Control Theory

The dependent variable is the rate of criminality in a particular community at a specific moment and its evolution over time. This rate refers either to an overall rate or to rates for specific types of criminal acts. The independent variables, as in the personal control theory, comprise six constructs: social structure, setting, social organization, cultural organization, opportunities, and direct controls. These constructs represent, respectively, the environment and the background contexts, the bonding, the modeling, the unfolding, and the constraining mechanisms (Fig. 1). Our community control theory assumes that a high rate of conformity to conventional standards of behavior persists when the social organization is sound and the cultural organization robust, when direct controls are efficient, and when there are sufficient legitimate opportunities. This regulation of conformity is conditional on the quality of the setting and on the position of the community in the social structure. Alternatively, a high rate of criminality exists when social disorganization and cultural disorganization are persistent, when direct controls are inappropriate, and when deviant subcultures and opportunities are numerous. These causes of a high rate of criminality in a community will be more efficient when the setting is physically degraded and when the social status of the community is low. The rate of conformity will vary over time and between communities according to changes in the position of the community in the social structure and the quality of the setting and to variations in the levels of social and cultural organization, direct controls, and opportunities. The construct of social structure refers to the population and urbanization dimensions, the socioeconomic composition of the community, the residential stability of the residents, and the racial and ethnic heterogeneity of the neighborhood. The setting is the isomorphic construct to biological capacity at the level of the individual. It is defined traditionally by the density and crowding dimensions and by the physical deterioration of the inner city. It could also involve such characteristics as the level of pollution, traffic, noise, and so on. Social disorganization refers to the weakness of informal networks; it is

541

C

the loss of community control over members and the erosion of informal networks, the quality of primary relational networks (intimate informal primary groups: family, friends, neighbors), and the strength of secondary relational networks (broader local interpersonal networks and the interlocking of local institutions). The distinction between structure and culture has long been established, particularly by Parsons. Various notions are present in the criminological literature that refers to cultural organization. There is Sutherland’s constructs of economic and political individualism, Sellin’s notion of culture conflict, Merton’s anomie construct, and Angel’s social integration notion. Cultural disorganization refers to the attenuation of societal cultural values as controls. The desirable state for any society is the presence of robust and influential cultural values. In a deteriorated setting and a low socioeconomic status community, various deviant models and numerous illegitimate opportunities are available to residents. The notion of opportunities includes the subcultures, their representative gangs, and the corresponding illegal markets (stolen goods businesses, drugs, prostitution, and so on). However, it also involves black markets, such as undeclared work, and deficiencies in resources for conventional activities, such as work, playgrounds, sport organizations, and art classes. The notion of opportunities also refers to the large availability of suitable targets because of the deficiencies in social and cultural organization and formal and informal direct controls. The existence of illegitimate opportunities and the scarcity of legitimate opportunities encourage the use of repressive direct controls, and they sustain a high rate of criminality. Kornhauser (1978, p. 74) defines direct controls as “. . .purposive efforts to ensure conformity or limit deviance....” Hunter (1985) proposes three levels of such controls. The private level refers to relationships among peers and adults. The parochial level points to the broader set of local interpersonal networks of neighbors and interlocking of local institutions, such as voluntary organizations, stores, schools, and churches. The public level involves the ability of the community to secure public goods and

C

C

542

services (health services, social services, policing, and so on). We argue that an integrative community control theory contains six constructs: social structure, setting, social organization, cultural organization, opportunities, and direct controls. In addition, based on past theoretical statements and empirical studies, we proposed many interdependencies between these constructs (Fig. 1). There is a consensus among the specialists of community control that the social structure of the community and its setting are exogenous factors to community processes. These sets of variables directly influence the social and cultural organization of the community, the nature of available opportunities, and the nature of direct controls. In consequence, the social structure and the setting do not have a direct impact on the rate of criminality. Communities, like humans, have two primary needs, self-conservation and integration to its larger society. First, the primary need of self-conservation has its origin in the setting and the social structure, and they are the main sources of community organization. As a result, the setting and social structural contexts encourage the bonding (social organization) and the unfolding (cultural organization). These mechanisms imply that, from a particular setting and a specific position in the social structure, the community develops its social and cultural organization. Secondly, the primary need of integration of the community into its larger society leads also to the nature of the social and cultural organization. There is also a consensus that the available opportunities and the nature of direct controls are consequences of the social and cultural organization of the community. The two primary needs of communities, self-conservation and integration, are persistent during history. This is not the case for the modeling (opportunities) and constraining (direct controls) mechanisms. Over time, the models change and the constraints alter with the evolution of society. For example, gangs of today are different from gangs of previous decades. During history, constraints have been transformed from repressive to humanitarian. As a consequence, Fig. 1 states that the modeling and the

Control Theory

constraining mechanisms are influenced by the social organization and cultural organization mechanisms. Figure 1 also indicates that this theory is dynamic and interactional. Reciprocal, directional, and retroactive arrows represent interactions. The Content of the Control Theory of the Criminal Event

An event control theory has common grounds with the personal and the community control theories. Hirschi (1986) argues that there is no fundamental opposition between rational choice and control perspectives. They share the same image of man as a self-seeking individual. Bursik and Grasmick (1993) state that the routine activities and the social disorganization perspectives complement each other. They argue that the community dynamics relate naturally to the offender/target/capable guardian convergence so important for routine activities theorists. In sum, a criminal event is a function of the community in which it takes place and of the individual who commits it. In these circumstances, it is natural to move to the micro-level explanation of the criminal phenomenon with a control perspective. The offense control theory assumes that conformity to conventional standards of behavior in a specific situation occurs when routine activities are conventional, when the individual presentness is low, when there is no occasion for the perpetration of a criminal act, and when guardianship is reliable. This regulation of conformity is conditional on the quality of community and personal controls. Alternatively, a crime is likely when the person’s presentness is high, when his routine activities are unconventional, when there are numerous occasions to commit crimes, and when possible targets are unprotected. These causes of the perpetration of an offense will be more potent when the person has a high propensity for crime (low personal control) and lives in a disorganized community (low community control) (Fig. 1). The dependent variable is a particular criminal event. This event can be various, characterized according to the nature of the crime: the mechanics of the perpetration of the act (planning, use of

Control Theory

instruments, accomplices, and so on) and the psychological reaction before, during, and after the event (Le Blanc and Frećhette 1989). The explanatory constructs of community control, personal control, presentness, routine activities, occasions, and guardianship are isomorphic with the constructs of the other two levels of explanation of the criminal phenomenon. Personal and community controls represent the exogenous variables. Routines’ activities delineate the bonding mechanism; the presentness component manifests the unfolding mechanism; the occasion’s construct defines the modeling mechanism; and the guardianship dimension represents the constraining mechanism. The construct of routine activities refers to habitually enacted public activities. This construct focuses particularly on the individual lifestyle, the daily activity patterns that disperse the person away from his family, the household situation, and the involvement in conventional activities. According to routine activity theory, these activities will bring the person in contact with numerous targets for the commission of a crime that is improperly guarded. It incorporates public and private and institutional and noninstitutional activities. The construct of presentness is defined as “the idea that people differ in the extent to which they are vulnerable to the temptation of the moment” (Gottfredson and Hirschi 1990, p. 87). It characterize people with low self-control as “. . .impulsive, insensible, physical, risk-taking, shortsighted, and nonverbal” (p. 90). The Chicago school tells us that in crime-prone communities, there are numerous occasions for the commission of criminal acts. The individual makes a rational choice about the suitability of the target; he evaluates the degree of effort involve, the amount and immediacy of the reward, the likelihood and severity of punishment, and the moral costs. We expect that the more a target seems suitable, the higher the probability of the commission of an offense, particularly if the person’s self-control is low and his routine activities are public and noninstitutional. In Cohen and Felson’s (1979) routine activity theory, three elements are necessary for the commission of a criminal act: a likely

543

C

criminal, a suitable target, and the absence of a capable guardian. The absence of guardianship is a proximal cause of a crime, and that situation is likely to be perceived as such when the person has high presentness, when his routine activities are dominantly unconventional, and when there are suitable targets. Our integrative offense control theory relies greatly on Felson’s (1986) discussion of the relationships between criminal choices, routine activities, indirect control, ecology, and social control theories. Figure 1 represents the interdependencies between the constructs from a dynamic point of view (the superimposed boxes) and in an interactional perspective (the directional, bidirectional, and feedbacks arrows). Le Blanc (1997a, 2006) discusses the interdependencies between the constructs of the community control and the personal and event control theories and the personal control and the event control theories. The Developmental Statement of the Integrative Multilayered Control Theory of the Criminal Phenomenon The structural statement of our multilayered control theory has defined its constructs and structure. Most criminological theories do not discuss the dynamics of continuity and change of the criminal phenomenon, and the criminological data are rarely longitudinal. Only recently, theories include a developmental perspective. We believe that the time is ripe to further these endeavors through our multilayered control theory. To do so, we used the frameworks provided by theoretical developmental psychology and chaos theory (Le Blanc 1997a, 2006). Crime is embedded in criminality and individual offending, and criminals are constituent parts of the crime rate. Event control is part of personal and community control, and personal control is a component of community control. Figure 2 represents embeddedness by three parallel spirals. Development is contextual and probabilistic. This means that the influence of the changing context on the trajectory of development is partly uncertain and that development must be defined in terms of “. . . organism-context

C

C

544

Control Theory

Event control Crime

Legend: Effects Reciprocal (contemporaneous) Causal (Lag) Feedback (lag)

Personal control Criminal

Initial (launch) Spiral: state dependent

Community control Time, age, moment

Criminality

Control Theory, Fig. 2 The developmental multilayered control theory of the criminal phenomenon

reciprocal or dynamic-interactional relations” (Lerner 1986, p. 69). However, the organization and the internal coherence of the organism limit the probabilities of different trajectories. These principles are also those of the chaos perspective (Gleick 1987; Briggs and Peat 1989). This perspective talks about structured randomness, complex systems, nonlinear dynamics, and inner rhythms. In sociology, Fararo (1989) develops that line of thinking under the notion of dynamic social systems and formalizes his theory in terms of the mathematics of nonlinear dynamics. According to the contextual perspective on development, we can state that the level of community, personal, and event control is specific to a particular time and space. According to chaos theory, we can state that changes at one layer of control will affect changes at the other layers of control. Spirals in Fig. 2 represent the coevolution of the changes at the various layers of control. Arrows from one spiral to the others represent that changes in event control will

modify the level of personal and community control and vice versa. Studies in criminology document contextual effects. Whatever the level of explanation of the criminal phenomenon, its developmental trajectory reflects the orthogenic principle stated by Werner (1957, p. 126): “. . .whenever development occurs it proceeds from a state of relative globality and lack of differentiation to a state of increasing differentiation, articulation, and hierarchic integration.” Psychologists will accept that proposition and sociologists will recognize the pertinence of that statement for societal change. The evolution toward more complexity is governed by two principles: sensitivity to the initial condition and probabilistic epigenesis. Criminologists demonstrated that past criminal activity explains subsequent offending (Le Blanc and Loeber 1998). Individual offending is not the only variable affected by this principle; a significant proportion of the explained variance of explanatory variables – attachment to parents, commitment to education,

Control Theory

association with delinquent peers, and others – is accounted for by the past level of the same variables (Le Blanc 2006). Superimposed boxes in Fig. 1 and turns in spirals in Fig. 2 represent this developmental mechanism. As a consequence, each subsystem is partly self-organizing and self-perpetuating. This is the case for the bonding, unfolding, modeling, and constraining systems and their subsystems. Concerning the principle of probabilistic epigenesis, results about individual offending clearly show that there are some normative stages but that the outcome of individual development is only probable, never certain. Finally, development is interactional as recognized by criminologists (Thornberry 1987) and contextual developmentalists (Lerner 1986). The development of event, personal, and community control implies interactions that are represented in Fig. 2. Interactions take various forms: reciprocal interdependencies among constructs at a specific time; causal relationships between constructs over time, such that constructs will become alternatively independent and dependent variables; state dependencies for each construct; and, retroactive, the impact of the criminal phenomenon at time 2 on the four control mechanisms at time 3. After reviewing the principles that govern the development of the control of the criminal phenomenon, it is necessary to specify the nature of its course. In our developmental criminology paradigmatic paper (Le Blanc and Loeber 1998), we propose a developmental analysis of individual offending. Le Blanc (2006) argues that we can apply the methods of study for the analysis of within-individual change to the course of event and community control. The course of controls can take the forms of quantitative and qualitative changes, a distinction fundamental for developmentalists and specialists of social change. Quantitative changes are usually termed “trends” at the community level and “growth curves” at the individual level. First, quantitative changes are the degree of change on any construct of our multilayered control theory. Second, they assess the direction of change, progression

545

C

or regression. Finally, they also refer to the rate of change, the relationship between the degree of change and time. We can measure the degree, direction, and velocity of change for community control, for example, concerning the social structure: variations in the racial heterogeneity, in its social organization; changes in participation in voluntary organization, in its direct control; and changes in the number of police patrols. We could assess the quantitative measures of change in attachment, self-control, and number of delinquent friends for the personal control. We could describe quantitative changes in guardianship, availability of targets. Qualitative changes refer to something that is different from what went on before, something that is more complex according to the orthogenetic principle. These changes in nature are habitually subdivided in a developmental sequence that comprises a certain number of stages. When sociologists talk about industrial, postindustrial, developing, postmodern societies, and so on, they refer implicitly to such stages. When social ecologists talk about a community moving from a middle-class status to a working-class status and from a homogeneous underclass status to a gentry’s status, they define implicitly stages of development. Psychologists refer explicitly to stages when they talk about sensorimotor or preoperational intelligence; oral, anal, or phallic functioning; and conventional or postconventional reasoning. In each of these examples, social and behavioral scientists call attention to a universal developmental sequence divided into a limited number of stages. At the level of analysis of events, there is probably a sequence in planning and organization, for example, between a forged check to a sophisticated credit card fraud, from an unplanned breaking and entering to a professional one, and from events of bullying during childhood to gang fights during adolescence and wife battering later on. We assume that a normative developmental sequence exists for the communities, individuals, and events constructs of our multilayered control theory. Figure 2 represents the dynamics of control of the criminal phenomenon. This figure integrates the mechanisms of the course of development

C

C

546

and the developmental processes to the structure of the multilayered control theory of the criminal phenomenon. The dynamics of control are of two categories. First, Fig. 2 represents continuity and change over time; in that case, the figure must be read from left to right. Second, Fig. 2 also shows the interactions between the layers of control; in that case, the figure must be read vertically. The horizontal reading of Fig. 2 refers to minutes and hours in the event control spiral; to days, weeks, months, and years in the personal control spiral; and to years and decades in the community control spiral. These spirals are metaphors that represent the action of the mechanisms we proposed to specify the course of development of the criminal phenomenon. The time dimension that is associated with the spirals shows orthogenesis. The beginning of the spirals represents the initial condition, while the rest of the spirals introduce the sensitivity to the initial condition. The independent spirals are there to indicate that event, personal, and community control are self-organizing phenomenon. Coevolution is indicated by the placement of the spirals on three dimensions. Finally, along each spiral, there are probabilistic quantitative and qualitative changes. The vertical reading of Fig. 2 implies that we think in terms of interdependencies between the layers of control of the criminal phenomenon. In our discursive statement of our theory, synergy implied embeddedness and reciprocal relations between the layers of control. In Fig. 2, embeddedness is represented by the fact that the crime spiral is placed in-between the criminal and the criminality spirals. This position indicates that an offense is a constituent part of individual offending and of the rate of criminality and that individual offending is part of criminality. In Fig. 2, interactions are shown by the large arrows; each arrow specifies these interdependencies. Synergy also exists at all points along the time dimension, and as a consequence, the multilayered control of the criminal phenomenon looks like a torus attractor such as proposed by chaos theorists to represent continuity and change (Briggs and Peat 1989; Le Blanc 2006).

Control Theory

Figure 2 also represents the interactional perspective. The discursive statement of our multilayered control theory identifies various types of relations: reciprocal, causal, state dependent, and retroactive. Large arrows indicate reciprocal relations, while thin arrows show causal relations and retroactions. A discursive statement of the relations showed in Fig. 2 would be the following. Insufficient community control and tenuous personal control will diminish the level of event control of a person and a crime is more likely to be committed. Conversely, the commission of a crime will alter the actual level of personal control for that person and modify the actual level of control in his community. Notwithstanding these relationships between layers of control, there are also some independent within layers changes relative to the initial condition. In addition, a change in a person level of event control will alter his subsequent level of personal control and the subsequent level of control in his community. A change in community control will also affect the subsequent level of personal and event control for the person. Finally, a change in the level of a person’s personal control will affect the subsequent level of event control for that person and the following level of control in his community. Notwithstanding these changes, the level of the criminal phenomenon also modifies the subsequent level of control.

Conclusion Our control theory covers three levels of definition and explanation of the criminal phenomenon: the crime, the criminal, and the criminality. We constructed an integrative theory with different words and terms from different theorists, but their theoretical meanings and operational definitions are synonymous. Our theory is fractal because of the similarity of structure from one level of definition of the criminal phenomenon to the other. Six generic constructs and two categories of exogenous factors are defined and organized in a similar structure. The control

Control Theory

mechanisms are bonding, unfolding, modeling, and constraining. We define a contextual perspective that allows for developmental principles of orthogeny, sensitivity to the original state, and epigenetic probability. The course of development has quantitative and qualitative dimensions and is nonlinear. As presented, this formulation of our multilayered control theory has several deficiencies. The first concern is its generic and discursive status. The statement of our control theory is not yet specific enough, and additional work is needed to facilitate an operational formulation of the theory. A theory is also never complete before a formal test identifies its logical inconsistencies. A formalization of the individual level control theory exists (Le Blanc and Caplan 1993), but the same operations should take place for community and event control and our generic control theory as a whole. Only then will it be possible to discuss its testability and start empirical tests. However, numerous limited tests of the structure our control theory do exist in the criminological literature.

Related Entries ▶ Age-Crime Curve ▶ Anomie and Crime ▶ Career Criminals and Criminological Theory ▶ Causes ▶ Criminal Careers ▶ Desistance from Crime ▶ Early Chicago School Theory ▶ General Strain Theory ▶ Genetic Basis to Self-Control ▶ Group-Based Trajectory Models ▶ Growth Curve Models with Categorical Outcomes ▶ Informal Social Control ▶ Integrating Rational Choice and Other Theories ▶ Interactional Theory of Delinquency ▶ Labeling and Deviance ▶ Lifestyle Theory ▶ Longitudinal Crime Trends at Places ▶ Measurement of Self-Control

547

C

▶ Modus Operandi of Sex Offenders ▶ Moffitt’s Developmental Taxonomy of Antisocial behavior ▶ Neighborhood Effects and Social Networks ▶ Pathways to Delinquency ▶ Poverty, Inequality, and Area Differences in Crime ▶ Psychopathy ▶ Rational Choice Theory ▶ Situational Action Theory ▶ Social Capital and Collective Efficacy ▶ Social Control ▶ Social Control and Self-control Through the Life Course ▶ Social Control Theory of Sexual Homicide Offending ▶ Synthesizing Biological and Social Theorizing

Recommended Reading and References Briggs J, Peat FD (1989) Turbulent mirror. Harper and Row, New York Bursik RJ, Grasmick H (1993) Neighborhoods and crime: the dimensions of effective community control. Lexington Books, New York Cohen LE, Felson M (1979) Property crime rates in the United States: a macrodynamic analysis 1947–1977, with ex ante forecasts for the mid – 1980. Am J Sociol 86:90–118 Durkheim E (1934) De l’e´ducation morale. Alcan, Paris. English edition: Durkheim E (1961) Moral education (trans: Wilson EK, Schnuer H). The Free Press, New York Empey LT (1978) American delinquency. Dorsey Press, Homewood Fararo TJ (1989) The meaning of general theoretical sociology, tradition and formalization. Cambridge University Press, New York Felson M (1986) Linking criminal choice, routine activities, informal control and criminal outcomes. In: Cornish DB, Clarke RV (eds) The reasoning criminal: rational choice perspectives on offending. Springer, New York Gibbs JP (1989) Control sociology’s central notion. University of Illinois Press, Chicago Gleick J (1987) Chaos, making of a new science. Vicking Penguin, New York Gottfredson MR, Hirschi T (1990) A general theory of crime. Stanford University Press, Stanford Hirschi T (1969) Causes of delinquency. University of California Press, Berkely Hirschi T (1979) Separate and unequal is better. J Res Crime Delinq 16:34–38

C

C

548

Horowitz FD (1987) Exploring developmental theories: toward a structural/behavioral model of development. Lawrence Erlbaum, Hillsdale Hunter AJ (1985) Private, parochial and public school orders: the problem of crime and incivility in urban community. In: Suttles GH, Zald MN (eds) The challenge of social control: citizenship and institution building in modern society. Ablex Publishing, Norwood Kornhauser RR (1978) Social sources of delinquency: an appraisal of analytic models. University of Chicago Press, Chicago Le Blanc M (1997a) A generic control theory of the criminal phenomenon, the structural and the dynamical statements of an integrative multilayered control theory. In: Thornberry TP (ed) Developmental theories of crime and delinquency, vol 7, Advances in theoretical criminology. Transaction Publishers, New Brunswick, pp 215–286 Le Blanc M (1997b) Socialization or propensity: a test of an integrative control theory with adjudicated boys. Stud Crime Crime Prev 6:200–224 Le Blanc M (2006) Self-control and social control of deviant behavior in context: development and interactions along the life course. In: Wikstrom P-O, Sampson R (2005) The social contexts of pathways in crime: development, context, and mechanisms. Cambridge University Press, Cambridge Le Blanc M (2009) The development of deviant behavior, its self-regulation. Mon Kriminol Strafr 91:117–136 Le Blanc M, Caplan M (1993) Theoretical formalization, a necessity: the example of Hirschi’s social control theory. Adv Criminol Th 4:329–431 Le Blanc M, Frećhette M (1989) Male criminal activity, from childhood through youth: multilevel and developmental perspectives. Springer, New York Le Blanc M, Loeber R (1998) Developmental criminology upgraded. In: Tonry M (ed) Crime and justice handbook, vol 23. University of Chicago Press, Chicago, pp 115–198 Lenski G (1988) Rethinking macrosociological theory. Am Sociol Rev 53:163–171 Lerner RM (2002) Concepts and theories of human development. Lawrence Erlbaum, Mahwah Lytton H (1990) Child and parent effects in boys’ conduct disorder: a reinterpretation. Dev Psychol 26:683–697 Messner SF, Krohn MD, Liska AE (eds) (1989) Theoretical integration in the study of deviance and crime: problems and prospects. State University of New York Press, Albany Pinatel J (1963) Traite´ de criminologie. Dalloz, Paris Reiss AJ (1951) Delinquency as the failure of personal and social controls. Am Sociol Rev 16:196–207 Shoemaker DJ (2005) Theories of delinquency: an examination of explanations of delinquent behavior. Oxford University Press, New York Short JF (1985) The level of explanation problem in criminology. In: Meier RF (ed) Theoretical methods in criminology. Sage, Beverley Hills

Controlled Experiments Thornberry TP (1987) Toward an interactional theory of delinquency. Criminology 25:863–892 Wagner DG, Berger J (1985) Do sociological theories grow? Am J Sociol 90:697–728 Werner H (1957) The concept of development from a comparative and organismic point of view. In: Harris DB (ed) The concept of development. University of Minnesota Press, Minneapolis

Controlled Experiments ▶ Randomized Experiments in Criminology and Criminal Justice

Co-offending Peter J. Carrington Department of Sociology and Legal Studies, University of Waterloo, Waterloo, ON, Canada

Overview This entry is a general introduction to the subject of co-offending. In-depth treatments of specific topics in co-offending are provided in the articles by Hochstetler (decision-making), van Mastrigt (attributes of co-offenders), and Weerman (theorizing co-offending). Co-offending is the commission of a crime by more than one person. It is often called “group crime,” although that term is somewhat misleading, as the vast majority of co-offenses are committed by only two offenders (see Size, below). Co-offending is common and has been observed and studied by criminologists for at least a century, since Breckinridge and Abbot (1912) remarked on its high frequency among delinquent boys appearing in the juvenile court of Cook County, Illinois. (Shaw and McKay 1931) also commented on the prevalence of co-offending among boys appearing in the Cook County juvenile court in 1928. The study of co-offending became

Co-offending

established in the 1960s, grew slowly through the rest of the twentieth century, and has begun to flourish in the early twenty-first century, partly because of a growing awareness of the applicability of the theories and methods of social network analysis (Carrington 2011). Co-offending is studied as a phenomenon of interest in itself and also because it can provide insight into other criminological phenomena, such as problems in the measurement of crime, social contagion of criminal attributes and behavior, and criminal networks, gangs, and organized crime.

Prevalence Writers on co-offending are unanimous that it is very common, especially among young offenders. However, empirical estimates of its prevalence vary widely – from 10 % of criminal incidents to over 80 % of offenders – probably because of variations in the definition of prevalence and in the types of data, as well as the variety of places, times, and samples, that have been used to measure it (Table 1). van Mastrigt and Farrington (2009) point out that in estimating prevalence, offenses (that is, criminal incidents) must be distinguished from offense participations, or the participation of one person in one offense. Since co-offenses involve multiple offense participations, prevalence estimates based on counting co-offenses will be lower than estimates based on counting offense participations. In addition, some authors report a third type of prevalence: the percentage of subjects in the sample who ever co-offended during the period of observation. Each definition of prevalence provides a different perspective on the same phenomenon. Estimates of prevalence also differ because they are derived from self-reported versus official (recorded) data (or, occasionally, victim reports), because the sample is limited to a certain age range, such as (commonly) young people, or because different types of crime are included (Table 1). Early estimates of the prevalence of co-offending tended to be relatively high.

549

C

Breckinridge and Abbot (1912) commented that lone offending was rare among the delinquent boys that they studied. Shaw and McKay (1931) reported that 82 % of the offense participations of a sample of boys who came before the Cook County juvenile court in 1928 were in co-offenses, and 81 % of the boys had at least one co-offense in their delinquent careers. Erickson (1971) reviewed 11 studies of juvenile delinquency in the United States prior to 1935 and found that the reported prevalence of co-offending ranged between 70 % and 80 % of offenses; his own research on self-reported delinquency found that 65 % of offenses were co-offenses. Of the offenses that were self-reported by a sample of 13–16-year-olds in the National Survey of Youth in 1967, 70 % were co-offenses (Warr 1996). Estimates based on more recent data are lower. Sarnecki (2001) reports that 56 % of the young offenders registered by the Stockholm police between 1991 and 1995 were involved in at least one co-offense during that period. Fifty-one percent of the recorded offenses committed between the ages of 10 and 32 by the boys in the Cambridge Study were co-offenses (Reiss and Farrington 1991), as were 37 % of recorded offense participations involving young persons in the British police data for 2002–2005 analyzed by van Mastrigt and Farrington (2009) and 25 % of offenses involving young persons in Carrington (2002a) Canadian police data for 1992–1999. The prevalence of co-offending is much lower in the very few studies that analyze population samples of offenders of all ages. van Mastrigt and Farrington (2009) report that only 10 % of offenses by offenders of all ages were co-offenses; Carrington (2002a) reports 12 %; and Becker and McCorkel (2011) report 18 %.

Co-offending and the Measurement of Crime The phenomenon of co-offending has important implications for the validity of simple crime

C

van Mastrigt and Farrington (2009) Carrington (2009)

Pettersson (2005)

Carrington (2002a)

Sarnecki (2001)

Erickson (1971) Warr (1996)e

Study (a) Young offenders Shaw and McKay (1931)

All All

5–17

Violent

0 (Boardman et al. 2006). NPV has a caveat; NPV does not necessarily identify the most efficient allocation of resources. Rather, NPV identifies the more efficient allocation of resources. For example, an intervention for which program alternatives vary with respect to potential output Q is considered in Fig. 3. The benefits and costs associated with alternatives are represented by the functions (BQ) and (BC), respectively. Moving from Q0 to Q* increases efficiency, but at a decreasing rate. That is, NPV(Q*) > NPV(Q2) > NPV(Q1) > NPV(Q0). However, as we move beyond the

C

C

646

Cost-Benefit Analysis, Fig. 3 Efficient resource allocation. Adapted from: Boardman et al. (2006)

Cost-Benefit Analysis

C(Q)

BQ/BC

B(Q)

Optimum resource allocation (efficiency)

Q0

Q1

optimal scale (Q*), the net benefits decrease. Net benefits remain positive until the point where the curves C (Q) and B (Q) intersect. The problem is that Q* may not have been selected as it was not included in the set of evaluated options. Therefore, NPV identifies the more efficient alternative among the group of alternatives evaluated but does not identify the most efficient alternative. The optimum alternative (Q*) may have been omitted because of bounded rationality problems (i.e., cognitive limitations of both knowledge and cognitive capacity) (Manning et al. 2011), the optimum level may not have been known, or budgetary or political constraints may have limited the range of alternatives considered for evaluation. This considered, however, NPV is still an appropriate method as it provides accurate answers with respect to the alternatives that are evaluated (Boardman et al. 2006).

Q2 Q∗

Q3

Output (Q)

Figure 4 provides a diagrammatic representation of the incremental cost curve, where ICa’, Q1 is the cost of option 1 at quantity (activity level) Q1, and ICb’, Q1 is the equivalent estimate for option 2. The incremental difference (cost) between option 1 and option 2 at quantity (activity level) Q1 is ICab, Q1. Figure 4 highlights that the incremental cost is equal to the change in total cost divided by the change in quantity. This may appear to be a marginal cost; however, the incremental cost is distinguished from the marginal cost by holding Q constant at Q ¼ 1 (Q1). Therefore, the alternatives (activities) are compared at Q ¼ Q1. The incremental cost is represented mathematically in Eq. 2: IC ¼ DTC=DQ where

Incremental Cost-Comparison Analysis

Any economic analysis should include an incremental cost-comparison analysis of the available options. The term incremental cost analysis refers here to a change in the scale of activity; in other words, the difference in cost or effect between two or more options being compared.

TC ¼

n X

ðPi QiÞ

(2)

i¼1

Note, IC ¼ incremental cost, Q ¼ quantity/activity, D ¼ change in activity, TC ¼ total cost, Pi ¼ price of inputs (derived from Table 3’) and Qi ¼ quantity of inputs.


647

C

Cost-Benefit Analysis, Fig. 4 Incremental cost analysis. Adapted from: Boardman et al. (2006) ICa⬘, Q2 Cost ($)

C

ICa⬘, Q1 ICa−b⬘, Q2 ICa−b⬘, Q1

ICb⬘, Q2 ICb⬘, Q1

Q1

The Optimal Choice of an Action

The optimal choice of an action for a decisionmaker is to maximize the benefits they receive less the opportunity cost. Comparing the addition to benefits with the addition to opportunity costs identifies the optimal choice. These are commonly referred to as the marginal benefit and the marginal cost of an action. The rule for optimal decision-making is that the program should be funded if the additional benefits from the program are equal to or greater than the additional costs. A full discussion of making the optimal decision and marginal CBA is available in Boardman et al. (2006).

Examples of Cost-Benefit Analyses Below, two economic analyses of two third party policing (TPP) interventions are discussed to illustrate the use of cost-benefit analysis and its potential limitations. For those not familiar with TPP, it is defined as . . .police efforts to persuade or coerce other regulators or non-offending persons, such as health and building inspectors, housing agencies, property owners, parents, and business owners, to take some responsibility for preventing crime or reducing crime problems (Buerger and Mazerolle 1998, p. 301).

Q2

Quantity

Ma˚nsdotter et al. (2007) conducted an economic evaluation of a multicomponent TPP alcohol prevention program implemented in licensed premises in Stockholm, Sweden. The study aimed to quantify the societal savings directly generated from the intervention. Savings were estimated as potential monetary consequences of victims of violence to the judicial system (e.g., police, public prosecutor, courts, prisons), production changes (e.g., sickness absence, other effects on working hours), health care (e.g., transport from crime scene, emergency treatment, inpatient care, outpatient care, pharmaceuticals), and other damage (e.g., technical support, personal assistance, and personal belongings). In addition, victims of violence were surveyed to ascertain the health effects of the intervention. The victim’s health state was measured before, 2 weeks after the event, and present using the EQ-5D questionnaire. The EQ-5D questionnaire is a multi-attribute health status classification system that includes five health attributes: mobility, self-care, usual activity, pain/discomfort, and anxiety/ depression. Each attribute consists of three levels: no problem, some problems, and major problems, defining 245 possible health states including the possible states of unconsciousness and dead (Drummond et al. 1987). The results were converted to a weighted state health index by using values of health profiles from a UK general

C

648


Cost-Benefit Analysis, Table 2 Total costs of DWI incidents by severity and cost Medical Emergency services Productivity Employer costs Administrative Legal Travel delay Property damage Monetary costs Quality of life Comprehensive costs

Fatal injury $6,693 $930 $665,453 $6,679 $48,337 $70,925 $387 $8,059 $807,473 $1,977,529 $2,785,002

Non-fatal injury $4,203 $194 $7,919 $530 $1,259 $1,703 $187 $3,231 $19,244 $48,367 $67,611

PDO vehicle $24 $35 $31 $127 $107 $1,157 $1,481 $1,481

Costs for all cases $4,127M $218M $19,391M $23M $2,474M $2,869M $543M $7,082M $37,490M $80,832M $118,322M

Source: McKnight and Streff (1994)

population study (Dolan et al. 1996). Health gains were reported in quality of adjusted life years (QALYs). QALYs is a generic outcome measure that is calculated by adjusting the length of time affected through the health outcome by the corresponding utility value of the resulting level of health status (Drummond et al. 1987). The costs of implementation included administration (e.g., salaries for project staff, offices, travel, and conferences), studies of alcohol serving practices (i.e., remuneration to actors), community mobilization (e.g., advisory group participation), responsible beverage service training (i.e., 2-day course on responsible beverage service for 572 participants), and stricter alcohol law enforcement (i.e., joint controls by the licensing board and police and increase of notification letters to licensees). Ma˚nsdotter and colleagues estimated the cost of the program to be 769,000 euros. The average cost of violent crime was estimated to be 19,049 euros, which equates to an approximate saving of 31.314 million euros. Savings were attributed to a reduction in judicial system costs (78%), production losses (15%), health-care issues (5%), and damages issues (2%). The cost saving ratio of the program was 1:39. Overall, health gains were estimated to be 236 QALYs (discounted QALYs), corresponding to 3–4 lives in normal health. As illustrated in the study, the intervention excluding savings was cost-effective in terms of health outcomes (assuming EUR 54,000 per quality of life – 795828/236 is approximately EUR 3,000 per QALY).

McKnight and Streff (1994) assessed the effects of a TPP intervention that enforced laws, using trained servers, to prohibit the service of alcohol to already intoxicated patrons. Specifically, the study addressed the following questions: (1) What is the effect of enforcement upon the service of alcohol to intoxicated patrons? (2) What impact does any change in service to the intoxicated have upon the relative numbers of DWIs coming from bars and restaurants? (3) How do the costs of enforcement relate to the estimated savings yielded by projected reductions in alcohol-related incidents? The benefits of the intervention were calculated using Eq. 3: B ¼ a b x ni ci

(3)

where B represents the benefits from reduced drinking and driving, a equals the proportional reduction in DWIs (from taverns) (0.361), b represents the proportion of current DWIs (from bars and restaurants) (0.35), x is a factor that controls for the percentage of alcohol-related incidents that could have occurred in the absence of alcohol consumption (0.853), and ∑ni ci represents the number of DWI incidents in a given year multiplied by the average costs per incident ($118.3 billion) (total comprehensive costs of all DWI incidents). Table 2 provides a summary of total DWI incidents by severity and cost. Results show a cost-savings ratio of 260:1; that is, each dollar invested in the intervention


649

Cost-Benefit Analysis, Fig. 5 Economic vs. accounting costs. Adapted from: Frank et al. (2007)

How an accountant views the firm

How an economist views the firm Economic Profit $

Revenue $20,000

Accounting Profit $

Implicit Costs $

Explicit Costs $

generated a savings of approximately $260. Data for calculating this figure was derived from a study by Levy and Miller (1992), where savings ($260) equals $12.8billion (total cost to everyone involved in alcohol-related incidents) divided by $48,400 (total cost of effort needed to bring about a decrease in alcohol service violations and DWIs from bars. A number of assumptions were made with respect to estimated savings: (1) only reduced traffic risk was used—this excluded assault, household injury, and other risks associated with alcohol consumption; (2) there was a low percentage of patrons who were refused drinks drinking elsewhere; (3) “traffic safety benefits from refusals of service are confined to reductions of those matching BACs in excess of 0.10%”; (4) Washtenaw County is representative of the United States; (5) the level of enforcement is consistent with the effort as that employed in the Washtenaw County study; and (6) program effectiveness and the benefits of the intervention are the same over time. Issues to Consider Ex Ante CBA To highlight the potential limitations of CBAs and emphasize important issues ex ante CBA, the CBA examples presented above are used as an illustration. This discussion is disaggregated by the cost-benefit equation, where the benefitcost ratio (BCR) is calculated by dividing the benefits of a program/intervention (B) by the

C

Total opportunity Costs $19,200

C Revenue $20,000

Explicit Costs $

costs of the program/intervention (C). The reader should note, however, that due to space limitations the discussion is restricted to a sample of limitations (specific to the CBA methodology) that may distort results. Boardman et al. (2006) provide a full discourse of limitations. The Cost Side of the Equation

Ma˚nsdotter et al. (2007) noted that their cost estimates were based on “. . .account statement, average cost estimates and various kinds of estimations by individuals” (p. 622). Costs should not be limited to accounting costs, which are based on an account statement. Rather, costs should be economic costs. Economic costs include the following: implicit costs (a cost that is represented by lost opportunity in the use of one’s own resources, excluding cash—these are sometimes referred to as intangible costs as they are not easily accounted for, and no actual payment is made) and explicit costs (a direct payment made to another party in the course of operation, e.g., wage, rent, and materials), where accounting costs only include explicit costs (see Fig. 5). Economic costs are important for a number of reasons: (1) the information contained in accounting budgets does not include all the cost information on all the resources used (e.g., volunteer time, donated equipment and services, and any other unpaid inputs); (2) inputs that have already been

C

650

paid for, or are not included in the budget, will not be discernable; (3) standard accounting budgets may distort the true cost of an input. That is, accounting budgets typically charge the cost of an input to the year in which the cost occurred. If the life of the asset is, for example, 20 years, then this cost should be distributed over the life of the asset; (4) costs are often embedded in a budget or expenditure statement that is larger than the project in question. For example, costs of an intervention (e.g., police time) may be embedded into the total operational costs of the enterprise, and as a consequence, these must be categorized according to economic classifications (e.g., variable costs); and (5) accounting documents often represent plans for how resources will be allocated as opposed to classifying expenditures after they have been used. That is, accounting statements often include budgeted accounts with actuals being left for inclusion until the end of the accounting period (Levin and McEwan 2001). Total opportunity costs (TC) (implicit + explicit costs) should be classified into fixed (FC) and variable costs (VC). FCs represent costs that are independent of output (e.g., buildings and machinery). VCs are variable with output. VC generally varies at a constant rate relative to capital and labor. That is, VCs change as the quantity of output changes (e.g., if output ¼ 0, then VC ¼ 0; if output ¼ 1, then output ¼ 50; if output ¼ 2, then output ¼ 100). With this information, we are then able to calculate total costs (i.e., TC ¼ FC + VC). The Benefits Side of the Equation

The studies by Ma˚nsdotter et al. (2007) and McKnight and Streff (1994) raise a number of general problems with respect to assessing the benefits resulting from interventions/policies: (1) The first limitation is low response rate (e.g., surveys of victims of violence in the Ma˚nsdotter study). This may be problematic as it potentially influences results in the opposite direction; Ma˚nsdotter and colleagues state, “. . .one may imagine that the most harmed victims are most inclined to respond, but also that the severely harmed hesitated most about answering the questionnaire” (p. 622). (2) The savings and health


gains attributed to an intervention are based on a decline in a given variable (e.g., violence). As such, there may be an underestimation in results, for example, spillover effects from not serving alcohol to underage patrons and staff turnover of RBS trained servers in the experimental area. On the other hand, there may be an overestimation where problem drinkers may be displaced to another area (Ma˚nsdotter et al. 2007). (3) Savings may be underestimated when an effectiveness study is limited to given times. For example, violence between 10pm and 6am and DWIs between 11pm and 5am Friday to Sunday. (4) Benefits of an intervention need to be comprehensive, that is, beyond the stakeholder perspective. For example, the Ma˚nsdotter study excluded benefits such as avoided damage and short-time closures. (5) Studies should include the long-term monetary benefits that are gained from an intervention. For example, the study by Ma˚nsdotter asked respondents to restrict their reports to 12 months after the violence—this represents only short-term benefits. (6) Studies should broaden the savings scope to include variables such as administration costs of sick leave and also medication provided by pharmacies (Ma˚nsdotter et al. 2007). (7) QALYs studies should expand their evaluation on a given independent variable. For example, examining whether unlawful threats involve more enduring violence could strengthen a study by potentially uncovering “more costs and health-damaging effects, than non-serious assaults” (Ma˚nsdotter et al. 2007, p. 622). (8) When assessing health gains (using the EQ-5D questionnaire) respondents in the Ma˚nsdotter study were asked to remember and report earlier and current health states. This may be problematic, particularly when respondents report better health at “present” than before the violence. This potential flaw certainly leads to an underestimation of health gains. (9) Studies should ensure that benefits of the intervention are assessed across all groups. For example, Ma˚nsdotter et al. (2007) did not include benefits to individuals who were prevented from purchasing alcohol. This may have uncovered both health gains and improvements in non-health-related quality of life

Counterterrorism

(e.g., changes in life events had the individual been arrested for an alcohol-related crime). (10) Studies should be aware of the two potential shortcomings of research that may compromise the internal and the external validity of results, for example, (1) what else, other than participation in the program/intervention may have caused these results and (2) how general or representative are these findings to other groups?

Final Remarks CBA is a technique that when used properly is a powerful policy decision-making tool. The technique is a little more subtle and sophisticated than most noneconomists recognize, and as such one should be careful in its application. Further, when using the results of CBA to inform public policy, one should understand that economic efficiency/effectiveness is not the only component of the decision-making protocol. Other goals such as equity and fairness should carry weight. Further, the maximization of social benefits that are identified in the economic study must be weighed against other competing interests such as moral and political objectives and how the goals of the system are conceived.

Related Entries ▶ Measuring Police Unit Performance ▶ Police-Led Interventions to Enhance Police Legitimacy ▶ Pulling Levers Policing ▶ Third Party Policing and School Truancy

651

C

Buerger M, Mazerolle L (1998) Third party policing: a theoretical analysis of an emerging trend. Justice Q 15:301–328 Dolan P, Gudex C, Kind P, Williams A (1996) The timetrade-off method: results from a general population study. Heal Econ 5:141–154 Drummond MF, Stoddart GL, Torrance GW (1987) Methods for the economic evaluation of health care programmes. Oxford University Press, Oxford Frank RH, Jennings S, Bernanke BS (2007) Principles of microeconomics. McGraw-Hill Irwin, Sydney Levin HM, McEwan PJ (2001) Cost-effectiveness analysis, 2nd edn. Sage Publications, London Levy D, Miller T (1992) A cost-benefit analysis of increased enforcement to reduce serving intoxicated patrons. University of Baltimore and the Urban Institute, Baltimore Manning M (2004) Measuring the costs of communitybased developmental prevention programs in Australia. Griffith University, Brisbane, Unpublished Masters (Hons) Manning M (2008) Economic evaluation of the effects of early childhood intervention on adolescent outcomes. Griffith University, Brisbane, Unpublished PhD Manning M, Homel R, Smith C (2006) Economic evaluation of a community-based early intervention program implemented in a disadvantaged urban area of Queensland. Econ Anal Pol 36(1 & 2):99–120 Manning M, Homel R, Smith C (2011) An economic method for formulating better policies for positive child development. Aust Rev Pub Aff 10:61–77 Ma˚nsdotter A, Rydberg M, Wallin E, Lindholm A, Andreásson S (2007) A cost-effectiveness analysis of alcohol prevention targeting licensed premises. Eur J Public Health 17(6):618–623 McKnight A, Streff F (1994) The effect of enforcement upon service of alcohol to intoxicated patrons of bars and restaurants. Accid Anal Prev 26:79–88 Mishan E (1972) The futility of pareto efficient distributions. Am Econ Rev 62(5):971–976 Zerbe R, Bauman Y, Finkle A (2006) An aggregate measure for cost-benefit analysis. Ecol Econ 58(3):449–461

Recommended Reading and References

Counterterrorism

Boardman AE, Greenberg DH, Vining AR, Weimer DL (1996) Cost-benefit analysis: concepts and practice, 2nd edn. Prentice Hall, Upper Saddle River Boardman AE, Greenberg DH, Vining AR, Weimer DL (2001) Cost-benefit analysis: concepts and practice, 2nd edn. Prentice Hall, Upper Saddle River Boardman AE, Greenberg DH, Vining AR, Weimer DL (2006) Cost-benefit analysis: Concepts and practice, 3rd edn. Pearson-Prentice Hall, Upper Saddle River Brent R (2008) Applied cost benefit analysis. Edward Elgar Publishing, Northhampton

Laura Dugan Department of Criminology, University of Maryland, College Park, MD, USA

Overview This entry divides terrorist countermeasures into deterrence-based, conciliatory-based, and

C

C

652

situational-based strategies. Deterrence-based strategies attempt to reduce terrorism through punishment and the threat of punishment. While politically popular, empirical evidence suggests that punishment or repression directed toward terrorists or their constituency can lead to increases in terrorist violence. Conciliatorybased counterterrorism strategies offer rewards for legitimate alternatives to terrorist behavior. Examples include actions directed toward terrorists, such as leniency that might be offered to them for publicly renouncing their organization or conciliation that might be directed toward the terrorists’ constituency, such as providing medical assistance at a time of need. The two studies that evaluate the effects of conciliation show that these actions are, indeed, related to less terrorism (i.e., in Israel and Turkey). Situational-based counterterrorism refers to efforts to reduce opportunities for terrorist attacks (e.g., installing metal detectors in airports). While evidence suggests that target hardening has effectively reduced some forms of terrorism (aerial hijacking), terrorists are often innovators who can develop alternative methods of attack. Finally, scholars are still learning what counterterrorism strategies have worked in the past to reduce terrorism as they begin to develop datasets that document the prior actions by governments to control terrorist violence.

Introduction Concerns about terrorism have grown substantially since the late 1960s when the Popular Front for the Liberation of Palestine (PFLP) hijacked an Israeli El Al flight that was in route from Rome to Tel Aviv, demanding the release of Palestinian prisoners and creating a drama that permeated major media capturing the world’s attention. It signaled to the world that terrorist violence is no longer contained in known conflict zones (Hoffman 2006). Prior to that event, most terrorist campaigns were localized, affecting only a single country or region and targeting only those individuals who were relevant to the terrorists’ grievances (e.g., Basque Fatherland and

Counterterrorism

Freedom [ETA] in the Basque region of Spain, Corsican National Liberation Front [FLNC] in the French island of Corsica, the National Liberation Front [FLN] in Algeria, the Irish Republican Army [IRA] in the United Kingdom). Once governments realized that terrorist violence was becoming a transnational problem, efforts began to develop more robust strategies for protecting citizens from being targeted by terror. This was especially apparent after the 1972 massacre of members of the Israeli Olympic team by the Palestinian terrorist organization Black Friday when Western Europe began to formulate antiterrorism policies and create counterterrorism units. In the more than 40 years since the PFLP hijacking, reports on terrorist attacks have regularly pervaded media outlets, leading to global concerns that most persons are potentially vulnerable targets of terrorism. Despite this increasing concern, it is surprising how little is known about effective strategies to reduce terrorist violence. Schmid and Jongman (1988) reviewed more than 6,000 published works that examine terrorist violence and found that most of it is impressionistic making broad generalizations based only on episodic evidence. This conclusion is disheartening given that officials had to regularly develop counterterrorism strategies based on anecdotal evidence rather than empirically sound evaluations. As of 2004, this problem had only marginally improved. Lum and colleagues (2006) conducted a thorough and systematic review of counterterrorism evaluations as part of the Campbell Collaborative program. From the more than 20,000 written documents on terrorism between 1971 and 2004, only seven studies met their criteria of being moderately rigorous evaluation studies. This is especially surprising because a similar review of criminal justice evaluations found more than 500 rigorous and scientifically sound program impact evaluations (Sherman et al. 1997), raising a stark contrast between what we know about what works in criminal justice and what little we know about controlling terrorism. To be fair, one of the primary reasons terrorism research had relied so heavily on case studies

Counterterrorism

rather than empirical analysis is because comprehensive data that documents terrorist activity simply did not exist. This changed shortly after the beginning of the twenty-first century when RAND began to chronicle both international and domestic attacks and the National Consortium for the Study of Terrorism and Responses to Terrorism (START) compiled the Global Terrorism Database (GTD). As data became more widely available and extensive, scholars began to evaluate the effects of known government efforts to stop terrorist violence. The following sections describe different strategies that governments have used to control terrorism and the studies that have evaluated their effectiveness.

Types of Countermeasure Approaches In this section, three major strategies are outlined to addressing terrorist violence. The first – deterrence-based counterterrorism – is by far the most appealing to officials because it is consistent with strategies to enforce criminal law and it demonstrates the strength of the government, increasing its popularity across constituencies. The second strategy – conciliatory-based counterterrorism – is typically less popular among officials and their constituents, but might be more effective in reducing violence. The third approach – situational-based counterterrorism – is similar to situational crime prevention, in that it hardens likely targets making it more difficult for terrorists to gain access to them. All three approaches are explained in greater detail below. Deterrence-Based Counterterrorism In his famous speech after the Iranian hostage crisis, President Ronald Reagan warned, “Let terrorists beware that when rules of international behavior are violated, our policy will be one of swift and effective retribution” (Stanik 2003: 33). This sets the tone for the Reagan administration’s and future US administrations’ response to international terrorism, as evidenced by the 1986 bombing of Libya, the 1993 bombing of Iraq’s military intelligence headquarters, and the 1998 missile attacks in Afghanistan and Sudan. While

653

C

rhetorically bold, Reagan’s proclamation was consistent with hundreds of years of practice under Western criminal law traditions, which calibrates levels of punishment to best deter lawbreaking. Given this history, the appeal of repressive responses to terrorism is unsurprising, especially after devastating attacks like that of 9/11. In fact, President Bush’s approval ratings soared as he assertively offered phrases like, “wanted dead or alive,” “smoke them out,” and “bring ‘em on” (CNN 2001; Knowlton 2001; Associated Press 2003). With the popular support of the US public, the United States invaded Afghanistan demonstrating its resilience and strength, while decimating al-Qaeda training camps and forcing the Taliban into hiding. Other efforts to deter terrorist behavior include apprehending terrorists and extending their prison sentences, passing antiterrorism laws, targeted assassinations, imposing curfews and containment, retaliating with violent military repression, invading territory, and indiscriminate repression. While deterrence is appealing to many, others have argued that repression can lead to backlash. For example, since the US invasion of Afghanistan, opinions about its effectiveness are mixed. It is true that the organization al-Qaeda, as it was in September 2001, no longer exists; however, its successors remain highly networked and now have offshoots planning attacks across the globe. This new global jihad reaches supporters through its teachings on various Internet sites and its regularly published online English magazine, Inspire. Despite the rise in the global jihad movement, to date, the “new” al-Qaeda has been unable to replicate any attacks as sensational as the 9/11 attacks in the United States or elsewhere in the world, suggesting that the US repression response may have perhaps been an effective deterrent strategy. However, it can be argued that the damages caused by the attacks in Afghanistan and Iraq are far worse than any benefits because they have undermined the legitimacy of the United States, as the collateral damage gives the impression that the United States places little value on Afghani and Iraqi lives. Further, the military

C

C

654

strikes seem to have increased al-Qaeda’s popularity throughout the region, attracting large numbers of recruits into the global jihad (Malvesti 2002). In fact, some go as far as to claim that Osama bin Laden’s intent behind the 9/11 attacks was to elicit a US response that would kill Muslims leading to retaliations (Malvesti 2002). This sort of “jujitsu” strategy is purposely designed to elicit a harsher response than the precipitating attack, increasing grievances and strengthening any broad-based loyalty to the terrorist organization (McCauley 2006). In other words, repressive policies adopted by the United States and other governments might actually cause more terror than they avert. Thus, despite the appeal of deterrence-based counterterrorism strategies, the above argument suggests that when directed at terror, punishment will sometimes fail to deter and could even lead to more violence. This sort of backlash effect has been found in responses to other criminal behaviors and has been well explained by criminological theory. For example, labeling theorists argue that punished offenders will begin to identify more thoroughly with their role as lawbreakers solidifying their criminal (or terrorist) identities (Farrington 1977). Others have claimed that punishment can compromise the punisher’s perceived legitimacy and consequently elicit acts of defiance (Tyler 2006), which is consistent with McCauley’s (2006) argument that terrorists sometimes deliberately use a “jujitsu” strategy in order to sabotage a government’s legitimacy. Evidence of this sort of backlash effect from repressive actions has also been found in scholarly research. Several studies of the Republican terrorist campaign in Northern Ireland find that efforts by the British government to suppress terrorist violence increased rioting (Peroff and Hewitt 1980) and terrorism (LaFree et al. 2009). In fact, one Irish Republican Army member has been quoted as saying, “The British security forces are the best recruitin’ officer we have” (Geraghty 2000: 36). Elsewhere, analysis has shown that during the Iranian revolution, repression led to a long-run increase in the frequency of protests (Rasler 1996). Piazza and Walsh (2009) found that countries that violate human rights are

Counterterrorism

more likely to suffer terrorism than countries that adhere to human rights. Bartholomew (2011) found that laws that increased punishment for abortion-related terrorism failed to act as a deterrent, and in certain instances, they were associated with increases in property destruction and violence. Given the above evidence, one can argue that terrorism emerges as a response to perceived injustices, such as government repression, that inspire groups to mobilize in retaliation. In fact, even in cases where repression seems to work, it rarely comes without some costs. For example, Eppright (1997) found evidence that Israel’s 1996 incursion into Lebanon led to a significant drop in Hezbollah’s rocket attacks in Israel. Yet, Israel’s actions in Lebanon also increased dramatically the local and international support for Hezbollah, undoubtedly strengthening the terrorist organization’s propensity for future attacks. Despite these mixed findings, one point seems clear from the research conducted thus far. Most terrorist organizations are unlikely to be deterred by traditional sanctions, especially since they are often wholly willing to exchange their lives or their freedom to strike a blow against their enemies. In fact, of the eight reasons that Cronin (2006) gives for a terrorist group’s decline, only two are directly related to deterrence: the capture or killing of the leader and military force or repression. Empirically, Dugan and Yang (2011) show that the risk of continued terrorist attacks by Sendero Luminoso dropped substantially after the Peruvian government captured its leader, Guzman. While this result is promising, Cronin (2006) warns that the effectiveness of capturing one organization’s leader depends upon its structure, the charisma of the leader, and the presence of a viable successor. This warning can be better heeded when we consider the effects of Israel’s 1992 deportation of top Hamas leaders. Afterward, more radical mid-level leaders took over which eventually produced more deadly attacks against the Israelis (Hoffman and Cragin 2002). Capturing or killing a leader can also backfire if the group raises its captured or killed leader to the status of martyr, motivating further attacks (Cronin 2006). To date, we have

Counterterrorism

yet to determine the effects of the killing of Osama bin Laden on the strength of the global jihad. Unsurprising, rhetoric suggests that attacks will increase, yet without effective leadership, the threats could be innocuous. Despite the empirical findings described above, it is difficult to draw strong conclusions about deterrence without accounting for alternative approaches to countering terror. For example, LaFree et al.’s (2009) study of Northern Ireland selected only six actions by the British government from 1968 through 1992 to test their impact on Republican terrorism. The six actions were chosen because they were repeatedly cited as important in news accounts and scholarly research. Yet, by only selecting six actions over a 25-year period, the study ignored all less publicized – and perhaps less repressive – attempts by the British government that could have also influenced the rise and fall of Republican terrorism. Without data on the more subtle actions by governments, scholars may be getting biased results. Moreover, without including conciliatory as well as repressive government actions, we cannot assess the impact of alternatives to repression on reducing terrorist violence. Conciliatory-Based Counterterrorism Conciliatory-based counterterrorism is a logical alternative to deterrence-based strategies because it is similarly sourced in rational choice theory (see Dugan and Chenoweth 2012). Both strategies appeal to the rational actor, but instead of punishing lawbreaking behavior, conciliatorybased strategies reward legitimate alternatives to terrorism. Examples of earlier conciliatory efforts are found in Europe in the 1980s when Spain pardoned imprisoned ETA members after they publicly renounced the organization and its use of violence. The “reinsertion” policy allowed ETA members to live normal lives, free from the terrorist organization. Similarly, the Italian government offered leniency to members of the Red Brigades when they provided information that lead to the apprehension of other members (Cronin 2006). In both examples, governments strategically provided rewards for good behavior rather than punishing bad behavior. More

655

C

recently, deradicalization programs have been aiding convicted terrorists by engaging them in religious dialogue to dismantle the ideological beliefs that justify terrorism. Concurrently, the programs work closely with detainees’ families to prepare them to lead nonviolent lives by providing financial support to educate their children, train their wives, and help to reintegrate detainees into the community (Kruglanski et al. 2010). Despite the promise offered by the theory behind conciliatory-based counterterrorism, little is known about its effectiveness because little is known about the types of actions that governments have engaged in to reduce terrorism. In an effort to remedy this empirical gap, Chenoweth and Dugan (2011) have been developing a comprehensive database that documents government’s terrorist-related actions ranging from fully conciliatory to fully repressive in select countries. Along with my colleague Erica Chenoweth, I have collected the Government Actions in Terror Environments (GATE) data for the years 1988 through 2004 in Israel, Turkey, Algeria, Lebanon, and Egypt, and we plan to expand the collection to cover more countries over more years. Because this is the only effort to systematically capture a range of counterterrorism activity, the collection method will be briefly described below. In order to attempt to capture all government actions that are relevant to any terrorist conflict for each country, we relied heavily on open sources. Only through media reports it is possible to learn about the more subtle, yet relevant government activity in a terror environment. The data were collected using Textual Analysis by Augmented Replacement Instructions (TABARI), which searches news articles and identifies observations that match the criteria of an extensive set of dictionaries designed to capture international and domestic activity (Schrodt 2006). TABARI is an automated text-coding program that codes the first sentence of news articles based on verb and noun pattern recognition. TABARI was used to code hundreds of thousands of Reuters articles downloaded from Factiva using country-level search terms (243,448 stories for “Israel*,” 67,107 for “Leban*,” 52,575 for

C

C

656

“Algeria*,” 109,694 for “Egypt*,” and 152,998 for “Turk*”) as the sole search criterion for the period June 1, 1987, to December 31, 2004 (Chenoweth and Dugan 2011). After TABARI identified relevant news articles, the output (i.e., actor, verb, and target) was filtered to select only those actions that government actors implemented toward substate targets. All verbs within these criteria were kept as relevant in order to ensure that all unexpected actions would be captured and that a wide range of actions were included even if they may not immediately seem like counterterrorism (e.g., Israeli attempts to allow developers to build more efficient water wells in the Palestinian territories). After autocoding the verbs to fall on a conciliatory-repression scale, every lead sentence was examined for proper coding (Chenoweth and Dugan 2011). During this cleaning process, each government action was attributed to politicians, the military, the judiciary, or the police. The resulting file contains the article’s lead sentence, the actor, action, target, and new codes measuring level of conciliation and repression, whether the action was material and whether the target was discriminate (or specific) or indiscriminate (or general) for 6,063 Israeli government actions, 1,856 Turkish actions, 680 Algerian actions, 307 Lebanese actions, and 624 Egyptian actions for the years 1988 through 2004. This dataset gives an action-by-action view of state attempts to resolve conflicts with various non-state actors. Chenoweth and Dugan (2011) present figures that compare conciliatory and repressive trends of government actions with that of terrorist attacks from the GTD for each country. In general, the three trends seem to track one another. For example, the GATE data show a rise in repressive actions during the First Intifada in Israel that was later combined with conciliatory actions. Further, during the Second Intifada, the Israeli government acted with a record high frequency of repressive actions. In Turkey, it appears as if repressive actions increased corresponding to terrorist attacks during the height of the Kurdish insurgency, which accounts for most of the terrorist attacks and most of the

Counterterrorism

Turkish government’s actions. It also appears that conciliatory actions only began to rise in the mid- to late 1990s, at the termination of the PKK-led Kurdish insurgency. In Algeria, the vast majority of repressive actions seem to accompany the Algerian Civil War through the 1990s, with repressive policies declining near the end of that decade as (primarily Islamist) terrorist attacks increased. In Egypt, the primary terrorist activity occurred in the mid-1990s, when a variety of Islamist groups launched an internal war against Hosni Mubarak’s regime. These acts were accompanied by massive increases in repression, while conciliatory acts remained virtually nonexistent in Egypt. Finally, the patterns in Lebanon appear to be more mixed, which is likely due to the relatively low number of terrorist-relevant government actions compared to the other countries. Thus far, the GATE data have only been used for two evaluations. In their study of Israel, Dugan and Chenoweth (2012) found, using a monthly time series analysis, that conciliatory actions are associated with decreases in Palestinian terrorism and repressive actions are related to increases in such violence. These findings suggest that conciliatory-based counterterrorism might be promising and affirms that repressive strategies should be used with caution. Preliminary findings on Turkey suggest a similar pattern, although the number of conciliatory actions is so few that the findings related to those actions fail the tests of statistical significance. Situational-Based Counterterrorism When we turn our attention away from the motivation of the offender and just assume that terrorists will always want to attack, we are free to consider traditional crime control strategies that are based on limiting offending opportunities. Clarke and Newman (2006) present this strategy of controlling terrorism by suggesting that we adopt a situational approach to preventing terrorism – similar to that used to control more common crimes. By systematically analyzing the opportunities that terrorists exploit, they suggest that we can direct our efforts toward

Counterterrorism

developing reasonable means to block those opportunities. Of course, their strategy is adaptive because terrorist organizations will always exploit new avenues of attack. For example, Enders and Sandler (1993) and others (Dugan et al. 2005) show that after metal detectors were installed in US airports, the number of hijackings decreased. However, Enders and Sandler (1993) also demonstrated that metal detectors also led to increases in other kinds of hostage-taking tactics, suggesting that terrorists adapt to changes in situational factors. Jackson (2005) argues that terrorist organizations are innovative and that with enough resources will evolve beyond current barriers to their success. Thus, while situational-based counterterrorism is an important strategy, it should be accompanied by a better understanding of terrorist tactics that have been used in the past to overcome protective barriers (Clarke and Newman 2006). Regardless of the sophistication of the perpetrators, it is good policy to make it more difficult for terrorists to successfully attack vulnerable targets.

Conclusions For this entry, I explored three strategies to control terrorism: deterrence-based, conciliatorybased, and situational-based counterterrorism. While it seems that the most popular approach is to rely on deterrence strategies, it is unclear from the empirical research that this is the most effective strategy. In fact, most evidence suggests that deterrence perpetuates more violence rather than facilitating its ending. Unfortunately, evaluations of the other two approaches are scant, making any strong conclusions premature. Regardless, it is safe to say that the problem of terrorism is complex and that a broad range of tactical strategies should be considered before taking action. The literature on terrorist countermeasures is currently in its infancy. However, several developments show the potential to better understand the effects of a broad range of approaches to reduce terrorist violence. First, the availability

657

C

of incident-based terrorism data, such as that of the GTD, allows scholars to finally have an objective empirical dependent variable to use in evaluation studies. Second, more scholars are developing datasets to document a broad range of efforts by governments and others to reduce terrorist violence. As these datasets are completed and made available, I expect an increase in the number of evidence-based studies, providing policy-makers with the much-needed substantiation to inform their decision-making.

Related Entries ▶ Deterrence ▶ Ecoterrorism ▶ Homegrown Terrorism in the United States ▶ Incarcerating Terrorists ▶ Innovation and Crime Prevention ▶ Preventing of Terror at Shopping Malls ▶ Rational Choice Theory ▶ Rational Choice, Deterrence, and Crime: Sociological Contributions ▶ Social Disorganization and Terrorism ▶ Strategies of Policing Terrorism ▶ Terrorist Organizations

Recommended Reading and References Associated Press (2003) Bush: ‘Bring on’ attackers of U.S. troops. 2 July 2003 Bartholomew B (2011) The utility of deterrence-based sanctions in the prevalence of violence against abortion providers: testing a blended model of deterrence and backlash. Doctoral Dissertation, University of Maryland Chenoweth E, Dugan L (2011) Exploring counterterrorism in the Middle East and North Africa: a new data set. Paper presented at the American Political Science Association annual meeting, Sept 2011 Clarke RV, Newman GR (2006) Outsmarting the terrorists. Praeger Security International, Westport CNN (2001) Bush: Bin Laden wanted dead or alive, 17 Sept 2001 Cronin AK (2006) How al-Qaida ends. Int Secur 31:7–48 Dugan L, Chenoweth E (2012) Moving beyond deterrence: the effectiveness of raising the expected utility of abstaining from terrorism in Israel. Am Sociol Rev 77(4) (forthcoming)

C

C

658

Dugan L, Yang S-M (2011) Introducing group-based trajectory analysis and series hazard modeling: two innovative methods to systematically examine terrorism over time. In: Lum C, Kennedy L (eds) Evidence-based counterterrorism policy. Springer, New York Dugan L, LaFree G, Piquero A (2005) Testing a rational choice model of airline hijackings. Criminology 43:1031–1066 Enders W, Sandler T (1993) The effectiveness of antiterrorism policies: a vector-autoregressionintervention analysis. Am Polit Sci Rev 87(4):829–844 Eppright CT (1997) Counter terrorism and conventional military force: the relationship between political effect and utility. Stud Confl Terrorism 20:333–344 Farrington DP (1977) The effects of public labeling. Br J Criminol 17:112–125 Geraghty T (2000) The Irish war. Johns Hopkins University Press, Baltimore Hoffman B (2006) Inside terrorism. Columbia University Press, New York Hoffman B, Cragin K (2002) Four lessons from five countries. RAND Rev 26:42–43, Summer 2002 Jackson BA (2005) Aptitude for destruction, volume 1: organizational learning in terrorist groups and its implications for combating terrorism. RAND Corporation, Santa Monica Knowlton B (2001) Terror in American/‘We’re going to smoke them out’: president airs his anger. New York Times, 19 Sept 2001 Kruglanski AW, Gelfand M, Gunaratna R (2010) Detainee deradicalization: a challenge for psychological science. Observer 23:1 LaFree G, Dugan L, Korte R (2009) The impact of British counterterrorist strategies on political violence in northern Ireland: comparing deterrence and backlash models. Criminology 47:501–530 Lum C, Kennedy LW, Sherley AJ (2006) Are counterterrorism strategies effective? The results of the Campbell systematic review on counter-terrorism evaluation research. J Exp Criminol 2:489–516 Malvesti ML (2002) Bombing Bin Laden: assessing the effectiveness of air strikes as a counter-terrorism strategy. Fletcher Forum World Aff 26:17–29 McCauley C (2006) Jujitsu politics: terrorism and response to terrorism. In: Kimmel PR, Stout CE (eds) Collateral damage: the psychological consequences of America’s war on terrorism. Praeger, Westport, pp 45–65 Peroff K, Hewitt C (1980) Rioting in northern Ireland. J Confl Resolut 24:593–612 Piazza JA, Walsh JI (2009) Transnational terrorism and human rights. Int Stud Q 53:125–148 Rasler K (1996) Concessions, repression, and political protest in the Iranian revolution. Am Sociol Rev 61(1):132–152 Schmid AP, Jongman AJ (1988) Political terrorism: a new guide to actors, authors, concepts, databases, theories and literature. North-Holland, Amsterdam

Counterterrorism and the Media Schrodt PA (2006) Twenty years of the Kansas event data system project. Unpublished manuscript, University of Kansas Sherman LW, Gottfredson D, MacKenzie DL, Eck J, Reuter P, Bushway S (1997) Preventing crime: what works, what doesn’t, what’s promising: a report to the United States congress. National Institute of Justice, Washington, DC Stanik JT (2003) El Dorado Canyon: Reagan’s undeclared war on Qaddafi. U.S. Naval Institute Press, Annapolis Tyler TR (2006) Psychological perspectives on legitimacy and legitimation. Annu Rev Psychol 57:375–400

Counterterrorism and the Media Revital Sela-Shayovitz David Yellin Academic College and The Hebrew university, Jerusalem, Israel

Overview The role of the media is particularly highlighted during critical periods in which there are extreme levels of terrorism threats, since the media are the primary source for gaining a perspective of the situation. The importance of the media was keenly demonstrated during the terrorist attacks on September 11, 2001, when the American public heavily relied on news reports and followed them more closely than before. Indeed, CNN and FOX News were the major news networks to learn about the events (Castro 2006). Furthermore, for the police, cooperation with the media is essential in order to successfully deal with a terrorism crisis. Media reports of anti-terror policing may help police navigate events, reduce feelings of panic, and raise the sense of confidence in society. At the same time, during high threat levels of terrorism, the media stress the coverage of public trust in police and contribute to the legitimacy by underscoring the effectiveness of police counterterrorist efforts. However, as the terrorism threat subsides, counterterrorism coverage becomes more complicated and may harm police legitimacy.

Counterterrorism and the Media

Faced with these realities, what can police organizations do in order to improve their cooperation with the media in their counterterrorist efforts? Fortunately, there is much that can be done. Police organizations must take into account the media when engaging in anti-terror policing and may develop various strategies to effectively respond to a terrorist strike with the media’s help. In addition, police organizations can take advantage of the media in order to secure public assistance for carrying out antiterrorism policing.

Fundamentals of Police and the Media The media are highly prone to focus their attention on the police in news programs as well as in the realm of fiction (e.g., in crime drama television series). Indeed, police work provides cases that are newsworthy and involve drama and violent action. Consequently, public perceptions of crime are distorted and characterized by an overrepresentation of severe and violent crime, and an underrepresentation of other types of offenses. However, the media’s tendency to concentrate on the police also derives from the perception that the police represents rule-governed order and even symbolizes the state’s moral authority. Furthermore, the police force is the most visible organization of the criminal justice system, and therefore the onuses of dealing with crime and social control tend to fall disproportionately within their mandate (Ericson 1991; Chermak and Weiss 2005). While the news media provide a wide forum in which police work is covered, they also present a conflicting portrayal of police. In some reports, the police image is highly professional and even heroic, while in others policing is portrayed as ineffective, marked by internal conflicts and corruption. In addition, the police image is also affected by the fictional depiction of television police serials (e.g., CSI and NYPD Blue), which profile the police in an unrealistic manner (Dowler and Zawilsk 2007). Overall, it seems that the media tend to represent police work

659

C

as unsuccessful and incompetent rather than professional and effective (Fox and Van Sickel 2001; Gallagher et al. 2001). Another important aspect is related to the role of the media in reshaping the legitimacy of the police in the public sphere. Police legitimacy plays an essential role in public compliance and cooperation with the police. Accordingly, if people perceive the police as legitimate and place trust in them, they are more willing to obey the law and collaborate with the police (Sunshine and Tyler 2003). In this context, the media assume a key function in “policing the police” by exposing cases of police wrongdoing and mismanagement. As a result, they stimulate public discourse regarding procedural propriety, which urges police accountability, and this in turn contributes to legitimacy. Therefore, the coverage of these cases serves as a mechanism of regulation of police work and the compliance of its members. Moreover, the very existence of journalists as vigilant watchdogs over police forces on behalf of the public interest also contributes to the legitimization of the police (Ericson 1991; Reiner 1985). However, the media can sometimes become very critical and may in fact erode police legitimacy for a long period of time. Even a single high-profile event of police misconduct may dramatically undermine the public trust and seriously damage police prestige. Findings indicate that following well-publicized incidents of police misbehavior cases, there was an immediate and dramatic increase in public disapproval of the police, which persisted for a long period of time (Sunshine and Tyler 2003). Furthermore, the media’s exposure of internal conflicts and management problems in the police organization may damage the latter’s legitimacy. As a result, police spokespeople consistently struggle to control the police’s image in the public sphere. There is a continuous debate on the relationship between the police and media organizations. Police-media interaction seems to serve the needs of both types of organizations. The police are dependent upon the media in order to gain public cooperation, while the media need the police in order to obtain information on crime and

C

C

660

terrorism, which are interesting and newsworthy items (Chermak and Weiss 2005; Lovell 2001). However, the dynamic interaction between these organizations may be depicted as coexisting relationships that ebb and flow in terms of dominance and control. This is because the police and the media organizations have different agendas: the police engage in controlling information in order to decipher crime, while the media are devoted to providing the “sensational” details of crime and the exposure of police misbehavior. Nevertheless, police organizations have increasingly developed a growing acknowledgement of the need to improve their interaction with the media. For example, police organizations have established structures and allocated appropriate resources to this end, such as public information officers (PIOs) who have trained in media communication in order to advance the organization’s goals. In some countries, the police have adopted media education-training programs for police officers in order to improve their performance skills in the media. Furthermore, some police organizations have collaborated with the media in solving crime, e.g., by partaking in investigation television programs which promote police-media cooperation. Thus, although the interaction between the police and the media is sometimes tense, in general, both organizations report high levels of satisfaction with respect to the quality of their relationship (Chermak and Weiss 2005). Public Opinion About Policing and the Media There is scholarly consensus in respect to the influence of the media on the public views of the police. However, there is considerable debate about the direction, course, and degree of the effect. In this regard, three main hypotheses have been offered. The hypodermic needle hypothesis assumes that coverage has a powerful and long-term effect, since people mainly rely on the media for their sources of information. The limited effects approach suggests that people evaluate information from the media about the police on the basis of the context of what they already know from other sources, such as direct contact with the police.


Therefore, reports about the police must contend with their preexisting knowledge. The subtle/ minimal effects hypothesis maintains that the effect of the media is neither dominant nor minimal, but in fact is achieved differently. To elaborate, the media guide the perceptions of people regarding the most important issues involving crime and policing by highlighting particular topics and certain crimes. Hence, by framing police work in a certain way, the media reshape public perceptions and expectations about policing. Studies have found that there is evidence for each of the three approaches. The overwhelming majority of the public consider the media as their main source of information about police and crime (Fox and Van Sickel 2001; Gallagher et al. 2001). At the same time, it was found that the media’s influence is moderated by race and preconceived attitudes toward the police. People who have negative views about the police or belong to minority groups are more likely to consume the coverage of police misconduct cases, which in turn may reinforce negative perceptions and lead to generalize these cases as representative of police work. Thus, the impact of media coverage depends on people’s perceptions and their prior experience with the police (Weitzer and Tuch 2004). Media Framing Counterterrorism: International Perspectives Terrorist attacks offer a powerful narrative for the media while often providing significant ratings. News reports provide terrorist attacks with meaning by reconstructing them within a social and political context that can be understood in a simple way (Nacos 2002). However, not all terrorist events receive coverage, and two main characteristics were found to determine the likelihood of broadcasting a report: the death toll and the extent of destruction. Accordingly, terrorist events occasioning a high death toll or involving significant destruction will produce considerable amounts of news coverage (Chermak and Gruenewald 2006). Furthermore, the coverage of terrorist attacks is liable to be distinguished by several aspects: the report usually appears in


the headlines of the news broadcasts; the construction of newspaper headlines often becomes emotional and features dramatization; and news articles are longer and accompanied by further visual aspects, such as headline colors and photographs (Sela-Shayovitz 2007). While the framing of terrorism has been extensively studied, the coverage of police counterterrorism efforts has largely been neglected. The discussion on this topic can be divided into two main categories: the news framing of anti-terror policing under high levels of terrorist threat and the coverage of anti-terror policing as a routine activity. The news coverage of an extreme terrorist threat is often characterized by assigning a great priority to anti-terror policing. Journalists frequently tend to spotlight police performance in anti-terror events on the front pages of the newspapers rather than focus on the traditional roles of police work (Sela-Shayovitz forthcoming). Indeed, during times of a security threat, people are more concerned about the success of police performance in counterterrorist operations rather than take an interest in other aspects of police work (Tyler and Huo 2002). Thus, it appears that news editors are attentive to public concerns and, as a result, stress the police’s ability to combat the danger. A further central aspect is the role of the media in the rally ’round the flag effect. The rally effect relates to the impact of the security threat on the increase in public support for leaders and governmental institutions (Mueller 1973; Sigelman and Conover 1981). For example, following the September 11th terrorist attacks, the American public was united behind President George W. Bush, and his approval ratings soared to unprecedented levels. Furthermore, President Bush’s policies benefitted from high levels of support from all across the political spectrum at this time (Gaines 2002). Similarly, in research conducted in Israel, during a period of intense terrorism in the Second Intifada, there was an increase in the Jewish public’s positive evaluations of the Israeli police (Jonathan 2010). Several studies have demonstrated the key role of the media in the “rally effect.” For instance, the

661

C

CNN’s massive coverage of the Persian Gulf crisis in 1990–1991 substantially relied on official sources and reshaped public approval of US policy by providing justifications for a military response (Livingston 1994). Additionally, immediately following the 9/11 terrorist attacks, the ten largest American newspapers reflected Bush’s policy of “war on terror,” and editors refrained from reporting criticism about the practical consequences of military intervention (Ryan 2004; Ruigrok and Van Atteveldt 2007). In much the same manner, at these times, the media contribute to the “rally effect” and foster the legitimacy of the police in the public sphere. Research shows that during the period of an extreme terrorist threat in Israel, there was a significant increase in the favorable coverage of different aspects of police legitimacy, such as trust in police, police performance, interpersonal treatment, and the fairness of procedural processes. Journalists highly emphasized trust in police by frequently quoting statements of government ministers and citizens that reflected confidence in antiterrorism policing. In addition, the reporters refrained from engaging in criticism about police work. Yet, as the threat of terrorism declined, the media no longer remained positive and highly stressed their criticism of police (Sela-Shayovitz forthcoming). Another interesting issue is the media’s tendency to become patriotic when terror strikes. Immediately after the 9/11 terrorist events, the American media framing of the situation was repeatedly characterized by a network of associations among ideas, images, and symbols, such as war and homeland security, which stimulated popular support of the government (Eatman 2003). Furthermore, the American media tended to be more patriotic in comparison to the media in other countries that suffered from terrorism, such as Britain (Ruigrok and Atteveldt 2007). The coverage of anti-terror policing is also distinguished by the use of terms and symbols of war. During periods of severe terrorism threats, journalists often use phrases, symbols, and associations that foster patriotism and an affinity with the police forces in

C

C

662

counterterrorism, such as police’s struggle with terror or police combat with suicide bombers. In addition, the newspapers glorify the achievements of anti-terror policing, employing terms associated with heroism. For example, “police officers prevent terrorist attack with their own bodies” or “blues to those who wear the blue uniform” (Sela-Shayovitz forthcoming). Thus, the media tend to embrace the police forces by underscoring their efforts in combating terrorism. This manner of coverage policing may encourage public solidarity and cooperation with the police forces. However, it is also possible that it results from the tendency of reporters to mainly rely on police sources, which consequently draw attention to police successes. In this context, research indicates that during these times, press editorials tend to adopt the official point of view provided by the government (Dor 2004). Key Issues of Covering Anti-terror Policing Following the decrease in terrorism threat, the covering of anti-terror policing becomes much more complex, since it involves ideological and political aspects. Indeed, there is an ongoing debate regarding the implications of anti-terror activities on human rights (Tyler et al. 2010). Thus, the media may serve as a forum where different perspectives about counterterrorism policies are debated and contested in the public discourse. However, anti-terror strategies are less concerned with the infringements of human rights, whereas journalists are often devoted to the watchdog role of defending people’s rights (Chermak and Weiss 2005; Ericson 1991). Therefore, the report may weaken police legitimacy by exposing cases of human rights violations in counterterrorism. Following the 9/11 terrorist attacks, antiterrorism policing often became a justification for undermining civil liberties, and the exposure of these cases in the media led to public disapproval with counterterrorism (McCulloch and Pickering 2005). In a similar vein, heavily publicized cases of the abuse of human rights in Britain have considerably damaged police legitimacy (Loader and Mulcahy 2003; Reiner 2000).


It is important to note that the coverage of the violations of human rights in counterterrorism specifically causes harm to police relations with minority groups, since they are more likely to consume police misconduct cases in the media, which in turn reinforces their negative opinion of the police (Weitzer and Tuch 2004). Indeed, minority groups, such as Muslim communities, are often perceived as high-risk groups and become a focus of anti-terror policing due to their ethnic background. Consequently, their relations with the police are delicate and fragile (Tyler et al. 2010). On the other hand, the media can provide assistance for successfully dealing with terror both from an organizational aspect and in particular cases. From an organizational perspective, the media can help sustain public opinion regarding the necessity to persist in counterterrorism efforts, since this type of policing is an extension of police duties (Bayley and Weisburd 2009). Consequently, as the risk of terrorism subsides, it might be relegated to a lower priority because the public becomes once again more concerned with traditional policing (e.g., crime control and maintaining the social order). Therefore, the police can take advantage of the media’s power to promote public awareness in carrying out this mission. Furthermore, police organizations may use the media for lobbying to obtain additional resources for anti-terror activities. The media can help the police in specific cases in recruiting public cooperation and assistance. For instance, police organizations can use the media for issuing emergency warnings and other messages to the public, announcing clarifications about anti-terror operations (e.g., closing public areas for anti-terror drills), and promoting civil assistance in witnessing incidents, sharing suspicions, and providing information about terrorism.

Conclusions and Future Research During times in which there is a high threat of terrorist attack, police-media cooperation is crucial in order to deal effectively with the


situation and reduce panic. Additionally, media coverage of counterterrorism has a key role in police legitimacy and people’s readiness to obey police instructions. However, although the debate about crime and the media is long running, little scholarly attention has been paid to the coverage of police counterterrorist efforts. Recent findings indicate that under an intense terrorist threat, the media become favorable toward counterterrorism policing, and legitimacy significantly increases in the press (Sela-Shayovitz forthcoming). These findings highlight the importance of additional inquiries into the role of the media in profiling anti-terror policing. From a practical perspective, research also raises the question of whether police organizations can improve their collaboration with the media in regard to counterterrorism. In this context, several aspects are particularly noteworthy. Police organizations must take the media into account in the management of anti-terror policing, since police-media collaboration is crucial for gaining control over terrorist events. Moreover, the police should take advantage of a suspension in terrorism to train police officers about media performance in order to respond more effectively at the time of crisis. Second, it would be more beneficial to develop a framework of cooperation between the police and the media in order to formulate strategies of the management of emergency events and in handling sensitive information. In fact, several police organizations have developed models of cooperation with the media, such as the Scotland Yard and Israel (Castro 2006). Third, police forces may enhance their capacity for working with the media in counterterrorism efforts by engaging a specific spokesperson police officer who is a professional in this field. This is based on the fact that general police spokespersons cannot address the specific needs of this type of policing. Indeed, counterterrorism policing is a very complicated task, and police spokespersons who are proficient in their work are vital for cultivating public assistance and cooperation (Chermak and Weiss 2005; Lovell 2001).

663

C

Another important aspect is related to the necessity of fostering dialog with Muslim communities in respect to antiterrorism policing. In this context, police organizations can promote their relations with the local media and particularly those media outlets that are consumed by minority groups in order to enhance dialog and collaboration. However, due to the media’s scrutiny of the violation of human rights, the police must be well equipped with strategies to manage the threat and minimize damage to the public trust. Although anti-terror policing is less accountable than other types of traditional policing, the police can use a variety of strategies to deal with these cases. For example, police can issue public announcements about investigations the case in order to draw their own lessons and pacify critical voices. Furthermore, the police can generate a process of regulation concerning the violation of human rights, which in turn reinforces legitimacy. Finally, police organizations may use the recourse of dismissing policepersons who have been accused of the abuse of human rights. Thus, by engaging in several techniques under the watchful eye of the media, the police can preempt the likelihood of overall criticism and restore public confidence in the police. To conclude, future research should more closely examine the interaction between the police and the media in times of crises in order to develop strategies for effective cooperation during terrorist events. In addition, further research is required for a better understanding of the role of the media in shaping public perceptions about anti-terror policing. Most research in this field usually focused on one media outlet, whereas people from different social groups consume a diversity of media forms (national and local television, newspapers, radio, and the Internet). Thus, expanding the analysis to the impact of the various media genres on public opinion regarding counterterrorism may provide a deeper insight into this topic. It is also necessary to further examine the effect of the news coverage of police anti-terror activities and police legitimacy in order to enhance existing knowledge in this field.

C

C

664

Recommended Reading and References Bayley D, Weisburd D (2009) Cops and spooks: the role of the police in counterterrorism. In: Weisburd D, Feucht TE, Hakimi I, Felson ML, Perry S (eds) To protect and to serve: policing in the years of terrorism, and beyond. Springer, New York Castro CD (2006) The impact of police and media relations on a crisis. Naval Postgraduates School Monterey, California Chermak S, Gruenewald J (2006) Media coverage of police misconduct. Int J Police Strateg Manag 29:261–281 Chermak S, Weiss A (2005) Maintaining legitimacy using external communications strategies: an analysis of police-media relations. J Crim Justice 33:501–512 Chermak S, McGarrell E, Gruenewald J (2006) Media coverage of police misconduct and attitudes toward police. Polic Int J Police Strateg 29:261–281 Dor D (2004) Intifada hits the headlines: how the Israeli press misreported the outbreak of the second Palestinian uprising. Indiana University Press, Bloomington Dowler K, Zawilsk (2007) Public perceptions of police misconduct and discrimination: examining the impact of media consumption. J Crim Justice 35: 193–203 Eatman RM (2003) Cascading activation: contesting the White House’s frame after 9/11. Political Communication 20:415–432 Ericson RV (1991) Mass media, crime, law, and justice: an institutional approach. Br J Criminol 31:219–249 Fox RL, Van Sickel RW (2001) Tabloid justice: criminal justice in an age of media frenzy. Lynne Rienner Publishers, USA Gaines BJ (2002) Where’s the rally? Approval and trust of the president, cabinet, congress, and government since September 11. PS Polit Sci Polit 35:531–536 Gallagher C, Maguire E, Mastrofski S, Reisig M (2001) The public image of the police. International Association of Chiefs of Police, Alexandria Jonathan T (2010) Police involvement in counterterrorism and public attitudes toward the police in Israel: 1998–2007. Br J Criminol 50:748–771 Livingston S (1994) The terrorism spectacle. Westview Press, Boulder, CO Loader I, Mulcahy A (2003) Policing and the condition of England: memory, politics and culture. Oxford University Press, Oxford Lovell JS (2001) Police performances: media power and impression management in contemporary policing. Unpublished doctoral dissertation. The State University of Rutgers, Newark, NJ McCulloch J, Pickering S (2005) Suppressing the financing of terrorism: proliferating state crime, eroding censure and extending neo-colonialism. Br J Criminol 45:470–486 Mueller JE (1973) War, presidents and public opinion. Wiley, New York

Counterterrorism Strategies Nacos BL (2002) Mass-mediated terrorism: the central role of the media in terrorism and counter-terrorism. Rowman and Littlefield, Lanham Philo G, Berry M (2004) Bad news from Israel. Pluto Press, London Reiner R (1985) The politics of the police. Harvester, Brighton Reiner R (2000) Crime and control in Britain. Sociology 34:71–94 Ruigrok N, Van Atteveldt V (2007) Global angling with a local angle: how US, British and Dutch newspapers frame global and local terrorist attacks. Booksurge Publishers, Charleston Ryan M (2004) Framing the war against terrorism: US newspaper editorials and military action in Iraq. Gazette 66:363–382 Sela-Shayovitz R (2007) Female suicide bombers: Israeli newspaper reporting and the public construction. Crim Justice Stud 20:197–215 Sela-Shayovitz R (forthcoming) Police legitimacy under the spotlight: media coverage of police performance in the face of a high terrorism threat. J Exp Criminol Sigelman L, Conover PJ (1981) The dynamics of presidential support during international conflict situations. Polit Behav 3:303–318 Sunshine J, Tyler TR (2003) The role of procedural justice and legitimacy in shaping public support for policing. Law Soc Rev 37:513–548 Tyler TR, Huo YJ (2002) Trust in the law. Russell Sage, New York Tyler T, Schulhofer S, Huq AZ (2010) Legitimacy and deterrence effects in counterterrorism policing: a study of Muslim Americans. Law Soc Rev 44:365–401 Weitzer R, Tuch SA (2004) Race and perceptions of police misconduct. Soc Probl 51:305–325

Counterterrorism Strategies ▶ Strategies of Policing Terrorism

Courts ▶ Identification Technologies in Policing and Proof

Covert Policing ▶ Law of Undercover Policing

Crime and Justice on Television

Creativity ▶ Innovation and Crime Prevention

Credibility Assessment ▶ Evaluating Truthfulness: Interviewing and Credibility Assessment

Crime ▶ Situational Action Theory

Crime and Justice ▶ Victimization, Gender, and the Criminal Justice System

Crime and Justice on Television Nickie D. Phillips Department of Sociology and Criminal Justice, St. Francis College, Brooklyn, NY, USA

Overview The past two decades have seen the emergence of a variety of crime-based reality television shows that purport to give watchers an insider’s view of the criminal justice system. The convergence of private and public in a mediated context, particularly in the form of infotainment, is a consequence of late modernity. It is true that we are less able to separate our lives from the mediated reality in which we exist. As Mark Deuze (2011) points out, we are increasingly living life in media as opposed to with media. For example, it is becoming ever more

665

C

common for police to be active participants in the commodification of crime as entertainment. For crime-based reality television shows featuring police officers, law enforcement agencies maintain editorial control over the final product and, in the process, strive to present themselves in the best light. Unseemly police behavior has been edited from the shows before air date, leaving a rather sanitized version of police work. The merging of crime and entertainment on TV raises interesting questions. Among them is how “real” the depiction is of criminal justice and law enforcement in reality shows, with the acknowledgement that in a mass-mediated, latemodern society reality is a fluid concept. As the role of policing and media merge, questions arise as to what responsibility the police officers who appear in them, as well as the makers of the programming, have to produce a well-rounded and honest portrayal of the system. Generally, crime-based reality programming reinforces a “crime-control” ideology, which places its emphasis on reducing the crime in society through increased law enforcement, prosecutorial powers, and harsher sentences. The shows tend to portray law enforcement, and other components of the criminal justice system, as working as intended – successfully implementing justice for the good of local communities. However, at times, these shows depict due process as an obstacle and require the outside support of citizen-participants. Ultimately, however, the shows feature a narrative in support of the status quo, one that reinforces punitive attitudes while providing little, if any, critical analysis of the system itself or discourse conducive to questioning existing criminal justice policies.

Background Described on its website as “the rawest and the realest of all so-called reality shows,” COPS was a successful forerunner among crime-based reality programs. The show premiered during primetime on the Fox television network in

C

C

666

1989 and became one of their most successful and longest running programs. The COPS aesthetic approach, cine´ma ve´rite´ style camera work accompanied by a “Bad Boys” soundtrack, became an iconic representation of gritty, lifeon-the-line police work. COPS is notable for being the first crime-based reality television show in the United States “to use actual video footage as opposed to reenactments” (Doyle 1988, 96). The show is filmed in various cities around the country and features shaky, handheld camera footage of police officers responding to calls in progress and engaging in ordermaintenance policing. The viewer is on a virtual ride-along as the officers enter local communities to assist in such scenarios as spousal abuse, low-level drug busts, and alcoholinspired disorderly conduct. Crime-based reality television shows have proliferated since the late 1980s. America’s Most Wanted, a show that enlisted the help of its audience to capture fugitives, debuted in 1988 and aired for 23 years. Since then, entire networks and cable channels have devoted their air time to crime – and justice-related programming. These shows vary in format, from the “unscripted” to the news magazine-style formatting of Dateline. They also vary in focus, with some dedicated to highlighting only federal crimes, female offenders, high-profile or celebrity cases, or corrections. Launched in 1991, the cable channel Court TV was designed to provide live coverage of trials, largely criminal, with the intention of helping inform individuals about the operations of the courtroom. The extensive, round-the-clock coverage of high-profile court cases, with attention to irrelevant and often salacious personal details of the various participants involved, has led to a media-frenzied environment characterized as “tabloid justice” (Fox et al. 2007). More recently, Court TV has been supplanted by TruTV, a cable channel selfdescribed as “television’s destination for reallife stories told from an exciting and dramatic first-person perspective and features highstakes, action-packed originals.” The network features a slew of crime-based reality shows, including Police POV, hyped by the network


as a “groundbreaking” and “the most intense police series ever.” Discovery network’s Investigation Discovery Channel (known as ID) offers 24-h programming devoted to crime issues, featuring syndicated crime docudramas and original programming such as Deranged, Life of a Crime, and Most Evil. Further, about one-third of the shows now airing on the A&E network are crime-based reality shows such as Cold Case Files, The First 48, and Beyond Scared Straight. Understanding the ever-increasing popularity of these crime-based shows requires a broader look at the genre of reality television. Given the diversity in format and content, it is not surprising that scholars maintain that there is no singular, clear definition of the genre of reality television. In response, Nabi et al. offered the following characteristics of reality television shows (see also Baruh 2009; Nabi et al. 2003): (a) people portraying themselves (i.e., not actors or public figures performing roles), (b) filmed at least in part in their living or working environment rather than on a set, (c) without a script, (d) with events placed in a narrative context, (e) for the primary purpose of viewer entertainment. This definition excludes programs captured by other genres. (p. 304)

Although their definition excludes “news programming, talk shows, and documentaries, as well as programs featuring reenactments (e.g., America’s Most Wanted) and simple video clips not placed in a narrative context (e.g., America’s Funniest Home Videos)” (p. 304), other scholars hold a broader definition of reality programming. In their seminal work on crime-based reality television, Mark Fishman and Gray Cavender (1998) place reenactment shows that encourage audience participation squarely in the realm of reality television. Specifically, they distinguish two types of crime-based reality shows: reenactments and narratives encouraging audience participation (e.g., America’s Most Wanted) and ridealongs with officers (e.g., COPS). Whether scripted or unscripted, these shows share one defining feature: the blurring of fact and fiction in an entertainment context.


The “Infotainment” Factor Regardless of their specific content or format, the proliferation of crime-based reality shows is a testament to the prominence of “infotainment,” a blending of “informative” programming (often news-like, with the purported intent of informing viewers of important social issues) with entertainment. Ib Bondebjerg points out that reality television shows elevate “private life stories” into “public discourse” while blurring the boundaries between fact and fiction (Bondebjerg 1996, p. 28). Bondebjerg does not necessarily find this hybridization troubling as long as the conditions of communication are clear. For example, the title cards at the beginning of the show could clearly delineate that the stories are based on true accounts but that the scenes consist of reenactments. For Bondebjerg, our everyday experience is being merged with more traditional forms of public service information into a hybrid form that at once informs and entertains the viewer. Shows fall on a continuum with regard to the amount of useful information given to inform the viewer versus dramatization designed to sensationalize. Scholars have generally been critical of this blurring of boundaries, pointing out that although television and other new media may afford us access to much more information – particularly with regard to the operations of the criminal justice system – we are not necessarily more knowledgeable about how the system actually operates. Given that a functioning democracy depends on an informed citizenry, some scholars have pointed out that entertainment disguised as information is not just misleading but rather poses a potential danger to democracy (Gamson et al. 1992; Postman 2005). Crime-based reality television has been found to incorporate stylistic elements that blur the distinction between fact and fiction, with the intention of making the show seems more realistic. For example, some shows such as America’s Most Wanted feature logos and set designs that are similar to official law enforcement regalia and often feature hosts that seem to have credibility either through their celebrity or personal story of victimization (Cavender 1998). Some shows rely

667

C

on official police statistics, intimating that a close relationship exists between law enforcement and the hosts. (Cavender 1998) also notes that shows such as America’s Most Wanted and Unsolved Mysteries usher in a sense of urgency by encouraging viewers to “call in now,” which also gives the appearance of the shows being broadcast live even though they were prerecorded. In shows such as COPS, the action appears to be occurring in real time, even though the raw footage is heavily edited; sounds are blended to give the illusion that time is continuous (Doyle 1988). While reality television merges fact and fiction, viewers are keenly aware that the shows do not portray an unadulterated “reality.” Researchers have found that the public tends to view the reality claims made by such programs with a skeptical eye (Hall 2006; Nabi et al. 2003). Viewers recognize that although unscripted, the producers often manipulate circumstances with the intention of increasing dramatic effect. Police officers and other agents of the criminal justice system, like all participants in reality television shows, alter their behavior in the presence of cameras. In fact, in her study of viewer perceptions of reality programs, Alice Hall (2006) notes that realism is a fluid concept and that viewers acknowledge that the presence of the camera affects the behavior of participants, that circumstances are frequently staged, and that the final product has endured selective editing. Perhaps most significant, Hall found that while the viewers remarked that most situations were probably staged, they nevertheless contended the reactions of the participants were “real” and, as such, a reflection of their true nature and personality. Overall, though, scholars suggest that the more realistic the program is believed to be, the greater the effect on the viewer’s attitudes and beliefs and that viewers were significantly more likely to characterize crime-based reality shows as more realistic than fictional programming (Oliver and Armstrong 1998).

Crime-Control Ideology Scholars have suggested that the emergence and popularity of these shows are reflective of the

C

C

668

political climate that arose during the 1970s and continued into the following decades. This toughon-crime era is characterized by an escalating war on drugs, increased mandatory prison sentences, a shift away from rehabilitation, and the implementation of quality-of-life policing focused on reducing social and physical disorder. Crimebased reality television shows reinforce the status quo through support of crime-control policies and ultimately work to sustain crime myths. Further, like most media coverage of crime, the shows tend to overrepresent violent crime and are focused heavily on lower-class street crime to the exclusion of corporate crime, white-collar crime, and crimes committed by the state. For example, Oliver (1994) conducted a content analysis of 76 crimebased reality programs, including COPS, America’s Most Wanted, Top Cops, FBI, The Untold Story, and American Detective, and found that, consistent with prior research, the shows overrepresented violent crimes, along with an unusually high rate of successful resolution. Shows such as Dallas/Kansas City/Detroit SWAT and DEA feature paramilitary police units that are engaged in high-risk assignments such as enforcing the war on drugs, responding to barricade situations, and apprehending dangerous suspects. The SWAT franchise goes to great lengths to humanize the officers (most of whom are male), featuring on-screen mentions of their home life, including wives/girlfriends and children, and a website containing a feature where viewers can “meet the SWAT team.” The suspects, however, are portrayed as interchangeable and anonymous one-dimensional characters. There is little to no consideration given to the (continuing) failure of the war on drugs. Crimebased reality shows have long been criticized for presenting crime in a decontextualized manner, with a reliance on use of force as the go-to option for resolving conflicts (Doyle 1988; Grant 1992). By presenting crime solely from the perspective of law enforcement officers, there is little room for a critical assessment of the system and its operation. Crime-based reality television shows provide a neatly packaged narrative wherein good versus evil is on stark display, and law enforcement is on the side of the angels. That


police officers serve as the thin blue line protecting citizens and ultimately restoring social order is a compelling narrative for viewers. In her study examining the ways in which crime-based reality television shows contribute to our nation’s mythology surrounding crime control and the administration of justice, Jessica Fishman (1999) notes that the genre is not ideologically monolithic. While the worldview presented is one of dangerousness and where retributive justice prevails, the shows differ in the role of the state and civilian participants in the quest to eradicate crime. For example, in her study of COPS and America’s Most Wanted, Fishman finds that COPS portrays state agents as heroic social protectors. The restoration of social order is reserved for police officers with special knowledge, skills, and resources to tackle the crime problem. From this perspective, civilians are ill-equipped and inadequately prepared to deal with crime. Conversely, America’s Most Wanted provides viewers with a worldview in which law enforcement has initially failed in its mission to protect society and, therefore, citizens are empowered to actively participate in crime control. For America’s Most Wanted, the most important catalyst in fighting crime is the alert citizen (Fishman 1999). Some researchers have found that viewers who exhibit personality traits consistent with authoritarianism report greater enjoyment of reality programs, including crime-based shows (Nabi et al. 2003, p. 326). For example, Oliver and Armstrong (1995) surveyed viewers of crimerelated programs and tested the concept of “disposition theory,” the idea that “viewer enjoyment is strongest when liked characters are portrayed as winning and disliked characters portrayed as losing” (1995, p. 561). The researchers found that when compared to fictional programs, the crimebased reality programs were “most enjoyed by viewers who evidenced higher levels of authoritarianism, reported greater punitiveness about crime, and reported higher levels of racial prejudice” (1995, p. 565). Researchers have also found that attitudes toward police may be impacted based on consumption of crime-based reality television


shows, although the effect differs by race. For example, white viewers of crime-based reality shows reported higher levels of confidence in police compared to African-American viewers who demonstrated no increase in confidence (Escholz et al. 2002). These findings demonstrate that viewers are not simply passive recipients of manufactured ideological information. Rather, they actively interpret images of crime and justice, and as such, meanings surrounding criminal behavior and paths to justice are constantly negotiated within a cultural context. Some have postulated that reality television, particularly crime-related programming, may be attractive to viewers because of its voyeuristic appeal. Researchers investigating the idea, however, have found mixed results. In seeking to assess the validity of the claim, the researchers first made the distinction between the clinical definition of voyeurism, which contains pathological, psychosexual elements of viewing unknowing victims, and the more common usage of voyeurism to mean watching other persons as they live their lives (Nabi 2003). While researchers have found some evidence indicating a voyeuristic gratification in the more common usage related to watching reality-based television shows, this is not necessarily the most important motivator for the viewers. In fact, Nabi, Stitt, Halford, and Finnerty suggest that for crime-based reality television, viewers may be less likely to tune in for the purposes of voyeurism and are more likely interested in learning about “criminal behavior and the criminal justice system” (2006, p. 442). For crime shows such as COPS, the structure of the programming lends itself to what Baruh calls a “fly on the wall” approach in which camera positioning places the viewer in the position of a detached onlooker. This technique contributes to the idea that the viewer is bearing witness to a gritty reality that is often dangerous and always enveloped in larger notions of justice.

Race and Gender Stereotypes Crime-based reality television shows have frequently been criticized for perpetuating racial

669

C

and gender stereotypes. In COPS, for example, researchers found that the type of offense presented varied by the race of the offender: Whites were found to be more likely portrayed in offenses related to “cars and alcohol,” while African-Americans were more likely to be shown committing burglary and grand theft (MonkTurner et al. 2007). Researchers also found, after analyzing 8 h of episodes, that aside from the appearance of one female officer, all the police officers portrayed on COPS were white males. There was a conspicuous absence, on any episodes analyzed by the researchers, of any female minorities appearing as either police officers or offenders. Researchers have also found that while minorities remain underrepresented as cops, they are not necessarily overrepresented as suspects. However, when aggression was used by officers, minorities were “significantly more likely than white criminal suspects to suffer from unarmed physical aggression from police officers” (Oliver 1994, p. 9). In their analysis of COPS and World’s Wildest Police Videos, Prosise and Johnson (2004) suggest that the shows serve as a means to legitimize and justify questionable police behavior such as racial profiling. For example, the authors point out that police officers often rely on hunches, or “officer intuition,” to justify police stops. These hunches are portrayed as an effective means of making a good arrest and implicating a guilty suspect. Hunches that prove to result in bad stops or arrests are likely to end up on the editing room floor. Female police officers are more likely to be featured in their own spin-off shows, such as Female Forces and Police Women of Broward County, rather than fully incorporated into the more traditional, male-centric shows. RabeHemp (2011) examined the ways in which gender was negotiated on the show Female Forces, airing on Biography network. The show, described on the network website as “Brains, beauty, and a badge. . .,” follows female police officers in Naperville, Illinois, “as they fight a full gamut of big city crime in the suburbs of Chicago.” In contrast to other, male-dominated crime reality television shows, Female Forces was more likely

C

C

670

to show vignettes featuring the officers engaged in order-maintenance style policing as opposed to responding to calls regarding violent crimes. While this portrayal more closely aligns with the “reality” of crime (i.e., according to official reports, there are more calls for service for noncriminal and/or order-maintenance incidents than for violent crimes), it illuminates the extent to which the male-dominated shows tend toward a macho, aggressive style of policing focused heavily on crime fighting. In her analysis, Rabe-Hemp explores how stereotypes of female police officers are both reinforced and challenged. By virtue of portraying women in an authoritative position, the patriarchal structure that privileges males as the sole figures of authority is challenged. At times though, this resistance was negotiated through emphasized femininity. For example, Rabe-Hemp describes how the female officers insisted that it was possible to remain feminine and still be a competent police officer, all the while engaging in personal grooming such as hair brushing and applying makeup (RabeHemp 2011).

Current Issues and Controversies Future Directions Compared to the courts and corrections, generally, most media coverage of the criminal justice system is focused on law enforcement. The relative lack of attention to corrections is likely due to the lack of willingness on the part of correctional officials to open up the facilities to the media (Surette 2011). Surette (2011) noted that generally, prisons are presented in the media rather negatively, focusing on prison riots and misconduct by correctional officers, in portrayals that are unlikely to garner public support for the institutions. Perhaps not surprisingly, most research on crime-based reality television programs also centers on the front end of the criminal justice system, including the early, highly successful shows such as law enforcement-focused COPS and America’s Most Wanted. Other reality television shows that feature corrections, such as Lock Up


or Inside American Jails, are often neglected by researchers. Future research may examine the ways in which prison-focused reality shows feature inmates as spectacle and doomed to recidivate. Researchers may explore whether the shows provide a critical discourse on the failures of the system or engage in promoting a view of criminal behavior that is reduced to a simple issue of individual responsibility, while neglecting the insurmountable obstacles that most offenders face upon reentry. The Role of the Media and the Role of Law Enforcement Other future issues for criminologists to explore may center less on the extent to which shows are misrepresenting “reality” and more on acknowledging that criminal justice is increasingly a lived experience on-camera. It is in this context that the distinction between the role of the media and the role of law enforcement has become increasingly muddled. With the proliferation of crime-based reality shows, members of police departments are taking on a more proactive role as media, while members of the media are serving as pseudo-law enforcement agents. For example, police officers who have participated in COPS have reported to researchers that they felt that the television crew was on their side, even ready to jump into a fight if necessary (Hallett and Powell 1995) [emphasis added]. The consequences of the fusion between media and law enforcement are no more apparent than in the top-rated show To Catch a Predator. As scholars have noted, in that show, the delineation between civilian vigilante group, law enforcement, and the media has become dangerously blurred. The show featured Dateline NBC correspondent Chris Hansen as he confronted alleged sexual predators who have made contact with underage participants in stings set up by the program in cooperation with local law enforcement agencies. The alleged predators were lured to the decoy house and subject to Hanson’s interrogations. Law enforcement agencies partnered with a civilian group known as Perverted Justice, whose members were dedicated to stamping out pedophilia by posing online as underage victims


and exposing the adults who engage in explicit communication with the decoys. The program’s popularity stems in part from its reliance on shame and humiliation as public spectacle that has become a staple of reality television (Kohm 2009). The show found itself the recipient of unwanted media attention as a result of the suicide of Bill Conradt, an assistant district attorney, caught exchanging explicit internet chats with a presumed underage decoy in one of the show’s sting operations. Conradt shot himself as the SWAT team and camera crews closed in on his home. Investigations into the botched sting targeting Conradt revealed the extent to which media directs and influences law enforcement activity. Though Conradt had, for whatever reason, declined to actually show up at the decoy house, Hansen pressed further, requesting the police obtain arrest and search warrants to be executed at Contradt’s home the next morning. From the hastily written arrest and search warrants that were riddled with errors to the involvement of a SWAT team – likely influenced by the show’s usage of dramatic takedowns in situations in which less a show of force would suffice – it becomes apparent that concern for spectacle was among the primary motivations for the actions of law enforcement. In fact, when determining whether to press charges on the other 24 alleged perpetrators captured in the sting, the prosecutor was forced to drop the cases citing issues of illegality in police procedure and venue problems. A lawsuit filed by the show’s own producer alleged unethical relationships between the show, Perverted Justice, and law enforcement. The suit claimed that the Perverted Justice group did not always supply the producers with full transcripts of the chats and that the show covered up questionable police behavior. The producer alleged she was fired for speaking out about the unethical practices of the show. Although her lawsuit was eventually dismissed, a second wrongful death lawsuit filed by Conradt’s sister was settled for an undisclosed amount. The suit alleged that “the network interfered with police duties and then failed to protect her brother’s safety.” When

671

C

ruling that the wrongful death case could move forward, Judge Denny Chin pointed out “that NBC had ‘placed itself squarely in the middle of a police operation, pushing the police to engage in tactics that were unnecessary and unwise, solely to generate more dramatic footage for a television show.’” Future research could continue to document the ways in which crime and criminal justice is “dangerously confounded with its own representation” (Ferrell et al. 2008). Shame and Humiliation One of the hallmarks of crime-based reality television shows is the extent to which suspects are shamed and humiliated on-screen. Future research may continue to explore the ways in which moral sentiments are mass-mediated and the public spectacle of punishment has been revived in the context of reality television. Crime-based reality shows have managed to merge crime and entertainment into “humiliation TV,” often airing segments with no other end than to ridicule the offender. Shows that compile caught-on-tape moments such as Disorder in the Court and World’s Dumbest. . ., highlighting the misadventures of criminals and other ne’er do wells, are prime examples. For some, such as those implicated in To Catch a Predator, the consequences of shame and humiliation may prove deadly. In her analysis of COPS, Mariana Valverde (2006) points out that the show portrays police officers as engaged against the enemy “others” who represent stereotyped “criminal classes.” The social context from which crime emerges (i.e., larger structural constraints of poverty, race, class, and gender) is neglected in favor of a focus on the apprehension of lower-class criminals, frequently intoxicated, and virtually always surrounded by social and familial disorder. Humiliation and shame are part and parcel of the show as the camera focuses an inordinate amount of time on the incoherent speech and movements of those who are intoxicated for the viewers’ pleasure. Many subsequent shows such as Dallas/ Kansas City/Detroit SWAT, DEA, and Police Women of Broward County have replicated COPS cinema verite´ style along with other conventions and are subject to similar criticisms.

C

C

672

Increasingly, shame and humiliation have extended to reality shows that focus on ethical dilemmas that at times may overlap with criminal behavior. Shows such as Intervention, Celebrity Rehab, and Sober House focus on the shame and humiliation of addiction yet decouple drug use and addiction from criminal behavior, putting on full display our cultural ambivalence surrounding drug use. Cultural criminologists have suggested that part of the pleasure of viewing reality television is that it provides a means of privately transgressing. In a vicarious manner, viewers are relating to and experiencing deviance in a socially sanctioned way – they are offered a means of reaffirming social boundaries and for condemning deviance through the humiliation and shame of the other. In fact, Hall (2006, p. 204) found that the viewers’ enjoyment of reality television shows is, in part, “a strong element of schadenfreude.” Cultural criminologist Mike Presdee suggests that watching “real” people being humiliated calls up a “different order of emotions, pleasures and desires” than that of fictional humiliation (Presdee 2001, p. 80). However, given that there is such fluidity between viewers’ perceptions of what constitutes “reality” television from fictional television, it is likely that those emotions, pleasures, and desires are not so distinct. Reality shows then are part of a larger social context in which crime, shame, humiliation, and violence are sold as commodities to a willing public eager to be entertained.

Related Entries ▶ Cultural Criminology

Recommended Reading and References Baruh L (2009) Publicized intimacies on reality television: an analysis of voyeuristic content and its contribution to the appeal of reality programming. J Broadcast Electron 53(2):190–210. doi:10.1080/ 08838150902907678, doi: papers://227A529C-089D4314-B924-24AB50571CE2/Paper/p512

Crime and Justice on Television Bondebjerg I (1996) Public discourse/private fascination: hybridization in ‘true-life-story’ genres. Media Cult Soc 18(1):27–45. doi:10.1177/016344396018001003 Cavender G (1998) The shadow of shadows: television reality crime programming. In: Fishman M, Cavender G (eds) Entertaining crime: television reality programs. Aldine De Gruyter, New York, pp 79–94 Deuze M (2011) Media life. Media Cult Soc 33(1):137–148. doi:10.1177/0163443710386518 Donovan P (1998) Armed with the power of television: reality crime programming and the reconstruction of law and order in the United States. In: Fishman M, Cavender G (eds) Entertaining crime: television reality programs. Aldine De Gruyter, New York, pp 117–137 Doyle A (1988) Cops: television policing as policing reality. In: Fishmann M, Cavender G (eds) Entertaining crime: television reality programs. Aldine De Gruyter, New York Eschholz S, Blackwell B, Gertz M, Chiricos T (2002) Race and attitudes toward the police: assessing the effects of watching. J Crim Just 30:327–341, (12355527252820796286related:fqMK9ZKmd6sJ), doi: papers://227A529C-089D-4314-B924-24AB505 71CE2/Paper/p446 Ferrell J, Hayward K, Young J (2008) Cultural criminology: an invitation. Sage, Thousand Oaks Fishman J (1999) The populace and the police: models of social control in reality-based crime television. Crit Stud Mass Commun 16:268–288 Fishman M, Cavender G (1998) Entertaining crime: television reality programs. Aldine Transaction, New York Fox R, Van Sickel R, Steiger T (2007) Tabloid justice: criminal justice in an age of media frenzy. Lynne Rienner, Boulder Gamson WA, Croteau D, Hoynes W, Sasson T (1992) Media images and the social construction of reality. Annu Rev Sociol 18:373–393 Grant J (1992) Prime time crime: television portrayals of law enforcement. J Am Cult 15(1):57–68 Hall A (2006) Viewers’ perceptions of reality programs. Commun Q 54:191–211 Hallett M, Powell D (1995) Backstage with “Cops”: the dramaturgical reification of police subculture in American crime “info-tainment”. Am J Police 14(1):101–129 Jewkes Y (2010) Media & Crime. Sage, Thousand Oaks Kohm S (2009) Naming, shaming and criminal justice: mass-mediated humiliation as entertainment and punishment. Crime Media Cult 5:188, (related: woyeB3m3gTsJ), doi: papers://227A529C-089D4314-B924-24AB50571CE2/Paper/p80 Kooistra P, Mahoney J, Westervelt S (1998) The world of crime according to “Cops”. In: Fishman J, Cavender G (eds) Entertaining crime: television reality programs. Aldine De Gruyter, Hawthorne, pp 141–158 Kort-Butler L, Hartshorn K (2011) Watching the detectives: crime programming, fear of crime, and attitudes about the criminal justice system. Sociol Q 52:36–55

Crime and the Racial Composition of Communities Monk-Turner E, Martinez H, Holbrook J, Harvey N (2007) Are reality TV crime shows continuing to perpetuate crime myths. I J Criminol 6:132 Nabi RL, Biely EN, Morgan SJ, Stitt CR (2003) Realitybased television programming and the psychology of its appeal. Media Psychol 5(4):303–330 Nabi RL, Stitt CR, Halford J, Finnerty KL (2006) Emotional and cognitive predictors of the enjoyment of reality-based and fictional television programming: an elaboration of the uses and gratifications perspective. Media Psychol 8(4):421–447 Oliver M (1994) Portrayals of crime, race, and aggression in “reality-based” police shows: a content analysis. J Broadcast Electron 38(2):179–192 Oliver M, Armstrong G (1995) Predictors of viewing and enjoyment of reality-based and fictional crime shows. Journal Mass Commun Q 72(3):559–569, doi: papers://227A529C-089D-4314-B924-24AB50571CE 2/Paper/p414 Oliver M, Armstrong G (1998) The color of crime: perceptions of Caucasians‘ and African-Americans’ involvement in crime. In: Fishman M, Cavender G (eds) Entertaining crime: television reality programs. Aldine De Gruyter, New York, pp 19–36 Postman N (2005) Amusing ourselves to death: public discourse in the age of show business. Penguin (NonClassics), USA Presdee M (2001) Cultural criminology and the carnival of crime. Routledge, USA Prosise T, Johnson A (2004) Law enforcement and crime on cops and world’s wildest police videos: anecdotal form and the justification of racial profiling. West J Commun 68(1):72–91 Rabe-Hemp CE (2011) Female forces: beauty, brains, and a badge. Fem Criminol 6(2):132–155. doi:10.1177/ 1557085111398471 Surette R (2011) Media, crime and criminal justice: images, realities and policies, 4th edn. Wadsworth, Belmont Valverde M (2006) Law and order: images, meanings, myths. Rutgers University Press, New Brunswick

Crime and the Racial Composition of Communities Charis E. Kubrin Criminology, Law and Society, University of California, Irvine, CA, USA

Overview One of the most significant social facts about crime is that it is not randomly distributed across neighborhoods within a city. Rather, crime tends to

673

C

cluster in certain areas or neighborhoods. This is why most residents are able to identify the “bad section” of town. For over a century, scholars have attempted to understand the spatial patterning of crime. A common approach involves investigating the characteristics of neighborhoods with high crime rates. One frequently examined neighborhood characteristic is racial composition. Scholars ask, how is the racial and ethnic composition of a neighborhood related to its crime rate? This entry reviews the literature that attempts to address this question, the theoretical approaches that hypothesize a relationship between the two, and some challenges and obstacles associated with studying neighborhood racial composition and crime. In the concluding section, the entry offers some important future directions researchers can take to better understand crime and the racial composition of communities.

Key Findings in the Literature A consistent finding in the neighborhoods and crime literature is that racial composition matters. Although racial composition is a broad construct that can reflect a variety of things, with few exceptions, studies published to date focus primarily on one of two measures to capture community racial composition: racial heterogeneity and percent black. As far back as the early 1900s, researchers investigated the effects of racial and ethnic heterogeneity within communities (Shaw and McKay (1969 [1942]; Sellin 1938). In their classic work Juvenile Delinquency and Urban Areas, Shaw and McKay (1969 [1942]) argued that three structural factors – ethnic heterogeneity among them – led to the disruption of local community social organization, which in turn accounted for variations in rates of crime and delinquency. Nearly 100 years later, studies that analyze racial heterogeneity and crime continue to document a strong positive relationship between criminal violence and an area’s level of racial diversity, although it is fair to say the findings reveal a more complex relationship than previously assumed.

C

C

674

For example, using victimization data from 57 neighborhoods to examine the relationship between neighborhood characteristics, including racial heterogeneity, and rates of violent crime and burglary, Smith and Jarjoura (1988) show that poverty and ethnic heterogeneity, among other factors, predict neighborhood victimization rates, although their influence is more conditional than direct and varies by type of crime. Similar findings are reported by Warner and Pierce (1993) using data on calls to the police in 60 Boston neighborhoods. Warner and Pierce find that although racial heterogeneity has a positive effect on burglary rates, its effects are conditional on the level of poverty. Further evidence of interaction effects between racial heterogeneity and poverty are documented in both Warner and Rountree (1997) and Kubrin (2000) who examine neighborhoods in Seattle, Washington. Warner and Rountree (1997) show that heterogeneity has a significant direct effect on assault when poverty is at its mean, but its effect is greatest in nonpoor neighborhoods and is less strong in high-poverty neighborhoods. Likewise, Kubrin (2000) finds that in neighborhoods characterized by low poverty levels, racial heterogeneity increases violent crime, whereas in high-poverty neighborhoods, heterogeneity actually decreases crime. Far more studies have operationalized racial composition in a more limited way – as reflective of the percentage black in a community. As with racial heterogeneity, findings from this literature indicate that this measure of racial composition is associated with crime rates across communities. In their recent review of the neighborhood literature on race and crime, Peterson and Krivo (2010) conclude, “Rates of crime are higher in local areas where African Americans are more heavily concentrated and, conversely, lower where African Americans are constitute a smaller portion of the residents, whether in Atlanta, Baton Rouge, Chicago, Cleveland, Columbus, New York, Philadelphia, or Seattle” (p. 14; see this discussion for a list of relevant studies in this area). Peterson and Krivo note that often the differentials in crime across communities of different racial composition are

Crime and the Racial Composition of Communities

quite stark. For example, McNulty’s (2001) study of block groups in Atlanta reveals that murder rates were a startling six times higher in predominantly black neighborhoods than in predominantly white neighborhoods, while rape and robbery rates in African American areas were over three times those in white areas and aggravated assaults were about twice as common in the former than the latter. As this (admittedly brief and incomplete) overview of studies on crime and the racial composition of communities reveals, race, place, and crime are inextricably linked in the United States. What accounts for this linkage and what mechanisms underlie these relationships? In the remainder of the entry, I discuss three theoretical approaches – social disorganization, labor-market stratification, and cultural/subcultural explanations – that attempt to account for the racial composition-crime connection across neighborhoods.

Social Disorganization Theory Social disorganization theory has emerged as the critical framework for understanding the relationship between community characteristics and crime in urban areas. According to the theory, certain neighborhood factors can lead to social disorganization, defined as the inability of a community to realize the common values of its residents and maintain effective social control (Kornhauser 1978, p. 120). Communities with ineffective social control have higher crime rates. Social control serves two purposes: to ensure residents’ conformity to norms and to regulate deviant behavior. According to the theory, two types of social control – formal and informal – operate to achieve these ends. Formal control is exercised by institutions such as the police to maintain stability in the neighborhood. Informal social control is exercised by families, peers, and neighbors who “look out” for the community by discouraging deviant behavior. Informal social control, at the heart of the theory, is said to provide the first line of security for community members.


The dynamics of informal social control are discussed by Greenberg and her colleagues (1985) who identify three ways it may impact crime: (1) via informal surveillance, or the casual but active observation of neighborhood streets that is engaged in by residents during daily activities; (2) via movement governing rules, or the avoidance of areas in or near the neighborhood (or in the city as a whole) that are viewed as unsafe; and (3) via direct intervention, or through the questioning of strangers and neighborhood residents about suspicious activities. Direct intervention may also include chastising adults and admonishing children for behavior defined as unacceptable. From its inception, social disorganization theory has considered a community’s racial composition as among the most salient predictors of neighborhood crime rates. In particular, arguments center on the community’s level of racial diversity among residents. The theory proposes that racially heterogeneous communities have higher crime rates in large part due to diminished informal social control. It is argued that in communities with diverse racial groups living side by side, interaction between members will be low, or at least lower than in neighborhoods with individuals from the same racial background. It is also assumed that racial heterogeneity can undermine ties between neighbors, limiting their ability to agree on a common set of values or to solve commonly experienced problems. Reasons for both arguments point to cultural differences between racial groups, language incompatibility, and the fact that individuals prefer members of their own race to members of different races. As noted earlier, studies that explicitly examine racial heterogeneity’s impact consistently find it is significantly and positively associated with neighborhood crime rates (Kubrin 2000; Smith and Jarjoura 1988; Warner and Pierce 1993; Warner and Rountree 1997).

Labor-Market Stratification Theories A second theoretical perspective that attempts to account for the racial composition-crime

675

C

connection across neighborhoods focuses on labor-market characteristics in neighborhoods with significant numbers of racial minorities. Researchers ask. What are the key labor-market characteristics associated with high rates of violence that help to explain high crime rates in minority neighborhoods? Theorists interested in answering this question draw heavily on dual labor-market theory, whose underlying assumption is that not all jobs are created equal. The theory differentiates between primary sector jobs, or those desirable jobs with high wages, good work conditions, and job stability, among other advantageous features, and secondary sector jobs, or those less desirable jobs with low wages, poor work conditions, job instability, and other unattractive features. The stability of primary sector jobs, it is argued, fosters strong social relationships with peers in the workforce. In contrast, in secondary sector jobs, as a result of job instability, workers fail to develop strong ties to their coworkers and the workplace. The implications of this for crime and violence are critical, according to scholars such as Crutchfield (1989) and Crutchfield and Pitchford (1997), who theorize a link between labor-market characteristics and violence. They identify two mechanisms by which labor-market characteristics may be associated with crime. The first occurs at the individual level among workers and reflects core arguments of social control theory. Secondary sector jobs, they argue, have low wages, low skill levels, and bleak promotion prospects. As such, they are less likely to produce the kinds of job commitments (commitments to conventional lines of behavior) and interpersonal ties to coworkers (attachments to conventional others) enjoyed by primary sector occupations. Moreover, secondary sector workers have more idle time and are often unconcerned with losing their jobs or losing respect among coworkers. In contrast, primary sector workers receive higher wages, have more complex jobs, and have promotion and career prospects. As such, they are likely to develop strong job commitments and ties to coworkers and are less likely to engage in behavior

C

C

676

that will jeopardize their jobs or employment relations (Crutchfield et al. 2006, p. 203). Given the nature and characteristics of their employment, and in line with social control theory, secondary sector workers are more likely to engage in crime. The second mechanism linking labor-market characteristics and crime operates at the neighborhood level and occurs when large numbers of marginally employed individuals live in proximity to one another. Residential segregation and other forces such as discrimination produce high concentrations of jobless and secondary sector male workers, who move in and out of the labor force, work less than full time when employed, and develop few commitments to jobs. Freed from these commitments and constraints, these individuals have time on their hands with little to do. It is argued that such workers frequently pass the time by hanging out on street corners, in pool halls, bars, and nightclubs – places that can be staging grounds for violent encounters. In some neighborhoods, then, concentrations of marginally employed secondary sector workers result in a “situation of company” in which high concentrations of young men experiencing unstable employment are conducive to violence (Crutchfield et al. 2006: 204). At this point, race has not factored into the equation. Where does race enter in and how does this all relate to crime and the racial composition of communities? Race becomes salient because, due to discrimination, inequality, lower education levels, and less social capital, blacks are disproportionately employed in the secondary sector labor market. Crutchfield and his colleagues (2006) argue that, collectively, these factors and processes produce neighborhoods with high concentrations of African Americans, secondary sector workers, and unemployed workers and are at risk for high rates of violence. They describe this as a four-step process: First, macro processes of dual labor markets and residential segregation give rise to neighborhoods with concentrations of marginally unemployed Black adults and youth. Second, high rates of job instability and the accompanying low levels of

Crime and the Racial Composition of Communities economic resources in such neighborhoods give rise to a “situation of company,” in which high concentrations of jobless and marginally employed Black males have time on their hands and weak institutional commitments. Third, labor instability and low levels of economic resources result in increased violent crime through the presence of these “situations of company.” Finally, collectively these processes produce the association of neighborhood racial composition with violent crime. (p. 204)

Empirical research largely supports the proposition that labor instability and violent crime rates are related (Crutchfield 1989; Crutchfield et al. 1999). In particular, researchers find that relationships between poverty and violent crime and income inequality and violent crime depend heavily on the distribution of workers in the primary and secondary occupational sectors and upon levels of unemployment (Crutchfield 1989, p. 505). What has not been empirically demonstrated, however, is whether (and to what extent) the effect of racial composition on crime across neighborhoods can be accounted for by labor-market characteristics – until very recently. In their analysis of Seattle neighborhoods, Crutchfield et al. (2006) examine the relationships among neighborhood racial and ethnic composition, labor-market indicators, and violent crime rates, examining the extent to which labor-market concentration accounts for racial differences in neighborhood rates of violence. Among the many findings, they show that the effects of racial composition are, to some extent, explained by labor-market characteristics in line with the theoretical arguments described above. An important caveat must be noted here. Although Crutchfield et al. (2006) find that labor instability helps to account for the racial composition-crime relationship across Seattle neighborhoods, it does not completely explain why rates of violence are higher in African American (and Latino) communities than elsewhere. They argue, “Even after taking into account labor instability and social disorder, African American and Latino neighborhoods still have higher rates of violence than other neighborhoods” (p. 217). This raises an important


question: What else may account for these racial differences in levels of violence? Crutchfield and his colleagues (2006) encourage researchers to consider the role of culture and cultural processes, which brings us to a final theoretical approach related to crime and the racial composition of communities.

Cultural and Subcultural Theories Both the social disorganization and labor-market perspectives constitute structural theories that address the link between racial composition and crime rates across neighborhoods by focusing on the crime-producing characteristics of communities such as poverty and joblessness. Cultural and subcultural theories take a different approach. Scholars advocating a cultural approach tend to go in many directions but perhaps the most popular are arguments related to a “black subculture of violence.” Cultural theories of this sort range from the more liberal, oppression/ power/anomie-based theories to those that criticize black culture, attacking permissive values which have undermined personal responsibility and character. The more liberal theories, which I discuss below, identify a subculture of violence where, due to a variety of structural conditions, residents resort to violence to defend status, honor, or reputation. Perhaps the most recognized culturally based perspective related to crime and the racial composition of communities comes from Elijah Anderson (1999). In Code of the Street, Anderson’s landmark ethnography of inner-city Philadelphia communities, he argues that as a result of worsening conditions in inner-city communities over the last several decades, black youth in disadvantaged communities have created a local social order – or street code – comprised of their own set of norms and rituals of authenticity. This street code articulates norms and characterizes social relations among residents, particularly with respect to violence. The street code helps us understand why rates of violence are so high in many impoverished African American communities.

677

C

According to Anderson (1999), among those who ascribe to the street code, respect typically forms the core of the person’s self-esteem, especially when alternative avenues of self-expression are limited. One way respect is acquired is by developing a reputation for being violent, by creating a self-image based on “juice” (p. 72). On the streets, the image one projects is paramount and must communicate the message that one is capable of violence when the situation requires it. At the top of the street code hierarchy is the “crazy,” “wild,” or “killer” social identity. In Anderson’s study, he found that youth often created altercations with the sole purpose of building “juice” and to be known as having a reputation for being quick tempered. According to this perspective then, violence is considered the single most critical resource for achieving status among those who participate in street culture. Creating a credible violent reputation not only commands respect but also serves to deter future assaults. However, when challenges do arise or do transgressions occur, violence is viewed as an acceptable, appropriate, and even obligatory response. If a person is assaulted, for instance, it is essential in the eyes of his peers and others to seek revenge. As a result, youth who ascribe to the code often feel constrained to pay back, or seek revenge, after a transgression. Their identity, self-respect, and honor are tied up with the way they perform on the streets during and after such encounters. Violent social control, or what Black (1983) long ago referred to as self-help, is directly related to the availability and effectiveness of authoritative parties of dispute resolution such as the police. Self-help crimes are more likely where law is less accessible, for example, in poor minority communities where residents are often at odds with the police. When called, the police may not respond, which is one reason, Anderson argues, many residents feel they must be prepared to defend themselves and their loved ones against others. Kubrin and Weitzer (2003) document this in their study of extremely disadvantaged communities in St. Louis. They found that problems confronting residents were

C

C

678

often resolved informally – without calling the police – and that neighborhood cultural codes supported this type of problem solving, even when the “solution” was a retaliatory killing. It is often quite difficult to empirically test the core cultural arguments laid out by theorists such as Anderson. However, work by Stewart and Simons (2006) attempts to do just that. Using two waves of data from 720 African American adolescents in 529 neighborhoods, they investigated whether neighborhood context, family type, and discrimination influenced adoption of the street code among youth. More relevant to the discussion above, they also assessed whether adoption of the street code mediated the effects of neighborhood context, family characteristics, and racial discrimination on violent delinquency. Consistent with Anderson’s arguments, they found that neighborhood disadvantage, living in a “street” family, and discrimination significantly predicted adopting street code norms. Moreover, the street code mediated about one-fifth of the effect of racial discrimination and about 4 % of the effect of family characteristics on violent delinquency. In short, the results of their study were generally consistent with Anderson’s code of the street thesis. In understanding the linkages among race, place, and violence, and in explaining why rates of violence are higher in African American communities, perspectives such as Anderson’s emphasize the importance of local culture in the production of violence. It is important to stress, however, that neighborhood structural conditions generate this subculture; in this sense, cultural differences are themselves adaptations to broader structural inequality, which itself is an outgrowth of social, political, and economic forces such as globalization and deindustrialization, redlining and residential segregation, punitive criminal justice policy, and discrimination.

Challenges and Future Directions Despite significant advances in research on crime and the racial composition of communities over


the past several decades, much of which is linked to the three theoretical perspectives just discussed, obstacles and challenges remain. As is obvious from the review of the literature in this area, apart from those studies which measure racial heterogeneity and a handful of others that incorporate additional racial and ethnic minorities into their studies, the vast majority of research refers to the effects of percent black on crime rates, ignoring other racial categories. To some extent, this is understandable given that most large cities in the USA have had only small numbers of other minority groups such as Asians or Latinos. In such situations, then, percent black – by default – has reflected the racial heterogeneity in the area. However, in recent decades with increased mobility and migration, the population composition in communities throughout the USA is changing. Beyond black and white differences, many larger (and some smaller) cities are becoming racially diverse. Further, this change is magnified in some regions of the country such as the Southwest and West Coast, where percentages of Hispanics and Asians have increased dramatically. For these reasons, scholars increasingly underscore the need to move beyond black-white comparisons in neighborhood crime studies (see, e.g., Peterson et al. 2006; Peterson and Krivo 2010). An obvious next step, then, would be to broaden the racial focus as a means to deepen our understanding of just how race, ethnicity, and crime are related across communities. Yet this may be difficult without first addressing a more fundamental (though often neglected) issue in neighborhood research on racial composition and crime: how “race” and “ethnicity” are defined. Although these terms are frequently used interchangeably, in fact, race and ethnicity is not the same thing. Definitions of race include populations that are believed to be distinct in some ways from other populations based on real or imagined physical differences such as skin color or facial characteristics. An individual is typically externally classified (i.e., someone else makes the classification) into a racial group rather than the individual choosing where they belong as part of their identity.


For many social scientists, though, race is more than just biology – it is a social construct or invention that changes as political, economic, and historical contexts change. That is, race is considered a socially significant category, one that is vital to understanding the organization and consequences of social relationships. In this sense, race is, among other things, a sorting mechanism for marriage, dating, and friendship and an organizing device for mobilization to maintain or challenge systems of racial stratification. Although related to race, ethnicity refers not to physical characteristics but social traits that are shared by a human population. These social traits include things like nationality, tribe, religious faith, shared language, shared culture, and shared traditions. Unlike race, ethnicity is not usually externally assigned by other individuals. The term ethnicity focuses more upon a group’s connection to a perceived shared past and culture. Definitions of race and ethnicity are critical because they identify classifications of offenders (and victims), and thus impact conclusions about the linkages among race, ethnicity, and crime across communities. Moreover, difficulties created by the absence of a uniform definition of race are compounded by the data collection practices of government agencies and programs. As LaFree and Russell (1993) explain by way of example, “. . . both the US Census Bureau and the National Crime Survey allow respondents to provide data on their racial origin. Neither organization applies uniform criteria (e.g., mother’s race) or requires external verification. Likewise, the Uniform Crime Reports (UCR) do not provide law enforcement officials with standard guidelines for coding racial origin” (pp. 281–282). Also in UCR data, ethnicity is ignored all together as Hispanics are classified as white, nonwhite, or other race. Of course compounding all of this is the fact that multiracial Americans are one of the fastest growing demographic groups in the country. How are these individuals to be defined and classified? Another important challenge for researchers who study crime and the racial composition of

679

C

communities deals with teasing out race versus class effects in neighborhood crime studies. This is due, in large part, to the significant overlap between race and socioeconomic status in the United States. Because of racial inequality, neighborhoods with high concentrations of African Americans are also the most disadvantaged, experiencing greater social isolation from mainstream society, joblessness, family disruption, and less exposure to conventional role models and fewer workingand middle-class families to serve as social buffers in the community. In fact, it is argued that racial differences in concentrated disadvantage are so strong that the “worst” urban contexts in which whites reside are considerably better than the average context of black communities (Sampson 1987, p. 354). According to many scholars, the structural conditions in black communities are central to understanding the large observed racial differences in urban crime and likely explain why racial composition – especially as reflected in percent black – and crime are associated in neighborhood studies. The argument follows that if similar conditions prevailed in white neighborhoods, they too would exhibit high crime rates. As plausible as it may be, this thesis has gone largely untested because whites rarely live in extremely disadvantaged communities, rendering it difficult to obtain appropriate samples of black and white neighborhoods for comparison. One exception to this is a study by Krivo and Peterson (1996), who examine racial composition and neighborhood crime in Columbus, Ohio (one of only a handful of cities that has a sufficient number of poor black and poor white neighborhoods for comparison). They compare crime levels in extremely poor and disadvantaged black and white neighborhoods with crime in their counterparts. They find that racial differences in structural disadvantage largely account for black-white differences in crime rates across communities. In other words, the fact that black communities are more likely than white communities to experience poverty, social isolation, joblessness, and

C

C

680

a lack of conventional role models explains higher rates of crime in those communities. This finding implies that neighborhood racial composition may not matter as much as previously believed. At the same time, however, they also find that structural disadvantage does not completely explain the racial gap in violence. Crime rates for racially distinct neighborhoods generally approach one another when structural conditions are controlled but they do not become equal – similar to what Crutchfield and his colleagues (2006) found with respect to racial composition, labor-market conditions, and rates of violence in Seattle. That notable gaps in violence remain unaccounted for by racialized community conditions suggests that a focus on differentiation in structural conditions within white, African American, and other-race neighborhoods may be limited. In this context, in subsequent work, Peterson and Krivo (2010) question the following: How much do the distinct spatial contexts in which white, African American, and other-race neighborhoods are located contribute to varying crime rates within these communities? Using crime and census data for 9,593 neighborhoods in 91 large cities for the year 2000, Peterson and Krivo (2010) examine how inequality in nearby neighborhoods contributes to patterned racial and ethnic differentials in crime. They argue that a common feature of many African American neighborhoods, whatever their internal character, is proximity to communities with characteristics typically associated with higher crime rates, such as disadvantage and residential instability. In contrast, white areas are often surrounded by neighborhoods where crime-promoting conditions are absent and factors that discourage crime, such as external community investments, are prevalent. In line with these arguments, a key finding of their study is that white neighborhoods benefit from the dual privileges of low internal disadvantage as well as embeddedness within a context of other white and advantaged areas, whereas African American (and other minority) neighborhoods suffer


a double jeopardy: they are at risk of greater violence stemming from their own internal – often highly disadvantaged – character and they bear the brunt of isolation from violencereducing structures and processes because they are surrounded by disadvantaged areas. Peterson and Krivo’s (2010) study begins to address the challenges associated with teasing out race versus class effects in neighborhood crime studies and offers an additional explanation for why racial composition and crime remain associated across neighborhoods, even after controlling for disadvantage, labor-market characteristics, and other related factors. Yet researchers still have a long way to go before we are able to fully understand the dynamics that underlie the race, place, and crime nexus in US communities.

Related Entries ▶ Disadvantage, Disorganization and Crime ▶ Neighborhood Effects and Social Networks ▶ Race and Ethnicity in Social Disorganization Theory ▶ Race, Ethnicity, and Youth Gangs ▶ Social Network Analysis and the Measurement of Neighborhoods ▶ Spatial Models and Network Analysis

Recommended Reading and References Anderson E (1999) Code of the street. Norton, New York Black D (1983) Crime as social control. Am Sociol Rev 48:34–45 Crutchfield RD (1989) Labor stratification and violent crime. Soc Forces 68:489–512 Crutchfield RD, Pitchford SR (1997) Work and crime: the effects of labor stratification. Soc Forces 76: 93–118 Crutchfield RD, Glusker A, Bridges GS (1999) A tale of three cities: labor markets and homicide. Sociol Focus 32:65–83 Crutchfield RD, Matsueda RL, Drakulich K (2006) Race, labor markets, and neighborhood violence. In: Peterson RD, Krivo LJ, Hagan J (eds) The many colors of crime: inequalities of race, ethnicity, and crime in America. New York University Press, New York, pp 199–220

Crime Investigations by the International Criminal Court Gans H (1968) The balanced community: homogeneity or heterogeneity in residential areas? People and plans: essays on urban problems and solutions. Basic Books, New York, pp 166–181 Greenberg S, Rohe WM, Williams JR (1985) Informal citizen action and crime prevention at the neighborhood level: executive summary. US Department of Justice/National Institute of Justice, Washington, DC Kornhauser R (1978) Social sources of delinquency. University of Chicago Press, Chicago Krivo LJ, Peterson RD (1996) Extremely disadvantaged neighborhoods and urban crime. Soc Forces 75:619–648 Kubrin CE (2000) Racial heterogeneity and crime: measuring static and dynamic effects. In: Chekki DA (ed) Research in community sociology, vol 10. JAI Press, Stamford, pp 189–218 Kubrin CE, Weitzer R (2003) Retaliatory homicide: concentrated disadvantage and neighborhood culture. Soc Probl 50:157–180 LaFree G, Russell KK (1993) The argument for studying race and crime. J Crim Justice Educ 4:273–288 McNulty TL (2001) Assessing the race-violence relationship at the macro level: the assumption of racial invariance and the problem of restricted distributions. Criminology 39:467–490 Peterson RD, Krivo LJ (2010) Divergent social worlds: neighborhood crime and the racial-spatial divide. Russell Sage, New York Peterson RD, Krivo LJ, Hagan J (eds) (2006) The many colors of crime: inequalities of race, ethnicity, and crime in America. New York University Press, New York Sampson R (1987) Urban black violence: the effect of male joblessness and family disruption. Am J Sociol 93:348–382 Sellin T (1938) Culture conflict and crime. Social Science Research Council, New York Shaw CR, McKay H (1969 [1942]) Juvenile delinquency and urban areas. University of Chicago Press, Chicago Smith DA, Jarjoura RG (1988) Social structure and criminal victimization. J Res Crime Delinq 25:27–52 Stewart EA, Simons RL (2006) Structure and culture in African American adolescent violence: a partial test of the ‘code of the street’ thesis. Justice Q 23:1–33 Warner BD, Pierce GL (1993) Reexamining social disorganization theory using calls to the police as a measure of crime. Criminology 31:493–517 Warner BD, Rountree PW (1997) Local social ties in a community and crime model: questioning the systemic nature of informal social control. Soc Probl 44:520–536

Crime Concentration ▶ Prediction and Crime Clusters

681

C

Crime Investigations by the International Criminal Court Michel De Smedt and Anja Schro¨der Office of the Prosecutor, International Criminal Court, The Hague, The Netherlands

Synonyms ICC; International Criminal Court

Overview The International Criminal Court (ICC) was established by the Rome Statute and aims to deal with the most serious crimes of concern to the international community as a whole. There are five elements rendering the investigation of war crimes, crimes against humanity, and genocide at the ICC distinct from common criminal investigations: first, the gravity of international crimes creates challenges to cover all crimes at stake and to meet the expectations of all victims. Second, the size and type of the organizations responsible for crimes impacts in different ways on the focus of the investigation. Third, protecting persons at risk due to their interaction with the Court is a particularly challenging requirement. Fourth, the Court highly depends on cooperation from national jurisdictions, given the lack of enforcement mechanisms inherent to the Rome Statute. Fifth, investigations are often performed in the context of ongoing conflicts where the “peace versus justice” debate impacts on the willingness to support investigations. Criminology could support international investigations and prosecutions in various ways: it could assist in measuring scale and impact of crimes. Furthermore, it could develop models to analyze and prove the role of the most responsible perpetrators. In addition, it could map links between organized crime and conflict economies. Finally, criminology could develop tools to assess the effectiveness of the justice system created by the Rome Statute and the ICC as part of that system.

C

C

682

Fundamentals International criminal justice has come a long way from the International Military Tribunals of Nuremberg and Tokyo via the ad hoc International Criminal Tribunals for Rwanda and the former Yugoslavia, and the hybrid international tribunals for Sierra Leone and Cambodia to the final emergence of the International Criminal Court (ICC), the first permanent international criminal court in the world (Schabas and Bernaz 2011). The ICC, which is based on the Rome Statute, an international treaty adopted in 1998 in Rome, became operational in 2002. The ICC’s capacity of being the first permanent international criminal court in the world brings with it one major difference compared with previous tribunals: the former focused primarily and are still focusing a posteriori on events well defined and limited in terms of time and place. By contrast, the ICC has been assigned a dual role: through having jurisdiction over the most serious crimes of concern to the international community, the Court shall not only end impunity by bringing the perpetrators of past crimes to justice. In addition, the Court is intended to deal with ongoing crimes and contribute in this dual way to the prevention of crimes. These crimes may comprise a broad variety of offenses, often occurring in the context of prolonged armed conflicts or protracted violence in different parts of the world. As a consequence, the ICC’s Office of the Prosecutor (OTP), with a workforce of almost 300 staff from over 80 countries, deals with numerous “situations” in most diverse countries at the same time. As of April 2012, seven situations are under investigation before the Court (Uganda, Democratic Republic of the Congo [DRC], Central African Republic [CAR], Darfur, Kenya, Libya, Coˆte d’Ivoire). Fifteen cases have been brought before the Court, of which six are currently at trial stage. Eight situations are currently under preliminary examination in four different continents (Colombia, Honduras, Guinea, Georgia, Afghanistan, Nigeria, Korea, Mali), meaning that the OTP is analyzing information to determine whether an investigation should be opened or not. In addition, as of 3 October 2011,

Crime Investigations by the International Criminal Court

the OTP had received 9,303 communications in which individuals or organizations have submitted information on alleged crimes potentially falling within the jurisdiction of the Court. While the ICC’s scope may appear relatively broad at first glance, there are important limitations as to when and where the Court can intervene. First, the Court has jurisdiction only over crimes having occurred since 1 July 2002. Next, the Preamble of the Rome Statute proscribes that the ICC only deals with “the most serious crimes of concern to the international community as a whole.” These are genocide, war crimes, and crimes against humanity. In addition, the Court will be able to exercise jurisdiction over the crime of aggression once the provision adopted by the Assembly of States Parties enters into force in the future. Furthermore, the ICC can only investigate and prosecute individual persons, as opposed to organizations as a whole. Finally, other than where the Security Council refers a situation, the Court’s treaty-based personal and territorial jurisdiction does not extend its reach to all 193 countries in the world. While a number of major States such as the USA, Russia, and China have not yet joined the Rome Statute, the ICC has witnessed a steep increase from 60 States Parties in 2002 to 121 States Parties as of April 2012. There are three ways for the OTP to become active. First, a State Party can refer a situation to the Court, as happened in the Democratic Republic of the Congo (DRC), Uganda, and the Central African Republic (CAR) (Rome Statute, Art. 13(a), 14). Second, the Prosecutor can open an investigation proprio motu subject to the authorization of the Pre-Trial Chamber (Rome Statute, Art. 15). For both a State Party referral and proprio motu authorization, the Court can only exercise jurisdiction if the alleged crimes occurred on the territory of a State Party or were committed by a national of a non-State Party or on the territory of a non-State Party which has lodged a declaration accepting the Court’s jurisdiction as in the case of Coˆte d’Ivoire (Rome Statute, Art. 12). Third, the UN Security Council can refer any situation to the Court pursuant to its Chapter VII powers under the UN Charter, including in relation to a non-State Party, as


happened in 2005 with Darfur, Sudan, and in 2011 with Libya (ICC Statute, Art. 13(b)). Irrespective of how a situation is brought before the Court, in all three instances, the Prosecutor must be satisfied that the information available provides a reasonable basis to proceed with investigations (Rome Statute, Art. 15(3), 53(1)). In any other situation not covered by any of the above mentioned, the Court cannot become active. The Rome Statute also sets as an admissibility requirement that any case brought forward for prosecution must be of sufficient gravity (Rome Statute, Art. 17(1)(d)). In addition, the ICC can only proceed with a specific case if the State concerned is either inactive or otherwise unwilling or unable to carry out genuine proceedings (Rome Statute, Art. 17(1)(a)–(c)). This touches on one of the most fundamental principles of the Rome Statute, namely, complementarity. The Court is designed to complement national criminal justice systems, not replace them. Therefore, the ICC functions as a court of last resort, which should only become active in the exceptional situations foreseen in the Statute. All these aspects and additional ones to be discussed in-depth in the following render investigations and prosecutions by a permanent international tribunal distinct from those conducted by either national or hybrid tribunals. This entry will identify five main peculiarities: the gravity of crimes, the size and type of alleged responsible groups, the protection challenges, the international cooperation, and the “peace versus justice” debate. It will also assess to what extent criminology could become more involved in addressing these challenges, despite the fact that it has been rather reluctant to study international crimes so far (Haveman and Smeulers 2008). Finally, one remark is in order: the fact that the jurisprudence is still developing creates a specific challenge as such. However, this aspect will not be dealt with here, as it is not idiosyncratic to international investigations, but rather related to the emergence of a new legal framework. Four fundamental principles define the OTP’s strategy: positive complementarity, focused investigations and prosecutions, addressing the

683

C

interests of victims, and maximizing the impact of the Office’s work (OTP 2010a). The OTP’s action is further guided by three key notions: independence, impartiality, and objectivity (OTP 2010b). That is, the OTP shall act independently of “instructions from any external source” according to Art. 42(1) Rome Statute. Furthermore, from the notion of impartiality implicit in Art. 21(3) of the Rome Statute, it follows among others that the Office shall “apply consistent methods and criteria irrespective of the States or parties involved or the person(s) or group(s) concerned” (OTP 2010b). Finally, objectivity in accordance with Art. 54(1) Rome Statute means that the OTP must investigate incriminating and exonerating circumstances equally. As ICC Prosecutor Fatou Bensouda has stated, objectivity aims at establishing the truth, as opposed to securing “convictions at any cost” (Bensouda 2010).

Key Issues Gravity of International Crimes According to the Rome Statute, the cases selected for investigation and prosecution have to be of sufficient gravity. The OTP considers four indicators for assessing gravity in determining whether an investigation shall be opened. These are scale, nature, manner, and impact of the crimes committed. In the case of ICC crimes, the scale tends to be massive. For instance, in the DRC, the OTP announced in September 2003 that it had identified the crimes in the Ituri region as warranting the focus of an investigation, with reports indicated the particular gravity of the crimes, resulting in inter alia an estimated 5,000 civilian deaths since 2002. The massive scale of the precursor to this violence in the DRC is further evident in the reported total number of both direct and indirect conflict deaths in the whole DRC between 1998 and 2002, which has been estimated at 3.3 million people (Geneva Declaration on Armed Violence and Development 2008). In the case of the postelection violence in Kenya in 2007–2008, which happened over a period of only a few weeks, 1,200 persons were allegedly killed, in addition to at least 1,000

C

C

684

reported rapes and 450,000 persons displaced. Apart from the scale of the crimes, the manner of commission is considered for the assessment of gravity. In Ituri, the Office considered information on summary executions, burning of people alive, physical mutilation, and the specific targeting of vulnerable groups, in particular women and children. Examples of extreme brutality can also be found in other situations. In the Bemba case, victims were “raped at gunpoint, in public or in front of or near their family members” (Pre-Trial Chamber II 2009). In Darfur, babies were reportedly burnt in the presence of their families. Finally, impact refers to the consequences of a given crime. This includes social, economic, and environmental damage of crimes; the intention of spreading terror among the civilian population; or crimes committed with the aim or consequence of increasing the vulnerability of civilians, as alleged by the OTP in the Banda and Jerbo case concerning an attack against peacekeepers (OTP 2010b). Nonetheless, gravity can sometimes be difficult to measure. Quantitative data, such as figures of people killed or raped, are often rare and, if available, may be difficult to ascertain. This is due to a variety of factors. First, some areas may not be covered by any monitoring mechanism. This may be the case for certain locations not accessible in times of ongoing conflict. Potential bias may also result from a possible interest of the parties to a given conflict to underreport crimes by one side and overreport those by another. Moreover, there may be a reluctance to report crimes altogether, which is particularly true for incidents of sexual violence. Here, an assumed systematic underreporting can be traced back to a number of reasons, including social stigma attributed to rape victims in certain communities. This is the case, for instance, in Libya, where victims of sexual violence have reportedly been at risk of reprisals and stigmatization. As a consequence, it can often be difficult to assess the full scale of sexual violence allegedly committed during the armed conflict of 2011. While the credibility of information, particularly numeric information, may often require scrutiny, the reliability of sources may often be equally


hard to assess. In this context, methodologies derived from criminology could help to further refine indicators for assessing gravity, including models for measuring scale and impact on societies affected by international crimes. See for an example thereof the work of John Hagan (Hagan et al. 2005) who won the Stockholm Prize in Criminology in 2009 for his work. In well-functioning national jurisdictions, it is hard to imagine that serious crimes are not investigated or that they go unpunished. In situations of massive violence that the OTP faces, however, there will often be large amounts of information concerning multiple perpetrators at varying levels of responsibility committing large-scale and widespread crimes against numerous victims. It is not the responsibility or role of the Prosecutor to investigate and prosecute each and every criminal episode within a situation. This is not only practically unfeasible: it would also run counter to the notion of a system of courts combating impunity through complementary action at the international and national level, as foreseen in the preamble of the Statute, and the primary responsibility of those States for the investigation and prosecution of such crimes (Rome Statute, Preamble; OTP 2003). The OTP, thus, must exercise discretion in determining which cases should be selected and prioritized for investigation and prosecution. As a consequence, the OTP must strategically focus its activities on a limited number of cases at the highest levels of responsibility and on the most serious crimes. This selection process comes with positive and negative implications. On the one hand, maintaining focus allows expeditious investigations and prosecutions while limiting the number of persons put at possible risk due to their interaction with the Court. It also prevents excessive spending of resources by investigating too broadly. On the other hand, a limited pool of witnesses may give rise to problems for a case should a witness cease cooperating at a later stage. Therefore, it is necessary to balance the use of documentary and other evidence with the value of witness testimony and the reliability and stability of such witnesses. Focused investigations may further mean that the cases will not capture the full extent


of all crimes suffered by all victims nor enable prosecution for all possible perpetrators. While it remains to be seen how the ICC judges will respond to the OTP’s focused approach, it is safe to assume that the OTP cannot meet the expectations of all victims. However, it does certainly seek to ensure in the cases it brings that the incidents selected for prosecution are reflective of the main types of victimization. In the case of massive crimes, the number of direct victims and indirectly affected communities demanding justice from the Court is high with victims approaching the Court in a single situation often in thousands. Expectations must thus be managed from the beginning not only by clarifying the limitations of the mandate (such as temporal jurisdiction) but also by explaining the prosecutorial policy of the Office. In this context, it is important to note that under the Rome Statute, victims have for the first time in the history of international criminal justice the possibility of taking an active role as witnesses, participants in the proceedings, and beneficiaries of reparations. However, due to the Office’s policy of focused investigations and prosecutions, there may be victims who suffered harm as a result of crimes other than those actually included in the charges selected for prosecution. The interests of these victims, who therefore do not qualify for participation in the pretrial or trial proceedings in a particular case, need to be addressed by the Office in other ways. In addition to outreach sessions in affected communities in situation countries such as Uganda or DRC, victims may present their views early on through submitting information on crimes or participation in town hall meetings, which have at times contributed to the definition of incidents and charges brought forward by the Prosecution. Second, through considering the broader impact of crimes, the Office seeks to address the interests of a wider community of victims. Third, at the reparations stage, the OTP favors a wider approach to allow participation of victims and representations from or on behalf of victims and other interested persons who suffered harm as a result of crimes other than those included in the charges selected for prosecution (OTP 2010a). In this context, it

685

C

should be noted that the ICC, as the first international criminal tribunal awarding reparations, will have to decide on how to provide former child soldier victims with compensation in the case of the former Commander-in-Chief of the Forces patriotiques pour la libe´ration du Congo, Thomas Lubanga Dyilo. In addition, stimulating supplementary solutions to close the impunity gap will be necessary, either through national proceedings or alternative mechanisms such as truth and reconciliation commissions. A comprehensive strategy is needed to determine how situation countries can (be helped to) deal with all crimes committed. Perpetrators of International Crimes Massive crimes require extensive planning and usually involve a broad variety of direct and indirect perpetrators from a State apparatus or an organization. For instance, three State armies and 20 armed groups were involved in Congo Wars over the last two decades, the remnants of which can be witnessed in the ongoing conflict in eastern DRC. In the Darfur conflict, one State army, eight government-allied forces, and 14 armed opposition groups have been active. The strength of such groups can range from 50 to 50,000 fighters. Besides quantitative criteria, groups vary in their nature, particularly their level of organization, which may range from network-like, temporary structures to militias up to a fully fledged state apparatus. The share in the commission of gravest crimes needs to be carefully investigated and evaluated for each group against the gravity threshold to identify the most responsible perpetrators. In the situation of Kenya, for instance, the Office has investigated simultaneously persons affiliated with both the government and opposition side. In the situation of Libya, the ICC’s first Prosecutor, Luis Moreno-Ocampo, pointed to allegations of crimes not only by pro-Gaddafi forces but also by NATO forces and forces affiliated with the National Transitional Council. In the situation of Coˆte d’Ivoire, the Office has emphasized that the investigation of alleged crimes committed by different parties to the conflict will be conducted irrespective of political affiliation. However, both

C

C

686

focused investigations and resource limitations may also require prioritization. For instance, regarding the complex conflict in Ituri, in which many militias have allegedly committed crimes, the OTP focused its investigation on those militias allegedly responsible for the most serious crimes, which resulted in cases concerning the Union des Patriotes Congolais (“UPC”) and its military wing, the Forces Patriotiques pour la Libe´ration du Congo (“FPLC”), and the Force de Re´sistance Patriotique en Ituri (FRPI) and the Front de Nationalistes et Inte´grationnistes (FNI). In Uganda, upon analyzing the gravity of crimes allegedly committed by different armed forces and groups, the OTP found that alleged crimes by the LRA were of highest gravity. The OTP therefore started with an investigation of the LRA. Case selection is thus based on a number of legal and policy principles, including overriding principles based on objectivity, impartiality, and independence, as well as considerations of gravity focused on the most serious criminal episodes. A similar process is applied once a specific organization or group has been identified for investigation. The OTP has stated in its 2003 policy paper that “as a general rule, the Office of the Prosecutor should focus its investigative and prosecutorial efforts and resources on those who bear the greatest responsibility, such as the leaders of the State or organisation allegedly responsible for those crimes” (OTP 2003). Accordingly, individuals investigated and prosecuted before the ICC have included senior civilian leaders and military commanders, two Heads of State, and one former Head of State. In some cases the focus of an investigation by the Office may go wider than high-ranking officers, if investigation of certain type of crimes or those officers lower down the chain of command is necessary for the whole case. Nonetheless, as a general rule, the Office encourages national prosecutions, where possible, for the lower-ranking perpetrators or otherwise seeks to work with the international community to ensure that the offenders are brought to justice by some other means. The goal is to develop comprehensive anti-impunity strategies combining international and national efforts to combat impunity. If former perpetrators


are not dealt with through comprehensive strategies, which may involve a variety of transitional justice tools, there is a risk that they may rejoin criminal subcultures or migrate to third countries to form newly emerging criminal networks elsewhere. Here, criminology could research further offender integration in relation to perpetrators of international crimes at different levels of responsibility. To link suspects to a given crime, the guiding model for investigations by the OTP puts emphasis on a comprehensive crime analysis enabling one to piece together crime patterns and chains of command and to collect the type of evidence needed to establish nature and degree of responsibility (OTP 2003). While a fundamental requirement in this context is to prove charges beyond reasonable doubt, the criminal responsibility of leaders is not always easy to establish. For instance, in many conflict-ridden environments, sexual violence is a frequently reported phenomenon. However, the context of leadership responsibility for individual incidents is not always clear: rape may be committed as an opportunistic crime by private citizens or members of an armed group or state apparatus against the backdrop of a general breakdown of the rule of law or disciplinary failures, or rape may be committed or condoned as an organized campaign by a state apparatus or organization to further specific policy objectives. A further challenge arises from proving criminal responsibility at the highest level. The more sophisticated a group, the more its members will cover their traces in order to make issues of attribution difficult. In other situations, a criminal network may benefit from a conflict in undetected or underreported manners. Ignoring such a link may result in organized crime reinforcing itself in a conflict area, rendering it also more difficult to fight it at the national level, particularly in countries far removed from the conflict. In this area, criminology has a twofold task: first, it can assist in bringing together various disciplines including organizational thinking, organized crime models, and military organizational models to better understand how the most responsible perpetrators fulfill their role in


a given organization and to define different ways of proving criminal responsibility. Second, criminology could build on the work of Edwin Sutherland and other criminologists to apply the study of crimes by corporations and State crime to the linkage between organized crime, international crimes, and armed conflict. Protection of Persons at Risk The OTP often operates in areas of ongoing conflict or post-conflict situations. This comes with a variety of challenges including poor infrastructures coupled with often remote and inaccessible locations, volatile security situations, and a precarious information security environment. In addition to operational challenges, cultural differences need to be taken into account. For instance, the Office has to communicate effectively with witnesses in local languages, some of which have no corresponding words for legal terminology, while finding qualified professional translators requires exceptional efforts. However, while many of these challenges do not differ from those encountered by other agencies working in the same environments, the Court faces one distinct challenge. This is the legal duty to protect persons at risk on account of their interaction with the court, predominantly victims and witnesses. As the OTP has increased its activities and judicial proceedings have evolved from opening investigations towards actual trials and conviction, the number of persons threatened due to their interaction with the Court has also increased, as well as the capabilities of suspects to track, monitor, and influence witnesses. The risk of threats and actual victimization of witnesses implies that the OTP has to approach witnesses without exposing them, which includes identifying safe sites for interviews, securing discrete transportation for investigators and witnesses, or checking for possible relationships between drivers and hotel owners with suspects. While witness protection is a challenge in national jurisdictions as well, it becomes even more complex when one cannot rely on a wellestablished law enforcement structure. Take the case of Darfur, in which two arrest warrants have been issued against the Head of State, President

687

C

Bashir. Due to security concerns, the OTP had to conduct investigations into this situation without being able to secure access to witnesses and crime scenes inside Darfur. Instead, it collected information from witnesses in third countries. This scenario arises when one operates in ongoing conflicts and post-conflict situations without state support or when one is operating globally in multiple places with limited resources. A particular difficulty is related to the fact that the OTP is usually investigating the powerful, i.e., members of armed groups controlling a given territory or members of a state apparatus. Here, the Office is dealing with alleged perpetrators who see themselves as (and are de facto) in control of the law. In addition to challenges posed by powerful suspects, individuals at risk due to their interaction with the Court such as witnesses sometimes may not fully appreciate the dangers posed by the security situation or have a different sense of risk due to their ingrained behavior in living in a conflict zone, which may ultimately lead them to undermine protective measures. In addition, there is a risk that certain individuals may see protection through the Court as a way to improve their own circumstances. Inevitably, security concerns may influence the speed of an investigation. While theoretically speaking, quickly securing an arrest warrant could have a deterrent impact in restraining the further commission of crimes, unfortunately the execution of an arrest warrant is not always automatic. Cooperation with National Jurisdictions and Other Institutions Cooperation is the fourth aspect rendering investigations and prosecutions by the OTP distinct. Cooperation is needed for a number of reasons. Next to the general support to the Court’s decisions, operational cooperation from national authorities and international organizations is a core requirement for conducting investigative activities such as gathering information and collecting evidence, dealing with logistic challenges in the field, responding to security issues, or building capacity several thousand kilometers away from the ICC Headquarters in The Hague. Overall, OTP operations benefit from

C

C

688

cooperation with a broad range of partners, including referring States, third States involved in conflicts related to the situation under investigation, States Parties, international organizations, war crimes units of national jurisdictions, and other experts. For instance, in the situation of Libya, as of 3 October 2011, the OTP had conducted 50 missions to 15 different countries. In other situation countries such as the DRC, to facilitate missions, cooperation mechanisms had to be established with other partners including the peacekeeping mission and other relevant organizations. In addition to support needed on the ground, the OTP has benefited in the past from its network with national law enforcement agencies and other specialized organizations such as national forensic institutions. At the same time, the OTP has also provided assistance where possible and subject to witness security considerations to national jurisdictions, such as in the case of the Forces De´ mocratiques pour la Libe´ration du Rwanda (FDLR), an armed group operating in the DRC, where the OTP successfully contributed to the investigation by the German authorities of two senior leaders living in Germany, while the OTP itself investigated a third senior leader residing at the time in France. Also, as part of its policy of positive complementarity, the OTP has provided assistance to the Ugandan authorities in preparation of their first domestic war crimes case before the newly established International Crimes Division of their High Court, against an alleged midlevel LRA commander. While these are positive examples of cooperation, the interaction with national jurisdictions is hampered at times by a number of factors: first, there are the ordinary challenges of law enforcement cooperation. These constraints are equally known to other types of international investigations such as organized crime investigations. A recent unpublished study conducted under the guidance of Interpol together with the ICC identified major challenges to international investigations including different legal frameworks, different capabilities and resources, incompatibility of procedures, lack of trust, and cultural obstacles. The study further identified factors


that can improve such cooperation including the ability to assess the impact of transnational crime, strategic investment into international cooperation including training programs, and other ways of promoting interpersonal contact as well as joint planning (Police Academy of the Netherlands 2011). The level of support by a given country or organization can have considerable impact on the ability of the OTP to investigate and prosecute. While the duration of an investigation depends on a number of factors such as focus and complexity of the investigation, the assistance from national jurisdictions or absence thereof can be decisive. Nonetheless, the OTP has also demonstrated that it can pursue its mandate in the face of intransigence or active opposition from the State concerned: for example, in both Darfur and Libya, where it showed that it was able to pursue investigations against the highest level of responsibility without the assistance of the territorial State. The most critical challenge for prosecuting international crimes, however, is the execution of an arrest warrant. As often pointed out, the ICC lacks an own police force. In fact, the Court depends entirely on the assistance of States to ultimately fulfill its arrest mandate. The level of support in the area of arrests depends on both the ability of a given country to control its own territory and its willingness to surrender a suspect to the Court. There have been positive examples, such as three nationals of the DRC who were surrendered to the Court by the DRC’s national authorities, or Belgium, which implemented an arrest warrant against the President and Commander-in-Chief of the Mouvement de libe´ration du Congo, Jean-Pierre Bemba Gombo. Another positive example is the surrender of former President Laurent Gbagbo by the national authorities of Coˆte d’Ivoire following a warrant of arrest issued under seal in November 2011. Yet warrants remain pending against Sudanese President Bashir, regarding which the Pre-Trial Chamber has issued several decisions regarding noncompliance following the failure by some States Parties to cooperate with the Court. With regard to Libya, the OTP has received support


from Interpol through three “red notices” for the arrest of ICC suspects Muammar Gaddafi (until his death), Saif Al-Islam Gaddafi, and Abdullah Al-Senussi. The OTP has also been able to secure the voluntary appearance of nine suspects before the Court, namely three rebels from Darfur and six public figures in Kenya. Despite concerted efforts, however, a total of 11 suspects sought by the ICC were still at large as of April 2012. The execution of arrest warrants, nonetheless, is a perennial challenge for all international courts and tribunals. The arrests of Radovan Karadzic and Ratko Mladic´ and their surrender to the International Criminal Tribunal for the Former Yugoslavia, for example, occurred 13 and 16 years after their first indictment in July 1995, respectively. Nonetheless, they were ultimately brought to face justice. While the OTP’s timing of an arrest warrant application will in the first instance depend on the status of the evidence collected, other factors will also come into play. These include the window of opportunity to actually arrest a suspect. Despite such challenges, international criminal justice has witnessed the gradual emergence of a paradigm shift away from the rule of power towards the rule of law. Needless to say, such a shift creates tensions. As first ICC Prosecutor Luis Moreno-Ocampo has pointed out, “Reality has demonstrated that the Office’s independent decisions have triggered conflicts of interests for States. Leaders who are using crimes to retain power have criticized the Court and managed to mobilize some international support to this end. States Parties have struggled to prioritize their commitment to international justice over more immediate economic or political interests” (Moreno-Ocampo 2011). Against the backdrop of a general lack of knowledge of what the Court can and cannot do, coupled with high expectations about justice being brought about, insufficient levels of cooperation may ultimately lead to the perception of double standards. Peace Versus Justice? Finally, the fact that ICC investigations and prosecutions often take place against the backdrop of

689

C

ongoing conflict routinely triggers an old debate revolving around a difficult question: what is more important – peace or justice? This debate is inevitable considering that the OTP may be investigating and prosecuting in an arena where other actors are focused on political settlements to end conflicts, on securing access for the provision of humanitarian assistance, or on enabling implementation of cease-fire agreements or the deployment of peacekeeping personnel. It is often presented as a dilemma difficult to resolve, but the dilemma itself may also often be illusory and intentionally instrumentalized by those in power to avoid justice. While supporters of the paradigm “peace first, justice second” argue that peace is more important to achieve security, proponents of the paradigm “no peace without justice” emphasize that if justice is not ensured at all levels, tensions may often continue to fester and foster to new violence based on previous patterns of impunity for violent crimes. The long-term logic of favoring short-term solutions needs also to be considered: what deterrent message is conveyed to the perpetrators of genocide, crimes against humanity, or war crimes by signaling that they can ensure impunity for their crimes by holding peace a hostage to threats to commit more crimes? It is worthwhile mentioning here that the UN Security Council referrals of two situations to the ICC, namely, Darfur and Libya, have both stressed the need to account for past crimes. As described in the OTP policy paper on the Interests of Justice of September 2007, “The issue is no longer about whether we agree or disagree with the pursuit of justice in moral or practical terms: it is the law. Any political or security initiative must be compatible with the new legal framework insofar as it involves parties bound by the Rome Statute” (OTP 2007). Moreover, in an April 2009 report on mediation, UN Secretary-General Ban Ki-Moon advised all mediators that they must take into account and respect the ICC jurisdiction and action in accordance with the Rome Statute: “Where serious crimes have been committed, pursuing international justice during mediation can generate considerable tension and affect the outcome, since indicted parties may cease cooperation and

C

C

690

actively obstruct the process. Ignoring the administration of justice, however, leads to a culture of impunity that will undermine sustainable peace. Now that the International Criminal Court has been established, mediators should make the international legal position clear to the parties. They should understand that, if the jurisdiction of the ICC is established in a particular situation, then, as an independent judicial body, the Court will proceed to deal with it in accordance with the relevant provisions of the Rome Statute and the process of justice will take its course” (UN Security Council 2009).

Future Directions In light of the numerous challenges, it seems that the ICC is still at the beginning of a long-term process to bring an end to impunity. While the Court has recently completed its first trial in the case against the former Commander-in-Chief of the Forces patriotiques pour la libe´ration du Congo, Thomas Lubanga Dyilo, and nears completion in two others, many of the challenges the OTP faces will continue to exist. This includes high expectations towards and knowledge gaps regarding the Court, protection challenges, and cooperation issues, as well as the recurring peace and justice debate. Furthermore, at a time of an already high caseload, the Court must effectively respond to demands by some States Parties for budget restraints in an overall stretched economic climate. Both the first and current prosecutor have noted that the OTP has surpassed the limit of resources in comparison to the number of cases. Despite numerous challenges outside of its control, the ICC and with it the Rome System, which celebrated their 10th anniversary in 2012, will and must be ultimately judged based on the ability to achieve the goal of the preamble of the Rome Statute, namely, the determination to end impunity and contribute to the prevention of the most serious crimes in the world. In this context, criminology could play a leading role in assessing the effectiveness of the two complementary protagonists of the Rome System, i.e., the Court on


the one hand and States Parties on the other. Here, academic research may assist in developing performance indicators to find answers to a number of open questions, for instance, as follows: is the Court having a deterrent effect in modifying behavior? Has the existence of the Court started to lead to an increase of national proceedings, as a result of complementarity? Is the Court being perceived by the broader public as bringing and safeguarding justice? Successful investigations and prosecutions of genocide, crimes against humanity, and war crimes require the integration of a broad variety of academic disciplines and fields of work usually not found. Those most relevant in the current context are international relations, investigative work, security studies, prosecutorial approaches, and organizational studies. Criminology is the field which, due to its idiosyncratic, interdisciplinary orientation, would be well suited to assist in improving our understanding of international crimes and their perpetrators, our ability to prove responsibility of the most responsible ones, including criminal networks whose operations are sustained and fuelled by conflicts around the world. Finally, if the ultimate goal of the ICC is, according to the principle of positive complementarity, not to intervene due to genuine national proceedings, criminology will be needed to conduct research on how to prepare national jurisdictions not only to deal with gravest crimes but also to prevent them from happening in the first place. As noted in the OTP’s policy paper of 2003, “the absence of trials by the ICC, as a consequence of the effective functioning of national systems, would be a major success” (OTP 2003).

Related Entries ▶ Crimes of the Powerful ▶ Development of International Criminal Law and Tribunals ▶ Genocide ▶ International Sentencing ▶ Network Analysis in Criminology

Crime Location Choice

▶ Organized Crime, Types of ▶ Resistance to State Crime ▶ Scientific Evidence Before International Criminal Tribunals ▶ Social Network Analysis of Organized Criminal Groups ▶ State Crime ▶ Transnational Exchange of Forensic Evidence ▶ Victims of State Crime ▶ Victims and the International Criminal Court

Recommended Reading and References Bensouda F (2010) Challenges related to investigation and prosecution at the international criminal court. In: Bellelli R (ed) International criminal justice. Law and practice from the Rome statute to its review. Ashgate, Farnham/Burlington, pp 131–142 Geneva Declaration on Armed Violence and Development (2008) Global burden of armed violence. http://www. genevadeclaration.org/fileadmin/docs/Global-Burdenof-Armed-Violence-full-report.pdf. Accessed 4 Jan 2011 Hagan J, Rymond-Richmond W, Parker P (2005) Criminology of genocide: death and rape in Darfur. Criminology 43(3):525–561 Haveman R, Smeulers A (2008) Criminology in a state of denial – towards a criminology of international crimes: supranational criminology. In: Smeulers A, Haveman R (eds) Supranational criminology: towards a criminology of international crimes. Intersentia, Antwerp/Oxford/Portland, pp 3–26 Moreno-Ocampo L (2011) Address to the assembly of states parties, tenth session of the assembly of states parties. New York. http://www.icc-cpi.int/NR/ rdonlyres/7F1DD418-2E1A-4216-BFA5-D8F618AF AE95/284107/ASPLMOFinalrevised_2_.pdf. Accessed 4 Jan 2012 OTP (2003) Paper on some policy issues before the office of the prosecutor. http://www.icc-cpi.int/NR/ rdonlyres/1FA7C4C6-DE5F-42B7-8B25-60AA962E D8B6/143594/030905_Policy_Paper.pdf. Accessed 4 Jan 2012 OTP (2007) Policy paper on the interests of justice. http:// www.icc-cpi.int/NR/rdonlyres/772C95C9-F54D-4321BF09-73422BB23528/143640/ICCOTPInterestsOfJusti ce.pdf. Accessed 3 Jan 2012 OTP (2010a) Policy paper on victims’ participation. http://www.icc-cpi.int/NR/rdonlyres/9FF1EAA1-41C44A30-A202-174B18DA923C/281751/PolicyPaperonVictimsParticipationApril2010.pdf. Accessed 4 Jan 2012 OTP (2010b) Draft policy paper on preliminary examination. http://www.icc-cpi.int/NR/rdonlyres/ 9FF1EAA1-41C4-4A30-A202-174B18DA923C/2825

691

C

15/OTP_Draftpolicypaperonpreliminaryexaminations 04101.pdf. Accessed 4 Jan 2011 Police Academy of the Netherlands (2011) Summary report: pearls in policing, The Hague. http://spl.politie academie.nl/Portals/0/docs/Internationaal/PEARLS% 20SUMMARY%20REPORT%202011.pdf. Accessed 13 Apr 2012 Pre-Trial Chamber II (2009) Decision pursuant to Article 61(7)(a) and (b) of the Rome statute on the charges of the prosecutor against Jean-Pierre Bemba Gombo (ICC-01/05-01/08). Public document. http://www. icc-cpi.int/iccdocs/doc/doc699541.pdf. Accessed 4 Jan 2012 Rome Statute of the International Criminal Court (1998) UN doc. A/CONF.183/9, 2187 UNTS 90, entered into force 1 July 2002 Schabas WA, Bernaz N (eds) (2011) Routledge handbook of international criminal law. Routledge, London/New York UN Security Council (2009) Report of the secretarygeneral on enhancing mediation and its support activities (S/2009/189). http://daccess-dds-ny.un.org/ doc/UNDOC/GEN/N09/278/78/PDF/N0927878.pdf? OpenElement. Accessed 10 May 2012

Crime Linkage Analysis ▶ Linkage Analysis for Crime

Crime Location Choice Wim Bernasco1,2 and Stijn Ruiter1 1 Netherlands Institute for the Study of Crime and Law Enforcement (NSCR), Amsterdam, The Netherlands 2 Department of Spatial Economics, VU University Amsterdam, Amsterdam, The Netherlands

Synonyms Discrete spatial choice

Overview Most behavior of interest to social scientists is choice behavior: actions people commit while

C

C

692

they could also have done something else. In geographical and environmental criminology, a new framework has emerged for analyzing individual crime location choice. It is based on the principle of random utility maximization as developed in economics. It integrates the study of spatial crime distributions with journey-to-crime research, and it is used to explain the offender’s choice of where to commit an offense. It allows the analyst to simultaneously assess the role of location attributes and the role of these attributes in relation to offender characteristics, such as their age, ethnicity, gender, criminal experience, and where they live. Initial applications of the model have shown that the decision of where to commit an offense can successfully be described as a function of characteristics of the decision-maker (i.e., the offender), the potential crime target locations, and their interactions, including the distance that separates them. Other applications have established that physical and social barriers inhibit the journey to crime, that railways facilitate the journey to crime, and that offenders are more likely to offend near former anchor points (past homes). New developments in the area of crime location choice include a focus on small spatial units of analysis and the assessment of spatial spillover effects.

Introduction Many research questions in the social and behavioral sciences, including criminology, deal with understanding and predicting individual choices. Political scientists aim to understand why people vote and what makes them choose a particular political party (Palfrey and Poole 1987). Sociologists want to understand what makes people decide in favor of a particular education, occupation, or marriage partner (Jepsen and Jepsen 2002). In marketing research, understanding and predicting consumer choice is core business (McFadden 1980). Transportation researchers aim to find out what it is that makes commuters choose to travel by bicycle, car, bus, or train (Train 1980). Behavioral ecologists try to find out what influences a nonhuman animal’s choice of where to forage, rest, or reproduce (Krebs and Davies 1993).


Making choices requires agency, that is, the capacity of actors to make decisions. That capacity does not necessarily imply consciousness of the choice process on the part of the decisionmaker. Behavioral ecologists, for example, study nonhuman animals’ choices but do not make any assumptions about the mental processes that give rise to these behaviors. Choice is also a central concern in criminology, victimology, and criminal justice research. The following questions provide some examples. How do judges or juries decide on whether a suspect is guilty or not, and on the type and the severity of a sanction? What makes people decide in favor of perpetrating crime? What makes them select a particular victim? This entry is about crime location choice, which addresses the questions where offenders go to commit crime and why they go there instead of somewhere else. The aim of the chapter is to provide a concise but complete discussion of the various issues involved in using the discrete choice framework to understand the spatial decision making of criminals. The remainder of this text first discusses how the issues of crime location choice have been addressed in the literature. It subsequently outlines random utility maximization theory and the discrete choice framework. The section that follows discusses what has been learned substantively from recent applications of the discrete choice framework to crime location choice. The final section discusses pending issues that have yet to be resolved and most of which require additional empirical research.

Different Approaches to Study Crime Location Choice Crime location choice addresses the questions where offenders commit crime, and why there instead of somewhere else. Before the discrete choice framework was introduced in the geography of crime, there were three separate approaches to the study of crime location choice (Bernasco and Nieuwbeerta 2005). Each of these approaches is characterized by a specific unit of analysis and a specific dependent variable.


The offender-based approach uses either the offender or the offense as the unit of analysis. The dependent variable is the length of the journey to crime, which has always been operationalized as the distance between the home of the perpetrator and the place where the offense was perpetrated. The most basic question answered is the question how far from home offenders perpetrate their offenses. A somewhat more complicated question is whether and how steep the offending intensity decreases with the distance from home. More generally, this question addresses the form of the distance decay function, because it has also been claimed that this distance function is upward sloping near the offender’s home (within a “buffer zone”) and only starts to slope downwardly at greater distance. Still more complicated questions compare the distance between different types of offenders and different types of offenses. Empirical research has shown that most offenders commit their offenses rather close to their homes, that there is distance decay, and that the average distance traveled varies across offense types and across offender types. The theoretical basis of the offender-based approach is small. It starts from the assumption that travel is costly and that for that reason offenders prefer to travel as little as possible to the locations of their offenses. Because this implies that most offenders would perpetrate crime on their doorsteps, and this appears not to be the case, an additional hypothesis has been introduced which states that offenders do not commit crimes close to their own homes for fear of recognition by victims or bystanders. The offender-based approach is limited for establishing where offenders perpetrate crime and for understanding that choice. It is limited because at any given distance, there are a number of alternative locations (all located on the circle around the offender’s home) that can be selected, and the offender-based model does not provide any further help in understanding the actual location that was chosen. It is also limited because it assumes that distance (and thus travel cost) is the main criterion that determines where offenders perpetrate crime (in addition, offenders would avoid perpetrating very close to their homes for

693

C

fear of being recognized by victims or bystanders). Thus, the offender-based approach ignores that location choices may be driven by geographic variations in the availability of criminal opportunities (i.e., the presence of suitable victims and targets and the absence of formal and informal guardians) or by characteristics of the offender (e.g., awareness space). The target-based approach uses potential target locations as units of analysis. The dependent variable is the crime rate or the victimization rate at the target location. This approach actually includes the bulk of studies in geographic criminology in which the crime locations are aggregated to larger areas and related to the characteristics of these aggregated areas, in particular characteristics that signal the presence of suitable victims and targets (e.g., entertainment areas or shopping centers) and the absence of formal and informal guardians (e.g., police presence and likelihood of social control exercised by residents). Thus, while the target-based approach does analyze crime location as a choice outcome, it typically does not consider offender mobility and thus ignores that some areas are vulnerable because they are more accessible to offenders than other areas that are identical in every other respect. For example, a shopping strip may be an ideal place for a late-night street robbery for motivated offenders who live nearby, but if the nearest motivated offender lives 5 miles away, it may not be a suitable location for any offender. The third approach is the mobility-based approach. The mobility-approach uses pairs of geographical locations as the units of analysis. The dependent variable is the number of crime trips from one of both locations to the other. This approach applies spatial interaction models, also known as gravity models (Haynes and Fotheringham 1984), to model volumes of offender travel within and between the neighborhoods, traffic zones, or other geographic entities in the study area (Elffers et al. 2008; Reynald et al. 2008; Smith 1976). These studies estimate regression models of trips from a particular zone of origin to a particular zone of destination. The variables include the number of observed trips, which is the dependent variable, and crime

C

C

694

attractors (services and people that attract criminals from elsewhere to the destination), crime producers (services and people that produce criminal motivation in the origin), and impedance variables. Impedance variables measure the “friction” or obstacles that must be overcome to move from the origin to the destination, including the distance, the travel time, or the cultural or ethnic similarity between the origin and the destination. In the mobility-based approach, distance and other impedance indicators are accounted for as choice criteria: They are independent variables that determine, among other things, how likely an offender from a particular origin location is to perpetrate a crime in a particular destination location. Spatial interaction models analyze aggregate flows of crime, but are not able to appropriately model variation across categories of offenders or among individual offenders. For example, to answer the question whether distance differently affects the crime location choice of juveniles and adults, one would have to estimate separate models for juveniles and adults. This is not a tractable option when the analysis is to include more than just a few categories. The discrete choice framework constitutes the fourth approach to the issue of crime location choice. The dependent variable in this approach is the choice outcome, that is, which alternative of a countable set of alternative locations does the offender select, and the unit of analysis is the individual decision-maker. There have thus far been published eight empirical studies that use the discrete choice framework to analyze crime location choice, including Bernasco and Nieuwbeerta (2005); Bernasco (2006); Clare et al. (2009); Bernasco and Block (2009); Bernasco (2010a); Bernasco (2010b); Bernasco and Kooistra (2010); and Bernasco et al. (2012).

Random Utility Maximization and Discrete Choice In many disciplines, the discrete choice framework (Ben-Akiva and Lerman 1985) has emerged as a powerful and elegant approach to


theorize about choice, to statistically model it, and to predict its outcomes. The discrete choice framework is a combination of random utility maximization (RUM) theory and a statistical model of the family of discrete choice models (e.g., conditional logit, multinomial logit, nested logit, and mixed logit models). The models can be used without RUM theory for other purposes than choice modeling, but the major strength of the discrete choice model is that the statistical model is directly derived from RUM theory. Because the theory is a formal (mathematical) theory, its predictions are quite precise and can be tested quantitatively. The intimate link between the theory and the statistical model makes it relatively straightforward to test new elements and conditioning clauses when they are added to the theory. While these features may sound like obvious requirements, most theories in the social sciences are informal and are only loosely linked to the statistical models that are used to test them. The discrete choice framework is a set of assumptions and methods to model a decisionmaker’s choice among a set of alternatives that are mutually exclusive and collectively exhaustive. This means that the decision-maker must select exactly one alternative, and that the alternatives do not overlap, so that by choosing an alternative all other alternatives become unavailable. Most of the assumptions in the discrete choice framework are based on RUM theory (McFadden 1973). RUM theory is the microeconomic theory of behavior in which a random component is added to the utility function. The random component represents incomplete information on the part of the analyst, not on the part of the decision-maker. By making certain assumptions about the distribution of this random component, the theory is directly linked to a statistical model that allows probabilistic statements to be formulated and tested. The discrete choice framework was developed in the 1970s by McFadden and others working in the field of travel demand, and the first applications of discrete choice were in the study of travel mode choice (i.e., the choice between train, bus,


695

car, or airplane). Later, the model was also applied to the choice of a travel routes and travel destinations (Ben-Akiva and Lerman 1985). The discrete choice framework defines four elements of a choice situation (Ben-Akiva and Bierlaire 1999): 1. Decision-makers. The decision-maker is the person or agent that makes a choice. 2. Alternatives. The decision-maker must choose one alternative from the choice set, that is, the set of available alternatives that are mutually exclusive and collectively exhaustive. 3. Attributes. Alternatives have attributes that affect the utility that the decision-maker derives from them when they are chosen. The decision-maker evaluates the utility of all alternatives. The decision-makers themselves can also have attributes that may affect the utility they derive from the alternatives. 4. Decision rule. According to RUM theory, the decision-maker chooses the alternative that maximizes (expected) utility (net gain, profits, satisfaction) when chosen. In further discussing discrete choice modeling, the notation of Train (2009) is followed. For stylistic reasons and clarity of argument, the decision-maker is referred to as a “he” and the researcher is referred to as a “she.” A decision-maker, labeled n, must make a choice among the J alternatives in the choice set. Decision-maker n obtains a level of utility (profits, satisfaction) Uni from alternative i if that alternative is chosen. The decision rule of utility maximization theory asserts that the decisionmaker decides in favor of the alternative i if and only if he expects to derive more utility from alternative i than from any other available alternative. Thus, if he decides in favor of alternative i, he must expect to derive less utility from each of the other alternatives. Uni > Unj 8j 6¼ i:

(1)

The utilities are assumed to be known by the decision-maker, but not by the researcher. The researcher only observes the actual chosen alternative i, the set of J alternatives, some attributes

C

ani of the alternatives, and some attributes dn of the decision-maker; and she can specify a function V, often called representative utility or systematic utility, that links these observed attributes to the decision-maker’s utility: Vni ¼ Vðani ; dn Þ8i

(2)

The researcher incompletely observes utility, so that generally Unj 6¼ Vnj . The utility can be written as the sum of representative utility Vnj and a term enj that captures the factors that determine utility but are not observed by the researcher and that is treated as random. Uni ¼ Vni þ eni

(3)

The probability Pni that decision-maker n chooses alternative i is the probability that the utility associated with choosing i is greater than the utility associated with any other alternative in the choice set: Pni ¼ PrðUni > Unj 8j 6¼ iÞ

(4)

Substituting Eq. 3 in Eq. 4 yields Pni ¼ PrðVni þ eni > Vnj þ enj 8j 6¼ iÞ Pni ¼ Prðenj eni < Vni Vnj 8j 6¼ iÞ

(5)

This is the most general formulation of the discrete choice model. Statistical models that implement this include not only the workhorses of this family, the conditional and the multinomial logit models, as special instances, but also various others, such as nested logit, mixed logit, and multinomial probit. Details of these models are described in Train (2009). All applications of the discrete choice model to crime location choice have thus far used the conditional logit model, which is also referred to as “the (multinomial) logit model with variables that vary over alternatives.” This model assumes that the unobserved utility term eni follows an extreme

C

C

696


value type 1 distribution, from which the following formula for the probability that decisionmaker n chooses alternative i can be derived. Pni ¼

eVni J P eVnj

(6)

j¼1

This formula relates the utility derived the alternative that is chosen (the numerator) to the total utility of all alternatives in the choice set (the denominator). For computational convenience, representative utility Vni is usually assumed to be linear in the K parameters: Vni ¼ b0 Xni ¼

K X

bk Xkni

(7)

k¼1

Application of Discrete Choice Models to Crime Location Choice The application of discrete choice models to crime location choice implies that the four elements of choice situations are specified. A collection of potential crime locations constitutes the choice set. In most applications of the model thus far, this collection of locations is formed by the set of “neighborhoods” in a single city or region. Thus, the problem addressed can be phrased as “given that an offender will commit an offense in the study area, in which of the neighborhoods in the area will s/he commit it?” Substantive theory is needed to specify what the characteristics of potential targets are that make them attractive to offenders. Thus far, the discrete spatial choice model has been applied mostly to burglary and to robbery. Studies on burglary include Bernasco and Nieuwbeerta (2005); Bernasco (2006) and Bernasco (Bernasco 2010a) in The Hague, the Netherlands; and Clare, Fernandez, and Morgan (Clare et al. 2009) in Perth, Australia. Bernasco (2010a)

and Bernasco and Kooistra (2010) studied commercial robbery in the Netherlands, and Bernasco and Block (2009) and Bernasco et al. (2013) studied street robbery in Chicago, USA. Bernasco (2010b) studied burglary robbery, theft from vehicle, and assault. Target Characteristics The discrete choice framework was introduced in criminology by Bernasco and Nieuwbeerta, who used data on 548 burglaries committed by solitary (non-co-offending) offenders in any of the 89 neighborhoods of the city of The Hague, the Netherlands, by offenders who were themselves residents of the city. They argued that burglars would be attracted to targets located near their homes (indicated by distance and by distance to the city center) that were affluent (indicated by property value) and accessible (indicated by percentage single-family dwellings) and where they were least likely to be disturbed by residents or bystanders (indicated by low residential mobility and low ethnic heterogeneity). They found that offenders were attracted to neighborhoods located near their own homes and near the city center, to ethnically mixed neighborhoods, and to neighborhoods with a relatively high percentage of singlefamily dwellings. Bernasco (2006) extended this analysis by adding data on co-offender groups, by covering a longer period, and by using factor analysis to group neighborhood characteristics into the sets of variables indicating affluence, physical accessibility, and social disorganization. Only the physical accessibility of neighborhoods was found to significantly increase their odds of being targeted. The spatial anchor point of offenders structures their choices because it defines the distances that they must overcome to reach target locations. Because offenders live in different locations, the distances between their homes and potential targets vary between potential targets (for the same offender) as well as between offenders (for the same potential target). An important difference between journey-tocrime research and criminal location choice


research is that in the former, variation in distance to crime is the explanandum; distance is the dependent variable. In criminal location choice, distance is part of the explanans; the distance to a potential target is one of the arguments that offenders consider when they decide on where to commit an offense. Barriers The findings on burglary in The Hague mentioned above already indicate that distance plays an important role in criminal location choices. Distance is actually a poor measure of the impedance encountered when traveling between two places. Arguably, better measures include monetary cost, travel time, and lack of comfort. Studying 1761 burglaries in 292 suburbs in Perth, Australia, Clare et al. (2009) demonstrated that physical barriers (such as rivers or highways that form obstacles for cross-travel) and connectors (such as public transport lines) affect offenders’ journeys to crime. They also found that the influence of impermeable barriers increases with proximity of these obstacles to the offender’s point of origin. More abstract barriers to mobility are social in nature. A social barrier deters offenders to commit crime in a potential target area because the potential target area is socially (culturally, ethnically, economically) different from the offender or from the area where the offender lives. A study on Chicago robbers (Bernasco and Block 2009) demonstrated that, controlling for distance from home, offenders displayed a strong tendency to commit robberies in places where the majority of the population was of their own ethnic category. There are two plausible reasons for this finding. First, because birds of a feather flock together, ethnic minorities tend to live together in areas or congregate elsewhere. For this reason, most individuals spend a considerable amount of time in the proximity of their peers, and it is only logical that they commit crime around the places they normally visit. The second reason is that offenders like to disguise their motives and that they may be much less conspicuous in areas where people live who are like they are.

697

C

Awareness Space Crime pattern theory (Brantingham and Brantingham 2008) suggests that for an individual to commit a crime at a place, the place must not only offer criminal opportunities but must also be part of the individual’s awareness space, which is shaped by current and recent routine activities. Recent research utilized the discrete spatial choice model to investigate whether offenders are more likely to target areas near their former residences (Bernasco 2010b; Bernasco and Kooistra 2010). It was demonstrated that offenders were not only likely to offend near their current residence, it was also demonstrated that they were more likely to offend near former homes, especially if they had lived there for a long time and had moved away only recently. This type of idiosyncratic spatial knowledge need not be restricted to offender’s home locations but may include the homes of their friends and family members, as well as other major anchor points such as schools, workplaces, or entertainment and business areas they regularly visit during presumably legal activities. Offenders must also learn about the criminal opportunities of places where they commit crimes. Indeed, it has been suggested that the phenomena of repeat victimization and near repeat victimization (whereby in the wake of a crime, victimization risk is elevated in the vicinity of the crime site) may be explained by the postulated tendency of offenders to return to their prior targets (Bernasco 2008; Bowers and Johnson 2004). This tendency might be lower for locations where offenders were arrested.

Small Spatial Units of Analysis From geography, it is well known that physical and social processes are not constant across spatial scales. The finding that the outcomes of both descriptive and explanatory analyses strongly depend on the size and shape of the spatial units of analysis has been coined the modifiable areal unit problem (or, more neutrally, phenomenon) (Openshaw 1984). As most theories

C

C

698

of crime seem to focus on small spatial units, criminologists are urged to collect and analyze data on small spatial units (Weisburd et al. 2009). In spatial discrete choice, the MAUP has not been given much attention, and studies of location choice, whether in human migration, recreation, or firm location, have typically based the scale decision on convenience and data availability. In crime location research, Bernasco (2010a) and Bernasco et al. (2013) discuss the complexities involved in using small geographical units. The first challenge is mainly computational. Small spatial units of analysis typically imply a large or very large choice set, which may generate computational challenges because the matrix that is used in the maximum likelihood estimation of the parameters has N J records, where N is the number of decision-makers and J is the mean number of alternatives available per decision-maker. Bernasco et al. (2013) demonstrate that in their analyses of street robbery at census block level, they would have to join a set of 12,000 offenders with a set of 25,000 census blocks, yielding a matrix of 300 million records to be included in the maximum likelihood estimation procedures. Fortunately, the coefficients of a conditional logit model can be estimated consistently on a random subset of the alternatives (McFadden 1978). Therefore, sampling from alternatives has become an agreed standard in applications with large numbers of alternatives, including location choice. Guevara (2010) developed methods to sample from alternatives for more complex discrete choice models, such as nested logit. The second challenge is theoretical. Small spatial units imply that spatial spillover effects may be more salient, especially if spillover effects display strong distance decay. For example, a retail store may not only attract street robbers to the exact location where it is located but also to blocks across the street and around the corner. A robber may follow a customer from a store in one block to a more isolated place in an adjacent block before he attacks, or wait in ambush around the corner. In analyses at higher levels of aggregation, such as census block


groups, census tracts, or neighborhoods, such short-distance effects are much weaker because the large majority of the short-distance spillover takes place within the boundaries of the larger geographical units.

Related Entries ▶ Agent-Based Models to Predict Crime at Places ▶ Co-offending ▶ Crime Mapping ▶ Crime Science ▶ Econometrics of Crime ▶ History of Geographic Criminology Part I: Nineteenth Century ▶ Rational Choice Theory

Recommended Reading and References Ben-Akiva ME, Bierlaire M (1999) Discrete choice methods and their applications to short term travel decisions. In: Hall RW (ed) Handbook of transportation science. Kluwer, Norwell, MA, pp. 5–34 Ben-Akiva ME, Lerman SR (1985) Discrete choice analysis: theory and applications to travel demand. MIT Press, Cambridge, MA Bernasco W (2006) Co-offending and the choice of target areas in burglary. J Invest Psychol Offend Profil 3:139–155 Bernasco W (2008) Them again? same offender involvement in repeat and near repeat burglaries. Eur J Criminol 5(4):411–431 Bernasco W (2010a) Modeling micro-level crime location choice: application of the discrete choice framework to crime at places. J Quant Criminol 26(1):113–138 Bernasco W (2010b) A sentimental journey to crime; effects of residential history on crime location choice. Criminology 48:389–416 Bernasco W, Block R (2009) Where offenders choose to attack: a discrete choice model of robberies in Chicago. Criminology 47(1):93–130 Bernasco W, Kooistra T (2010) Effects of residential history on commercial robbers’ crime location choices. Eur J Criminol 7(4):251–265 Bernasco W, Nieuwbeerta P (2005) How do residential burglars select target areas? a new approach to the analysis of criminal location choice. British J Criminol 45:296–315 Bernasco W, Block R, Ruiter S (2013) Go where the money is: modeling street robbers’ location choices. J Econ Geogr 13(1):119–143. doi:10.1093/jeg/lbs005

Crime Mapping Bowers KJ, Johnson SD (2004) Who commits near repeats? a test of the boost explanation. Western Criminol Rev 5(3):12–24 Brantingham PJ, Brantingham PL (2008) Crime pattern theory. In: Wortley R, Mazerolle L, Rombouts S (eds) Environmental criminology and crime analysis. Willan, Cullompton, pp 78–93 Clare J, Fernandez J, Morgan F (2009) Formal evaluation of the impact of barriers and connectors on residential Burglars’ macro-level offending location choices. Aust N Z J Criminol 42:139–158 Elffers H, Reynald D, Averdijk M, Bernasco W, Block R (2008) Modelling crime flow between neighbourhoods in terms of distance and of intervening opportunities. Crime Prevent Commun Safety 10:85–96 Guevara CA (2010) Endogeneity and sampling of alternatives in spatial choice models. PhD Thesis. MIT, Department of Civil and Environmental Engineering, Cambridge, MA Haynes KA, Fotheringham AS (1984) Gravity and spatial interaction models. Sage, Beverly Hills Jepsen L, Jepsen C (2002) An empirical analysis of the matching patterns of same-sex and opposite-sex couples. Demography 39(3):435–453 Krebs JR, Davies NB (1993) An introduction to behavioural ecology, 3rd edn. Blackwell, Oxford McFadden D (1973) Conditional logit analysis of qualitative choice behavior. In: Zarembka P (ed) Frontiers in econometrics. Academic Press, New York, NY, pp. 105–142 McFadden D (1978) Modeling the choice of residential location. In: Karlkvist A, Lundkvist L, Snikars F, Weibull J (eds) Spatial interaction theory and planning models. North-Holland, Amsterdam, pp 75–96 McFadden D (1980) Econometric models for probabilistic choice among products. J Business 53(3): S13–S29 Openshaw S (1984) The modifiable areal unit problem. Geo Books, Norwich Palfrey TR, Poole KT (1987) The relationship between information, ideology, and voting behavior. Am J Polit Sci 31(3):511–530 Reynald DM, Averdijk M, Elffers H, Bernasco W (2008) Do social barriers affect urban crime trips? the effects of ethnic and economic neighbourhood compositions on the flow of crime in the Hague, the Netherlands. Built Environ 34:21–31 Smith TS (1976) Inverse distance variations for the flow of crime in urban areas. Soc Forces 54(4): 802–815 Train KE (1980) A structured logit model of auto ownership and mode choice. Rev Econ Stud 47(2): 357–370 Train KE (2009) Discrete choice methods with simulation. Cambridge University Press, New York, NY Weisburd D, Bernasco W, Bruinsma GJN (eds) (2009) Putting crime in its place: units of analysis in geographic criminology. Springer, New York

699

C

Crime Mapping Spencer Chainey Jill Dando Institute of Security and Crime Science, University College London, London, UK

Synonyms Geographical analysis; Geography of crime

Overview Crime mapping is the direct application that comes from considering two particular characteristics about crime – it has an inherent geographical quality and crime does not happen randomly. That is, all crime occurs at some location, and it is the features of this location and the nature of the crime that has been committed, including the decision-making of an offender and their interaction with a victim (or other target such as a building), that make the study of its geography useful for law enforcement and public safety. This can include responding to calls for service, identifying crime hot spots, and supporting the investigation of a crime series. This section begins by discussing the fundamental theoretical principles that underpin crime mapping, offers a brief history of its development, and then illustrates examples of its application. The final section identifies future directions for its development.

Theoretical Fundamentals The inherent geographical quality of crime can be observed in many theoretical forms. For example, the four dimensions of crime state that for a crime to occur it must have a legal dimension (a law must be broken), a victim dimension (someone or something has to be targeted), an offender dimension (someone has to do the crime), and a spatial dimension (it has to happen somewhere) (Brantingham and Brantingham 1981). This

C

C

700

therefore highlights the value that an assessment of the location of the crime can play, alongside the study of the offender and their behavior, and the vulnerability of the person or property that was targeted. The nonrandom geographical nature to crime can also be explained theoretically by drawing from Routine Activity Theory, Rational Choice Theory, Crime Pattern Theory, and principles of least effort. Routine Activity Theory (Cohen and Felson 1979) states that for a crime to occur, three components must be present. There must be the presence of a likely offender, the presence of a suitable target, and the absence of a capable guardian. These three components must meet in time and space, to formulate the necessary “chemistry” for crime (Felson 1998). Routine Activity Theory therefore states that the risk of crime changes over time with the movement of people throughout their daily routine activities, meaning that the geography relating to these activities helps to determine features relating to victim vulnerability and offender motivation. Most researchers take the existence of likely offenders as a constant in our society and attempt to explain and prevent crime by examining the two remaining components. Victim and target suitability can change over space and time, and the possibilities for introducing suitable guardianship into an area as a way to prevent crime, or to make existing guardians more suitable, offer a means for considering ways for reducing crime opportunities. The consideration of these three components of crime has recast Routine Activity Theory into the Crime Problem Analysis Triangle (Eck 1995) – stating that all crime problems have an offender, victim, and a place component (incorporating both the geographical location and temporal features of the crime), with the notion of guardianship in this model acting as a handler to help reduce the vulnerability of the victim or target. Rational Choice Theory (Clarke and Felson 1993) provides a framework to consider offender decision-making (and its relationship with space) when a crime opportunity is presented. That is, the commission of crime by a person is a (fairly) rational decision, with the aim being to achieve

Crime Mapping

some sort of desire or goal. This goal may be to derive personal financial gain (e.g., as in a theft of cash), personal pleasure (e.g., stealing a car for joyriding), or through the imposition of power (e.g., through the extension of bullying by stealing possessions off a person or through violence). This criminal decision-making is considered to involve two parts. There is a long-term, multistage decision to become generally involved in criminal activity (criminal involvement decision) and a shorter-term, more immediate decision (the criminal event decision) to grasp an opportunity that is presented (Cornish and Clarke 1986). In a geographical sense, factors relating to social and economic disadvantage and social disorganization (Burgess 1925; Shaw and McKay 1942) may have a role to play in the long-term decision-making process, whereas the immediate decision-making of an offender is primarily based on the environmental cues from the prospective target that can change from place to place. Crime Pattern Theory provides a useful convergence of Routine Activity Theory and Rational Choice Theory by helping to explain the geographical behavior of an offender, their interaction with potential targets, and offense distribution. That is, it helps examine the “relationship of the offence to the offender’s habitual use of space” (Bottoms and Wiles 2002, p. 638). Crime Pattern Theory suggests that offenders are influenced by the daily activities and routines of their lives, so that even if they are searching for a criminal opportunity, they will tend to steer toward areas that are known to them (Brantingham and Brantingham 1984). In their day-to-day activities, they will be watching for targets that have no guardians or place managers (Eck 1995). Crime Pattern Theory considers the notion of awareness space (the areas that are familiar, through carrying out routine activities in these areas), opportunity space (those places which offer a target-rich environment, i.e., where there is a concentration of suitable targets), and the places where these two overlap. This is illustrated in Fig. 1, showing a hypothetical model of an offender’s awareness space (consisting mainly of the places they

Crime Mapping

Crime Mapping, Fig. 1 A hypothetical model of the creation of criminal occurrence space where offender awareness space and opportunities coincide (Source: Chainey and Ratcliffe 2005)

routinely frequent and the routes between these places) and the overlap with opportunities for crime. For this offender, they are most likely to commit crime in those locations where their awareness space and the opportunity space overlap. The opportunity space extends to areas outside of their awareness, but for this offender, crime that they commit is unlikely in these areas because they are not familiar with the opportunities, nor comfortable with the higher risk taking that may be involved (i.e., lack of familiarity of escape routes and a higher perceived risk of being caught). The final fundamental theoretical consideration behind crime mapping is the least effort principle. This states that most journeys we take in our daily activities are short and that the frequency of these activities decays with distance. That is, while we may make long distances to carry out activities, the majority of things we do involve us seeking out trips that minimize this distance. For instance, we may need a carton of milk. While there are many places that sell milk, most people will purchase milk by going to a local store, rather than expending additional effort and travelling to a store some additional distance away. This least effort principle is also evident with offending behavior. Research into offenders’ journeys to crime shows that the majority of trips are short (within a few miles) (see Rossmo 2000

701

C

for a comprehensive review of journey to crime literature). This concept is also captured in Fig. 1 through the illustration of boundaries to the awareness space. If there was no concept of the least effort principle, the awareness space would extend to infinity in all directions. These theoretical principles provide the backdrop to the application of crime mapping. However, while these theories help to conceptualize the spatial behavior and interaction between offenders and their targets, there will always be exceptions, e.g., an offender may not commit crime within their own awareness space due to the influence of peers who introduce the offender to new areas. For the majority of offenders, broad behavioral tendencies do exist and can be explained as aggregate criminal spatial behavior. Crime mapping therefore draws from geographical and environmental criminology concepts to help explain why crime occurs, and makes use of geospatial tools (such as Geographical Information Systems – GIS) to capture, analyze, and visually interpret these geographical qualities. Examples of its application are provided in a following section. Before this a short history of GIS and crime mapping is provided to help capture how its use has evolved.

GIS and a Brief History of Crime Mapping A GIS is “a computer system for capturing, managing, integrating, manipulating, analysing and displaying data which is spatially referenced to the Earth” (McDonnell and Kemp 1995, p. 42). A GIS provides the computing environment within which crime data can be layered with base maps and other geographic data that relate to the landscape and conditions of the area under study. This other geographic information may include population data from a census, the locations of bars and nightclubs, or data describing the local land use patterns. These data can be represented as individual layers that can be manipulated, analyzed, or displayed as separate entities, or could be combined with other layers to be displayed together, integrated to provide a new

C

C

702

perspective of the area that they represent, or analyzed against each other to explore relationships. GIS began to emerge as a discipline in the 1960s, with its origins being in the development of land use applications in Canada, growing then through applications that included the automation of the cartographic drafting process used by mapping agencies such as the British Ordnance Survey and acting as a platform to aid military intelligence gathering, display and analysis from satellite, and other remote sensing imagery (e.g., aerial photography and radar). The use of GIS in policing did not begin to emerge until the 1980s, in large part due to computer hardware prices being too prohibitively expensive before this time, the lack of suitable GIS software that could be used to support policing requirements, and the lack of electronic recording of police records. Even then, it took until the late 1990s for the use of GIS in policing to gather pace, mainly due to being held back by organizational and management problems, issues with sharing information, technical problems, and geocoding issues (see Chainey and Ratcliffe 2005 for a more detailed assessment). Many of these problems are still apparent today, but through developments in the practical use of GIS, several innovators have identified how these can be overcome, and have shared this knowledge across a network of users who engage via forums, conferences, training courses, and publications that since 1998 have begun to capture evidence of how crime mapping can be effectively used to support policing and crime prevention.

Crime Mapping Applications Crime mapping contributes to policing, law enforcement, and crime reduction by helping to generate a real understanding of criminal activity and the direction in tackling it (Chainey and Ratcliffe 2005). The further reading section provides details of many examples that have captured its operational use – both for tactical and strategic purposes. In this section, examples

Crime Mapping

of its use are provided relating to several of the main operational functions in law enforcement and public safety: • Performance and accountability – CompStat • Response – directing calls for service • Prevention – the analysis of crime hot spots • Detection – geographic profiling • Evaluation – through the use of analysis that explores the impact, displacement, and diffusion of benefit from a geographically targeted policing or crime prevention initiative. Performance and Accountability: CompStat CompStat (short for Computer Statistics) is an operational management process that combines computer technology, operational strategy, and managerial accountability to determine the provision of crime reduction policing (Walsh 2001). From its early origins with the New York City Police Department, the CompStat idea has been adopted in many police agencies across the world. Figure 2 captures the CompStat process in operation at Thames Valley Police Force in England. The basic aim of CompStat is to provide detailed intelligence to operational commanders on a regular basis so that they can determine an appropriate crime reduction and law enforcement strategy and then subsequently be held accountable for the success of the strategy. The computer system to analyze and map crime patterns is at the core of the CompStat process (Walsh 2001, p. 352). Crime maps form a useful basis around which those in attendance can visualize operational decisions, resource allocation, and managerial accountability. This can include the use of live and interactive crime mapping as a way to help focus and direct discussions around particular problems, deployed tactics, and performance in relation to the current local crime reduction priorities. Response: The Use of GIS for Calls for Service Pinpointing the location of an incident and ensuring that it is identified with minimal ambiguity is vital for a quick and direct response. Many police agencies use gazetteers that are integrated into

Crime Mapping

703

C

Crime Mapping, Fig. 2 Thames Valley Police (England) CompStat meeting

C

their mapping systems to search for and view the location of an incident when a call comes in to their command and control center. Increasingly, police forces are also utilizing Global Positioning Systems (GPS) to provide a real-time link to the command center to show the location and details of response teams that are on patrol (e.g., see Home Office 2005). The use of mobile computing technology also helps police agencies organize responses to incidents. In-car computer technology has become the norm in law enforcement agencies in the United States, permitting live access to central databases, allowing for the display of recent events, and permitting the recording of details about incidents while on out on patrol. Crime Prevention: Hot Spot Analysis Hot spot mapping is a popular analytical technique used by law enforcement, police, and crime reduction agencies to visually identify where crime is highest, aiding decision-making that determines where to target and deploy resources. Its application has been used to support the operational briefing of police patrols (Home Office 2005; LaVigne and Wartell 2000; Harries 1999), informs the creation of intelligence products and problem solving (Home

Office 2005; Chainey and Ratcliffe 2005), has been used as a tool to capture the measurement and analysis of crime patterns for crime auditing purposes in the UK’s crime reduction partnerships (Chainey and Ratcliffe 2005; Chainey 2001), and supports performance analysis (e.g., see Chainey and Ratcliffe 2005; Home Office 2005; Walsh 2001). In essence, hot spot mapping is a technique that is used to help determine where crime may happen next, using data from the past to inform future actions. In this sense it acts as a basic technique for predicting where crime may occur, using the premise that retrospective patterns of crime are a useful indicator for future patterns. There are many different mapping techniques that can be used for identifying and exploring patterns of crime. Attempting to do this via point mapping has become outdated since the proliferation of GIS software and the increasing sophistication of mapping techniques. Three of the most common techniques include the following: • Thematic mapping of geographic administrative areas (also referred to as choropleth mapping): this approach uses geographic units that are defined for administrative or political use (e.g., police beats, census blocks, wards, or

C

704

districts). Crimes that are geographically referenced as points can be aggregated to these geographic unit areas and then shaded in accordance with the number of crimes that fall within them. These types of maps are quick to produce and allow for additional information relating to the geographic unit (such as population) to be linked to explore relationships or calculate a crime rate – increasing their versatility for analysis. However, the varying size and shape of most geographical boundaries can lead the map reader to misinterpret the true distribution of crime due to issues associated with the distribution of crime across the geographical division of boundary areas (Chainey and Ratcliffe 2005) and due to issues associated with the Modifiable Areal Unit Problem (MAUP). This is where changes in the boundaries themselves can directly affect the patterns shown on the map. Thematic mapping of boundary areas continues to see widespread application, from being used for comparing the different volumes of unique and repeat burglaries across a study area’s census zones (Ratcliffe and McCullagh 2001) to being used for comparing vehicle theft in relation to land use in Overland Park, Kansas (Harries 1999, p. 47). An example of a thematic map, generated for census areas, is shown in Fig. 3a. • Grid thematic mapping: in order to address the problems associated with different sizes and shapes of geographical regions, uniform grids can be drawn in a GIS and used as the unit of aggregation for point geo-referenced data (see Fig. 3b). This approach does have some limitations (e.g., it suffers from the same MAUP problems outlined above), but its use has been illustrated for identifying vulnerable residences where target hardening was then implemented (Bowers et al. 2001) and for mapping the volume of emergency calls and violent offenses per square mile in North Carolina (LeBeau 2001). • Kernel density estimation: this technique has become increasingly popular due to its now wide availability in standard GIS software products, its accuracy for hot spot

Crime Mapping

identification (Chainey et al. 2008), and the aesthetic look of the resulting map in comparison to other techniques (Eck et al. 2005). Point data (offenses) are aggregated within a user-specified search radius and a continuous surface that represents the density of crime events is calculated. A smooth surface map is produced, showing the variation of the crime density across the study area, with no need to conform to geometric shapes such grid units. Examples of the use of KDE are now widespread, with popular crime mapping texts showing its use for a variety of applications (see Chainey and Ratcliffe 2005; Eck et al. 2005; Harries 1999). An example of a KDE crime hot spot map is shown in Fig. 3c. The use of hot spot analysis continues to advance, particularly in terms of exploring how geographic information that represents the underlying causal factors of crime can be modeled alongside recorded incidents of crime. Risk Terrain Modeling is one of these techniques and involves combining separate layers of risk variables (e.g., transport termini, schools, bars, and nightclubs) to produce a composite map showing the presence, absence, or intensity of all risk factors across a study area (Kennedy et al. 2011). Detection: Geographic Profiling Geographic profiling is an investigative technique that uses the locations of a linked series of crimes to determine the most probable area of offender residence (Rossmo 2000). Geographic profiling is a method of strategic information management that prioritizes the offender search efforts to maximize the efficiency of serial crime investigations. Until recently, this technique’s application has very much focused on serious violent or sexual crimes such as rape and murder. However, it is also now more widely applied to nonviolent, high-volume crime (e.g., criminal damage and burglary) where a series is evident. Geographic profiling is based on research from the fields of criminology, geography, forensic and environmental psychology, mathematical modeling, statistical analysis, and criminal investigation.

Crime Mapping

705

C

Crime Mapping, Fig. 3 Common hot spot mapping techniques. (a) Thematic mapping of administrative units, (b) grid thematic mapping, and (c) kernel density estimation

In a criminal context, each site associated with a single offender (e.g., victim encounter site and attack site) represents a location that is most likely to be present within the offender’s awareness

C

space. Using a series of sites linked to the same offender, a probability map can be constructed that predicts the location of an offender’s anchor point – usually the offender’s residence.

C

706

A geographic profile is constructed using a mathematical algorithm that models the offender’s likely journey to crime distances and the relationship between crime location sets and offender residence. The process assigns scores to the various points on a map that represent the offender’s hunting area. These scores are based on the crime site locations and certainty in the linkages between crimes. The map output of a geographic profile identifies the most probable area of offender residence. The geographic profile can then be used to help recommend investigative strategies that will support the detection of the offender. This could include suspect prioritization based on the cross-referencing of the geographic profile with suspect lists, the targeting of DNA swabbing, or patrol saturation and operational stakeouts. Daniell (2008) provides an example that illustrates the use of geographic profiling that led to catching a serial sex offender in Bath, England. Evaluation Many crime prevention initiatives and police operations are area-based: resources are targeted to a particular area in an attempt to deal with its crime problems. Crime mapping offers a useful tool to help monitor if the initiative or operation has been a success by analyzing the before and after picture of the crime levels in the area. With the use of techniques such as the Weighted Displacement Quotient (Bowers and Johnson 2003), an analysis can also be conducted to discover if there has been any displacement or diffusion of benefit as a result of an area-based targeted initiative. For example, in an analysis of alley-gating schemes in Merseyside, England (involving the installation of security gates across alleyways or footpaths that run behind properties), Bowers et al. (2004) used concentric buffer rings around the alley-gating areas to explore if and how far any burglary was displaced (see Fig. 4). The results showed there to be a diffusion of benefit to properties in the surrounding areas. The evaluation also enabled a rudimentary cost-benefit analysis to be calculated, showing that in the areas where the gates had been installed, the

Crime Mapping

cost-benefit ratio was around 1.86 (i.e., a saving of £1.86 for every pound spent).

Future Directions Advances in computing technology come with it opportunities to apply more sophisticated analysis of geographical patterns and apply new technology that supports improvements in policing, law enforcement, and crime prevention. For example, greater use of Internet-based crime mapping tools can help support the publication of neighborhood-level crime statistics to the public to provide positive messages of reassurance and alert residents to spates of crime in their neighborhood that may place them at a heightened risk of being victimized and the precautions they can take to minimize this risk (Chainey and Tompson 2012). To date, geographical analysis of crime patterns in policing has focused on the generation of descriptive material that helps to identify geographical patterns of crime (such as crime hot spots), but has yet to fully embrace more sophisticated forms of geographical information analysis. New developments that are expected in crime mapping include the following: • Analysis of the significance of geographical patterns – spatial significance testing identifies geographical patterns that are considered to be particularly unusual by applying the principles of significance testing (e.g., see Chainey 2011). • A harmonization of the geographical and temporal exploration of crime, so that each is explored together and as a continuum rather than in isolation. • The development of local spatial regression techniques to help examine relationships between crime patterns and other variables – to explain why crime occurs at certain places. • The development of techniques that predict where and when crime will occur in the future. • Spatial modeling processes that test “what if” scenarios such as examining the impact that a crime prevention initiative may have on crime levels (e.g., see Birks et al. 2012).

Crime Mapping

707

C

C

3,500

1,750

0

3,500

7,000

10,000

14,000 Meters

Legend Areas with alleygates N

Merseyside police beats Buffer zones 0.00–0.20

0.40–0.60

0.80–1.00

0.20–0.40

0.60–0.80

1.00–1.20

1.20–1.40

Crime Mapping, Fig. 4 Alley-gated areas in Merseyside and concentric buffers used to detect displacement or diffusion of benefit (Source: Bowers et al. (2004))

Related Entries ▶ Agent-Based Models to Predict Crime at Places ▶ Applied Geographical Profiling ▶ Biological Geographical Profiling ▶ Compstat

▶ Crime Location Choice ▶ Crime Science ▶ Criminal Investigative Analysis ▶ Early Chicago School Theory ▶ Effectiveness of Situational Crime Prevention ▶ History of Geographic Criminology Part I: Nineteenth Century

C

708

▶ Hot Spots and Place-Based Policing ▶ How to Make Crime Mapping More Inferential ▶ Information Technology and Police Work ▶ Longitudinal Crime Trends at Places ▶ Longitudinal Studies in Criminology ▶ Near Repeats and Crime Forecasting ▶ Outdoor Serious Violence: The Role of Place ▶ Place-Based Randomized Trials ▶ Prediction and Crime Clusters ▶ Predictive Policing ▶ Problem-Oriented Policing ▶ Rational Choice Theory ▶ Routine Activities Approach ▶ Spatial Models and Network Analysis ▶ Spatial Perspectives on Illegal Drug Markets

Recommended Reading and References Birks D, Townsley M, Stewart A (2012) Generative explanations of crime: using simulation to test criminological theory. Criminology 50(1):221–225 Bottoms AE, Wiles P (2002) Environmental criminology. In: Maguire M, Morgan R, Reiner R (eds) The oxford handbook of criminology. Oxford University Press, London, pp 620–656 Bowers K, Johnson SD (2003) Measuring the geographical displacement and diffusion of benefit effects of crime prevention activity. J Quant Criminol 19(3): 275–301 Bowers K, Newton M, Nutter R (2001) A GIS-linked database for monitoring repeat domestic burglary. In: Hirschfield A, Bowers K (eds) Mapping and analysing crime data – lessons from research and practice. Taylor and Francis, London Bowers KJ, Johnson SD, Hirschfield AFG (2004) Closing off opportunities for crime: an evaluation of alley-gating. Eur J Criminal Policy Res 10:285–308 Brantingham PJ, Brantingham PL (1981) Environmental criminology. Waveland Press, Prospect Heights Brantingham PJ, Brantingham PL (1984) Patterns in crime. Macmillan, New York Burgess EW (1925) The growth of the city: an introduction to a research project. In: Park RE, Burgess EW, McKenzie RD (eds) The city. University of Chicago Press, Chicago, pp 47–62 Chainey SP (2001) Combating crime through partnership; examples of crime and disorder mapping solutions in London, UK. In: Hirschfield A, Bowers K (eds) Mapping and analysing crime data – lessons from research and practice. Taylor and Francis, London Chainey SP (2011) Advanced hotspot analysis – spatial significance mapping using the Gi* statistic. In: Presentation at the 2011 international crime and

Crime Mapping intelligence analysis conference. Available at: http:// www.ucl.ac.uk/jdi/events. Accessed 17 Apr 2012 Chainey SP, Ratcliffe JH (2005) GIS and crime mapping. Wiley, London Chainey SP, Tompson L (2012) Engagement, empowerment and transparency: publishing crime statistics using online crime mapping. Policing J Policy Pract Chainey SP, Tompson L, Uhlig S (2008) The utility of hotspot mapping for predicting spatial patterns of crime. Security J 21:4–28 Clarke RV, Felson M (eds) (1993) Routine activity and rational choice. Advances in criminological theory, vol 5. Transaction Books, New Brunswick, NJ Cohen LE, Felson M (1979) Social change and crime rate trends: a routine activity approach. Am Sociol Rev 44:588–608 Cornish D, Clarke R (1986) The reasoning criminal: rational choice perspectives on offending. Springer, New York Daniell C (2008) Geographic profiling in an operational setting: the challenges and practical considerations, with reference to a series of sexual assaults in Bath, England. In: Chainey SP, Tompson L (eds) Crime mapping case studies: practice and research. Wiley, Hoboken Eck J (1995) A general model of the geography of illicit retail marketplaces. In: Weisburd D, Eck JE (eds) Crime and place. Criminal Justice Press, New York, pp 67–93 Eck JE, Chainey SP, Cameron JG, Leitner M, Wilson RE (2005) Mapping crime: understanding hot spots. National Institute of Justice, USA. Available online at www.ojp.usdoj.gov/nij Felson M (1998) Crime and everyday life: impact and implications for society. Pine Forge Press, Thousand Oaks Harries K (1999) Mapping crime: principle and practice. United States National Institute of Justice, Washington, DC. Available online at http://www.ojp.usdoj. gov/nij/maps/pubs.html Home Office (2005) Crime mapping: improving performance. Home Office Police Standards Unit, London. http://www.ucl.ac.uk/scs/people/academicresearch-staff/spencer-chainey/Slides/Home_Office_ CrimeMapping. Accessed 17 Apr 2012 Kennedy LW, Caplan JM, Piza E (2011) Risk clusters, hotspots, and spatial intelligence: risk terrain modeling as an algorithm for police resource allocation strategies. J Quant Criminol 27(3):339–362 LaVigne N, Wartell J (eds) (2000) Crime mapping case studies: successes in the field, vol 2. Police Executive Research Forum, Washington, DC LeBeau JL (2001) Mapping out hazardous space for police work. In: Bowers K, Hirschfield A (eds) Mapping and analysing crime data – lessons from research and practice. Taylor and Francis, London McDonnell R, Kemp K (1995) International GIS Dictionary. GeoInformation International, Cambridge, UK

Crime on Public Transport Ratcliffe J, McCullagh M (2001) Crime, repeat victimisation and GIS. In: Bowers K, Hirschfield A (eds) Mapping and analysing crime data – lessons from research and practice. Taylor and Francis, London, pp 61–92 Rossmo DK (2000) Geographic profiling. CRC Press, Boca Raton Shaw CR, McKay HD (1942) Juvenile delinquency and urban areas. University of Chicago Press, Chicago Walsh WF (2001) Compstat: an analysis of an emerging police managerial paradigm. Policing An Int J Police Strat Manage 24(3):347–362

Crime on Public Transport Andrew Newton Applied Criminology Centre, University of Huddersfield, Huddersfield, UK

Synonyms Mass transit; Public transit; Public transportation; Rapid transit

Overview This entry examines how crime concentrates and is potentially specialized within and around the public transport system. It begins by defining public transport for the purposes of this discussion and outlines why this is an area of importance. It also discusses some of the reasons why there is a paucity of evidence into crime on public transport and the unique challenges this system poses for identifying and analyzing concentrations of crime. It describes how public transport networks provide a number of unique settings (places and times) across which crime and disorder can occur. These include in and around stations and stops and on board moving vehicles. The public transport network brings increased accessibility to places, and this creates distinctive patterns of offending. This entry reviews the international evidence for the manifestations of crime in these situations and considers the theoretical reasons and explanations for

709

C

such events and the extent to which these may be specialized. It then examines the prevention opportunities offered by the crime concentrations and specializations identified, and some emerging trends and future avenues for research.

Introduction This entry considers crime and disorder on public transport, the ways in which it concentrates and the extent to which it is specialized. It begins by defining public transport for the purposes of this discussion, and emphasizes the importance of reducing crime and disorder and associated fears of personal security on the public transport network. It will then review the empirical evidence for the concentrations and specializations of crime and disorder evident on public transport systems, and assess the theoretical perspectives that help explain these patterns. It will describe how the public transport system provides unique settings (places and times) that subsequently create distinctive crime patterns that encompass a variety of changing environments. These include stations, stops, and interchanges and “en route” and “line of route” offenses, and therefore, a description is given of how the public transport network serves settings that incorporate both static and dynamic situations of crime and disorder (Newton 2005). This entry will explore how concentrations of crime manifest across each of these settings and highlight some of the unique challenges faced when attempting to explain and prevent crime on the public transport network. It will conclude by discussing some emerging trends and avenues for further research in this field. For the purposes of this entry, the term public transport is used to capture what North American readers may think of as “public transit,” “mass transit,” “rapid transit,” or “public transportation” systems. There is no clear consensus of a definition of public transport, and therefore the following are adopted as forming key components of the public transport system: • Public transport is used to describe a system used by the public, often a means of

C

C

•

•

•

•

•

•

710

transporting passengers in mass numbers, generally a for-hire system that occurs across a fixed route or line. It consists of a range of transport modes, including railway (railroads, light rail, metro/ subway/underground railway, high-speed rail, and intercity rail), buses, trolleybuses, and trams; ferries; coaches; airlines; water taxis, gondolas; and pedicabs. Bicycle hire schemes are becoming more popular in urban areas and could be included as part of the public transport system although they are not discussed here. In some regions “collective transport” is also considered as a form of public transport, for example, a minibus or fixed group taxi (e.g., South America and Russia). “Paratransit” is an expression sometimes used in areas of low demand and for people who need a door-to-door service. There are distinctions between multimodal and intermodal systems: a multimodal system accommodates different modes of transport, whereas an intermodal system ensures strategic connections between different modes. There is a debate as to whether or not taxis are part of the public transport system although they are not discussed within the context of this entry.

Why Examine and Reduce Crime and Disorder on Public Transport? There are a number of reasons why understanding and preventing crime on public transport should be of interest. Firstly, public transport has an important role to play in reducing social exclusion, by providing access to facilities such as work, health, leisure, and employment. In Great Britain, for example, the 2011 National Travel Survey estimated that about one quarter of households do not have access to a car. There are also obvious environmental benefits in promoting public transport as a means of sustainable travel, and hence, the use of public transport may be expected to grow over time rather than decrease. A number of surveys have identified that fear of crime and personal security is a major inhibiting factor to the use of public

Crime on Public Transport

transport, second only in many surveys to reliability and accessibility. Indeed, a report in the UK by the then Department for Environment, Transport and the Region (DETR) suggested that reducing fear of crime could increase patronage by 3 % at peak and 10 % at off peak times (Newton 2004). One of the clear messages from this is the significance of making passengers actually feel safe, and Smith and Clarke (2000) suggest this should be a mandate shared across all transport systems. It is evident in the literature concerned with the fear of crime, however, that reducing actual levels of crime does not necessarily equate to reductions in people’s perceptions of crime risk. The factors that contribute to worries over personal security on public transport are not disparate from those associated with fear of crime outside of the transport network, and the dynamics of vulnerability on public transport include variation by gender, ethnicity, age, familiarity with, and levels of public transport usage (Yavuz and Welch 2010). However, there are some features of the public transport network that are perhaps distinctive. As will become more apparent in this entry, the dynamic nature of the public transport system creates unique environments, through which specific modes of transport traverse, transporting potential targets and victims, on a system that passes through areas with different levels of crime risk, and therefore continuously receives different inputs and outputs over time. This creates a unique, potentially specialized, and certainly concentrated arena within which crime and disorder may occur. Context and Scope While the definitions of public transport alluded to earlier suggest that it is a for-hire system of mass transportation used by the public, in its broadest sense the public transport journey should actually be viewed in its entirety, what the then DETR called the “whole journey approach” (Newton 2004). This not only consists of travel on actual vehicles or different modes of transport, but when waiting at a stop, station, or interchange, and also walking to and from stops and stations at the start and end of journeys, or


even potentially transferring between different modes of transport. As suggested by the DETR, negative experiences across any part of the journey might actually influence fears of personal security and change travel patterns, perhaps to the extent that passengers may stop using public transport altogether. While the notion of the whole journey approach is crucial for examining crime on the public transport journey, the majority of the limited empirical evidence is focused on the “at stop or station,” “en route,” and “line of route” offenses. Therefore, the content of this entry will naturally focus on these situations in detail – there being a paucity of research into the walking parts of the journey. In addition, as much of the literature is focused on bus and rail travel, the scope is also limited to these two modes of transport. However, many of the concentrations and potential specialism’s identified and the theoretical concepts used to explain these can be readily applied to other parts of the public transport network and other modes of public transport vehicles. Finally, this entry does not examine security at a wider level than crime and personal security, and therefore, terrorism is excluded from this discussion.

The Nature and Extent of Crime on Public Transport There is limited information available as to the extent of crime and disorder on public transport, and furthermore, discrepancies exist as to the actual levels of crime and disorder on public transport. The most likely reasons for this include the following: the multiple agencies responsible for maintaining and operating the various systems, the lack of standardized reporting and recording of crime and disorder on public transport, difficulties in analyzing the available data, the different policing and security arrangements in place across the network, and the likely levels of underreporting on the system (Newton 2004). One of the definitive references in this area (Smith and Cornish 2006) identifies six categories of crime that are typically evident on the public transport network, these being antisocial

711

C

behavior; crimes against passengers including theft, robbery, and assault; crimes against employees including assault and robbery; vandalism and graffiti; line of route crimes which are offenses along routes that cause delay or affect safety; and a final category of “other offences.” In addition, it is useful to distinguish “en route” offenses from those at stations and stops (Newton 2005). This entry will review the empirical evidence of the concentrated nature of these six types of offenses, both in time and place. It demonstrates the spatial and topological patterns of crime evident on the public transport network, and it is contended that these are a result of the movements on the system in terms of offender and targets, and the increased accessibility it provides to the places the network serves. A question this raises is whether this leads to unique opportunities for offenders to specialize on and near to the public transport network. There are sizeable differences in published figures regarding the extent of crime and disorder on public transport. A study in the UK by the Department for Transport (DfT 2010) suggested that during 2008–2009, there were 12 crimes per million passengers on the bus network and 13 crimes per million on the underground and overground services. This also indicated that the rate of offenses varies by type of crime and analysis of the 2006/2007 figures estimated that the number of offenses (per million passenger journeys) were as follows: on the rail network, 11.9 for violence against the person offenses against passengers, 4.5 against staff, and 39.1 for theft from person and robbery offenses. On the bus network, the rates were 9.0 per million passenger journeys for violence against the person offenses against passengers, 4.1 against staff, and 15.01 for theft from person and robbery offenses. By contrast, a study in Los Angeles suggested an average of 1.55 crime incidents per 100 rides (Loukaitou-Sideris et al. 2002). A 1990s passenger study in England and Wales by Easton and Smith indicated 5 % of passengers had been threatened with violence and 4 % had been the victim of theft, and in Victoria in Australia, a 1990s study by Carr and Wright identified 690 crimes against persons on public transport over

C

C

712

12 months in the context of 300 million passenger journeys per year. While discrepancies will exist in crime rates across countries and different cities, it is suggested those found with the figures presented here are due to differences in recording and reporting practices rather than international or even national variations. There are additional difficulties with estimating and identifying the levels and extent of crime and disorder on public transport. Levels of underreporting are unknown, and Levine, Wachs, and colleagues examined this in California in the 1980s and found figures may underestimate actual levels of crime by 25–30 times. There are clear difficulties in identifying true levels of crime and disorder on public transport, and, for reasons that are explored next, this is perhaps more problematic than the well established levels of underreporting found in police-recorded crime figures. However, it is believed that levels of crime on public transport have, for the most part, followed the internationally declining levels of crime rates in general (Transit Cooperative Research Program 2009). The Limited Evidence Base There are a number of reasons why there have been limited studies into the nature of crime and disorder in public transport. These in part relate to the fragmented nature of regulation across the various systems, the number of agencies and organizations who are responsible for providing and regulating public transport services, difficulties in capturing reliable and accurate information on crime and disorder incidents, the different bodies involved in policing and the security of public transport systems, the levels of underreporting of crime and disorder on these networks, and the analytical challenges faced when attempting to investigate patterns of crime and disorder on public transport (Newton 2004). While the following examples relate to the organization of public transport in the UK, similar fragmented systems exist internationally. An examination of public transport on the rail network in the UK reveals the wide range of organizations responsible for delivering this system nationally, these include the following:


National Rail, the Train Operating Companies who operate under Strategic Rail Franchise Agreements, Network Rail who are responsible for national infrastructure, and the Rail Safety and Security Board (set up to improve the health and safety performance of the railways). There are also some privately owned systems such as light rail (e.g., the London Docklands and the Manchester Metrolink). However, on the rail network there is a consistent set of information collated on crime and disorder as the British Transport Police (BTP) are responsible for policing the rail networks, and they provide monthly and annual figures on crime on the network (both on trains and at stations). On the bus network, this system is more fragmented in the UK. Outside of London (which is regulated by Transport for London), the deregulation of buses which came into force in 1986 as part of the Transport Act 1985 has resulted in a number of operators who can provide commercial or subsidized services. In the large metropolitan areas, these are coordinated by the seven Passenger Transport Executives (PTEs), and outside of these areas Local Authorities regulate buses. In one of the large metropolitan areas, there were over 50 bus operators providing services, and outside of London there is no requirement to report crime incidents on buses, and this often relies on the goodwill and time of both operators and police forces. Indeed in 1998 only 19 police forces (less than 40 %) provided such information, and only 16 bus companies out of over 100 made returns on staff assaults (Newton 2004). While this information is dated, there is no evidence to suggest that the situation has improved considerably, and a key factor is that in the UK (and elsewhere) there is no statutory dedicated category for “crime on public transport” to be recorded as an incident in its own right by the police. Some operators retain the services of private security contractors to maintain safety and security on their systems which adds a further layer of incident reporting (which may or may not be passed onto the police). A third and final complication that contributes to the paucity of evidence on crime and disorder


on public transport is the unique analytical challenge that it poses, particularly when attempting to identify spatial concentrations of crime. In addition to the lack of consistent information available, the range of settings and environments in the transport network within which crime can occur presents a major analytical challenge. Examples include the following: criminal damage at a stop or station at an unknown time, graffiti or the slashing of a seat on a vehicle at both an unknown time and location as it may only be reported when a bus or train returns to the depot at the end of the day, and pocket picking that occurs as part of a passengers journey, somewhere between departure and arrival. In addition it is very difficult to pinpoint the location of a crime that occurs on a moving vehicle, although there are studies that attempt to analyze dynamic crime events (Newton 2005), for example, the development of “hot routes” or “hot lines” (Newton 2008; Tompson et al. 2009).

713

C

• What places and times (settings) within the public transport network create juxtapositions of offenders and victims that increase or reduce crime opportunities, and who can act as place managers for these?

Theoretical Perspectives: Why and How Is Crime Concentrated on Public Transport?

Victims, Targets, and Guardians There are a number of potential victims or targets that may experience crime and disorder on the public transport system. The most obvious of these are passengers and staff, and it is important to consider this in the context of all peripatetic staff who may work on the system. The risk to each of these varies across different sections of the transport network. For example, relative to the risk of crime for passengers during busy periods, the risk of crime is likely to be quite different for lone passengers at night and lone staff (at day or at night) who may or may not be carrying cash. There is also risk to the infrastructure of the transport system which includes criminal damage and graffiti of vehicles, stops, and stations, or damage to fixed lines (e.g., obstacles placed on routes which cause damage or obstruct routes) or even theft of copper lines (discussed later under emerging trends).

The environments of the public transport networks are in some ways distinct from other areas of public space due to the locations they serve, the dynamic nature of the system, and the different inputs and outputs to it. However, there is support within the literature for adopting a number of well-established theoretical criminological perspectives to examine crime on public transport, and this entry now adopts some of these perspectives as a framework for examining the manifestations and concentrations of crime and disorder evident on public transport. A useful starting point is to consider the problem analysis triangle (Clarke and Eck 2005) to conceptulalize patterns of crime and disorder on public transport, and three questions arise here: • Who are the potential targets and victims, and who can act as capable guardians? • Who are the likely offenders, and who can act as handlers to restrain the otherwise motivated offenders?

Offenders and Handlers There is a paucity of empirical evidence about the nature of offending on public transport, in particular the characteristics of the offenders, and whether they perhaps specialize in crime committed within the public transport network. Who are the potential offenders and why they may be present at particular places and times, and do they use the public transport network as part of their journey to crime? This is an under-researched area that warrants future research as discussed at the end of this entry. A 1990s study by Belanger, one of the few conducted on this topic, examined the extent to which public transport may facilitate the movement of criminals between areas. This study looked at the travel behavior of a sample of serious offenders across the New York City subway system and found that the distance travelled from home or anchor point did vary and that 56 % of offenders travelled within their own borough to commit crime, 29 % travelled to

C

C

714

Manhattan from another borough, and 12 % travelled from one borough to another excluding Manhattan. This raises the question of whether the distance travelled to commit crime differs for those trips that public transport is used and those for which it is not. Places, Times, and Place Managers? When examining the places on the public transport network where crime might occur, it is useful to categorize these settings by the three components of the transport journey: the walking, waiting, and en route settings. There is empirical evidence that crime is concentrated at a small proportion of transport stations, stops, and hubs (Smith and Clarke 2000; Loukaitou-Sideris et al. 2002; Newton 2004; Smith and Cornish 2006; Newton and Bowers 2007) and “en route” at particular sections of public transport journeys (Newton 2008; Tompson et al. 2009). These concentrations vary by both crime type and time of day, as will be discussed in more detail below. The waiting environment may also include transfer between transport modes within a transport interchange or hub, and this is perhaps distinctive from transfer between transport modes that occurs outside of transport interchanges as this may be classified as part of the walking environment. These external transfers and the walking environment more generally, for example, from point of departure to stop, or stop to place of arrival, are unlikely to be recorded as part of a public transport journey, and limited information is available on crime and disorder in these components of the public transport journey. The above discussion has considered who are the likely targets and potential offenders on the public transport and the places, times, and situations where these are more likely to coincide. The opportunities this offers for prevention will be discussed in more detail later but include consideration of the following: who can act as capable guardians for these different targets, who may be an appropriate handler for offenders, and the importance of place management and environmental in reducing crime and disorder on the public transport system.


Crime Pattern Theory and Routine Activities Two theoretical perspectives that can readily be applied to the public transport network are crime pattern and routine activity theories. Crime pattern theory defines people’s activity spaces in terms of three main components: nodes, paths, and edges (Clarke and Eck 2005). The term “node” is used to describe where people (both offenders and victims) travel to and from, and the idea of personal activity nodes closely resembles the ideas of a person’s routine activities. These nodes are linked by paths, which would include roads or other transport routes. Edges define the boundaries around nodes, within which people reside, work, or are routinely active in. According to crime pattern theory, crimes are likely to concentrate where offender routine activity spaces intersect with those of suitable targets of crime. Nodes and paths are particularly relevant in the context of the public transport network, with nodes represented by rail stations and bus stops (and so on), while transport routes correspond to the paths that connect the transportation nodes. Edges may even delineate the boundaries of the public transport network, and there may be regulated and controlled entrances and exits to the system (Newton 2004). Edges can be considered the boundaries of the public transport network, but these become less geographically distinct when considering the walking component of the whole journey approach discussed earlier. Another concept that has been applied to transport nodes is the notion of risky facilities (Clarke and Eck 2005), further discussed below. The transport network can be thought of as having the potential to shape the activity spaces (and hence the spatial and temporal concentrations) of offenders and potential targets and thus to increase the potential of crime occurring on the network. There is clear evidence, reviewed below, of crime concentrations on the network at two of the critical sections of the network, in and around stations and stops (nodes) and crime en route (paths). Moreover, the way in which the transport network influences the potential for crime is not limited to where it occurs but also


what types of crimes are most likely to occur and potentially which offender specialism’s may occur in and around the system. Crime Attractors, Crime Generators, and the Public Transport Network A final key theoretical perspective is notion of “crime attractors,” “crime generators,” and “crime neutral” areas (Clarke and Eck 2005). A crime generator is an area that attracts large numbers of people for reasons other than to commit a crime. At particular times and places, this concentration of victims and offenders produces an unplanned opportunity for offending. Crime attractors are places that offenders visit due to knowledge of the area’s criminal opportunities, such as bars and prostitution areas, in order to commit a crime. A key question of relevance here is do places on the public transport system act as crime attractors or crime generators. Research has suggested that transport hubs can act as both attractors and generators and that this varies by crime type and by time of day (Smith and Cornish 2006). Research by Levine and colleagues in California in the ‘1980s identified that for some crime types, for example rape, homicide, and robbery, the risk was greater when guardianship was likely to be low or where it was impeded by the design of the location (Newton 2004). Examples of this include when there were low levels of pedestrian traffic, low levels of surveillance, and many concealed areas. A study of graffiti and vandalism by Wilson and Healy in the 1980s on the rail network in New South Wales, Australia, found offenders were principally juveniles and that most damage occurs in unsupervised areas during off peak hours. Burrell (2007) highlights that violent crime on public transport tends to happen in the late evening/nighttime when there is less supervision, whereas pocket picking and purse snatching are more frequent during the rush hour. The above examples all demonstrate how transport hubs may act as attractors and or generators of crime, at different times of the day, and for different types of crime. In order to examine these theoretical perspectives in more detail and also highlight the

715

C

empirical evidence that exists as to how crime is concentrated and potentially specialized on the transport network, two of the key settings of the public transport network will now be examined in more detail: the first of these is crime in and around stops and stations, and the second is “en route” crime on a moving (and sometime stationary) vehicle.

Crime at Stops and Stations There are a number of studies that have examined how the design of transport stations can influence levels of crime. Examples include the Paris Metro (Clarke 1996), the Metro in Washington DC (La Vigne 1997), and the metro in Gothenburg, Sweden (Ceccato et al. 2011). Similar studies have examined over ground rail stations in the country of Wales(UK) (Cozens et al. 2004). A clear finding from these studies, particularly in relation to the Washington DC example, was how unusually low crime rates (compared to other systems and local crime rates) could be explained by reference to aspects of environmental design. In addition to the internal design of stations, there have been studies that examine the link between rail stations and their external environments or nearby surroundings. Studies in Chicago (Clarke 1996; Block and Block 2000) into street robbery in the vicinity of rapid transit stations found that street robbery is concentrated near (but not immediately outside) rapid transit stations and that robbery varied temporally and concentrated late at night (between 11:00 and 12:00 p.m., with a peak time of 2:00 a.m.). Block and Block contend that although the behavior of potential victims and offenders and the time of day and other individual features of each situation will affect risk, it is the place and the surrounding or nearby space that is crucial in bringing all these factors together in what they term the environs of rapid transit stations. A recent doctoral thesis by Herrmann (2011) who examined robbery hot spots in New York provides support for this notion, and this research demonstrated that although all identified hot spots were near subway stations, those that

C

C

716

occurred during the daytime (3:00 p.m. peak time) were near high schools but that those occurring in the nighttime (1:00 a.m.) were not, instead being around late night businesses. In addition to studying crime near railway stations, there have been a number of studies that have examined crime and disorder in and around bus stops and bus stations (LoukaitouSideris et al. 2002; Newton 2004; Newton and Bowers 2007; Vu 2009). Loukaitou-Sideris examined the link between criminal activities at bus stops and associated nearby environmental characteristics in a North American city and found that the ten bus stops with the most crime accounted for 18 % of all crimes, and although passenger levels at these stops were high, other nearby stops with high patronage exhibited little or no crime. They identified that there were an abundance of “negative” environmental factors and a general lack of “defensible space” near these high-crime bus stops, including liquor stores, bars, check cashing establishments, seedy motels, pawnshops, vacant lots/buildings, and adult book stores and movie theaters. This is supported in the UK by Newton and Bowers (2007) who found criminal damage to bus shelters was concentrated at a small number of bus shelters. Examining the criminogenic and socioeconomic characteristics of places, they also identified that bus shelter damage was related in known and predictable ways to characteristics of local neighborhoods. Indeed shelter damage risk was higher in areas with high levels of antisocial behavior, a lack of capable guardianship, and the presence of youths. For example, they found a positive relationship between bus shelter damage and proximity to parks, children’s play areas, and schools (particularly those whose unauthorized truancy levels were above the national average), and a negative relationship with the presence of pubs and bars, nightclubs, and off-licenses (late night liquor stores). An interesting feature of the two above studies if that some of the so-called negative environmental features found to increase crime in the US study, namely, the presence of off-licenses (liquor stores), actually reduced the risk of crime in the UK study. This finding is explained best


when considering that the US research examined crimes such as assault and violence, whereas the UK study was concerned with criminal damage to bus shelters. The late night land use associated with these features increased violence and assault in one study, but actually served as a deterrent for criminal damage in another study, by increasing the number of persons in the area, whose presence therefore may have actually served as a form of guardianship against criminal damage to bus shelters. It also serves as a clear example of how spatial crime concentrations vary by type of offense and time of day.

“En Route” Crime Few studies have examined crime that occurs while a public transport vehicle is moving, predominantly due to the analytical challenges this poses. A 1970s Home Office study by Mayhew examined damage to buses in Manchester (UK) and found that damage was greatest on buses without a conductor, more prevalent on the upper decks of the bus, and especially at the rear of the bus. On buses with a rear staircase, graffiti and vandalism was more prominent upstairs and at the front of the bus. This study concluded that a lack of supervision was an important factor in the occurrence of vandalism and graffiti on buses, supporting the idea that a lack of guardianship or place management on the transport network acts a contributory factor to criminal damage. This study considered the internal environment of a vehicle, but not the external environments it may pass through. There are, however, some studies which have examined how the external environments a moving vehicle passes through may influence crime risk on that vehicle. Crime and disorder is concentrated on particular routes and sections of routes on the public transport network. A 1980s study by Pearlstein and Wachs examined crime on the bus network in California and concluded that only 88 out of 233 routes (less than 40 %) experienced serious incidents of crime and that crime mostly occurred on routes that traversed areas with high crime rates. They also found crime was disproportionately


high during late evenings when violent crimes were more prominent and that theft and robberies were most likely during the rush hour (Newton 2005). Additional studies, such as those by Levine and colleagues in California in the 1980s, support the notion that incidents of bus crime are highest for bus routes that pass through highcrime areas. Moreover, Newton (2008) identified that not only is bus-related crime positively correlated with levels of crime in the areas a bus traverses, and that the risk of crime on those routes that traverse high-crime areas is greater than on other routes, but furthermore that the risk propensity is heightened on routes that go through high-crime areas and have a higher numbers of stops within these areas (multiple entry and exit points). It should be noted that this risk is not just confined to the passengers and staff who travel on public transport but also to the actual infrastructure of the public transport network. An analysis of criminal damage to vehicles using data of objects that were projected or thrown at buses (including for example rocks, stones, bricks, and eggs) also found that risk was highest to buses passing through areas that experienced high crime rates (Newton 2004). In this situation the theoretical edges or boundaries to the public transport network become more blurred, and the interaction between the internal (inside the vehicle) and external environments (that a vehicle passes through) is different to the inputs and outputs of passengers and staff to the system (which occurs at transport stops and stations or nodes). The latter can be regulated by controlling the entrance and exit points to the system (by offenders of victims), while the former cannot, and this provides an additional layer of complexity to the study of crime on the public transport system.

Crime Concentrations on Public Transport and Prevention Implications Perhaps a key significant feature of the concentrations and potential specializations of crime and

717

C

disorder that are evident is the prevention opportunities these offer. Crime and disorder has been shown to be concentrated within certain components of the public transport network, and these include along particular route sections of journeys, nearby and within certain stations and stops, and within particular parts of stations and interchanges, and these all vary by offense type and time of the day. There is also a clear interaction between the movement of vehicles and the environments through which they traverse, and regulating access to and from the system (the entry and exit points) can serve as a useful mechanism for reducing crime and disorder on the network. The design, environment, and management of stations and stops can influence crime rates, and there is an established evidence base of strategies and measures that can be adopted to improve the design of public transport infrastructure in order to reduce crime levels (Clarke 1996; LaVigne 1997; Smith 2008; Ceccato et al. 2011) including better lighting and illumination and removing dark corners and hiding places. Furthermore, effective place management of stations (Clarke 1996), the introduction of capable guardians (Newton et al. 2004), and the use of effective handlers (Van Andel 1989) can all significantly reduce risk on the public transport network. Many aspects of the public transport network have regulated entrance and exit points that connect the internal environment of a transport journey with its external environment. These offer clear prevention opportunities but these must be contextualized within an intelligence-led or problem-solving approach (Clarke and Eck 2005), and as already demonstrated in this entry, these differ significantly for offense type and time of the day. The design of transport systems and the introduction of new technologies can also influence crime rates. For example, a study on the London Underground in the 1980s by Clarke and colleagues revealed that the introduction of subway slugs (foil wrapped around 10 pence coins to create fake 50 pence coins) became widespread as new automated vending machines were introduced. Remedial measures were highly effective

C

C

718

in reducing this, but greater anticipation and investment at the design phase may have prevented this. Interviews by Ekblom with offenders on the London Underground (Clarke and Eck 2005) revealed that offenders would stand by signs warning passengers of “pickpocketing in operation,” as after reading these signs passengers would frequently pat their own pockets, which was of considerable assistance to the offender in identifying which pockets a wallet was likely to be in.

Emerging Issues and Future Debate There are a number of recent developments that should be mentioned, as they offer insights into potential avenues for further research into crime patterns on the public transport network. One of the issues that was raised but perhaps not answered within this entry is whether public transport systems act as a generator or attractor of crime, or perhaps both. A recent paper (Bernasco and Block 2011) investigated the distribution of street robberies in Chicago and how this was influenced by crime generators, crime attractors, and offender anchor points. They suggested that stations provide an increased level of accessibility to areas and found that street blocks with a station had four times as many robberies as similar street blocks without them. However, they suggested that the impact of stations alone was not the significant factor, and the combined impact of a number of additional features near to stations was also contributory. Indeed, research in both Vancouver and New Jersey (Sedelmaier 2003) found no evidence that, on their own, the introduction of a new station to an area increases crime risk. Block and Block (2000) describe the environs of transport stations and importance of the nearby environment, and Robinson and Giordano (2011) discuss the importance of what they term spatial interplay, the interaction of land uses in relation to crime incidents around transport stations. As described in the Herrmann (2011) study, it is not just the location of subway stations that influenced when and where street robbery


occurred but also what was near to the station. A clear avenue for future research is to examine this spatial interplay in more detail, of the interaction between crime risk; and the environs of public transport stops and stations and associated nearby land use types. Newton et al. (forthcoming) examine this interaction by considering pocket picking and theft on the London underground, and the relationship between what happens above the ground and below the ground, which could be termed the transmissibility of crime on public transport. This research shows that there are distinctive patterns of theft, both outside underground stations and on the network itself. The interesting issue here is that these are policed separately (the underground by British Transport Police and aboveground by London Metropolitan Police), yet offenders do not distinguish between these two settings, in the same way they do not distinguish between the geographical boundaries of different police forces. This demonstrates the need for information sharing between these policing units and also potentially the benefit of joint operations. It is clear that the environs of public transport and the potential transmission of crime around this system is an area for further research, particularly when examining how offenders may potentially use the public transport as part of their journey to and or from crime. This is a key question in examining the spatial and temporal patterns and concentrations of crime evident on public transport and perhaps, moreover, in determining the extent to which offenders may specialize on public transport networks. There is little information at present as to whether offenders specialize on the public transport network, and if this is different to or perhaps an extension of their usual offending patterns, and to contextualize this within the journey to crime research. The recent work of Reynald (2011) into the relevance of guardianship raises new issues that are particularly relevant to public transport. One of the findings of her research is that the three key aspects of capable guardianship are the willingness to supervise, the ability to detect potential


offenders, and the willingness to intervene. Familiarity of context was found to be a very important component of this, but the public transport network provides a dynamic environment which is constantly changing. This therefore makes transport nodes highly dynamic contexts for which perceptions of familiarity may not be time stable. When on board a moving transport vehicle, new people may enter or leave at each stop, and a vehicle may move through both familiar and perhaps unfamiliar environments. This is certainly an area that warrants future research activity. More recent technological innovations, including the use of mobile apps such as barcode scanners and the potential for introducing new payment forms (such as scanning mobile phones to pay for entrance/exit to transport systems), may also alert offenders to more attractive handsets and smartphones, akin to the Ekblom interviews with offenders who stood near signs warning of pocket picking. Gentry (2012) examined theft of electronic products on the Massachusetts rapid transit subway and found that theft of electronic devices was concentrated at a small number of stations. As new technologies are introduced to the system, these will offer new opportunities for offending that should be researched. A final issue that has become increasingly important on the public transport network is line of routes offenses, not en route offenses that occur on a moving vehicle, but those that disrupt lines and delay journeys, for example, damage and obstructions to lines and routes (Smith and Cornish 2006). The increasing value of metals has resulted in a growing increase in the number of metal thefts, and the rail infrastructure in particular has seen a dramatic rise in the number of copper cable thefts over recent years (Sidebottom et al. 2011). This is an example of an emerging and specific type of crime that may be expected to be concentrated spatially, particular to only a few types of environments, the transport system being a key one, and a crime that is hence likely influenced by criminal opportunities associated with the configuration of the transport network.

719

C

Related Entries ▶ Crime Location Choice ▶ Crime Prevention Through Environmental Design ▶ Crime Science ▶ Criminology of Place ▶ Hot Spots and Place-Based Policing ▶ Intelligence-Led Policing ▶ Routine Activities Approach

Recommended Reading and References Bernasco W, Block R (2011) Robberies in Chicago: a block-level analysis of the influence of crime generators, crime attractors, and offender anchor points. J Res Crime Del 48(1):33–57 Block R, Block C (2000) The Bronx and Chicago: street robbery in the environs of rapid transit stations. In: Goldsmith V, McGuire P, Mollenkopf J, Ross T (eds) Analyzing crime patterns; frontiers of practice. Sage, Thousand Oaks, pp 137–152 Burrell A (2007) Violence on and around public transport. UCL Jill Dando Institute, University College, London Ceccato V, Uittenbogaard AC, Bamzar R (2011) Safety in Stockholm’s underground stations: the importance of environmental attributes and context. Secur J Advance Access 26 December 2011: 1743–4645 http://www. palgrave-journals.com/sj/journal/v26/n1/abs/sj201132a. html Security Journal (2013) 26, 33–59. doi:10.1057/ sj.2011.32; published online 26 December 2011 Clarke R (1996) Preventing mass transit crime. In: Crime prevention studies, vol 6. Willow Tree Press, Monsey Clarke R, Eck J (2005) Crime analysis for problem solvers in 60 small steps. Office of Community Oriented Policing Services, United States Department of Justice, Washington, DC Cozens P, Neal R, Hillier D, Whitaker J (2004) Tackling crime and fear of crime while waiting at Britain’s railway stations. J Pub Transport 7(3):23–41 Department for Transport (DfT) (2010) Estimated costs to society of crime on public transport in England in 2006/07: final report on findings. DfT, London Gentry K (2012) Smartphones and safe subways: preventing cell phone theft within the MBTA subway system. Paper presented at the Academy of Criminal Justice Sciences (ACJS) Conference, New York City, 13–17 Mar 2012 Herrmann C (2011) Risky businesses: a micro-level spatiotemporal analysis of crime, place, & business establishment type. Unpublished Ph.D. thesis, The City University of New York, New York LaVigne N (1997) Visibility and vigilance: metro’s situational approach to preventing subway crime. National Institute of Justice, Washington, DC

C

C

720

Loukaitou-Sideris A, Liggett R, Hiseki I (2002) The geography of transit crime: documentation and evaluation of crime incidence on and around the green line stations in Los Angeles. J Plann Educ Res 22(2):135–151 Newton A (2004) Crime and disorder on buses: towards an evidence base for effective crime prevention. Unpublished Ph.D. thesis, University of Liverpool, Liverpool Newton A (2005) Crime on public transport: static and non-static (moving) crime events. West Criminol Rev 5(3):23–40 Newton A (2008) A study of bus route crime risk in urban areas: the changing environs of a bus journey. Built Environ 34(1):88–103 Newton A, Bowers K (2007) The geography of bus shelter damage: the influence of crime, neighbourhood characteristics and land use. Int J Criminol http://www. internetjournalofcriminology.com/; http://www.internetjournalofcriminology.com/Newton%20and%20Bowers %20-%20The%20Geography%20of%20Bus%20Shelter%20Damage.pdf Newton AD, Johnson SD, Bowers KJ (2004) Crime on bus routes: an evaluation of a safer-travel initiative. Policing Int J Police Strat Manag 27(3):302–319 Newton A, Partridge H, Gill A (forthcoming) Above and below: examining the transmissibility of theft risk on the London underground Reynald D (2011) Guarding against crime: measuring guardianship within routine activity theory. Ashgate, Surrey Robinson J, Goridano L (2011) Spatial interplay: interaction of land uses in relation to crime incidents around transit stations. In: Andresen M, Kinney B (eds) Patterns, prevention, and geometry of crime. Routledge, London, pp 175–199 Sedelmaier C (2003) Railroaded: the effects of a new public transport system upon local crime patterns. Unpublished Ph.D. thesis, Rutgers University, New Jersey Sidebottom A, Belur J, Bowers K, Tompson L, Johnson SD (2011) Theft in price-volatile markets: on the relationship between copper price and copper theft. J Res Crime Del 48(3):396–418 Smith M (2008) Addressing the security needs of women passengers on public transport. Secur J 21:117–133 Smith M, Clarke R (2000) Crime and public transport. Crime Justice A Rev of the Res (27) pp 169–233. Smith M, Cornish B (2006) Secure and tranquil travel: preventing crime and disorder on public transport. UCL Jill Dando Institute of Crime Science, London Tompson L, Partridge H, Shepherd N (2009) Hot routes: developing a new technique for the spatial analysis of crime. Crime Mapp J Res Pract 1(1):77–96 Transit Cooperative Research Program (TRCP) (2009) Chapter 2: Passenger perception of crime and terrorism. TCRP synthesis 80: transit security update. Transportation Research Board, Washington, DC, pp 10–15

Crime Prevention Van Andel H (1989) Crime prevention that works: the care of public transport in the Netherlands. Brit J Criminol 29(1):47–56 Vu Y (2009) Bus stops and crime: do bus stops increase crime opportunities in local neighbourhoods? Unpublished Ph.D. thesis, Rutgers University, New Jersey Yavuz N, Welch EW (2010) Addressing fear of crime in public space: gender differences in reaction to safety measures in train transit. Urban Stud 47(12):2491–2515

Crime Prevention ▶ Situational Crime Prevention and the Wild West

Crime Prevention Through Environmental Design Rachel Armitage Applied Criminology Centre, University of Huddersfield, Huddersfield, UK

Overview This entry examines an approach to crime reduction which differs from many others in that it focuses not on the offender or their reasoning for committing an offense but upon the environment in which an offense takes place. This approach also differs in its consideration of who should be held responsible for the reduction of crime, with a focus not solely upon the traditional criminal justice system agencies but also upon planners, architects, developers, and managers of public space. The approach is based on the presumption that offenders will maximize crime opportunities, and therefore, those opportunities must be avoided (in the first place) or removed (following the emergence of a crime problem). In the 2001 publication “Cracking Crime Through Design,” Pease introduces the concept of design as a means of reducing crime, but more importantly the premise that it is the moral

Crime Prevention Through Environmental Design

responsibility of many different actors and agencies to improve the lives of those who may fall victim to crime, those who live in fear of crime, and (less obviously) those who will, through the presentation of unproblematic opportunities, be tempted into offending. In the case of crime prevention through environmental design (CPTED), it is the planners, designers, developers, and architects who risk acting (as Pease paraphrases the poet John Donne) as the gateway to another man’s sin. CPTED is an approach to crime reduction (that may be described as a measure, program, or intervention) which aims to reduce crime through the design and manipulation of the built (and sometimes natural) environment. It focuses predominantly upon “designing out” opportunities for crime before they occur – for example, at the preplanning or planning stage – although some interventions take place post-development, in response to a crime problem which has emerged. A commonly used formal definition is that used by Tim Crowe who defines CPTED as: “The proper design and effective use of the built environment, that can lead to a reduction in the fear of incidence of crime and an improvement in quality of life. . .The goal of CPTED is to reduce opportunities for crime that may be inherent in the design of structures or in the design of neighbourhoods” (Crowe 2000, p. 46). Ekblom (2011) proposes a redefinition and presents the following alternative, which introduces several points not included within Crowe’s definition, including the balance between security and contextually appropriate design and the possibility of intervening at different stages between preplanning and post-construction. Ekblom states that CPTED is: “Reducing the possibility, probability and harm from criminal and related events, and enhancing the quality of life through community safety; through the processes of planning and design of the environment; on a range of scales and types of place, from individual buildings and interiors to wider landscapes, neighbourhoods and cities; to produce designs that are ‘fit for purpose’, contextually appropriate in all other respects and not ‘vulnerability led’;

721

C

whilst achieving a balance between the efficiency of avoiding crime problems before construction and the adaptability of tackling them through subsequent management and maintenance” (Ekblom 2011, p. 4). Early proponents of CPTED largely approached the subject from a planning, urban design and architecture perspective (Jacobs 1961; Newman 1973) with Jeffery coining the phrase “crime prevention through environmental design” in his 1971 book of the same title. While this period saw the development of what are still considered to be the key principles of CPTED, the main focus of these studies was upon the influence of the environment on human behavior in general (Coleman 1986) as opposed to the reduction of specific crime types. The move towards a practical application of the principles, focused largely upon the prevention of acquisitive crimes (such as burglary and car crime) and tailored towards crime reduction practitioners (through training and guidance), emerged in the 1980s with (among others) work conducted by Brantingham and Brantingham (1981), Poyner (1983), and Poyner and Webb (1991). More recently, research within the field of CPTED has focused upon the effectiveness of both the individual and collectively applied principles of CPTED measures in reducing crime and the fear of crime (by authors such as Armitage, Hillier, Pascoe, Cozens, and Kitchen), the process of applying CPTED principles within the police and planning environments (by authors such as Kitchen and Monchuk), the development of CPTED-based risk assessment tools to predict (and prevent) risk (by authors such as Winchester and Jackson, Van der Voordt and Van Wegen, and Armitage), and a wider approach to the potential benefits of such interventions including the impact upon environmental and social sustainability (by authors such as Cozens, Kitchen, Dewberry, Pease, and Armitage and Monchuk). Given a widening of the focus to include the process of application and consideration of benefits beyond crime reduction, such as social and environmental sustainability, a more appropriate definition of CPTED might be as follows: the

C

C

722


design, manipulation, and management of the built environment to reduce crime and the fear of crime and to enhance sustainability through the process and application of measures at the micro (individual building/structure) and macro (neighborhood) level. This definition is represented in Fig. 1 (below).

Principles of CPTED The principles of CPTED have been presented by several authors, including, but not exclusively, Poyner (1983), Crowe (2000), and Cozens et al. (2005), and adapted across different countries to form the attributes of safe places/environments within planning policy and guidance (e.g., Safer Places within England and Wales). Poyner (1983) outlined the principles as surveillance, movement control, activity support, and motivational reinforcement. Cozens et al. (2005) extended this to include the seven principles of defensible space, access control, territoriality, surveillance, target hardening, image, and activity support. The following summary presents a basic introduction to these terms. Defensible Space Defensible space is the creation of buildings/ enclosures/spaces which allow/facilitate/help the residents of that space to keep potential offenders out. The term was coined by Oscar Newman (1973) who suggested that the physical design of a neighborhood can either increase or inhibit people’s sense of control over the spaces

Design, manipulate, maintainance of built environment

Micro (building) - Macro (neighbourhood)

Reduce Crime

Reduce fear of crime

Maximise sustainability

Crime Prevention Through Environmental Design, Fig. 1 Crime prevention through environmental design

in which they reside. Newman categorized space into public (e.g., the road in front of a property), semipublic (e.g., the front garden), semiprivate (e.g., the back garden), and private (inside the property). He argued that if space is defensible, it will be clear to the owner/user of that space, and to non-legitimate users, who should and who should not be in this space. CPTED interventions ensure that space is clearly demarcated, that it is clear who has control/ownership/rights over that space, and that potential offenders have no excuse to be in that space (see discussion of “permissions” in Wortley and Tilley, this volume). CPTED interventions would rarely achieve this through the installation of physical barriers; rather, interventions would include the more subtle measures such as a change in road color and texture or a narrowing of the entrance to the development to mark the area as private. Territoriality Territoriality involves the human emotion/ response to the space which they define as their own. Physical responses to territoriality might include a resident marking an area as their own through the installation of a house sign or gate. Emotional responses to territoriality would include a resident’s feelings of intrusion or infringement should a person enter what they consider to be their space. Thus, territoriality refers to the human motivation to control the space which they believe is theirs, be that through the legal ownership of that space or through their adoption and management of that space. While Cozens et al. (2005) separate defensible space and territoriality, a more concise summary of CPTED principles might categorize defensible space alongside territoriality, given that the physical creation of defensible space aims to create territorial control over that space. Access Control Access control refers to the design of buildings and space to actively keep people out. While this principle has traditionally been referred to as “access” control, perhaps due to its routes in more traditional situational crime prevention measures to restrict entry into buildings and rooms within


723

C

buildings, within CPTED the aim is much wider. What has been referred to as access control encompasses the following aims: (1) to limit the likelihood that offenders will become aware of that area as a potential target (see discussion of awareness space within Wortley and Tilley’s summary of Crime Pattern Theory), (2) to make it more difficult for offenders to navigate into, out of, and within an area should they select it as a target, (3) to increase the physical difficulty of entering a building/space should offenders become aware of the area as a target, (4) to increase the difficulty psychologically for offenders to enter and move around an area without feeling conspicuous, and (5) to remove any excuse for potential offenders to be within a private or semiprivate space and maximize the legitimate users’ confidence in challenging non-legitimate users of space. Given the wider aims of this principle, “access control” would appear too limited a definition. A more appropriate term might be the “limitation of access, egress, and through movement.”

tasks of active (formal) and passive (informal) surveillance, the surveillability (Ekblom 2010) of that space, and the creation of the perception among offenders that they are being observed.

Surveillance Surveillance refers to the way that an area is designed to maximize the ability of formal (security guards, police, employees) or informal (residents, passersby, shoppers) users of the space to observe suspicious behavior. These formal and informal users are referred to in Routine Activity Theory as capable guardians. Within situational crime prevention more generally, surveillance may include the installation of CCTV or the use of formal security guards. Within CPTED, surveillance rarely relates to formal measures but refers more to the informal surveillance created through measures such as ensuring that dwelling entrances face the street, that rooms facing the street are active (such as the kitchen or living room), and that sightlines are not obstructed by shrubbery or high walls. Linked with territoriality, the principle of surveillance requires users of that space to realize that an individual is behaving in a suspicious manner (be that through their behavior or simply their presence within a private/semiprivate area) and to have the confidence to challenge them or intervene. Therefore, the term “surveillance” includes the operational

Activity Support Activity support relates to the creation of an environment which increases the likelihood that legitimate users will make use of space and subsequently act as additional surveillance. Although activity support is included by many as a distinct principle of CPTED, the ultimate aim is to enhance surveillance and so the two principles can be combined. Given this discussion of definitional issues and a desire to condense the list presented by Cozens et al. (2005), the principles of CPTED might be summarized as: • Physical security – securing buildings and spaces to a level which is appropriate to risk and where possible products which are tested to the relevant security standards should be utilized • Surveillance – designing building and space to allow both formal and informal surveillance from users of that space and to create a feeling of unease among non-legitimate users of the space • Movement control – limiting access, egress, and through movement

Target Hardening Target hardening is often referred to as physical security and includes the initial design, or retrofit upgrade, of doors, windows, fences, and other physical structures to increase the difficulty for offenders in entering a building or space. Image Cozens et al. (2005) use the term “image,” while others have used “management and maintenance” to cover the principle of creating buildings/spaces which are physically free from litter, graffiti, vandalism, and damage but are also areas without stigma or a poor social reputation. It is difficult to allocate a specific label to these concepts as image refers to a state and management and maintenance to the activities which create that state.

C

C

724

• Management and maintenance – ensuring that buildings and the surrounding spaces are designed to create a positive image and to ease future maintenance of the space and ensuring that programmed management and maintenance systems are in place • Defensible space – ensuring that spaces have a clearly defined ownership, purpose, and role to enhance feelings of territoriality among residents and legitimate users

Empirical Support for CPTED Ekblom (2009) presents an excellent discussion of where CPTED needs to be improved both conceptually and practically, and these are addressed in the conclusion to this entry. One weakness identified by Ekblom is the need to improve the evidence upon which CPTED is based. While it is acknowledged that there are still areas which require clarification, the evidence base has been strengthened over the last decade, with several rigorous examinations of the impact of individual design features on crime levels. Using the five principles defined above, the following section outlines the evidence base to support the efficacy of CPTED in reducing crime. Physical Security Research on security measures as a means of preventing residential burglary is mixed, with several studies suggesting that the actual home break-in requires little in the way of technical sophistication and that physical security is a low priority for burglars when searching for targets (e.g., Reppetto 1974). However, several studies suggest that with all other factors being equal, burglars would prefer to offend against properties with lower levels of physical security (Cromwell et al. 1991). Budd’s (2001) analysis of the British Crime Survey found that security devices are very effective in reducing the risk of burglary victimization. Budd’s study found that, in England and Wales in 1997, 15 % of households without security measures were burgled, compared to 4 % of households with basic measures in place and 3 % with higher levels of security.


Surveillance Research suggests that surveillance and visibility play a major part in offenders’ decision-making processes when selecting properties to offend against. Reppetto (1974) interviewed 97 convicted burglars and found that the most common reason for avoiding a target was that there were too many people around. Offenders stated that the possibility of neighbors watching them deterred them from selecting a property and that they would select targets where they felt less conspicuous and where there was less visual access from neighboring properties. When assessing the design characteristics of victimized properties, several studies have identified a lack of surveillance or poor levels of visibility as key features of crime-prone homes (e.g., Armitage 2006; Brown and Altman 1983; Winchester and Jackson 1982). Movement Control The efficacy of the principle of limiting movement is less clear-cut. Although the majority of research suggests that limiting through movement will reduce crime, some studies suggest that encouraging pedestrian and vehicular movement will provide informal surveillance of the area – something that Jacobs (1961) refers to as “eyes on the street.” The mechanisms through which limiting movement might reduce crime are as follows: firstly, that an area with high levels of through movement provides ease of entry/ escape for offenders, and therefore, reducing through movement would decrease the risk of offending. This supposition is supported by interviews with offenders and analysis of victimization levels within residential areas (e.g., Poyner and Webb 1991; Taylor and Gottredson 1987). The second mechanism suggests that an area with high levels of through movement is more likely to be within an offender’s awareness space. Offenders are likely to have passed through the area while conducting their day-to-day activities and to have become familiar with the suitability of targets. Limiting through movement would thus decrease the likelihood of offenders being aware of suitable targets. This supposition is supported by interviews with offenders and


analysis of patterns of crime within residential areas (e.g., Poyner and Webb 1991; Wiles and Costello 2000). The third rationale for limiting movement is based upon the premise that an area with high levels of through movement will allow offenders to feel less conspicuous and to blend in with the activities of legitimate users of the space. This is supported by several research studies (e.g., Poyner and Webb 1991; Taylor and Gottredson 1987). Although there have been many studies into the impact of through movement on crime levels within residential areas, two particularly rigorous studies, conducted in England, have found that burglary is higher where properties are located within an area with a higher number of connections to other areas and lower on what are referred to as “true” culs-de-sac – those with no connecting roads of pathways (Armitage et al. 2010; Johnson and Bowers 2010). In a review of the evidence relating to the impact of through movement on crime, Taylor (2002) concludes that “Neighbourhood permeability is . . . one of the community level design features most reliably linked to crime rates, and the connections operate consistently in the same direction across studies: more permeability, more crime” (Taylor 2002, p. 419). However, this assertion is not universally supported as there is some research – particularly research conducted in the last decade and using space syntax techniques – that has concluded that increased levels of through movement have a beneficial impact upon crime. Several studies have concluded that crime is concentrated in more isolated and less accessible streets (e.g., Hillier and Sahbaz 2009). One explanation for the disparity in research findings is that, although space syntax allows a greater number of properties to be analyzed (Hillier and Sahbaz looked at 101,849 properties), it also means that, for some aspects of layout (including particularly footpaths and through movement), presumptions are made about movement and patterns. For example, the remote assessment of a neighborhood may show a cul-de-sac development with no connecting footpaths, and that would be labeled a true culde-sac. However, on physically assessing

725

C

a development (i.e., completing an assessment on-site – as was the case with research completed by Armitage et al. 2010), it may become apparent that, although there are no official footpaths, residents and users of the space have created informal footpaths because they connect the development to nearby shops. Management and Maintenance Several studies have suggested that if low-level disorder such as vandalism and litter are not addressed, they can act as a catalyst for more serious crimes. Skogan (1990) refers to this as the contagion theory, suggesting that the “presence of vandalism stimulates more vandalism” (p. 39). Wilson and Kelling (1982) refer to this contagious effect as the “broken windows theory” (p. 16). Broken windows suggests that an area with existing deterioration such as graffiti and vandalism conveys the impression that (a) nobody cares so apprehension is less likely and (b) the area is already untidy so one more act will go unnoticed. This conclusion is supported by Taylor and Gottredson (1987) who found that physical incivilities indirectly influence offenders’ perception of risk in that they portray residents’ level of care or concern for the area in which they live and thus act as an indicator for the likelihood that the residents will intervene if they detect an offense taking place. Armitage (2006) assessed 1058 residential properties in England and found that those which showed signs of poor management and maintenance had experienced higher levels of prior burglary. Defensible Space As was discussed above, defensible space is a term used to describe the design features of an area that increase territorial behavior among residents and users of that space. Specific defensible space measures include maximizing the perception that a space is private or semiprivate through subtle design features such as a narrowing of the road entrance or a change in road color or texture. These environmental features are sometimes referred to as symbolic barriers as they do not physically keep people out. Brown and Altman (1983) and Armitage (2006) found that,

C

C

726

compared with non-burgled houses, properties which had been burgled had fewer symbolic barriers, as well as actual barriers such as fences and locked gates, protecting private territory from public access. Brown and Bentley interviewed offenders, asking them to judge (from pictures) which properties would be more vulnerable to burglary. The results revealed that properties showing signs of territorial behavior (such as the installation by of a gateway at the front of the property or a sign on the gate/door marking the area as private) were perceived by offenders to be less vulnerable to burglary.

Practical Application While CPTED is founded on an agreed set of theories and assumptions, the way that CPTED is applied varies across and even within different countries. It is beyond the scope of this entry to cover the different international approaches to implementing CPTED. Therefore, the focus in this section is upon the three countries England/ Wales, Australia, and the Netherlands. England/ Wales has been chosen because it has developed a holistic approach to incorporating CPTED into the planning system – through the publication of specific planning policy and guidance, the promotion of an award scheme, and the provision of specific police resources to ensure that CPTED is considered within local development. Australia, and specifically the state of New South Wales (NSW), is considered for two reasons. The first is that in NSW there is a legislative requirement for new developments to be assessed for crime risk; the second is that NSW has adopted a very different approach to implementing this requirement, one which involves very little police resources or even involvement, with crime risk assessments largely conducted by private crime reduction consultants. The Netherlands is considered for its comprehensive model of delivering CPTED within the planning system. The model includes legislative requirements, incentives, and process of delivery. Across England and Wales there are 43 police forces, and within each of these there is at least


one individual whose role involves reviewing the planning applications which are submitted to the local planning authority (within the local council) and offering CPTED advice to mitigate any potential crime risks associated with the proposed development. This role is referred to as Architectural Liaison Officer (ALO) or Crime Prevention Design Advisor (CPDA). The distinction is generally geographical, with northern police forces using the term “ALO” and southern forces using the term “CPDA.” Even within England and Wales (which share a government and associated laws and policies), the role of ALO/CPDA varies between police forces, with some ALO/CPDAs dedicated entirely to this role, while others have numerous additional roles. The role can also vary in terms of process, with some local planning authorities requiring preplanning consultation (e.g., the local authorities within Greater Manchester), while other forces have a more reactive response, with the consideration for crime prevention being entirely dependent upon the ALO/ CPDA seeking out current planning applications and contacting the planning office to offer CPTED advice. Nationally, within England and Wales, the consideration for crime prevention within planning decisions is not a requirement. The planning system in England and Wales is guided by national policy – at the time of writing the National Planning Policy Framework. This policy states that local planning policies and decisions should aim to create developments which are (among other considerations) safe and where crime, disorder, and the fear of crime do not undermine quality of life. This should be considered in local and neighborhood planning decisions but is not a legislative requirement. England and Wales also has planning guidance which directs local planning authorities, and those working within the built environment profession, as to how to develop safe neighborhoods. This guidance is entitled Safer Places – The Planning System and Crime Prevention. While this is a typical model of delivery of CPTED within England and Wales, there is one police force which has taken a very different approach – this is Greater Manchester Police (GMP) in North-West England. The GMP approach to


delivering the ALO/CPDA role involves a team of consultants who, while being civilian, are based within GMP Police Headquarters and have access to police recorded crime data, Neighbourhood Policing Teams (NPTs), and other relevant police intelligence. As was suggested above, the GMP approach also differs from other police forces in England and Wales in the emphasis placed upon preplanning consultation between the planning authority, developers, and the GMP team. Within the ten Greater Manchester local authorities, where a client wishes to apply for planning permission, the application which is made to the local planning authority must adhere to national planning policy but also to the requirements made by each local authority (through what is referred to as a validation checklist). Crucially, for Greater Manchester Police, each local authority requires that the submission of a major planning application be accompanied by a Crime Impact Statement (CIS) – a document which includes an analysis of crime statistics, reports from site visits, and local policing knowledge of the area to highlight potential crime risk and to make CPTED recommendations to mitigate those risks. The final unique element is that the GMP team is able to charge for their service. This has helped to protect the provision of this service within the current economic climate, and while other ALO/CPDA services are seeing cuts, this team has managed to sustain its services and even expand. In addition to policy, guidance, and allocation of police resources, England and Wales also implement an award scheme to encourage developers to design out crime at the planning or preplanning stage. The Secured by Design (SBD) scheme is managed by the Association of Chief Police Officers Crime Prevention Initiatives (ACPO CPI) and run on a day-to-day basis by local police ALOs or CPDAs whose role is to ensure that developments are designed and built to certain specifications. SBD is based upon the key principles of CPTED, and the standards and guidance follow those principles of physical security, surveillance, access/egress, territoriality, and management and maintenance. There have been five published evaluations of the effectiveness of the SBD scheme (see Armitage and

727

C

Monchuk 2011 for a review), each concluding that SBD confers a crime reduction advantage. In countries such as Australia, delivery varies dramatically from state to state. The state of New South Wales is selected as an example for this entry because of the model of delivery which includes a legislative requirement for a crime risk assessment to be conducted for developments considered by the local council to pose a crime risk. While this legislation shows a clear commitment to the importance of CPTED, the process of embedding this within the planning and policing system differs greatly to England and Wales. In New South Wales there is no equivalent of the ALO/CPDA role, and the closest position to this is the Crime Prevention Officer. In a similar vein to the Crime Reduction/ Prevention Officer role in England and Wales, the post includes a variety of roles and responsibilities. Within New South Wales, this post also has the additional burden of covering a large geographical area. This means that in practice, the Crime Prevention Officer cannot systematically assess all planning applications from a crime prevention perspective. Therefore, the role of conducting the required crime risk assessment and recommending alterations based upon crime risk is conducted either by private crime prevention consultants, planning companies, or the developers themselves. Clancey et al. (2011) conducted a review of 33 crime risk assessments submitted between January 2007 and October 2010 and found that these were conducted by 24 companies – 11 of which were planning firms, eight were social planning firms, seven were development companies, five were private crime prevention consultants, and two were engineering firms. The review also identified that while the guidance specifies that crime risk assessments should measure risk using police recorded crime data, hot-spot analyses, and socioeconomic data, only 16 of the 33 contained any reference to crime data, and for those which did, the analysis was broad and shallow with little indication of specific crime risk in terms of crime type, location, or modus operandi. The New South Wales model of delivery has greater legislative strength than that of England and Wales, with

C

C

728

a requirement in legislation for crime risk assessments to be conducted where a local council considers there to be a crime threat from a potential development. The process of delivering this also differs to the predominantly policebased model of England and Wales, with a greater responsibility placed upon external private consultants. Strengths of this method include legislative power to require consideration of crime as well and less pressure on police resources. Potential weaknesses relate to the threat to independence where a Crime Risk Assessment is conducted by a company who may have a vested interest in seeing the development go ahead with little or no alterations to the planning application. The Netherlands has one of the most comprehensive approaches to embedding CPTED within the planning process, and this applies to regulation, award schemes, and the process of delivery. In terms of regulation all new-built homes in the Netherlands have to comply with specific security regulations for windows and doors, and from the 1st of January 1999, planning permission could only be obtained if the application met the legal requirements for built-in security. The Netherlands also has an award scheme (similar to the UK’s SBD scheme) entitled Police Label Secure Housing. Unlike the SBD scheme, this award (which was originally owned and managed by the police) is managed by the Dutch government who adopted the police label into their planning policy guidelines, and (since 2004) every new estate or dwelling must be built in accordance with the police label or an equivalent label. Although the award was modeled on SBD, there are several distinctions which mark the two schemes apart. The first is that the label is split into three different certificates – Secured Dwelling, Secured Building, and Secured Neighborhood. These can be issued separately, but together they form the Police Label Secure Housing award. The label is also less prescriptive than SBD with more flexibility for developers aiming to achieve a secure development. The list of requirements is set out under five categories (urban planning and design, public areas, layout, building, dwelling), and these include performance


requirements (what) and specifications which indicate the way in which those requirements will be met (how). As a means of encouraging creativity and avoiding the risk of developers “designing down” to specific requirements, where a developer offers a solution which differs from that set out in the “how,” but can still demonstrate the same preventative effect, then this will be considered. The scheme also differs in that it is valid for ten years only, and after this period, a reassessment is required. In terms of the delivery of the scheme, the system is very similar to that within England and Wales. Until 2009, each police region had a number of Building Plan Advisors (Bouwplanadviseur) whose role was very similar to the ALO/CPDA role. As a response to budget cuts, the role has been civilianized and is run by the municipalities either through the employment of external consultants or civilian Building Plan Advisors located in-house.

A Critique of CPTED While many of the wider criticisms of situational crime prevention in general apply to CPTED, there remain some fundamental, unresolved weaknesses in the theory and application of this method of crime reduction. This entry concludes with a brief presentation of the main criticisms of CPTED and a discussion of how these might be addressed. Lack of flexibility: One area where CPTED has remained open to criticism is in the lack of flexibility in its principles and guidance as well as the application on the ground. This lack of flexibility may, in part, relate to the agencies traditionally involved in delivering CPTED (police, security consultants, ex-police) whose professional background and training focuses upon the requirement to follow and not challenge instructions. It may also relate to a lack of confidence, training, or experience among those carrying out the role of delivering CPTED that, in practice, leads to the rigid application of standards as opposed to adapting the design to fit a particular context. Failure to clarify confusion on specific CPTED principles: This criticism relates to the


confusion which has surrounded one particular CPTED principle – that being the impact of through movement on levels of crime. The debate centers upon the benefits of facilitating movement within an area weighed against the risks of potentially criminogenic design. For those who advocate increased connectivity, the rationale does not necessarily relate to crime reduction. The primary purpose of designing connected developments is to ensure that people can get from A to B without the need of a vehicle, thus reducing carbon emissions and the visibility of the car, and to avoid the need for residents to take unnecessarily lengthy routes. While the cul-desac layout is favored in the majority of the criminological literature on the subject, urban designers would argue that there are many negative features of this layout. It increases travel distance and therefore reliance upon the motor vehicle, it is an inefficient use of land, and it increases the difficulty of ensuring that public transport can travel close to these residential properties. While research will always present differing findings and subsequently viewpoints, this debate has led to polarized and often overstated statements regarding the crime risk of culs-de-sac versus through roads. Not only has such simplification proved unhelpful for those tasked with reducing crime through the design and manipulation of the environment, it has also led to unnecessary confusion regarding a subject for which the academic evidence appears to be relatively unambiguous. Non-standardized delivery: Perhaps linked to the lack of professionalization of the role of implementing CPTED at ground level, there is a concern regarding the considerable variation in the process of delivery, both nationally and internationally. There is not only a difference in who is in the CPTED delivery role (England and Wales being predominantly warranted police, the Netherlands civilians based within municipalities, and New South Wales security consultants) but also a difference in how CPTED is being applied. Some countries have taken the approach of introducing legislation or building regulations to require the specific security standards within residential dwellings (the Netherlands and

729

C

Scotland); others have introduced legislation to require the consideration of crime risk (New South Wales) or incentivized the consideration for crime prevention with awards such as SBD (UK) and Police Label Secured Housing (the Netherlands). While it is understandable that the delivery will differ between countries and even between states, there still remains a lack of consistency within countries such as England, which share the same legislation and regulations. Lack of clarity in scope: Perhaps less of a criticism than a note of caution that the boundaries and scope of CPTED must be made clear. While the principles cover design, building, and future management and maintenance of an area, the extent to which CPTED interventions can realistically influence so many factors is debatable. As Ekblom (2009) clearly highlights, “There is a tendency to use the label CPTED indiscriminately to cover everything that aims to prevent crime in the built environment. . .this is not conducive to focused thinking” (Ekblom 2009, p. 9). An example which highlights this is the principle of management and maintenance. The SBD scheme has historically stated that developments must have a programmed system in place to manage and maintain the area. Yet, unless this is social housing, how is this program established and, more importantly, maintained? It appears to be a principle which fits well in theory, but cannot be consistently applied in practice. While warning of the risks of extending its net too wide, there would be some merit in CPTED integrating the wider social approach addressed by what has been labeled as second-generation CPTED (see Saville and Cleveland 2003). Failure to align with other agendas: Although this criticism has begun to be addressed through work conducted by (among others) Cozens, Pease, Armitage, and Monchuk, the CPTED community has been very slow to adapt their focus to fit with contemporary issues such as social, economic, and environmental sustainability. While the two agendas of minimizing crime and maximizing sustainability may appear distinct, achievement of one (reducing crime) ultimately contributes to the other (maximizing sustainability). Crime is carbon costly, including

C

C

730

the carbon costs of police mileage in response to a crime, the replacement of stolen and damaged property, the health and other costs to victims in high crime, the costs of moving home in response to crime or fear of crime, and the maintenance and refurbishment of void properties. Crime has a huge carbon footprint, and given the prominence which the carbon reduction agenda has been given in government policy, media attention, funding for research, and the priorities of the general public, it would appear that there has been a missed opportunity to enhance the priority afforded to the importance of the consideration for crime reduction within the built environment. Failure to innovate and adapt to change: The final criticism relates to the failure of CPTED to respond to the social and economic environment in terms of its model of delivery, the principles upon which it is based, and its focus. The first example of the failure to innovate relates to the current economic crisis and how this has impacted upon funding for public services such as the police. Of the 43 police forces within England and Wales, only one has adapted its model of delivery not only to survive these cutbacks but to thrive and grow through innovation. There has also been slow progress among CPTED research, policy, and practice to adapt to the changing nature of crime, with the focus remaining on acquisitive crimes, while more common disorder issues have been sidelined.

Conclusions and Future Research While CPTED may lack some consistency in process and application, research suggests that the principles upon which it is based can work, both alone and combined with other interventions, to reduce crime and the fear of crime and to maximize social, environmental, and economic sustainability. In moving forward, CPTED must evolve, but in the words of Ekblom (2009), it must lose its historical baggage first. While there is always room for further research, the CPTED community can begin to confidently challenge some of the debates which have dominated this field and which extensive,


independent, and methodologically rigorous research has clarified. Attention should now be focused upon building upon examples of good practice both in terms of by whom and how CPTED is delivered on the ground. Implementation should be adapted to context and designed to suit the social and economic challenges of different communities. Future thinking should focus upon new models of delivery which can be implemented with limited public funding and within political environments which favor restricted legislation, regulation, and governmental interference. CPTED must also adapt to changing concerns regarding crime. The traditional focus upon acquisitive crime must widen to address public concern regarding low-level crimes and antisocial behavior and also governmental priorities such as terrorism and violent extremism. But these challenges can be seen as opportunities. Where there are problems, there is scope to develop solutions, and CPTED is a practical, cost-effective crime reduction measure which, research has shown, can adapt to many different problems and contexts.

Related Entries ▶ Agent-Based Models to Predict Crime at Places ▶ Applied Geographical Profiling ▶ British Police ▶ Broken Windows Thesis ▶ Co-offending and Offender Decision-Making ▶ Crime Mapping ▶ Crime Science ▶ Designing Products Against Crime ▶ Effectiveness of Situational Crime Prevention ▶ History of Geographic Criminology Part I: Nineteenth Century ▶ Innovation and Crime Prevention ▶ Public Housing and Crime Patterns ▶ Rational Choice Theory ▶ Residential Burglary ▶ Routine Activities Approach ▶ Situational Crime Prevention ▶ Theories for Situational and Environmental Crime Prevention

Crime Preventive Effects of Incapacitation

Recommended Reading and References Armitage R (2006) Predicting and preventing: developing a risk assessment mechanism for residential housing. Crim Prev Community Saf Int J 8(3):137–149 Armitage R, Monchuk L (2011) Sustaining the crime reduction impact of secured by design: 1999 to 2009. Secur J 24(4):320–343 Armitage R, Monchuk L, Rogerson M (2010) It looks good, but what is it like to live there? Assessing the impact of award winning design on crime. Spec Vol Eur J Crim Policy Res 17(1):29–54 Brantingham PL, Brantingham PJ (1981) Environmental criminology. Sage, Beverley Hills Brown BB, Altman I (1983) Territoriality, defensible space and residential burglary: an environmental analysis. J Environ Psychol 3:203–220 Brown B, Bentley D (1993) Residential burglars judge risk: the role of territoriality. J Environ Psychol 13:51–61 Budd T (2001) Burglary: practice messages from the British crime survey. Home Office, London Clancey G, Lee M, Fisher D (2011) Do crime risk assessment reports measure crime risks? Curr Issues Crim Justice 23(2):235–254 Coleman A (1986) Utopia on trail: vision and reality in planned housing. Hilary Shipman, London Cozens P, Saville G, Hillier D (2005) Crime prevention through environmental design (CPTED): a review and modern bibliography. Prop Manag 23:328–356 Cromwell PF, Olson JN, Avary DW (1991) Breaking and entering: an ethnographic analysis of burglary. Sage, Newbury Park Crowe T (2000) Crime prevention through environmental design: applications of architectural design and space management concepts, 2nd edn. ButterworthHeinemann, Oxford Ekblom P (2009) Redesigning the language and concepts of crime prevention through environmental design. http://reconstructcpted.wordpress.com/publicationsand-other-papers/. Accessed 21 Aug 2012 Ekblom P (2011) Deconstructing CPTED. . .and Reconstructing it for practice, knowledge, management and research. Eur J Crim Policy Res 17:7–28 Hillier B, Sahbaz O (2009) Crime and urban design: an evidence-based approach. In: Cooper R, Evans G, Boyko C (eds) Designing sustainable cities. WileyBlackwell, Chichester Jacobs J (1961) The death and life of great American cities. Random House, New York Johnson S, Bowers KJ (2010) Permeability and burglary risk: are Cul-de-Sacs safer? Quant J Criminol 26(1):89–111 Newman O (1973) Defensible space: people and design in the violent city. Architectural Press, London Pease K (2001) Cracking crime through design. Design Council, London Poyner B (1983) Design against crime: beyond defensible space. Butterworth, London

731

C

Poyner B, Webb B (1991) Crime free housing. Butterworth, Oxford Reppetto TA (1974) Residential crime. Ballinger, Cambridge, MA Saville G, Cleveland G (2003) An Introduction to 2nd generation CPTED: Part one. CPTED Perspect 6(1):7–9 Skogan WG (1990) Disorder and decline: crime and the spiral of decay in American neighbourhoods. University of California Press, California Taylor R (2002) Crime prevention through environmental design (CPTED): yes, no, maybe, unknowable and all of the above. In: Bechtel RB, Churchman A (eds) Handbook of environmental psychology. Wiley, New York Taylor R, Gottredson SD (1987) Environmental design, crime and prevention: an examination of community dynamics. Crime Justice Annu Rev Res 8:387–416 Wiles P, Costello A (2000) The ‘road to nowhere’: the evidence for travelling criminals, home office research study 207. Home Office, London Wilson JQ, Kelling GL (1982) Broken windows: the police and neighborhood safety. Atlantic Mon 127:29–38

Crime Preventive Effects of Incapacitation Hilde Wermink1, Robert Apel2, Arjan A. J. Blokland1,3 and Paul Nieuwbeerta1,4 1 Department of Criminology and Criminal Law, Leiden University, Leiden, The Netherlands 2 School of Criminal Justice, Rutgers University, Newark, NJ, USA 3 The Netherlands Institute for the Study of Crime and Law Enforcement (NSCR), Amsterdam, The Netherlands 4 Utrecht University, Utrecht, The Netherlands

Overview Among the manifold goals of penal confinement, incapacitation aims to impose a period of “time out” from the criminal career, by removing the opportunity for an individual to commit crime in This chapter is largely based on two articles that are published elsewhere (Blokland and Nieuwbeerta 2007; Wermink et al. 2012). Some parts are incorporated integral.

C

C

732

the community for the duration of his or her sentence. After a period of empirical stagnation, questions concerning the incapacitation effect of imprisonment have recently become salient. This recent empirical attention has been spurred, in part, by the unparalleled growth in the use of incarceration in Western society over the last three decades. This growth is most stark in the United States. Yet in Europe as well, two-thirds of the 35 countries surveyed experienced growth in their incarceration rate during the first half of the 2000s (Aebi et al. 2006). A growing literature has become attentive to the effects of incapacitation. This entry gives an overview of the current literature and pays attention to the (a) effects of general incapacitation and (b) effects of specific incapacitation of frequent offenders.

Effects of General Incapacitation Effects of Prison Approaches to Estimation of General Incapacitation Effects of Prison A key goal for a study of incapacitation is to estimate the number of crimes that an incarcerated offender would have committed, where he/she is free in the community rather than confined in prison. This is taken as an estimate of the incapacitation effect, defined as the number of crimes averted by physically isolating an offender from society at large (Blumstein et al. 1978). This task would seem straightforward at first glance, but the problem on closer inspection is that this quantity is a counterfactual which can never, even in principle, be directly observed. Estimation of the incapacitation effect of imprisonment is fundamentally an actuarial exercise. The empirical challenge is one of overcoming the counterfactual problem, that is, providing a credible estimate of behavior that is not observed because an individual is locked up in jail or prison rather than being free in the community. The key quantity for incapacitation is known as lambda, l, representing the annual offending frequency conditional on active


offending, which can be taken as an estimate of the number of crimes avoided through incarceration. Zimring and Hawkins (1995, p. 81) observe that there are two approaches to estimating lambda: [T]o determine the level of crime that would have occurred if a particular group had not been confined, one must either study the criminal activity of the same group at a different time in their lives to estimate what that group would have done if not confined, or one must study the behavior of persons other than those confined to approximate the crimes avoided by imprisonment in the past. It can be referred to the first as a within-person counterfactual approach, and to the second as a between-person counterfactual approach. Point of departure for both approaches mentioned is that the crimes the offender would have committed while free are actually prevented by incarcerating the offender. As such, these methods do not account for the mitigating effects of replacement – other offenders filling in the position left vacant by incarcerating the original offender – or the fact that co-offenders might have committed the particular crime regardless of the absence of the incarcerated offender. Replacement however seems especially relevant for drug offenses and is deemed largely irrelevant for the large majority of crimes. Group offending in turn might be more relevant for juveniles than for adults since juvenile offenders more often co-offend. Within-Person Approach The most influential incapacitation studies are universally of the first kind in Zimring and Hawkins’ (1995) typology. These studies provide within-person counterfactual offending rates from self-report surveys of arrestees or prison inmates. With this approach, targeted individuals are questioned about their criminal activity during the months leading up to their arrest or confinement. These estimates are taken as the number of crimes they committed on an annual basis when they were free (prior to incarceration) and by implication, the number of crimes they would have committed per year during the time that they were incarcerated. Therefore, the counterfactual offending rate for incarcerated


individuals is their own offending rate in the months preceding their confinement. The best-known source of a within-person counterfactual is the Rand Corporation’s second inmate survey, a study of over 2,000 male inmates in California, Michigan, and Texas (Chaiken and Chaiken 1982). This study, for example, revealed that annual offense frequencies for ten different crimes were highly skewed, with active offenders at the median committing 15 offenses and offenders at the 90th percentile committing 605 offenses (these figures exclude drug offending). They estimated the mean offending rate to be between 187 and 278 crimes per year, depending on how ambiguous survey responses were treated. Based on these figures, incarceration was shown to have the capacity to substantially incapacitate criminal behavior. In fact, on the basis of this research, policy-oriented criminologists began to advocate selective incapacitation of high-rate criminal offenders as an explicit penal policy. These incapacitation studies had limitations, however. First and foremost is the question of the reliability of offender self-reports of their criminal behavior (Spelman 1994; Zimring and Hawkins 1995). Specifically, it is unclear whether incarcerated offenders, particularly high-rate offenders, can accurately recall their prior criminal activity. Even if they can, they may not be motivated to report it honestly to interviewers. Spelman (1994), in fact, found evidence of both overreporting and underreporting in the second Rand inmate survey, but by individuals at different locations in the distribution of offending rates. Second, the presence and nature of crime spurts can introduce serious distortions in estimates of lambda (Blumstein et al. 1986) and therefore in estimates of incapacitation effects. Research has shown that offenders experience relatively short periods of high-rate offending immediately prior to incarceration, meaning that incapacitation effects will be severely overstated, especially if reporting windows are comparatively narrow. Moreover, a portion of the pre-incarceration crime spurt appears to be artifactual rather than behavioral, that is, a function of the way that individuals are

733

C

filtered through the criminal justice system and selected for custodial sentences rather than reflective of genuine growth in offending frequency. Finally, these early incapacitation studies were guilty of overly optimistic assessments of the incapacitative benefits of incarceration. The studies showed self-report offending rates to vary widely from one jurisdiction to another. For example, Chaiken and Chaiken (1982) reported the median total nondrug crime frequency among prisoners (not including jail inmates) in California, Michigan, and Texas to be 42, 17, and 9, respectively. The median offender reported committing 4.2 nondrug offenses per year in Nebraska (Horney and Marshall 1991), 4.4 nondrug offenses in New Orleans (Miranne and Geerken 1991), and 12 nondrug offenses in Wisconsin (DiIulio 1990). Relatedly, the incapacitation estimates from this research tend to assume constant offending during the reference period, whereas Horney and Marshall (1991) demonstrated that most offenders were actively involved in crime only intermittently and that failure to account for this intermittency would inflate estimates of the incapacitative benefits of imprisonment when offense rates are annualized. Between-Person Approach Until recently, within-person counterfactual studies exhausted all empirical research in the incapacitation tradition. Three studies in the recent years have adopted the second approach in Zimring and Hawkins’ (1995, p. 81) typology, that of estimating the offending rate of persons other than those confined to approximate the crimes avoided by imprisonment. Sweeten and Apel (2007) relied on propensity score matching to select incarcerated and non-incarcerated individuals who closely resembled each other on a wide variety of background variables, including criminal history. Using data from a contemporary, nationally representative sample of American youth, they limited their attention only to individuals who were not incarcerated prior to the reference conviction. They estimated lambda among the non-incarcerated sample to be about nine offenses per year

C

C

734

among 16–17 year olds and about six offenses per year among 18–19 year olds. Owens (2009) took advantage of a change in the sentencing guidelines in Maryland that lowered the age at which an offender’s juvenile record could be used to add criminal history points for the purpose of criminal sentencing. This policy change effectively reduced the recommended sentence faced by certain groups of young adults with a juvenile criminal history. In her study, individuals who were sentenced after the policy change accumulated 2.8 arrests per year (with an implied Index offending rate of 1.5 offenses per year) during the time that they would still have been confined under the old sentencing regime. Wermink et al. (2012) relied also on the between-person approach, but combined it with the within-person approach using a sample of all offenders convicted of a crime in the Netherlands in 1997. Using the between-person counterfactual approach, offenders who were convicted but not imprisoned in 1997 were matched to those individuals convicted to imprisonment in 1997. To estimate the incapacitation effect, they investigate the criminal behavior of the matched controls after their matched counterpart was sent to prison. Using the within- and betweenperson counterfactual approach, their best estimate is that between 0.17 and 0.21 convictions are prevented per year of first-time incarceration. Distinguishing for different social groups, their results show larger first-time incapacitation effects for male offenders compared to female offenders and for juveniles compared to adults. Their incapacitation estimates are considerably smaller than those of earlier studies. Remarks on Effects of General Incapacitation In conjunction with these prior findings, the current results inform both penal policy and the public discourse surrounding the use of imprisonment as the societal reaction to crime. The current results suggest that in as far as imprisonment is used with the primary goal of reducing crime, a general increase of imprisonment as the sanction of choice is not likely to yield high crime


control benefits. While the incapacitative effects of imprisonment are likely to be larger when specific offender groups are selectively targeted, studies examining the effects of such selective policies mostly find rather disappointing effects. In addition, prior studies indicate that even a relatively short prison sentence may have long-lasting detrimental effects in terms of increased offending. Given the incapacitation effect of first imprisonment is not that large to begin with, it is therefore likely to be offset – if not overruled – by the long-term collateral effects of incarceration. This is not to say that imprisonment does not have any function in the criminal justice system. It does however clearly point to its limitations and downsides, underscoring that imprisonment is to be regarded not as a panacea to the crime problem, but merely as an ultimum remedium.

Selective Incapacitation of Frequent Offenders Criminal justice policies also aim for effects of selective incapacitation, that is, imprisonment policies specifically targeting some predefined group of offenders. Selective incapacitation policies are increasingly being implemented, both in the USA and in Europe. The rationale behind these policies is straightforward: reducing crime at lower cost. The more skewed the distribution of offense frequency in the offender population, the more offenses can be prevented by selectively incapacitating those offenders that are at the high end of this distribution. If frequent offenders are also more serious and violent offenders, as both theory and empirical research suggests, the benefits of removing these offenders from society may be even greater (Piper 1985). Selective incapacitation however is not uncontroversial. Ethical objections have been raised to the sanction disparity inherent to these selective policies (e.g., Moore 1986). Others have stressed the “stochastic” selectiveness that is already built-in to the judicial system, predicting low yields of an additional focus on frequent offenders (Blumstein et al. 1993). Finally,


researchers have pointed to the rising prison costs, wastage of prison capacity, the costly care of the aging prison population, possible problems in monitoring growing groups of inmates serving long-term sentences, and the growing number of appeals lodged by those offenders meeting the selective criteria that may result from selective policies. Research into the effects of incapacitation has either been at a macro-level, combining aggregate data on prison populations and crime rates, or at the micro-level using models combining information on individual offense rates, probability of conviction, and sentence length. Aggregate research is faced with the problem of separating the effect of imprisonment on crime from the effect crime has on imprisonment. Additionally, this type of research can only address the combined deterrent and incapacitative effects of imprisonment. Micro-level research does focus solely on imprisonment’s incapacitative effect, but finds that outcomes vary widely depending on the estimates of the various criminal career characteristics used. Also, thus far, these latter studies pertain to adult offenders only. Macro-level Studies on Effects of Selective Incapacitation Studies into the effects of selective incapacitation on crime have used one of two approaches: macro- and micro-level studies. Macro-level or, as Spelman (2000a) refers to them, top-down studies have relied on aggregate data on societylevel crime rates and prison populations. These studies typically estimate the relation between crime rates and prison populations, while controlling for other variables that may affect crimes rates like the population’s age distribution or its unemployment rate using multivariate regression techniques. Researchers using aggregate data have specifically focused on the implementation of selective policies, most often the “three strikes” laws (Clark et al. 1997; Vitiello 1997; Zimring et al. 2001). As with the studies on imprisonment in general cited above, many of the studies into selective imprisonment did not control for simultaneity. This may lead effects of these selective

735

C

policies to be underestimated, since outcome measures may also be affected by preexisting increasing crime trends that gave rise to implementing the selective policy in the first place. Many studies also have a short follow-up period. However, as Kovandzic (2001) notes, the incapacitative effects of selective policies most probably do not occur for many years after the policy is implemented, since most offenders selectively imprisoned under the new policy would have been imprisoned for at least some period even under prevailing nonselective policy. Using long-term data from Florida, Kovandzic (2001) finds little evidence for a general deterrent effect of Florida’s habitual offender laws on crime and concludes that its limited effects on crime are mostly due to incapacitation. It must be noted, however, that while policy shifts on imprisonment in the USA have been both quick and dramatic, these data are far from experimental (Spelman 2000a). Decisions to expand imprisonment or to implement a selective policy did not occur randomly, and therefore results of these changing conditions are confounded with existing between-state differences in crime trends and prison rates and all other factors that may have influenced either one of these variables. Results, even of the most complex models, are therefore still potentially subject to bias. Micro-level Studies on Effects of Selective Incapacitation Instead of relying on aggregate data, micro-level, or bottom-up, studies try to model the workings of the criminal justice system in more detail. Making estimates of the offense rate per year, the probability of arrest and conviction, and the average sentence length, these studies calculate the proportion of the criminal career that a typical offender will spend imprisoned. This proportion then represents the benefits of imprisonment in terms of decreased crime rates. Unlike most aggregate studies, these micro-level studies focus solely on the incapacitative effect of imprisonment. Recent studies by DiIulio and Piehl (1991), Spelman (1994), and Piehl and DiIulio (1995)

C

C

736

have yielded elasticity estimates ranging from 0.16 to 0.26. However, recognizing the still limited precision of the estimates used, Spelman (2005) concludes that it is more reasonable to expect a 0.1–0.3 % decrease in crime for a 1 % prison expansion. Micro-level studies necessarily make assumptions about several criminal career characteristics. Small – and plausible – changes have been demonstrated to considerably affect the studies’ outcomes (Spelman 1994). While over the years these studies have gained validity, as a result of more detailed data on criminal career becoming available, they still need to rely on simplifications regarding the distribution and interrelationships of the various criminal career dimensions (Spelman 1994, p. 110). In an effort to circumvent uncertainties and simplifications tied to the use of criminal career estimates, Bernard and Ritti (1991) designed a rolling cohort methodology that allowed them to estimate the incapacitative effect of selective policies using actually observed data. Their analyses were based on count data from the Philadelphia birth cohort. In brief, they simulated the implementation of some hypothetical selective policy and then counted the actual number and type of offenses recorded after the youth had been incapacitated under the hypothetical policy. These offenses would make up the incapacitative effect of the selective policy since they would not have occurred under the selective policy, but actually did occur under existing, nonselective policy. Based on only the “chronic” offenders in the Philadelphia cohort (those offenders who accumulated at least five police reports), a selective policy which entailed imprisoning a youth after his second arrest until his eighteenth birthday resulted in a 25 % decrease in all reported crimes and a 35 % decrease in the serious crimes reported for this cohort. A similar study using Swedish birth cohort data showed that imprisoning every recidivist in this cohort for a period of 2 years would reduce registered crime for this cohort by 28 % (Andersson 1993). However, while both studies found that selective incapacitation would reduce crime, both studies also showed implementation of a selective policy would markedly


inflate prison populations by as much as 22 times the rate under nonselective policy. Finally, Blokland and Nieuwbeerta (2007) use data from the Netherlands Criminal Career and Life-course Study to estimate the incapacitative effects of alternative selective prison policies. Using the rolling cohort method, implementations of various penal scenarios differing in selection rate, sentence disparity, and selective accuracy are simulated. Results show that it is hard for selective policies to yield a positive societal result: costs of imprisonment typically exceed benefits gained from crimes prevented. Selectively incapacitating presumed frequent offenders on the basis of the number of offenses they have committed in the past – or in the preceding 5 years – leads to a substantial decline in crime. This decline is the greatest in the first years after the introduction of the selective policy as a rising share of offenders are selectively imprisoned. After some time, depending on the duration of the standard prison term, some of the offenders having completed their prison term rejoin the population, thus causing crime to rise again. The ultimate result of our simulations – assuming only incapacitative effects – is a 25 % decrease in crime under the strictest regime. However, these benefits are offset by the fact that the introduction of a selective policy causes the prison population to rise considerably – up to 45 times the population under prevailing nonselective policy for a very strict selective regime. As a result, if a selective policy based on the total number of prior convictions is to be efficient, the average conviction prevented by the selective incapacitation of frequent offenders would have to generate four to eight times as much in benefits than it would cost detaining an offender for 1 year. The study by Blokland and Nieuwbeerta (2007), however, also shows that on the basis of the total crime costs for 2004, the benefits from preventing an average conviction are only approximately twice as high as the costs of 1 year of prison. Under the assumption of equal probability of conviction across all offenders, this would mean that even the selective incapacitation scenario with the most favorable CBR


(five prior convictions in the entire criminal history, 5 years of prison) is only efficient if the costs per prisoner can be reduced from 190 to 58 Euros per day, which cannot be considered a realistic option. Allowing disparity in the probability of conviction, however, may – at current prison costs – result in a breakeven situation for some selective policies if frequent offenders are less than half as likely as non-frequent offenders to be convicted given an actual crime. Finally, risk estimation merely on the basis of criminal past generates high wastage of prison capacity. A lower selection rate (employing a stricter selection criterion) reduces this problem, but the resulting policy has hardly any added crime reduction value over prevailing nonselective policy. Remarks on Selective Incapacitation Selectively incapacitating offenders on the basis of an estimation of their future behavior rooted in their past criminal behavior points to the growing influence of risk-oriented thinking in criminal legislation and judicial policy. Frequent offenders belong to the category of offenders with a high risk of recidivism, and society thinks that they therefore ought to be removed from its midst. Offenders with a sufficiently extensive criminal history are thus penalized not for what they have done but for what they are expected to do in the future. This implies that a prison term is converted from a punitive into a preventive measure, one that our results suggest is used unnecessarily in many cases. Moreover, under a selective policy, there is no longer any relationship between the offense committed and the duration of the sentence. Selective incapacitation conflicts with a number of fundamental principles of criminal law systems. On the basis of our research results, which were mainly negative about the selective detention of frequent offenders, one may well wonder whether such a fundamental change to criminal law systems is actually opportune and even whether this policy has any empirical legitimacy. It is possible and even likely that criminological research would point to other policy alternatives, primarily those that do find support in criminological

737

C

research, as a more obvious choice – for example, investing more in the socioeconomic functioning of prisoners by improving and expanding the probation and aftercare service.

Related Entries ▶ Estimating the Effects of Incapacitation ▶ Incapacitation ▶ Penal Philosophy and Sentencing Theory ▶ Theories of Punishment

Recommended Reading and References Aebi MF, Aromaa K, de Cavarlay BA, Barclay G, Gruszczyn˜ska B, von Hofer H, Hysi V, Jehle J-M, Killias M, Smit P, Tavares C (2006) European sourcebook of crime and criminal justice statistics, 3rd edn. Boom Juridische, Den Haag Andersson J (1993) A longitudinal simulation study of incapacitation effects. Project Metropolitan No. 35. Department of Sociology, University of Stockholm, Stockholm Bernard TJ, Ritti RR (1991) The Philadelphia birth cohort and selective incapacitation. J Res Crime Delinq 28(1):33–54 Blokland AAJ, Nieuwbeerta P (2007) Selectively incapacitating frequent offenders: costs and benefits of various penal scenarios. J Quant Criminol 23:327–353 Blumstein A, Cohen J, Nagin D (eds) (1978) Deterrence and incapacitation: estimating the effects of criminal sanctions on crime rates. National Academy of Sciences, Washington, DC Blumstein A, Cohen J, Roth JA, Visher CA (eds) (1986) Criminal careers and career criminals, vol 1. National Academy Press, Washington, DC Blumstein A, Canela-Cacho JA, Cohen J (1993) Filtered sampling from populations with heterogeneous event frequencies. Manage Sci 39:886–899 Chaiken JM, Chaiken MR (1982) Varieties of Criminal Behavior. Report No. R-2814-NIJ. Rand, Santa Monica Clark J, Austin J, Henry DA (1997) “Three strikes and you’re out”: a review of state legislation. National Institute of Justice, Washington, DC DiIulio JJ Jr (1990) Crime and punishment in Wisconsin. Wisconsin Policy Research Institute, Milwaukee DiIulio JJ, Piehl AM (1991) Does prison pay? The stormy national debate over the cost-effectiveness of imprisonment. Brookings Rev 9:28–35 Horney J, Marshall IH (1991) Measuring lambda through self-reports. Criminology 29:471–495 Kovandzic TV (2001) The impact of Florida’s habitual offender law on crime. Criminology 39(1):179–203

C

C

738

Miranne AC, Geerken MR (1991) The New Orleans inmate survey: a test of Greenwood’s predictive scale. Criminology 29:497–518 Moore MH (1986) Purblind justice: normative issues in the use of predictive or discriminating tests in the criminal justice system. In: Blumstein A, Cohen J, Roth JA, Visher CA (eds) Criminal careers and ‘career criminals’, vol 2. National Academy Press, Washington, DC, pp 314–335 Owens EG (2009) More time, less crime? Estimating the incapacitative effect of sentence enhancements. J Law Econ 52:551–579 Piehl AM, DiIulio JJ (1995) Does prison pay? Revisited: returning to the crime scene. Brookings Rev 13:20–25 Piper ES (1985) Violent recidivism and chronicity in the 1958 Philadelphia cohort. J Quant Criminol 1(4):319–344 Spelman W (1994) Criminal incapacitation. Plenum Press, New York Spelman W (2000a) What recent studies do (and don’t) tell us about imprisonment and crime. Crime Justice 27:419–494 Spelman W (2000b) The limited importance of prison expansion. In: Blumstein A, Wallman J (eds) The crime drop in America. Cambridge University Press, Cambridge Spelman W (2005) Jobs or jails? The crime drop in Texas. J Policy Anal Manage 24(1):133–165 Sweeten G, Apel R (2007) Incapacitation: revisiting an old question with a new method and new data. J Quant Criminol 23:303–326 Vitiello M (1997) Three strikes: can we return to rationality? J Crim Law Criminol 87:395–481 Wermink H, Apel R, Nieuwbeerta P, Blokland AAJ (2012) The incapacitation effect of first-time imprisonment: a matched samples comparison. J Quant Criminol. doi:10.1007/s10940-012-9189-3 Zimring FE, Hawkins G (1995) Incapacitation: penal confinement and the restraint of crime. Oxford University Press, New York Zimring FE, Hawkins G, Kamin S (2001) Punishment and democracy: three strikes and you’re out in California. Oxford University Press, New York

Crime Profiling ▶ Criminal Investigative Analysis

Crime Repetition by Youth ▶ Specialization in Juvenile Offending

Crime Profiling

Crime Scene Analysis ▶ Criminal Investigative Analysis

Crime Scene Interpretation ▶ Criminal Investigative Analysis

Crime Scene Profiling ▶ Behavioral Investigative Advice

Crime Science Gloria Laycock Jill Dando Institute of Security and Crime Science, University College London, London, UK

Overview With detection, arrest, and prosecution as the universal response to crime, despite their manifest failure to reduce the problem, the time is right for a complete rethink about crime and crime reduction. This entry considers the need for a new discipline – crime science – which it is argued is necessary if crime control is to be improved. The entry suggests that greater emphasis on experimentation and a more scientific approach to crime problems under the banner of “crime science” will make a significant difference. Examples are provided of the advantages of scientists working together to address crime problems but the entry also argues for greater emphasis on ethical and aesthetic issues if we are to avoid potential pitfalls of this new approach.

Crime Science

739

C

Introduction

Background Description

This entry discusses aspects of crime science, which is a new approach to the control of crime, although its roots go back a further 30 or more years. It was formally established in 2001 at the suggestion of Nick Ross – a journalist and broadcaster who with others established the Jill Dando Institute of Crime Science in memory of his murdered colleague. A discussion of crime science in an encyclopedia on criminology may seem out of place but criminology and crime science are close in background, interests, aims, and objectives which makes them close bedfellows, and followers of each discipline should see each other as potential collaborators and close associates rather than as rivals for students, grants, or kudos (Smith and Tilley 2005). Indeed many criminologists could, if they so chose, also describe themselves as crime scientists rather as a rocket scientist could also call himself a physicist or just a scientist. The next section describes crime science – it looks at some of the history of the subject and the reasons for its development, what it is, what it is not, and how it relates to criminology and other mainstream disciplines. The following section describes the “state of the art,” briefly outlining some of the key findings of crime science most of which derive from the roots of the subject in environmental criminology (which is not directed at the study of crimes against the environment; it is concerned with the immediate “environmental” determinants of criminal behavior). After that, we look at some controversies – as with any science, there are a number of issues relating to the subject matter itself but in the case of crime science there are also questions around whether there really is a need for the discipline and how it can achieve one of its aims which is to adopt a multidisciplinary approach to the problem of crime control. The final section considers some “open questions” associated with crime science needing further thought or greater academic attention.

It is suggested that there are four major and related elements linking science and crime (Laycock 2005). Firstly, science can help us by improving our understanding of crime and its causes. This is the domain of the social sciences in general and criminology, psychology, and sociology in particular. Secondly, it can help us by making crimes more difficult to commit through the technologies that are developed on the back of it. Examples come from CCTV systems, car deadlocks and immobilizers, luggage scanners at airports, and so on. Thirdly, it can help us in catching offenders more quickly and bringing them to justice. The obvious example here is forensic science including forensic psychology, fingerprinting, and genetics and DNA technologies. Finally, it can help us in thinking like scientists think, that is, in testing hypotheses and developing a body of knowledge enabling the articulation of theory. It is this last point which is common to all those who see themselves as crime scientists and in this there are commonalities with experimental criminologists (Sherman 2009), which are expanded upon further below. Crime science grew from environmental criminology and crime analysis, both of which have a historical focus upon crime prevention, although crime science takes a somewhat broader view of prevention and includes the detection and disruption of crime. It is thus an outcome-focused approach rather like medical science, with its emphasis on the control of disease through prevention, diagnosis, and treatment. Also like medical science, a range of professions is associated with the achievement of the aims. This is particularly so in preventive medicine where parents teach their children to wash their hands, local authorities provide drains and sewage systems, and the government provides inoculation programs, all of which contribute to the maintenance of good health in society. Similarly with crime prevention, individuals lock their homes and vehicles, look after their children and families, and support their communities in

C

C

740

Crime Science

Crime Science, Fig. 1 Relationship between crime science and other disciplines

preventing crime and disorder, while the central and local government provide a police service and criminal justice system and might also encourage commercial organizations and industry to play their part in crime prevention through the better design of goods, services, and management systems. Medical science also makes use of a range of scientific approaches in addressing its problems, so we have medical physics and engineering where products are designed to deal with specific medical problems such as heart monitors and limb replacements. Chemistry also plays a major part in medical science in the development of drug treatments, as increasingly do biology and genetics. Again, crime science takes the same approach in encouraging other scientists to think about the application of their disciplines to the control of crime. Figure 1 below shows diagrammatically what we have in mind. There is no significance to subjects being in the same box and it is not an exhaustive list, but note that the social sciences are included, particularly criminology and psychology which have an obvious special relevance to the control of crime. Crime science is primarily about the testing of hypotheses associated with the understanding, prevention, disruption, and detection of crime. In the social context, this is a much more challenging task than in a laboratory, where control of extraneous variables is perhaps more easily achieved. Indeed some might argue that experimentation in society is neither possible nor desirable. Crime scientists would disagree and, as is demonstrated in the section below, not all experimentation has to be conducted in the real world. In order to carry out experiments or to speculate about cause and effect, we start with the collection of data. These data then need to be

manipulated logically and rationally drawing on the currently best available evidence. In this way hypotheses can be developed and tested, theories articulated, and understanding increased. Fundamentally, this is the scientific method (Townsley et al. 2003) which has been promoted to police practitioners and their partners under the acronym SARA (Eck and Spelman 1987) standing for scanning, analysis, response, and assessment. At each stage in this process, we are developing and discarding hypotheses. Take, for example, a police commander who may feel that in his police area there is a problem of theft of vehicles from outside residences overnight. This is his starting hypothesis, which it is possible to test using police data. His investigation (part of the scanning process) may show that there is indeed such a problem but that it is in only in the northern part of the police area and there is a much larger issue of theft of motor bikes and scooters in the south. It is then possible to formulate various hypotheses about why the thefts may be happening there (the analysis), to test them, and to implement a response or responses on the basis of the subsequent analysis (the response element of SARA). We then have the hypothesis that our action will reduce the problem by firing a particular mechanism (Pawson and Tilley 1997), which can then be refuted or supported (the final assessment stage of SARA). So we see that at each stage of the SARA process, hypotheses are being developed, tested, and discarded or, if supported, then acted upon. The formulation of hypotheses is not as simple or straightforward as it might sound. It involves asking questions about what might be causing what, in what context, and then testing the resulting hypotheses. Experience of teaching crime science to practitioners over the past decade has shown it to be one of the most

Crime Science

challenging aspects of the process. Articulating hypotheses involves asking clever questions, which is arguably what the best scientists do. Newton’s simple question – why did the apple fall – which led to the discovery of gravity is a stunning question. Indentifying hypotheses is the point at which science involves judgment, which is not in itself a science. Judgment can be thought of as common sense, which as we all know is not too common; indeed Huxley is quoted as saying that “Science is simply common sense at its best, that is, rigidly accurate in observation, and merciless to fallacy in logic” (Attributed to Huxley: see http://quotes.prolix. nu/Science/). It is through the articulation and testing of hypotheses that theories can be developed. The theoretical basis of crime science is that developed by the environmental criminologists and draws upon four “theories” or perspectives on crime. These are described in detail elsewhere in this volume [insert page number and reference]. In summary, they are the rational choice perspective, routine activity approach, crime pattern theory, and social/environmental and behavioral psychology. They are all concerned with the reduction of crime and the first three are sometimes called “opportunity theories” or the criminologies of everyday life (Garland 1996). Their focus is on the effects of the immediate environment on behavior, and although crime science would include the study of more remote causes of crime involving longitudinal studies, for example, it is largely concerned at this stage in its development with achieving more immediate reductions. Another feature of crime science, perhaps obvious from its focus on the reduction of crime, is its practical nature. Although concerned with theory development, it is also concerned with practical outcomes. There are implications from this for the ways in which research is carried out and particularly for the ways in which it is presented. In this, crime scientists can take lessons from the work produced by the British Home Office Police Research Group which for over a decade from 1992 published short research reports which were intended to be read by

741

C

intelligent practitioners, particularly the police, who may not have had specific research training and who would not have ready access to academic journals or even the interest in reading them if they did. Similar short summaries are now also available from a number of other agencies including the Australian Institute of Criminology and the US COPS Office website (www. popcenter.org). One of the most widely publicized themes from the body of work carried out under the auspice of the Home Office Crime Prevention Unit and later the Police Research Group was related to repeat victimization (Laycock 2001), which specifically tested a variety of hypotheses over a 15–20-year period, all intended to develop understanding of the statistically observable fact that crime concentrates on victims, places, and products and that by concentrating on their subsequent protection crime can be reduced. In keeping with the desire to influence police practitioners, much of this work did not immediately feature in the academic literature, and this in itself causes problems for academics who are often rewarded in the universities not for the practical influence of their work but for its publication in top journals. A specific example of this is the booklet written by Clarke and Eck and first published as “Become a Problem-Solving Crime Analyst” in 2003 and later, with minor changes, published as a US version. It has since been translated into almost 20 languages and has been hugely influential in supporting the development of police problem solving and crime analysis: But it does not feature in the academic literature and most academics would not, therefore, be aware of it.

State of the Art This section presents some of the more recent research carried out by crime scientists and which are at what we would currently see as the frontier of the research agenda. Examples are chosen from the four areas suggested as those on which crime science draws – social sciences which help our understanding of crime, physical

C

C

742

sciences including computer science which can assist in the prevention of crime, forensic sciences which support detection, and the use of scientific method to test hypotheses, which all the sciences do but which are discussed here in the context of crime control. Perhaps one of the most significant observations from criminology over the past few decades has been the extent to which crime concentrates in space (Sherman et al. 1989). This began as not much more than an interesting observation. Sherman’s early work, for example, showed the extent to which there were “hot spots” of crime in major cities of the United States and which he suggested might direct police patrol. More specifically, as crime scientist Ken Pease (Pease 1998) demonstrated, crime also concentrates on individuals, and as noted above this phenomenon of repeat victimization has become the focus of a number of hypotheses about ways in which to prevent crime (Laycock 2001). The conclusion being: Protect victims and crime goes down. This appears to be the case in relation to a number of different offenses and can be observed in a range of different jurisdictions (Farrell and Pease 2008). Recent work has investigated the way in which these spatial crime clusters are built up. For example, Bowers et al. (2004) have demonstrated that crime clusters in space and time. Domestic burglary hot spots are built up in ways similar to those seen in the foraging of animals (Johnson et al. 2009). The burglar will “hit” an area and then move on before he or she can be arrested. The researchers draw on the statistical methods used by epidemiologists in analyzing their data. They argue that rather like an outbreak of disease, the crimes occur in spates, which are often small enough for the police to patrol. In a series of related studies, Johnson and Bowers have shown that repeat victimization, which is partially explained by their work described above, has also led to near neighbors of burglary victims being at heightened risk for a short period following the initial offense. Working with international colleagues, they have demonstrated that the same effect can be found in other jurisdictions (Johnson et al. 2007).

Crime Science

All of this work has clear implications for crime prevention and detection. We can see from this substantial body of work the extent to which initial statistical observations from criminologists provided fertile ground for the development of hypotheses by crime scientists, which could then be developed to both inform practice and further the development of theoretical understanding of the phenomenon of crime itself. It is noted here that Johnson, Bowers, and the majority of their international collaborators have relatively little formal training as criminologists. There are a number of innovative approaches to crime control that might be discussed on the basis of technologies developed from various scientific discoveries. One of the more recent and certainly more controversial is the introduction of full-body scanners at airports and other transport hubs, which are intended to reduce the terrorist threat or to address smuggling and other offenses which involve carrying contraband on the body. The majority of these scanners are based upon either backscatter X-ray technology, which detects radiation reflected from the body, or passive or active millimeter wave technology which interprets natural radiation from bodies (in the case of passive millimeter scanners) or which reflects extremely high-frequency radio waves off the body (for active millimeter scanners) in order to detect illicit material that may be hidden underneath clothing. The potential threat to health is an obvious issue, which is still the subject of discussion, although the extent of the threat depends upon the system that is used. A problem common to all such systems is, of course, the potential invasion of privacy. The fact that these scanners can operate relatively quickly and do not require the subject to remove clothing is likely to increase their use, without any particular “just cause” as might otherwise be expected. This is an example of the importance of alerting the scientists and technologists to the social consequences of some of their inventions. Although it could be argued that innovators are not responsible for the use to which their innovations are put, there is increasing interest in encouraging scientists to think through some of

Crime Science

these ethical issues as is discussed more fully below. Computer science offers a particularly interesting contribution to crime control first through the realization that prevention must be paramount. Given the ubiquity of the use of computers and the (apparent) anonymity offered by the World Wide Web, effort spent in detecting crimes, which may have been committed by offenders in different countries, is easily seen as futile under most circumstances. So it is very clear that the design of the system itself must be such as to reduce the opportunities for crime and other abuses. We thus see arrangements for online banking, shopping, e-mailing, etc. continuously subject to improvements in the security arrangements associated with them. One observation in relation to the development of computer security systems is the need to take greater account of so-called wetware, that is, humans. Designers of computer password systems make a point of ensuring that the systems will refuse to recognize easily remembered passwords, and as a consequence people tend, for example, to write down passwords on bits of sticky paper or in other insecure places, which of course defeats the purpose. Having the designers of Internet security systems work together with psychologists in the development of these systems should improve the systems themselves and reduce the risk to security. This is a clear example of the advantage of individuals from more than one discipline working together to address a common problem. Forensic science is now a well-established approach to the detection of crime, and with the developments associated with fingerprint technology and DNA, we see some powerful techniques becoming available to the police and other security experts. Forensic science has a long history of scientists from different disciplines working together, all of whom might call themselves crime scientists but who might also benefit from some of the additional perspectives offered by the social sciences. So, for example, in deciding how to apportion scarce forensic resources when investigating domestic burglary, priority might be given to repeat crimes where we know that

743

C

there is around an 80 % chance that the same burglar has returned. Such a burglar would almost certainly have a record on the UK National DNA Database and we also know that although a DNA trace is found in only about 5–6 % of burglary offenses, when it is found, there is an over 40 % chance of finding a link to a known offender (Burrows et al. 2005): This percentage is likely to rise in the case of repeat instances of burglary. So a little knowledge of criminology on the part of the forensic scientist would arguably improve their day-to-day performance. Turning to what is new in forensics, we might look at forensic geosciences, which offer the promise of a whole new branch of science in support of crime detection (Morgan and Bull 2007). Drawing on insights from geology, they explore the distribution of soils, sediments, and dust from crime scenes and address the extent to which they transfer from one place or context to another. It is a fundamental of geography that every contact leaves a trace, and this is what much forensic science is based upon. It has been recently demonstrated, for example, that sand and soil particles have a “fingerprint” or characteristic profile depending upon where they are found. It is thus possible to challenge an assertion by a suspect that he or she was not present at a particular site by comparing the profile of granules found at a crime scene with those on the footwear of the suspect. All of the sciences test hypotheses but as noted earlier this can be problematic in the real world. It may be unethical, impractical, or politically unacceptable to carry out controlled experiments in the social context. Recent developments in computer science have pointed to a possible way out of this dilemma. Groff and Birks (2008) suggest that new crime prevention approaches could be tested in an artificial computergenerated world before being tried in reality. In a more recent article, Birks et al. (2012) provide a good introduction to computational modelling or more specifically agent-based modelling (ABM). They tested whether the hypothesized mechanisms of environmental criminology were sufficient to explain the observed characteristics

C

C

744

of residential burglary – particularly its spatial concentration (hot spots), its concentration on victims (repeat victimization), and the characteristic journey to crime curve. The results showed that the propositions of the routine activity approach, rational choice perspective, and crime pattern theory provide a viable generative explanation of those independent crime characteristics. This work is an exciting first step in what looks like becoming a rich research vein for crime scientists. The article goes on to provide a number of further projects that might be conducted and through which we might explore the theories and mechanisms associated with crime reduction. One of the advantages of the approach is that it forces the experimenter to be explicit about the mechanisms that are presumed to lead to the reduction in crime and to be highly specific in the articulation of the hypotheses that are being tested (Sidebottom and Tilley 2011).

Controversies To call this section “controversies” might be overstating the case, but there are a number of issues with various degrees of contention, which merit some discussion. First, and perhaps most controversially, why not stay within the criminological discipline? Why do we need crime science as a separate enterprise? There are several of reasons. Perhaps the best has been well said by Ken Pease (Pease 2008) in a discussion of crime science in which he reviews the psychological concept of “framing” and the way in which it can be used to describe the partitioning of academic disciplines. Such partitioning, while helpful to universities and employers in facilitating the organization of teaching and in informing assumptions about subsequent knowledge, is also contributory to constraining thinking into “silos” which do not help the development of science in general or problem solving in particular. As Pease puts it: The cognitive structures into which thinking is organized are known to psychologists as frames. Frames limit cognitive commerce. Thus the primary disadvantage of organizing the study of the

Crime Science natural and social worlds into disciplines is that the frames come to act as artificial barriers to thinking. It is difficult to overstate the importance of framing in the organization of knowledge. Problems are “framed” in ways that make them susceptible to the skills of the framer. (Pease 2008, p. 4)

He goes on to argue that criminology is rooted in sociology (with links to psychology and law), and in line with the notion of framing, criminology’s “solutions” to crime problems and indeed the conceptualizations of crime itself are similarly (over-)influenced by those frames of reference. Crime science is thus an explicit attempt to overcome the negative effects of framing and to make explicit the advantages of working across disciplines. In his further justification for the emergence of crime science, Pease is critical of criminology. To quote Pease again: The decision to champion such a discipline shift is rooted in the perception. . ..that the status, progress, and influence of the understanding of crime and criminality achieved by scholarship labeled criminology has. . ..been disappointing both in itself and in its application to the urgent social problems to which it is addressed. (Pease 2008, p. 14)

Pease is also critical of some of the subgroups of criminology and singles out for particular comment the latest iteration of curious thinking from the so-called cultural criminologists. They, he argues, see crime as culturally defined and take scholarship to mean critical reading of the written word rather than empirical, data-driven investigation. If crime science is at one extreme on a continuum, then critical criminology appears to be at the other. So Pease argues for the establishment of crime science and a break from criminology on three main grounds: First, criminology suffers like most disciplines from the effects of framing; secondly, it has failed to deliver what is required in contributing to the control of crime. On this later point, he is joined by a number of criminologists including past presidents of the American Society of Criminology. They have chosen to try to change the discipline from within while Pease prefers to leave it. Thirdly, some (but Pease concedes not all) criminologists are in his view

Crime Science

(and mine as it happens) totally wrong in their approach to crime control. They are members of the criminology “club” and any club that accepts them as members is one to which Pease would rather not belong, as he puts it (paraphrasing Groucho Marx). The relationship with criminology described in the definition of crime science outlined in this entry is different from that of Pease, being better described as cherry-picking collaboration. As argued above, and as applies to other social sciences, criminology has much to offer the crime scientist in its insights about crime, ethics, and experimentation as exhibited, for example, by the experimental criminologists. But crime scientists argue that there is a need to do more than work with criminologists if the concern is the control of crime (its prevention, disruption, and detection) ethically and with due deference to good design (i.e., we do not like the Fort Knox option, effective though that might be). We would wish to embrace all the sciences and in particular to approach crime control using scientific methods, that is, to experiment. Why call it crime science? What is wrong, for example, with police science as suggested by Weisburd and Neyroud (2011)? The arguments for the label crime science have been rehearsed elsewhere, see, for example, Laycock (2012); but in a nutshell, while we share many of the methodological and philosophical ideas and approaches of the police scientists, we would include the operation of the courts, prisons, and probation services as of interest to crime science. We would also wish to address issues of concern to communities such as minor disorder or other problems, which would hopefully fall short of policing. So unless we also want to create prisons scientists, probation scientists, and criminal justice scientists (the list could be longer), we have opted for crime scientists. Furthermore, we draw an analogy with medical science, which is notably not called doctor science and which concerns itself with the prevention, detection, and treatment of diseases. In doing so, it does not rely solely on doctors but, as noted above, expects support from the central and local government, from parents, from schools,

745

C

teachers, and others in advocating healthy lifestyles. There are direct and obvious analogies with crime control. There are other issues associated with crime science stemming from its roots in environmental criminology, a substantial component of which is situational crime prevention (SCP). A classic criticism of SCP is that it is no more than target hardening and pays no attention to social issues but concentrates on turning social space into a fortress. This criticism is wrong on at least two counts. First, it is not the case that SCP is no more than target hardening. There are currently five major approaches to SCP, all of which have been demonstrated in different contexts to reduce crime – these are the mechanisms (Pawson and Tilley 1997) through which crime is reduced. They are: 1. Increase the perceived effort required of the potential offender to commit the crime (this includes but is not restricted to target hardening). 2. Increase the perceived risk to the offender (this includes, but again is not restricted to, increasing surveillance such as introducing CCTV). 3. Reduce the expected reward (e.g., putting dye capsules on clothing to deter shop theft). 4. Remove provocation (e.g., controlling taxi queues around night clubs to deter queue jumping). 5. Remove excuses (e.g., setting clear rules for children about the unacceptability of bullying in schools). Under each of these headings can be found a wide variety of tactics (Tilley and Laycock 2001) but their adoption should depend upon a proper understanding of the crime or disorder problem being addressed, an appreciation of the context within which the tactic is to be applied, and some judgment about its appropriateness such as the extent to which it might cause an oppressive environment, be unethical or an overreaction to an otherwise relatively minor problem. At this point, the decision whether or not to implement becomes one of judgment. It is also perhaps worth noting here that the default response from the police to the prevention

C

C

746

of crime is typically increased patrol, CCTV, or walls, barriers, and other target-hardening approaches. This is perhaps why SCP has become seen as so restricted. Experience of teaching problem solving and the use of the SARA process over a number of years suggests that it can be quite difficult to persuade the police to broaden their response options beyond increasing the perceived effort and risk to the offender. There have been a number of points so far when the notion of judgment has been mentioned. This has arisen in relation to deciding on which problem to address, which hypotheses to test, and which responses to adopt. Good judgment is an integral part of scientific development. Science is not the barren, formulaic process it is sometimes supposed to be. It should be a creative and vibrant activity that is exciting for the scientist and those who work with them. Some of the latter include the politicians whose job is to decide whether and if so when to adopt the innovations of science and technology and to implement them in society. This again is a judgment: It often requires balancing the rights of individuals with the need to keep society safe or the aesthetic appearance of a neighborhood with the need to reduce crime. In making these judgments, politicians or those who work on their behalf need a reasonable estimate and understanding of risk. It is often this that they are balancing. So, for example, if the risk of a terrorist attack is low, is the cost of implementing a potentially intrusive surveillance system defensible? For the politician, that cost is more than financial: It would include the political cost, that is, they can lose their job. They also would want to take account of precedent, public reaction and what the media might say. This is the policy context within which science and technologies are developed, and although it may be the responsibility of the scientist to ensure that the decision makers are aware of the scientific and technological options, it does not follow that decision makers will simply follow the most scientifically rational route. If scientists aspire to influence judgments (as crime scientists do given their wish to influence policy and practice), then it is perhaps appropriate if they understand this wider decision-

Crime Science

making context and how it can be used to further scientific endeavor; science does not sell itself. The final issue to be discussed here in the current debate on the development of crime science is the level within the universities at which it is taught. The first university to teach crime science was University College London, which is the home of the UCL Jill Dando Institute of Security and Crime Science (JDI). The teaching there has so far been largely although not exclusively restricted to postgraduate training. This has meant taking postgraduate students trained in an acknowledged scientific discipline and exposing them to the challenges of crime control. An advantage of this is that the students either come ready versed in scientific method or are positive about the approach. It compares with the early development of forensic science, for example. Social sciences differ in the extent to which they welcome experimentation, but even within those disciplines that do not have a hard science core, there appears to be a ready market for this more practical and scientifically based style. Other universities are now offering similar programs as crime science slowly develops with some providing courses at undergraduate level. There is a strong argument for keeping with the postgraduate option and not encouraging the development of undergraduate courses. To do so seriously risks establishing just another academic silo with all the framing difficulties as set out by Pease. We would basically be back to the beginning as far as the encouragement of multidisciplinary working was concerned. Nobody should underestimate the difficulty of establishing multidisciplinarity, which is one of the current buzzwords in research funding circles. While we might be able to point to outstanding examples of cross-disciplinary working, which have led to major scientific breakthroughs such as the work on DNA by Watson and Crick, those collaborations were close to serendipitous. Actually creating an academic context within which similar joint working might be fostered is a real challenge, particularly in today’s world when there is less opportunity in universities for the casual conversations in common rooms seen as a feature of the rosy academic past.

Crime Science

UCL has been extremely fortunate in this area. The Engineering and Physical Sciences Research Council in the UK, which is one of the major funders of science, provided funding to UCL’s JDI in 2009 to establish a Doctoral Training Centre in Security Science. This involved a £7 million grant to support 10 fully funded PhD students per year for 4 years. UCL undertook to double those numbers. The funding enabled the recruitment of scientists from a range of disciplines to be offered studentships and to work in the same physical space, with all the day-to-day interaction opportunities that this implies, and to share common courses about crime and the ways to control it. This is an ongoing program and its success or otherwise will be closely watched. So far, it has brought computer scientists, mathematicians, political scientists, linguists, and engineers together to carry out research.

Open Questions Following from the previous section, one of the open questions has to be the extent to which crime science can avoid the “silo pitfalls” of the past and genuinely produce novel solutions to today’s problems which demonstrate sensitivity to ethical concerns and good design. On this, the jury has to be still out. A second related question is whether or not crime science can add value to crime control through investment in a multidisciplinary approach – indeed at this stage, it is an open question whether different disciplines will be working together on common problems or whether their interactions will be restricted to coffee and biscuits. On this, the recent launch of a crime science open access journal might in the longer term prove helpful in providing an appropriate academic outlet for joint papers which might otherwise struggle for acceptance in the well-established mainstream disciplinary journals. Fundamentally crime science is attempting to generate a common language and understanding of the potential methods of crime control across disciplines that might not normally exist.

747

C

The aspirations of crime scientists to affect policy and practice are laudable, if potentially challenging, not least because there is a fundamental difference in the contingencies under which practitioners, policy makers, and researchers operate. For example, practitioners are not typically rewarded for reading research papers or for their understanding of science, while policy makers increasingly have an eye on the media representation of their policies rather than the extent to which they may reflect risk or rationality as a scientist may see it. Whether or not crime scientists can better influence policy and practice than, say, criminologists have managed to do is therefore an open question. The way to deal with this, however, is to change those contingencies not to complain about the lack of influence of research over policy and practice. The questions are as follows: How do we make practitioners (in the present case largely the police) more interested in what they perceive as irrelevant research? How do we motivate researchers better to communicate with the practitioners? How do we encourage policy makers to pay more attention to scientific research results? Until these questions are addressed, the lack of connection of the police with science, as described by Weisburd and Neyroud (2011), for example, will remain. There is of course a huge range of further open questions when we turn from what might be described as administrative issues to the more academic ones. First, much more work needs to be done on ethics and legitimacy specifically in relation to the implementation of various technological solutions to crime problems. A tendency to exaggerate the risk or threat of crime (particularly terrorism) leaves societies vulnerable to the unfettered implementation of “hard” solutions to crime problems, which may be effective but which in the longer term may also be undesirable. The most important research questions are ultimately a matter of opinion and inclination, but for the writer, as a crime scientist with an interest in crime prevention as a means of controlling crime rather than detection, arrest, sentencing, and ultimately imprisonment, the key question has to be how to design a social system which minimizes criminal opportunities

C

C

748

while protecting rights, freedoms, and aesthetics. This might remain an open question for some time to come.

Recommended Reading and References Birks D, Townsley M, Stewart A (2012) Generative explanations of crime: using simulation to test criminological theory. Criminology 50(1):2012–2221 Bowers KJ, Johnson SD, Pease K (2004) Prospective hotspotting: the future of crime mapping? Brit J Criminol 44(5):641–658 Burrows J, Hopkins M, Hubbard R, Robinson A, Speed M, Tilley N (2005) Understanding the attrition process in volume crime investigations. Home Office Research Study 295, Home Office, London Clarke RV, Eck J (2003) Become a problem solving crime analyst, UCL Jill Dando Institute. Available from Routledge publishers or www.popcenter.org. Accessed 23 May 2012 Eck J, Spelman W (1987) Problem-solving: problemoriented policing in newport news. Police Executive Research Forum, Washington, DC Farrell G, Pease K (2008) Repeat victimization. In: Wortley R, Mazerolle L (eds) Environmental criminology and crime analysis. Cullompton, Willan. ISBN 978-1-84392-280-3 Garland D (1996) The limits of the sovereign state: strategies of crime control in contemporary society. Brit J Criminol 36(4):445–471 Groff E, Birks DJ (2008) Simulating crime prevention strategies: a look at the possibilities. Policing J Policy Pract 2(2):175–184 Johnson SD, Bernasco W, Bowers KJ, Elffers H, Ratcliffe J, Rengert G, Townsley MT (2007) Near repeats: a cross national assessment of residential burglary. J Quant Criminol 23(3):201–219 Johnson SD, Summers L, Pease K (2009) Offender as forager? A direct test of the boost account of victimization. J Quant Criminol 25:181–200 Laycock G (2001) Hypothesis based research: the repeat victimization story. Crim Just Int J Policy Pract 1(1):59–82 Laycock G (2005) Defining crime science. In: Smith MJ, Tilley N (eds) Crime science: new approaches to preventing and detecting crime, Crime science series. Willan Publishing, Devon, UK Laycock G (2012) Happy birthday? Policing J Policy Pract 6(2):101–107 Morgan RM, Bull PA (2007) The philosophy, nature and practice of forensic sediment analysis. Prog Phys Geog 31(1):43–58 Pawson R, Tilley N (1997) Realistic evaluation. Sage, London Pease K (1998) Repeat victimisation: taking stock. Home office crime prevention and detection Series Paper 90. Available from http://webarchive.nationalarchives.

Crime Simulation gov.uk/20110218135832/http://rds.homeoffice.gov. uk/rds/prgpdfs/fcdps90.pdf. Accessed 24 May 2012 Pease K (2008) Crime science. In: Shoham SG, Knepper P, Kett M (eds) International handbook of criminology. CRC Press, Boca Raton Sherman L (2009) Evidence and liberty: the promise of experimental criminology. Criminol Crim Just 9(1):5–28, ISSN 1748–8958 Sherman LW, Gartin PR, Buerger ME (1989) Hot spots of predatory crime: routine activities and the criminology of place. Criminology 27:27–55 Sidebottom A, Tilley N (2011) Further improving reporting in crime and justice: an addendum to Perry, Weisburd and Hewitt (2010). J Exp Criminol. doi:10.1007/s11292-011-9128-6 Smith M, Tilley N (eds) (2005) Crime science. Willan, Cullompton, Devon Tilley N, Laycock G (2001) Working out what to do: evidence-based crime reduction. Crime reduction series paper 11. Home Office, London. ISSN: 1468–5205, ISBN 1-84082-792-0 Townsley M, Johnson SD, Pease K (2003) Problem orientation, problem solving and organizational change. Crime Prev Stud 15:183–212 Weisburd D, Neyroud P (2011) Police science: toward a new paradigm. National Institute of Justice, Washington, DC. Available from https://www.ncjrs.gov/ pdffiles1/nij/228922.pdf. Accessed 24 May 2012

Crime Simulation ▶ Simulation as a Tool for Police Planning

Crime Spates ▶ Prediction and Crime Clusters

Crime Specialization, Progression, and Sequencing Christopher J. Sullivan and Shaun M. Gann School of Criminal Justice, University of Cincinnati, Cincinnati, OH, USA

Overview This entry describes what is known about three key dimensions of the active criminal career:

Crime Specialization, Progression, and Sequencing

frequency, specialization, and progression/ sequencing. Each of these topics is considered in terms of its theoretical implications and the empirical research that has been conducted on it. A great deal has been learned about these topics since they emerged early in the discussion of criminal careers. There are some questions about the ability to measure and assess them and the implications of doing so, however, and those are reviewed here as well.

Background on Dimensions of Active Criminal Careers The Study of Criminal Careers Understanding the criminal career is important as it can offer insight into the nature of offending generally and help in explaining the behavioral paths of offenders over time. This can, in turn, inform policy and practice that relates to career criminals as well as those who offend for only a limited period of time. The idea of a criminal career was raised in early life history research with delinquents. Later on, a report by Blumstein and colleagues (1986) outlined a number of relevant characteristics of criminal careers that might be described or explained by criminologists, and the idea of a criminal “career” became a conceptual framework against which empirical findings could be judged and theories could be developed (Piquero et al. 2003). This began a trend toward quantitative analysis of the dimensions of longterm careers in offending. Among those that were initially defined in that report or have received attention since are (a) the age of onset of careers, (b) the length and continuity in careers, (c) the process of desistance from criminal behavior, (d) the frequency of offending, (e) specialization or versatility in the types of offenses committed, and (f) progression of careers in terms of possible escalation or stepping down in the seriousness of offending. Each of these can serve as a touchstone for theory or policy/practice with respect to crime and criminality. According to Smith et al. (1991:8), “the criminal career paradigm is not a theory of crime but rather a framework that permits more detailed

749

C

study of criminal behavior and its causes by identifying dimensions of individual criminal careers.” One advantage of separating the dimensions of the criminal career is that criminological theories can be applied to a specific dimension (or dimensions), allowing theorists to be precise in their propositions. For example, one theory may explain why offenders become involved in criminal activity, while another may explain the variation in frequency among active offenders. This entry will focus on the last three dimensions of crime described above (frequency, specialization, and escalation), which are generally designated as dimensions of active criminal careers. Blumstein et al. (1986) asserted that high-frequency offenders were worthy of study from a career perspective because they “contribute disproportionately to the total measured number of crimes” (18). They also suggested that, within those careers, it is important to investigate questions of how crimes mixed and were arrayed as a possible means of better understanding offenders. This entry first provides some definitions for these concepts before moving into a synthesis of what is known about them. That discussion also considers potential controversies or disagreements that have come up over time. Finally, some “open questions” in each of these areas are discussed in order to provide a sense of where research and practice might go in the future. Dimensions of the Active Criminal Career The dimensions of the criminal career discussed here are concerned with the nature of offending from the standpoint of relative seriousness (including frequency) or type of crime across an extended period of time (Piquero et al. 2003). In the current discussion, “frequency” of offending is concerned with individual levels of criminal activity during a given timeframe. Knowledge of the frequency of offending has important implications for the potential costs associated with an individual’s pattern of behavior. It has been at the heart of discussion of criminal careers because it is useful theoretically, and practically, to identify and explain those who offend at high rates as they tend to commit a large proportion of crime from an aggregate perspective as well (Piquero et al. 2007). For example, in their

C

750

Crime Specialization, Progression, and Sequencing, Fig. 1 Illustration of individual frequency of offending (California Youth Authority Data)

Crime Specialization, Progression, and Sequencing Total number of Offenses Per Year

C

8 7 6 5 4 3 2 1 0 15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

Age

Crime Specialization, Progression, and Sequencing, Fig. 2 Illustration of specialization in offending (California Youth Authority Data)

Number of Offenses (by Category)

10 9 8

Alcohol-Drug Offenses Property Offenses Violent Offenses

7 6 5 4 3 2 1 0

18–21

study of almost 10,000 adolescent males in Philadelphia, Wolfgang and colleagues (1972) concluded that 627 of the boys committed a majority of the total crimes. Figure 1 shows a career pattern for an individual case drawn from data on California Youth Authority parolees (see Sullivan et al. 2009) comprised of years with relatively high and relatively low offending frequency values. This pattern would be that of an offender with a fairly high frequency who offends across a significant part of the life course. Given this individual’s overall frequency of offending across the career, he might qualify as one of the minority of individuals who commit the majority of offenses identified in the Wolfgang study. Offending specialization is primarily concerned with the nature of the crimes committed by individual offenders and the degree to which they form patterns across a career

22–25

26–29

30+

Age Period

(or a portion of a career). To Osgood and Schreck (2007), specialization reflects systematic differences in the type of offenses committed by individuals. Systematic differences are those that are sizeable enough to distinguish one offender from another statistically. In most cases, specialization reflects some discernable pattern in form or type of offending across a defined period of time. Figure 2 offers a general sense of the contrast between specialization and versatility in offending. Over four time periods, this individual engages in no violent offenses and, although he starts with a mix of property and substance use offenses, is predominantly a property offender. The issue of “progression” in criminal activity also reflects the presence of offending patterns but generally considers the nature of the acts in terms of seriousness as opposed to just type. The ideas of “escalation” and “de-escalation” in the seriousness of offending fall into this general

Crime Specialization, Progression, and Sequencing, Fig. 3 Illustration of escalation in offending (California Youth Authority Data)

Number of Offenses (by Category)


C

751 # Violent

# Property

6 5

5

5 4 4 3 2

2

2 1 1 0

0

0 18–21

22–25

26–29

30+

Age Period

framework (Blumstein et al. 1986; Piquero et al. 2003). Although escalation can be framed in terms of the increased frequency of offending, it is more generally be defined as a tendency to commit more serious offenses as the criminal career unfolds (Piquero et al. 2003). The idea of escalation/de-escalation inherently asks the question of whether an individual’s criminal behavior is getting “better” or “worse.” If one assumes that violent offenses are more serious than property offenses, the notion of escalation can be observed in Fig. 3 by contrasting the relative prevalence of the two offense types over time. Initially, this individual is involved in far more property crimes, but these wane over the observed time period, while violent crimes increase over time.

State of the Art and Controversies Frequency in Offending Careers Measuring and making sense of the frequency of offending from a criminal careers perspective would appear to be a straightforward endeavor. Still, there are different ways of framing questions regarding the frequency of criminal behavior. One of the earliest controversies in research on criminal careers hinged around the distinction between total observed crime and individual frequency of offending (Blumstein et al. 1988; Gottfredson and Hirschi 1986). A society’s aggregated age-crime curve is necessarily composed of a number of different individual events. Sometimes these represent actions of individuals who are involved for the first and only time, but,

more often, they represent repeated acts by the same individuals (see discussion of Wolfgang study above). The objective of criminal career research rests partly in distinguishing the latter group from the former. The number of offenses per year for inactive offender has been given a label of Lambda (l) by some scholars (Piquero et al. 2007), and a great deal of the thinking and research in the study of criminal careers has focused on identifying and explaining the behavior of those high rate offenders. Because offending frequency research is concerned with active offenders, researchers employing the criminal career paradigm do not rely on aggregate crime rates. Per capita crime rates are calculated by dividing the total number of crimes committed in a specified area by the total population for the area. As such, they do not distinguish between those who commit crime (active offenders) and those who do not. Instead, frequency research relies on identifying and understanding Lambda. Gottfredson and Hirschi (1990) argued that a single latent construct, criminal propensity, was the main influence on all aspects of the criminal career, including frequency. According to this view, the frequency of offending increases as criminal propensity becomes more elevated. Contrary to Gottfredson and Hirschi’s argument, however, early criminal career proponents posited that different factors influenced each of the dimensions, specifically that different factors influenced participation and, among active offenders, frequency (Smith et al. 1991). As such, they called for research to determine

C

C

752

whether different variables were correlated with the separate dimensions of the criminal career and, if so, which of those affected each dimension. Smith and colleagues (1991) attempted to determine whether the factors associated with participating in crime and delinquency were different from those associated with the offending frequency of active offenders. The results were inconclusive, indicating that most of the correlates of frequency were also correlates of participation. There were, however, a few variables associated with one dimension and not the other. For example, age was associated with participation but not with frequency. Overall, the authors concluded that their data did not support the argument that a general propensity toward crime explained each of the dimensions of the criminal career. Lambda values vary significantly among individual offenders and are skewed to the right (Blumstein et al. 1986). In other words, the majority of offenders commit relatively few crimes each year, while a small portion of offenders commit crime on a more frequent basis. For example, using a representative sample of prison and jail inmates in California, Michigan, and Texas, (Chaiken and Chaiken 1982) found that among offenders who committed at least one burglary prior to their incarceration, the median Lambda value was relatively small at 5.5 burglaries per year, whereas offenders in the 90th percentile committed over 230 burglaries per year. Research has found a few factors that are consistently correlated with increased frequency rates, including early age of onset, frequent drug use (Rhodes 1989), unemployment, and previous high levels of criminal involvement (Blumstein et al. 1986). However, research on the relationship between frequency and certain demographic variables – age, sex, and race – is mixed. In their report, Blumstein et al. (1986) concluded that these variables were highly correlated with an individual’s participation in criminal activities but only marginally associated with individual frequency. For example, although an offender’s sex is typically a strong predictor of participation in criminal activity (Liu et al. 2011), research has shown that the offending frequencies of active


females are not significantly different from those of active males (Blumstein et al. 1986; Rhodes 1989). Like Blumstein and colleagues, Rhodes (1989) reported that active male and female offenders had almost identical offending frequencies. Regarding race, Blumstein et al. (1986) concluded that among active offenders, the frequency rates of black and white offenders were nearly identical. However, contrary to the conclusions from Blumstein and colleagues, Rhodes (1989) found that nonwhite offenders had significantly higher Lambda values than whites. The relationship between age and the frequency of offending has been heavily debated; in part, these discussions hinge on disentangling individual offending from aggregate patterns. Gottfredson and Hirschi (1986); Blumstein et al. (1988a, b) took differing perspectives in that the overall pattern observed in the age-crime curve might arise out of different processes, each with its own etiology and distinct implications for theory and practice. Gottfredson and Hirschi (1986) attributed the decline in the aggregate age-crime curve that occurs after its peak in late adolescence to decreased individual offending frequencies or individuals “maturing out of crime.” Proponents of the criminal career paradigm, however, argue that the causes of the decline are more extensive than simply maturing out of crime (Smith et al. 1991). They posit that the decline in aggregate crime rates may be influenced by both the number of offenders participating in crime and individual frequency rates. During early adulthood, many individuals terminate their criminal careers, while those that remain active commit crime at the same or elevated frequency. After reexamining the Glueck data, Blumstein et al. (1988a, b) found that annual arrest frequencies remained stable – ranging from 0.86 to 1.14 – for a sample of boys followed from ages 14 to 29. The idea of career criminals, or offenders who commit relatively frequent crimes over an extended time period, is important to policymakers. The belief is that if these career criminals can be identified, policies aimed at preventing, controlling, or curtailing their careers – either


through incarceration or other programs designed to prevent offending or recidivism – may produce a sizeable decrease in crime rates. One such policy that has been suggested (and incorporated in many jurisdictions) is selective incapacitation, which aims to identify and incarcerate serious career criminals in order to prevent the potentially significant number of crimes they would commit if free in the community (Gottfredson and Hirschi 1986). An example of selective incapacitation is habitual offender laws, or “three strikes” laws, in which repeat offenders are given a significantly longer sentence after their third felony conviction. In the original criminal careers report and other earlier works, Blumstein and colleagues (1986, 1988a, b) argued that additional research was needed to identify the legal variables correlated with high values of Lambda in order to rigorously consider and guide selective incapacitation policies. Another possible approach is to use some of the identified correlates of high frequency of offending to identify targets for prevention efforts. For example, early age of onset is a noted marker of frequent offending, meaning that targeting resources toward adolescents who have early contact with the justice system may be one avenue by which later crime might be prevented (Ayers et al. 1999; Liu et al. 2011). Additionally, research typically finds that violent offenders have higher Lambda values than their nonviolent counterparts (Piquero 2000). As such, Piquero argued that prevention programs aimed specifically at reducing violent crimes might not be necessary. Instead, general prevention programs aimed at reducing offending frequency would concomitantly reduce violent crime as well since offenders with high frequencies tend to commit violent crime at a disproportionate level. Although frequency research has become prominent since the original criminal careers report, there is still debate on which subset of offenders should be used in studying offender frequency. Studies typically draw their samples from one of two populations: current inmates or arrestees (Blumstein et al. 1986; Miles and Ludwig 2007). Inmate studies typically use

753

C

retrospective interviews to determine the number and type(s) of crimes that inmates committed during a specified time period. Conversely, studies that sample arrestees (whether incarcerated or not) use official arrest reports to determine the number and type(s) of crimes for which an offender was arrested. Because frequency in the criminal career paradigm measures the number of crimes committed per year as opposed to the number of arrests per year, some researchers have argued that the most accurate way to estimate Lambda is by using retrospective inmate interviews where the inmates can describe crimes for which they were never arrested. Both methods used to measure frequency rates have flaws. Self-reports suffer from problems with offenders accurately recalling past criminal acts and under- or overrepresentation of the number and type of crimes committed. However, the most serious potential problem with self-reports is selection bias. There may be a significant difference between offenders willing to share their criminal history and those unwilling to reveal their past criminal activities. In addition, selfreport data on offense history is almost exclusively collected from incarcerated offenders. This poses problems when trying to generalize results to the entire offender population, most of whom never serve time in prison (Rhodes 1989). Similarly, frequency values obtained from official arrest data may be biased due to the fact that arrests occur only for a small percentage of the number of crimes committed and/or the reality that part of an offender’s career may be spent in prison or jail. As such, the possibility of underestimating Lambda is a potential problem in using arrest data. One way to ensure more precise estimation of frequency rates is to combine the two methods by using arrest data to serve as a check on the accuracy of inmates’ selfreported criminal activities. Miles and Ludwig (2007) suggest that measuring Lambda is fraught with problems and, consequently, is of limited use in considering crime prevention policy. As discussed above, offending frequency research has led to discussion of controversial selective incapacitation policies. These policies may be difficult to implement in practice;

C

C

754

however. Some have argued that by the time these high-rate offenders are identified and trigger habitual offender laws, they may have done considerable damage but be on the downside of their criminal careers (Gottfredson and Hirschi 1990; Miles and Ludwig 2007). Limitations in empirical research and the very real possibility of false positives (i.e., predicting an offender will be a high-rate offender in the future and s/he does not) ensure that the controversy surrounding selective incapacitation laws will remain. Specialization/Versatility in Offending Careers Although typological and taxonomic identification of offender groups has long been a subject of criminological research and justice practice, this aspect of offending careers has also grown as a topic of interest since the 1986 “Criminal Careers” report. The degree to which offenders specialize or generalize in turn has implications for understanding the etiology of criminal offending in terms of explaining and responding to crime. The main interest from a criminal career perspective is whether, and to what degree, criminals concentrate their offending in particular offense types. In other words, do certain offenders engage predominantly, or exclusively, in drug offenses, sex offenses, violent offenses, or property offenses? If the answer to this question is yes, offenders might be placed in distinct groups based on their focus on certain crime types. If not, it may be futile, or even counterproductive, to develop strategies targeting particular subgroups of offenders based on the types of crimes in which they are presumed to solely or predominantly engage (e.g., sex offender treatment, drug offender treatment). In other words, absent specialization, criminologists should focus primarily on general explanations of crime, such as Gottfredson and Hirschi’s (1990) self-control theory, and it would be best if the system treated all offenders similarly no matter their instant offense – contingent on their frequency of offending. Conversely, the presence of specialization might suggest the use of more fine-grained explanations of offending that recognize the potential for qualitatively distinct


groups of offenders. Policy-makers and practitioners might then look to make use of such grouping information in developing sentencing and treatment strategies. Broadly speaking, two types of theories make implicit or explicit assumptions about the possibility of specialization. In contemporary criminology, general theories that make few distinctions between offenders in terms of the nature of their offending and the explanations derived from these theories typically suggest that offenders do not discriminate in the types of crimes that they commit. The General Theory of Crime suggests that offending is only one of a number of actions that might be undertaken in pursuit of short-term benefits at the expense of potential long-term consequences (Gottfredson and Hirschi 1990). Offenders are believed to have low levels of self-control which leads them to engage in a variety of criminal offenses without much regard for the form that they take. Consequently, there should be no recurring patterns of particular offense types. Offenders are unlikely to show specialization as this short-term, pleasureseeking orientation will manifest itself differently depending on the situation (Britt 1994). Conversely, taxonomic perspectives of criminal careers are open to the possibility that some distinctions among offenders exist – including those related to the types of crime committed. The criminal career paradigm defined by Blumstein and colleagues (1986) focuses on individual-level variation in offending across the life course. It also suggests that characteristics like specialization are worthy of theoretical explanation. This view holds that there are certain groups of offenders who warrant distinct explanations and may possibly require specialized treatment or intervention. For example, there may be a small group of serious, generalized offenders who engage in a continuous and varied track of antisocial behavior over time. At the same time, it is possible that certain individuals will engage in less serious forms of delinquency within the boundaries of adolescence before desisting around the time that they reach adulthood. This is the premise of Moffitt’s (1993) taxonomy of antisocial behavior, which is an example of


a theory that outlines subgroups of offenders that are distinct in terms of the nature and etiology of their offending. Moffitt’s adolescent-limited offenders may specialize in relatively less serious antisocial behavior (e.g., status offenses, substance use), while life-course persistent offenders engage in a variety of offense types across a longer portion of their lives. The notion of offender types, however, has been criticized by those who espouse general views of crime and criminals. Even the use of the term “career,” which implies some coherent pattern of behavior over time, has been the subject of questions in this regard (Gibbons 1988). Probably the biggest controversy in this area is whether those interested in understanding criminal behavior should bother to study specialization at all. Gibbons (1988) suggests that the typological schemes offered by both criminologists and criminal justice practitioners tend to fall short when tested against empirical data and, in general, the literature on this characteristic of the criminal career tends to show only modest levels of offender specialization (Piquero et al. 2003). Nevertheless, the fact that some important theoretical propositions in criminology have differing views on specialization, coupled with the reality that offenders are often grouped by type in policy initiatives and treatment programs, has led to continued empirical research. Some early research in this area found that information about the nature of an offender’s previous criminal behavior was not a good predictor of later offense types (Wolfgang et al. 1972). Farrington and colleagues (1988) assessed the potential for specialization in a sample of juvenile court records and found an average coefficient value of 0.10, which falls well short of the 1.0 that would be expected if there was complete specialization in measured offending. They also compared the observed number of specialized offenders to what might be found by chance alone; roughly 20 % of their cases were characterized as specialists in that analysis. Blumstein et al. (1988a, b) found slightly higher levels of specialization in adult offenders, but these individuals represented just a minority of the population. Identifying limitations in the

755

C

previous approaches, Britt (1996) found that offenders showed a much greater tendency to repeat the same offense than switch to another on their subsequent criminal act. Mazerolle et al. (2000) analyzed diversity index values, which were first used to account for similarity/dissimilarity among species, to assess specialization and found that their sample exhibited fairly diverse offending profiles. Later work with data from the Cambridge Study in Delinquent Development likewise identified little evidence of violent specialization (Piquero et al. 2007). Despite these findings, some recent research suggests that answering questions about the degree of specialization in offending may require a slightly more nuanced view. In his ethnographic research, Shover (1996) found that, although it may be obscured by the variety that is apparent across an entire criminal career, offenders seem to make short-term changes in their offending indicative of specialized behavior. Francis et al. (2004) also identified some offender “types” in their study of career-length crime patterns. Osgood and Schreck (2007) report consistent evidence of specialization in violent offending using general population data on self-reported delinquency. Like Shover (1996), these studies suggest that offenders may alter their preferences over time, which adds up to a conclusion of versatility when looked at across a lengthy career but also suggests that it might be useful to consider these short-term, specialized patterns in more depth. While these findings suggest some degree of specialization in offending, it is important that they are viewed in light of the larger body of research that has found that offenders tend to commit a variety of different criminal acts over their careers. Britt (1994) points out that even those studies that identify specialized offenders suggest that they comprise only a small portion of the overall population. Over time, research on specialization has drawn on a variety of methods, and, in some cases, it appears that the observed differences in results may be partly attributable to alternative approaches to measurement or analysis (Sullivan et al. 2009). Some studies that have identified support for the notion of specialization

C

C

756

(e.g., Francis et al. 2004; Osgood and Schreck 2007) have used measurement and analytic techniques that may provide a better test for the presence and nature of specialization than those used previously. Also, more recent research on specialization has begun to better incorporate theory which helps in moving past the blunt question of whether there is any specialization while also illustrating the potential utility of understanding this aspect of criminal careers. For example, a recent study considered the question of specialization within the context of family violence using clear theoretical and preventive foundations (Piquero et al. 2006). Synthesizing the results of empirical studies, it is clear that most offenders will commit a variety of different types of crime over their careers, but there may be parts of the career where they tend to engage in one type or another and those patterns should be considered in theoretical development around criminal careers (Armstrong and Britt 2004). Progression and Sequencing in Criminal Careers Research on progression and sequencing in criminal careers is also concerned with identifying and explaining patterns across criminal careers (Piquero et al. 2003). While the question of how careers progress or sequences unfold can relate to specialization, it is primarily concerned with the possibility that there are trends in seriousness across the career. Although implicit in a number of approaches to dealing with offending behavior (e.g., diversion programs that seek to intervene with less serious offenders to prevent more serious offenses later on), few perspectives have formalized the notion of escalation in offending theoretically. Loeber and Hay (1997) offered a pathways model that did consider the possibility of escalation to violent behavior after a period of time engaging in less serious forms of delinquency. Their expectation is consistent with the idea of heterotypic continuity where less serious forms of aggression are believed to later give way to more serious violent behavior at a later stage of development. This theory also laid out several precursors to later escalation in aggression (e.g., temperament, cognitive factors). There are also


some notions regarding escalation in certain subtypes of offenders. For example, sex offenders and men who engage in domestic violence are perceived to be at risk of escalation in the nature or frequency of their behavior over time. Of the three criminal career dimensions considered in this entry, progression has received the least amount of attention (Armstrong and Britt 2004; Liu et al. 2011). Piquero et al. (2003) reviewed the evidence on the potential for escalation in offending and found the results to be decidedly mixed. Any observed patterns appear to be age-graded in that the seriousness of offending often trends upward for juveniles. Farrington (2003) asserted that most juveniles begin their criminal career by committing relatively minor offenses but tend to commit progressively more serious offenses until the age of 20. Conversely, the typical progression of offense seriousness for adult offenders appears to be stable or declining over time (Farrington 2003; Piquero et al. 2003). Britt (1996) offered a new modeling strategy for identifying and testing specialization and escalation, conducting a series of analyses that were similar to those previously used to understand social mobility. Using data from a sample of Michigan offenders and a crime-specific breakdown, he identified statistical trends showing general support for a pattern of greater seriousness in offense type from one to the next. A somewhat similar recent study by Massoglia (2006) sought to understand offending patterns in the transition from adolescence to adulthood. Although this study only covered a portion of possible criminal careers, it provides insight into patterns of escalation and de-escalation. Massoglia identified groups that represented clear instances of escalation. Specifically, three subgroups of individuals in that study (26 %) had transition patterns that fit with notions of escalation (i.e., moved to a more serious offending class). This included cases that moved from normative deviance (e.g., truancy, disorderly conduct) to a predatory offending or illicit drug use class in early adulthood. Piquero and colleagues (2006) studied the question of escalation within the framework of spousal assault using data from victim interviews. This analysis focused on the


severity of injury experienced by the victim in a second assault. They identified a high probability of escalation in these successive incidents of domestic violence in one site, suggesting that this pattern is prevalent. In other sites, there was a mix of escalation and de-escalation in the injuries experienced by victims. In general, this contradicted the notion that escalation is endemic to that particular form of criminal behavior. Policy and practice around escalation are primarily focused on the fact that, if there is a progression from less to more serious offending, the earlier behavior can be used as a marker for intervention and the more serious and harmful behavior on the part of that individual may be prevented. For example, using conviction data from the England and Wales Offenders Index, Liu et al. (2011) examined the seriousness of offenses committed by over 4,000 offenders to determine whether escalation and/or de-escalation of offense seriousness is present over the criminal career. They concluded that offenders with few convictions (1–5) showed de-escalation over time, whereas offenders with more than five convictions tended to escalate in offense seriousness. As such, the authors argued that intervention policies should be aimed at persistent offenders because they are more likely to eventually escalate into committing serious, violent crimes, a view that is shared by other researchers (Ayers et al. 1999).

Conclusions and Future Research In general, criminologists now know considerably more about the criminal career than was known 25 years ago when research and discussion in this area really began to take off. At that point, the idea of a criminal career was one that made sense intuitively, and Blumstein and colleagues (1986) summarized a good deal of research to that point that supported the usefulness of the perspective. Still, at that time, there were questions in terms of its fundamental connection to observed data and its potential implications for a general understanding of crime and related policy and practice. For example, there

757

C

are those that asserted that longitudinal research of the type described above was too expensive and time-consuming and would prove to be of limited use to theory and practice (Gottfredson and Hirschi 1986). There are, however, some important conclusions that can be reached upon reflection a couple of decades later. It is clear that most crime is committed by a small pool of high-frequency offenders and a number of correlates of those high-frequency careers have been identified (e.g., early age of onset, career length, and frequent drug use). In terms of specialization, it is safe to say that most serious offenders will engage in a variety of different crime types over the course of a lengthy criminal career, but it would be an overstatement to say that there is no group of specialized offenders – particularly when disaggregating patterns of behavior that can extend years or decades. Similarly, the research on progression in offending that includes a pattern of escalation is fairly mixed. Some evidence has been found to support the idea of escalation, but it seems to be contingent on the context of the study. So it is the case that criminologists currently know more about the three dimensions of the criminal career considered here. At the same time, there appears to be some limitations in knowledge as well as some question around the utility of pursuing these research questions given the methodological challenges inherent in their study. For example, although prevention may be more desirable than longer-term incapacitation as a method of dealing with high-frequency offenders, the question of early identification continues to loom large. Resolving the question of false positives and false negatives is a difficult one as predicting behavior is a challenging endeavor – even with high-quality information on risk factors. Consequently, the identification and response to high Lambda offenders continues to be a difficult empirical and practical question. Additionally, while advances in measurement and analytic techniques have helped in considering the question of offender specialization and progression more thoroughly, efforts to consider patterns in criminal offending in a clearer

C

C

758

theoretical framework are also emerging with increasing regularity. It will be important in the future to determine whether the identification of a small group of specialists and those with clear patterns of offending escalation have important implications for theory, policy, and practice or are mainly just products of intuitively appealing notions about the nature of criminal behavior. This will be more helpful in terms of understanding criminal careers than a simple “yes/no” answer to the question of whether specialization or escalation in offending exist. In general, the next iteration of research on these topics should continue to explicitly embed an understanding of the dimensions of the active criminal career in the pursuit of knowledge about the processes by which criminal behavior is sustained over time while also attending to any implications for practice that might emerge.

Related Entries ▶ Age-Crime Curve ▶ Criminal Careers ▶ Career Criminals and Criminological Theory ▶ Criminal Careers and Public Policy Responses ▶ Onset of Offending

Recommended Reading and References Armstrong TA, Brit CL (2004) The effect of offender characteristics on offense specialization and escalation. Justice Q 21(4):843–876 Ayers CD, Williams JH, Hawkins JD, Peterson PL, Catalano RF, Abbott RD (1999) Assessing correlates of onset escalation de-escalation and desistence of delinquent behavior. J Quant Criminol 15(3):277–306 Blumstein A, Cohen J, Roth JA, Visher CA (eds) (1986) Criminal careers and “career criminals”. National Academy Press, Washington, DC Blumstein A, Cohen J, Das S, Moitra SD (1988a) Specialization and seriousness during adult criminal careers. J Quant Criminol 4:303–345 Blumstein A, Cohen J, Farrington DP (1988b) Criminal career research: its value for criminology. Criminology 26(1):1–35 Britt CL (1994) Versatility. In: Hirschi T, Gottfredson M (eds) The generality of deviance. Transaction, New Brunswick

Crime Specialization, Progression, and Sequencing Britt CL (1996) The measurement of specialization and escalation in the criminal career: an alternative modeling strategy. J Quant Criminol 12:193–207 Chaiken JM, Chaiken M (1982) Varieties of criminal behavior. Rand Corporation, Santa Monica Farrington DP (2003) Developmental and life-course criminology: key theoretical and empirical issues— the 2002 Sutherland award address. Criminology 41(2):221–255 Farrington DP, Snyder H, Finnegan T (1988) Specialization in juvenile court careers. Criminology 26:461–485 Francis B, Soothill K, Fligelstone R (2004) Identifying patterns and pathways of offending behaviour: a new approach to typologies of crime. Eur J Criminol 1:47–87 Gibbons DC (1988) Some critical comments on criminal types and criminal careers. Crim Just Behav 15(1):8–23 Gottfredson M, Hirschi T (1986) The true value of lambda would appear to be zero: an essay on career criminals, criminal careers selective incapacitation cohort studies and related topics. Criminology 24(2):213–234 Gottfredson M, Hirschi T (1990) The general theory of crime. Stanford University Press, Palo Alto Liu J, Francis B, Soothill K (2011) A longitudinal study of escalation in crime seriousness. J Quant Criminol 27(2):175–196 Loeber R, Hay D (1997) Key issues in the development of aggression and violence from childhood to early adulthood. Annu Rev Psychol 48:371–410 Massoglia M (2006) Desistance or displacement? The changing patterns of criminal offending from adolescence to adulthood. J Quant Criminol 22:215–239 Mazerolle P, Brame R, Paternoster R, Piquero A, Dean C (2000) Onset age persistence and offending versatility: comparisons across gender. Criminology 38:1143–1172 Miles TJ, Ludwig J (2007) The silence of the lambdas: deterring incapacitation research. J Quant Criminol 23:287–301 Moffitt TE (1993) Adolescent-limited and life-coursepersistent antisocial behavior: a developmental taxonomy. Psychol Rev 100:674–701 Osgood DW, Schreck CJ (2007) A new method for studying the extent stability and predictors of individual specialization in violence. Criminology 45:273–312 Piquero AR (2000) Frequency specialization and violence in offending careers. J Res Crime Delinq 37(4):392–418 Piquero AR, Farrington DP, Blumstein A (2003) The criminal career paradigm. In: Tonry M (ed) Crime and justice: a review of research, vol 30. University of Chicago Press, Chicago Piquero AR, Brame R, Fagan J, Moffitt TE (2006) Assessing the offending activity of criminal domestic violence suspects: offense specialization escalation and de-escalation evidence from the spouse assault replication program. Public Health Rep 121:409–418

Crimes Against Animal Life Piquero AR, Farrington DP, Blumstein A (2007) Key issues in criminal career research: new analysis of the Cambridge study in delinquent development. Cambridge University Press, Cambridge Rhodes W (1989) The criminal career: estimates of the duration and frequency of crime commission. J Quant Criminol 5(1):3–32 Shover N (1996) Great pretenders: pursuits and careers of persistent thieves. Westview, Boulder Smith DA, Visher CA, Jarjoura GR, O’Leary V (1991) Dimensions of delinquency: exploring the correlates of participation frequency and persistence of delinquent behavior. J Res Crime Delinq 28(1):6–32 Sullivan CJ, McGloin JM, Ray JV, Caudy MS (2009) Detecting specialization in offending: comparing analytic approaches. J Quant Criminol 25:419–441 Wolfgang M, Figlio R, Sellin T (1972) Delinquency in a birth cohort. University of Chicago Press, Chicago

759

C

Crime Trends

as these phenomena are discussed within green, eco-global criminology. The “wildlife” trade is one of the fastest increasing illegal transnational trades and a threat to a great number of species, as animals are used, for example, in Asian traditional medicine, as “pets” or as part of collections, for meat, for trophies, or for their fur. The trafficking causes tremendous suffering to the individuals which are abducted from their habitats to be trafficked, and great numbers die before they reach their destination. Offenders are both local villagers and large organized groups which either concentrate on this crime alone or combine it with other organized crime, for example, the drug trade. The “wildlife” trade is controlled by means of the CITES convention, which sees “wildlife” as natural resources, the trade in which must be regulated in order to preserve species from extinction. Through this convention, animals are not accorded rights, only value as “specimen” representing a species. This entry concludes by encouraging criminological research which includes a non-speciesist perspective, taking into consideration the harms of the animal trade, whether legal or illegal.

▶ Youth Homicide in the United States

Key Issues

Crime Sprees ▶ Prediction and Crime Clusters

Crimes Against Animal Life Ragnhild Sollund Department of Criminology and Sociology of Law, Faculty of Law, University of Oslo, Oslo, Norway

Synonyms Poaching; Wildlife trade; Wildlife trafficking

Overview This entry addresses the trade and trafficking in animals and discusses this multivariate crime through the lens of animal abuse and speciesism,

As the field of “green” (Lynch 1990; South 1998) or “eco-global” (White 2008, 2011) criminology has developed and expanded since the 1990s, so has the number of articles and publications related to so-called “wildlife” crime, both criminological, from other academic fields and from NGOs (e.g., Zimmerman 2003; Warchol et al. 2003; Lemieux and Clarke 2009; Westerhuis in press; Pires and Clarke 2011a, b; Traffic 2008; IWAF 2008; Gonzales 2003; Wyatt 2009, 2011; Sollund 2011; Wellsmith 2011). A reason is the growth of the trade, placing it variably as the second or third largest illegal trade worldwide, competing with the illegal drugs and arms trades and with human trafficking. As with other harm encompassed by green criminology, this phenomenon is far from one dimensional. It is local, regional, and transnational, organized or part of individual and culturally rooted exploitation,

C

C

760

reflecting its broad spectrum of actors, their varied motivations, the causes of the trade and also its impact, for example, the massive suffering of the victims and species loss. This entry will discuss some typical findings in the literature describing the field and see them through the lens of animal abuse and speciesism. First, some linguistic clarifications are required. The term “wildlife” is anthropocentric and antagonistic and implies that those animals who are not living under human control and domestication in fundamental ways are different from and dangerous to humans and that (often) they must be controlled for human protection. It also implies the “othering” of these animal(s) and species as “nature” in contrast to human “culture.” Generally, the word “poaching” is used to describe the act of illegally killing or abducting animals from their habitats. This implies that other animals are regarded as the property of humans, regardless of the negative impact this has on the well-being of each individual victim, such as pain and death. To be closer to the true character of killing and removing animals in and from their habitats, the terms theriocide (Beirne 1999, 2009) and abduction (Sollund 2011) are preferred. This emphasizes that nonhuman animals have equal interests to humans in living a full life free from pain and abuse in their natural, local domain (Singer 1975; Regan 1983). The word “animal” is a rejection of the fact that humans are also animals. Authors have solved this difficulty in different ways, for example, by applying the terms “nonhuman animals” and “animals other than humans.” These concepts can still be criticized for giving humans priority over other species, as humans are the yardstick other species are measured against, and thereby categorizing other animals as the same, despite the obvious interspecies differences. In the absence of better alternatives and for consistency of style, the term “animal,” when referring to animals who are not human, will still be applied. Animal trafficking is a splendid case to illustrate why a broader perspective than that represented by traditional, mainstream

Crimes Against Animal Life

criminology is necessary in addressing the transnational crime of animal trafficking. Green or eco-global criminology represents a perspective encompassing a variety of counter-hegemonic approaches that are concerned not only with the health of the environment but also with the animals living in environments, acknowledging their inherent value and that animals and the environment are mutually interdependent (Beirne 2011: 353). The introduction of CITES (the Convention on International Trade in Endangered Species of Wild Fauna and Flora) in 1973 (in force from 1975) established a divide in the perception of the animal trade and trafficking. It introduced a move from the legally free exploitation of “wild” nonhuman species to a requirement for regulation, recognizing the need to preserve species from extinction due to overexploitation from hunting and “poaching” (Hutton and Dickson 2000; Reeve 2002). Previously, the wildlife trade phenomenon, much as other “green crimes,” for example, those related to pollution, waste, and land degradation (Beirne and South 2007; White 2008, 2011), was neglected by criminologists, probably because the “wildlife” trade like much other harm was not a crime and the trade and trafficking in animals has a long history in which animals have been regarded as resources that can rightfully be exploited – “harvested” like a crop – by humans and used for a wide variety of purposes, for example, refined as cultural apparel, in traditional medicine, for adornment, clothing, and food. What determines whether trade and trafficking of a nonhuman individual is legal or criminalized is the degree of protection attributed to its species and the growing scarcity of individuals belonging to a species, defined by which of the three CITES appendices the species is listed in, if any (see below). The CITES convention seeks not to protect individuals from suffering, only as representatives of a species. Trafficking in “wildlife” is partly a consequence of speciesism and entails massive animal abuse, death, and species extinction. The three phenomena – “wildlife” trade, animal abuse, and speciesism – are thus interrelated and


inseparable, as animals cannot be traded without abuse and pain, and pain and abuse would not be inflicted upon them were it not for the ideological basis of speciesism permitting and legitimating it. Some space will therefore be dedicated to elaborate upon the phenomena of animal abuse and speciesism and how these are understood in a green criminology framework.

Background Description: Animal Abuse, Speciesism, and Animal Trafficking Beirne (1999) should be credited with seriously introducing the topic of animal abuse to criminology in an article published in 1999. He argued that criminology had been speciesist and anthropocentric and that animal abuse should be included as a legitimate field of criminological research because animal abuse entails pain and suffering, a violation of rights, and is one of several forms of oppression, like sexism and racism (Beirne 1999: 140). Beirne’s call for a nonspeciesist criminology is by no means fully responded to in mainstream criminology, but with “green criminology” (Beirne and South 2007; South and Beirne 2006) or “eco-global criminology” (White 2008, 2011) and the introduction of terms such as “speciesism,” “ecological justice,” and “species justice,” it appears to be a move toward a criminology that researches animal abuse, not because it may be a symptom of interhuman violence but because animals increasingly are regarded as victims when abused (Beirne 2008). This includes more than philosophical and ideological arguments for why animals should have rights (not to be abused) and proceeds to include empirical studies where animal abuse is studied in its own right. Beirne builds on Robert Agnew’s (1998) definition of animal abuse as any act that contributes to the pain, suffering, or death of an animal or that otherwise threatens its welfare. Animal abuse may be physical, psychological, or emotional; may involve active maltreatment or passive neglect or omission; and may be direct or indirect, intentional or unintentional. (1999:121)

761

C

With this broad understanding, there is hardly any area of society where animals can be found that would be exempt from research, yet to narrow it down, one could start by limiting the studies to human-inflicted pain. Undoubtedly many animals suffer as prey, and, for example, a mouse being victim of a cat’s playful torture will likely suffer both physically and psychologically. The suffering caused by “cruelty play” by one animal against another is beyond the scope of this entry, though interesting and meriting attention. Nonhuman animals and humans share an interest in killing for food. However, in the case of the 30 % of nonhuman animals that are carnivores, this is because of a need, while humans may hunt and kill for a number of other reasons, for example, for “sport” and status. It is highly debated whether humans need meat and rejected by many; nor do humans need to fight to position themselves as leaders of a pack or flock. Therefore, animal-inflicted suffering and death and human-inflicted suffering and death are differently motivated and not comparable. This discussion is therefore pursued no further here. Most of the harm inflicted on animals is legal, yet as equally harmful and fatal as the very few criminalized acts directed against animals. Areas of animal abuse that should be prioritized in research are those where animals are abused by humans through a deliberate act that the human should understand would entail discomfort, suffering, or pain for the animal, or through omission, regardless of the motivations that could be sought, excused, or neutralized by appeal to higher loyalties, for example, in the case of vivisection. One could/should, for example, research all kinds of systemic animal abuse, as in the meat industry, research laboratories, the fur industry, as well as unsystemic, domestic animal abuse. The aim would be to start to look beyond human-centered understandings and interpretations of pain, in which only pain attributed to animals that does not counter human interests is defined as abuse by animal welfare legislation. Thus green eco-global criminology entails an expansion of the traditional foci of criminology, from breaches of the law (whether from a control, offender, or victim perspective) to include harm

C

C

762

as a central category (Beirne and South 2007; White 2008, 2011). In this, the analysis of abuse must be broadened to include not only those animals who are ascribed a social status as companions to humans (and consequently whose status is thus “elevated” to imply that they may also be ascribed victim status) but all sentient beings, regardless of their actual or potential, direct, or indirect value for humans. In such a perspective, the topic mainly addressed in this entry – the legal and illegal abduction, trade, trafficking and theriocide of animals (theriocide is a term introduced by Piers Beirne, as a parallel to homicide, referring to the acts in which nonhuman animals are killed by humans) (Beirne 2009: 17), not only that which is criminalized, finds its place. It is estimated that nine and a half billion animals are killed annually in the USA alone, of which nine billion are killed for food, while twenty to one hundred million animals are killed in experiments (Agnew 1998: 180). Animals destined to become “meat” or “research tools” live under conditions that are equivalent to torture, confined in narrow cages, deprived of offspring, seldom allowed to ever move freely or feel the sun, and exposed to painful and fatal experiments. These are conditions which would never have been accepted if applied to humans and under such circumstances would be criminalized. Agnew developed a theory that could grasp not only the illegal abuse but also the vastly dispersed and culturally accepted legal animal abuse. In his model (1998: 182) are several factors that are important for a person’s involvement in abusive acts toward animals: (1) his/her social position – gender, “race” [sic] education, occupation, and place of residence, urban/rural; (2) individual traits, most importantly empathy and self-control, socialization models, nature of animals; and (3) ignorance re abusive consequences of behavior, beliefs justifying abuse and perceived benefits/costs of abuse. The last, ignorance, may be the most important factor, as it accounts for much of the systemic abuse taking place in industrial complexes, on which, for example, the consumption of meat and production of medicine and cosmetics rely. However, this ignorance may also be the consequence of a desired ignorance and


denial, as confronting oneself with the harmful effects of one’s consumption may be harder than to act upon the knowledge of the harm this consumption entails, and end it (Sollund 2008). Neutralization techniques are important in minimizing and justifying abusive acts – as appeals to higher loyalties, for example, in the case of vivisection (Agnew 1998; Regan 1983), and the condemnation of the condemners – for example, attacking animal rights activists as “extremists” and “terrorists.” These techniques of neutralization, however, would not make sense, if it were not for the fact that they are underpinned by a governing ideology and doxic practice (Bourdieu 1995) – speciesism.

Speciesism The term speciesism was coined by Richard Ryder in 1970 and has since been expanded and defined in various ways. Peter Singer made the term speciesism famous with his book Animal Liberation (1975), which also ignited the animal liberation movement. Speciesism can be understood as a prejudice or biased attitude favoring the interests of the members of one’s own species against those of members of other species. As with sexism and racism, speciesism rests on domination and subordination (Nibert 2002), rooted in patriarchal ideology, based on the fact that other species are not human and paralleling the way that women are seen as different and inferior to the white male (Donovan and Adams 1996; Noske 1997). The most important issue regarding a prejudice is not necessarily the prejudice itself but whether the prejudice is turned into practice. As an example, racists may not transform racist prejudice into racist acts if they find themselves in a social environment where racist acts are socially unacceptable. In that case, their prejudices, though racist, may not have much significance because informal and/or formal social control will usually prohibit them from coming out in the open. In the case of speciesism, ideology legitimates and upholds prejudice, thus more easily converting it into abusive acts and


exploitative practice Nibert (2002). Speciesism (like racism, sexism, and classism) is a set of widely and socially shared beliefs that result from and support oppressive social arrangements. These include the objectification of animals (e.g., as “meat” or “fur”), thereby obscuring animals as beings with individual selves, desires, and needs as well as characteristics based on a sex. Their exploitation is facilitated by the denial of any individuality. When considering the treatment of animals regarded as suitable for consumption as “meat,” it is interesting to question how this is compatible with the love and care directed toward domestic pet animals. However, as already noted, there is a difference in the perceptions of harm inflicted on humans and animals related to their status: [. . .] some harms are defined as criminal, others as abusive, but not criminal, and still others as neither criminal nor abusive (Beirne 2009: 200). This state of affairs is generally unquestioned and hence facilitates speciesism in a way that means it operates even without conscious prejudice. Critics of this status quo would argue that not to take a stand is also to take a stand, by refusing to reflect upon the consequences of one’s actions. Such refusal may be caused by emotional or other reactions to consideration of the harms done to animals and for that reason be connected to various forms of denial, but it is also facilitated by the meat industry which distributes advertisements of “happy cows in pasture” far remote from the reality of intensive and factory farming. All of this contributes to the cultural denial in which whole societies partake in collective denial of atrocities (Cohen 2001: 11). Speciesism can thus be defined either as ideology or practice – or as the two in tandem– supported by discourse that polices and maintains the human-nonhuman boundary. There are, for example, numerous derogatory terms referring to animals, such as “stupid as a hen,” “stupid cow,” “sly as a wolf,” “fat as a pig,” or “old dirty pig” when referring to an old male human with an appetite for young girls. In fact, such terms are so integrated into our language(s) that it takes a very conscious mind to avoid them. Terms are often sexist, thus combining prejudice

763

C

against animals and female humans, like “bitch.” Derogatory terms serve not only to denigrate the individual they are directed to but also to prepare the ground for use, abuse, and objectification of this individual and others of the same kind, and thus its legitimacy (Cohen 2001; Sollund 2008). When animals, reptiles, and birds are referred to as “specimens” in the CITES convention, this is another way of objectifying them and depriving them of individuality and victim status, through which their only value is as part of a specie.

State of the Art: The Trafficking and Trade in Animals The growing number of criminological articles about the abduction of animals from their habitats usually refers to these acts as poaching (Lemieux and Clarke 2009; Pires and Clarke 2011a, b; Warchol et al. 2003). This term includes both abduction of “wild” living animals from their habitats and also killing them on the spot – theriocide. They are subsequently trafficked and traded, and the purpose of the live abduction or killing determines whether their entire bodies are trafficked or only part of them. These actions are encompassed under the terms wildlife trafficking (Wyatt 2011; Warchol et al. 2003) or/and wildlife trade (in print; Zimmerman 2003; Westerhuis’ in print; Wyatt 2009), or the narrower wildlife crime (Wellsmith 2011; Zimmerman 2003), emphasizing that the trade in animals represents a crime. This indicates that the abduction, trade, and trafficking in animals are criminal acts; however, this is not always the case, as a significant part of the animal trade is indeed taking place in accordance with regulations. Actually, the legal “wildlife trade,” excluding “fisheries,” is variously estimated to be worth from 5 to 50 billion US dollars annually (Reeve 2002: 10) and to 160 billion US dollars per year (Duffy in White 2011: 55). Whether the trafficking and trade in an animal is legal or illegal will depend on how threatened his/ her species is. The CITES convention (also referred to as the Washington Convention) is an agreement that

C

C

764

was initiated in 1973, now signed by 175 states (parties to the convention) which through the convention are obliged to pass legislation meeting the requirements of the convention, and consequently also to punish those breaching this legislation. The aim of CITES is to regulate trade, not to prevent trade, so that the trade in animals can be sustainable and species conserved. As quoted from the CITES webpage, Its aim is to ensure that international trade in specimens (sic) of wild animals and plants does not threaten their survival. CITES can be criticized for legitimating trade and trafficking in animals and for prolonging a situation encouraging animal abuse and species decline through its anthropocentric approach to other species (Sollund 2011; Hutton and Dickson 2000). In the CITES convention, species are accorded protection depending on the threat of the trade, which is defined in the appendices. Those listed on Appendix I may be traded only in exceptional cases (as scientific research with the prospect of the significant harm this implies), because they are close to extinction. Appendix I species can only be traded if accompanied by an export and import certificate, or a reexport certificate issued for individuals who have previously been imported (Wijnsiekers 2001, here in Wyatt 2011). Appendix II lists those species which are not yet necessarily threatened with extinction but that may become so unless trade is closely controlled. It also includes so-called look-alike species, that is, species that look like those listed for conservation. International trade in Appendix II species may be authorized by the granting of an export permit or reexport certificate. No import permit is necessary for these species under CITES. On Appendix III are those species which there at the request of one of the parties to the convention, which may imply that the species is not necessarily threatened worldwide but in this state, and this state needs the cooperation of other parties to prevent the species from becoming extinct. Through the conferences of the parties, species may be suggested for inclusion on the appendices and also moved between them, for example, when a species is moved from Appendix I to Appendix II. There are also recommendations


on how large the quota for export of individuals can be and directions regarding the purpose of the abduction, theriocide, trade, or trafficking. As an example, at the conference of 2007, it was suggested that the Panthera pardus (leopard) in Uganda be transferred from Appendix I to Appendix II with the following annotation: 1. For the exclusive purpose of sport hunting for trophies and skins for personal use, to be exported as personal effects 2. With an annual export quota of 50 leopards for the whole country As this example shows, CITES does not regard the capture and killing of animals as a crime. However, from the perspectives of ecological justice and species justice (White 2008), though not breaching laws, these acts justify the use of the term crime (ibid), thereby emphasizing the serious nature of such actions. For the leopards involved, it makes no difference that killing them is not a crime. Whether legal or illegal, both entail tremendous harm for the victims and for the environment because removing species from their habitats ruins ecosystems (e.g., Herbig 2010). As mentioned in the introduction, there exists a body of literature on the animal trade with contributions from the natural sciences, such as biology and ornithology. These form a necessary basis for criminological discussion together with the work from Traffic and UWAF (Pires and Clarke 2011a, b; Sollund 2011; White 2011) although in the following the focus is on the main features and findings of a selection of studies from criminology.

From Local Abductors to Transnational Organized Crime A principal finding in the literature reporting research on animal trafficking is its multidimensional character. The markets are multiple; animals are trafficked for a large number of reasons, as rare objects for collectors, as trophies, for medicine (e.g., gall blathers and rhino horn), for experiments in western laboratories, for the “pet” market, for collectors, for their


fur, for meat, for ornament, and for falconry (Warchol et al. 2003; Wyatt 2009, 2011; Pires and Clarke 2011a, b; Herbig 2010; Sollund 2011; Reeve 2002; Hutton and Dickson 2000; Reeve 2002). Illegal animal trafficking is categorized as follows: first, local farmers who abduct and sell individuals to supplement their income; second, larger, mafia-like groups who buy from local peasantry and sell the animals at great profit; and third, major international smuggling rings which are often involved in other illegal trades (Zimmerman 2003: 1668). These rings tend to use violence, have considerable resources, are aware of smuggling routes, and consequently pose the largest threat to the regulation of the illegal “wildlife” trade. For example, there are three ways in which the drug trade may be linked with the “wildlife” trade: by parallel trafficking along shared smuggling routes, by the use of “legal” “wildlife” shipments to conceal drugs, and by the use of “wildlife” products to barter for drugs, and the exchange of drugs for “wildlife” to launder the proceeds of drug trafficking (Reeve 2002: 12–13; South and Wyatt 2011). Animals are used as receptacles for illegal drugs, whether they are alive or entailing their death, and planeloads of birds are exchanged for drugs; in Brazil the police estimate that perhaps 40 % of illegal drug shipments are combined with “wildlife” traffic (Reeve 2002: 13). With established distribution chains and buyers, such organized trade has far greater impact on the numbers of individuals traded than caused by unorganized individual abductors. Such groups are vigorous, for example, in Africa and Asia, in the illegal ivory trade (Lemieux and Clarke 2009), and in Colombia where the same routes can be used to smuggle primates as drugs (Zimmerman 2003: 1672). Estimates show that between 200,000 and 600,000 primates are illegally exported from Colombia every year, destined for laboratories and research centers (Zimmerman 2003). In addition to the harm done to the individuals that are victims of trafficking, this trade can also lead to infection of domestic species with contagious diseases (Herbig 2010: 126–128).

765

C

Whether on a small or large scale, the motivation is economic though the gain will vary with the position of the trafficker in the distribution chain. For example, poor indigenous villagers acting as abductors can earn $5–25 for each salmon-crested cockatoo, but by the time the birds reach the infamous bird markets in Indonesia, they will each cost up to $150 dollars (Metz 2007; Low 2003; Pires and Clarke 2011). A scarlet macaw or hyacinth parrot can retail for $10–15,000 in the USA or Europe (Speart 1993 in Warchol et al. 2003: 6); rare parrots can be sold for even 100,000 $ per pair (Reeve 2002: 11). The value of illegal ivory has grown significantly from US$200 per kilogram in 2004 to up to US$1,700 per kilogram in 2009 (Grossman 2009). The hunting of “bushmeat” now represents a considerable threat to biodiversity, especially in Central Africa, where the existing impact of hunting for sustenance by indigenous groups is being added to by commercial interests which abduct and export animals from the forests to markets and restaurants in Europe. Elsewhere, Australian native species are on the tables of restaurants in Europe and the USA (Westerhuis in print). There is a point to making a distinction between legal and illegal animal trafficking for whether an act is a breach of law or not may be central to discussion of how it can be combated, for example, by means of law enforcement strategies and by studying the market and exploring how CITES actually works (Warchol et al. 2003; Pires and Clarke 2010; 2011; Zimmerman 2003; Lemieux and Clarke 2009; Wellsmith 2010, 2011). However, in the case of animal trafficking, the legal/illegal distinction is insufficient, and it is necessary to build an approach from a more holistic, multidisciplinary basis in order to expand understanding of the field, and this entails embracing the concept of harm rather than (just) crime (Beirne and South 2007). It is also important that the underpinning rationale of cultural practices and ideology – speciesism – is also explored. Although they may be investigating animal trafficking, the perspective adopted by some researchers seems anthropocentric, and their

C

C

766

terminology implies that the animal trafficking is a crime more directed to humanity than to the animal victims themselves (see White 2011: 68–69). For example, by uncritically and dryly referring to abduction and theriocide as “poaching” (Lemieux and Clarke 2009; Pires and Clarke 2010, 2011), the suffering of the victim is not central, but instead the focus is on someone appropriating a “common good – a common property.” The term “wildlife” creates a social distance between local animals and humans and domesticated animals. Investigating animal trafficking appears important, not because of the pain and suffering it involves but because it is a “crime,” and thus fit for criminological analyses (Pires and Clarke 2011). The reason why this is an insufficient approach will prove evident in the following when some of the consequences of animal trafficking are sketched out.

The Suffering Caused by Animal Trafficking Trade and trafficking in combination with habitat loss and habitat degradation constitute the largest threat to a great number of species (Du Plessis 2000). It is maintained that the earth now faces mass extinction of species on a scale potentially akin to the disappearance of the dinosaurs, which will result in the loss of 50 % of the earth’s species by the end of the century if unhindered. UNEP (United Nations Environment Programme) states that 24 % of all mammals (1,130 species) and 12 % of birds (1,183 species) are globally threatened (Reeve 2002: 7, 8). Of the 90 parrot species at risk of extinction, 81 % are threatened by habitat loss, while 43 % by the trade in live birds for the “pet” trade (Du Plessis 2000: 21). The growth in the trafficking industry witnessed during the past two decades cannot be disconnected from the expansion of the World Wide Web and its significant role as a marketplace (IWAF 2008). These numbers, though great, say nothing about the suffering involved for the victims. For example, one report describes how antitank


weapons are used to blow the heads off elephants in the hunt for “bushmeat” and how mammals, birds, and reptiles which are seized by customs are found dead, malnourished, or starved, with broken limbs and bones, with lesions and frozen to death (Warchol et al. 2003: 8). Regarding parrots abducted for the “pet market,” a report about the parrot trade in Mexico revealed that 65,000–78,500 Mexican parrots are captured each year and [that] the overall mortality rate for trapped parrots exceeds 75 % before reaching a purchaser. This means that 50,000–60,000 parrots per year die in the trade (Weston and Memon 2009). Up to 75–90 % of illegally trafficked parrots die during transportation (Gonzales 2003). A study of the trade of parrots listed on the CITES appendices estimated that 1.2 million birds were exported between 1991 and 1996, and this is likely to be a gross underestimation of the real figures (Wright et al. 2001). Nearly a third of the 145 parrot species in the Neotropics (Mexico and Central and South America) are threatened, making them among the most endangered groups of birds worldwide (Reeve 2002). In the case of reptiles, studies show a mortality rate of 90 % during the first year in captivity (Herbig 2010: 120). Future Directions: How to Approach the Crime of Animal Trafficking? While entailing loss in biodiversity, destruction of ecosystems, and generally a far poorer world, the treatment of the animals traded for profit – whether on a small or big scale – entails unimaginable suffering and abuse. Or one can try to imagine what it is like to be abducted from the nest and the flock to have one’s eyes sewn closed (see Wyatt 2009, 2011), the arms (wings) strapped to the body, the mouth (beak) taped, and to be stuffed in a bottle just big enough to fit the body in this strapped condition, and then stuffed into another box or a suitcase with hundreds of other victims for days without food or water. If one survives the journey, the rest of life will be spent in a cage or on a leash. This important dimension of animal trafficking is rarely discussed. However, there are exceptions, as in Westerhuis’ examination of Australian


legislation and discussion of the moral question of the “wildlife” trade. Westerhuis concludes [. . .] the anthropocentric nature of environmental ethics in contemporary law and regulation allows the killing and trading of native wildlife, and does not recognize the sentience of wild animals. When tested, Preston’s hypothesis that a new ecological ethic would replace the traditional anthropomorphic ethic cannot be supported (Westerhuis in print). By carefully examining the criminological literature addressing animal trafficking, sympathy for the victims at times shines through – animal abuse and suffering is mentioned (Warchol et al. 2003; Wyatt 2011: 116), and may also be a motivation for the research (Wellsmith 2011). Generally, however, empathy does not seem to be an important reason for studies in this field. This may be related to speciesism. Those researchers who are not themselves speciesist and are therefore interested in the field but do not address the abuse may fear of collegial attitudes and repercussions, including getting funded or published. Instead, to enhance their case and broaden its perceived relevance, they may center discussion around the harm done to humans or to the environment through the loss of biodiversity. Or it may be agreed in principle that this is another environmental crime, like pollution and the trade in toxic waste, which is therefore interesting for a criminologist. Or trafficking in animals may be studied because it is an expanding global transnational crime, which should and could be studied by criminologists because studying crime, to simplify, is what criminologists do, and this is an under explored field. However, animal abuse is increasingly becoming a legitimate field of study in criminology and as it does one can ask whether it is possible to research the animal trade without discussing the abuse it entails? To draw a parallel, does it make sense, for example, to research genocide, as in the case of Rwanda, without taking into account death, suffering, and power relations? No matter what the motivation for research, further study is required in order to determine what measures must be taken to combat animalrelated crime. It is clear though that different

767

C

actors and differing motivations call for refined and differentiated preventative measures, reactions, and sanctions. While organized criminal groups may be deterred by long prison sentences and huge fines, in the case of local peasants, other measures such as teaching them the value of their victims and encouraging ecotourism which could provide employment as guides are more likely to succeed in combating the animal trade. It has been suggested that in the case of the illegal trade in raptors in Russia that breeding programs should be encouraged to reduce the suffering of the birds who are victims of illegal trafficking and to preserve species from extinction (Wyatt 2011). However, such legal interventions and breeding programs can bring their own risks as exemplified by the CITES convention which permits rather than prohibits harmful trade in animals. When such trades depend on certificates and export permits, these can lead to criminal enterprise producing forgeries (Zimmerman 2003; Weston and Memon 2009; Wyatt 2009), and it is up to customs to distinguish between “legal” and “illegal” species which is not always a dependable (or honest) process. By introducing a total ban on trade in “wild” animals, customs and law enforcement agents would not have to distinguish between forged and authentic certificates and permits and between look-alike and protected species. A total ban would, however, encourage collectors and traffickers as the price of rare species would increase, as occurs when species are moved from Appendix II to Appendix I (Reeve 2002). Nonetheless, by legitimating animal trafficking (e.g., as through CITES), speciesist ideology accepts and promotes animals as “natural resources” and as a “crop” to be harvested by humans (Sollund 2011). If research into animal trafficking is undertaken without paying attention to the victims of the trade and its ideological basis – speciesism – the trade will prevail. To increase awareness of the inherent value of other animals is to do them justice and acknowledge their place in the world and their intrinsic right to not suffer abuse, whether this be one-onone harm, institutionalized harm, or harm arising from human action (White 2011: 23).

C

C

768

Related Entries ▶ Green Criminology ▶ Organized Crime and the Environment

Recommended Reading and References Agnew R (1998) The causes of animal abuse: a social psychological analysis. Theor Criminol 2(2):177–210 Beirne P (1999) For a nonspeciesist criminology: animal abuse as an object of study. Criminol Interdiscipl J Official Publication Am Soc Criminol 37(1):117–149 Beirne P, South N (ed) (2007) Issues in green criminology. Devon: Willan Beirne P (2009) Confronting animal abuse. Rowman and Littleman, Lanham Beirne P (2011) Animal abuse and criminology: introduction to a special issue. Crime Law Social Change 55(5):349–357 Bourdieu P (1995) Outline of a theory of practice. Cambridge, Cambridge University Press Cazaux G (1999) Beauty and the beast: animal abuse from a non-speciesist criminological perspective crime. Law Soc Change 31(2):105–125 CITES convention http://www.cites.org/eng/disc/text. shtml Accessed 14 June 2011 Cohen S (2001) States of denial. Polity Press, Cambridge Donovan J, Adams C (1996) Beyond animal rights. A feminist caring ethic for the treatment of animals. New York: Continuum Du Plessis MA (2000) CITES and the causes of extinction. In: Hutton J, Dickson B (eds) Endangered species threatened convention. Earthscan, London Gonzales JA (2003) Harvesting, local trade, and conservation of parrots in the Northeastern Peruvian Amazon. Biol Conserv 114(2003):437–446 Grossman A (2009) Tusks and trinkets: an overview of illicit ivory trafficking in Africa. Afr Secur Rev 18(4):50–69 Herbig J (2010) The illegal reptile trade as a form of conservation crime: a South African criminological investigation. In: White R (ed) Global environmental harm: criminological perspectives. Willan Publishing, Devon Hutton J, Dickson B (eds) (2000) Endangered species threatened convention. Earthscan, London IWAF (International fund for animal welfare) (2008) Killing with keystrokes. An investigation of the Illegal wildlife Trade on the world wide web. http://www. ifaw.org/Publications/Program_Publications/Wildlife_ Trade/Campaign_Scientific_Publications/asset_upload_ file64_12456.pdf. Accessed 14 June 2011 Lemieux AM, Clarke R (2009) The international ban on ivory sales and its effects on elephant poaching in Africa. Br J Criminol 49:451–471 Low R (2003) The worldwide trade in wild-caught parrots. In: Proceedings to the International

Crimes Against Animal Life Aviculturists Society. http://www.funnyfarmexotics. com/IAS/2003Proceedings/Trade_Parrots_Low.pdf. Accessed 8 June 2010 Lynch M (1990) The greening of criminology: a perspective on the 1990s. Crit Criminol 2:1–5 Memon (2009) The illegal parrot trade in Latin America and its consequences to parrot nutrition, health and conservation. Bird Popul 9:76-83 Metz S (2007) Rehabilitation of Indonesian parrots from the Illegal Wild Bird Trade: early experiences on Seram island, Indonesia. In: Proceedings of Association of Avian veterinarians, Australian comittee. http:// www.indonesian-parrot-project.org/pdf_files/AAV_ 07_6-29-07.PDF. Accessed 29 June 2010 Nibert D (2002) Animal rights/human rights. Rowman and Littlefield, Lanham Noske B (1997) Beyond boundaries. Humans and animals. Montreal: Black Rose Books Pires S, Clarke R (2011a) Sequential foresting, itinerant fences and parrot poaching in Bolivia. Br J Criminol 51:314–335 Pires S, Clarke R (2011b) Are parrots CRAVED? An analysis of parrot poaching in Mexico. J Res Crime Delinq, online 15 March p 1–22 Reeve R (2002) Policing international trade in endangered species: The CITES treaty and compliance. Earthscan, London Regan T (1983) The case for animal rights. University California Press, Berkeley Singer P (1975) Animal liberation. Pimlico, London Sollund R (2008) Causes for speciesism: difference, distance and denial. In: Sollund R (ed) Global harms. Ecological crime and speciesism. Nova science publishers, New York Sollund R (2011) Expressions of speciesism: the effects of keeping companion animals on animal abuse, animal trafficking and species decline. Crime, Law and Social Change, p 1–15, Online ISSN 0925-4994 South N (1998) A green field for criminology: a proposal for a perspective. Theor Criminol 1998(2):211–233 South N, Beirne P (2006) Green criminology. Aldershot: Ashgate South N, Wyatt T (2011) Comparison of the Illicit wildlife and Drug trades: an exploratory study. Deviant Behav 32:538–561 Traffic (2008) What’s driving the wildlife trade? A review of expert opinion on economic and social drivers of the wildlife trade and trade control efforts in Cambodia, Indonesia, Lao PDR and Vietnam Traffic (2011) http://www.traffic.org/home/2011/6/10/ experts-urge-better-regulation-of-bushmeat-trade.html Accessed 14 June 2011 Warchol GL, Zupan LL, Clarke W (2003) Transnational criminality: an analysis of the illegal wildlife market in Southern Africa. Int Crim Justice Rev 13:1–26 Wellsmith M (2011) Wildlife crime: the problems of enforcement. Eur J Crime Policy 17:125–148 Westerhuis’ D (in press) Trade in wildlife: a moral question reviewed

Crimes of Globalization Weston MK, Memon MA (2009) The illegal parrot trade in Latin America and its consequences to parrot nutrition, health and conservation. Bird Popul 9:76–83 White R (2008) Crimes against nature: environmental criminology and ecological justice. Willan Publishing, Devon White R (2011) Transnational, environmental crime: toward an eco-global criminology. Routledge, London Wright T et al (2001) Nest poaching in neotropical parrots. Conservation Biology 15(3):710–720 Wyatt T (2009) Exploring the organization of Russia Far East’s Illegal wildlife trade: two case studies of the illegal fur and illegal falcon trades. Global Crime 10(1):144–154, Vol. 10 (1–2) 144–154 Wyatt T (2011) The Illegal raptor trade in the Russian federation. Contemp Justice Rev 14(2):103–123 Zimmerman ME (2003) The black market for wildlife: combating transnational organized crime in the Illegal wildlife trade. Vanderbilt J Transnational Law IVOL 36:1657

Crimes Against Humanity ▶ Human Rights Violations in Criminal Court

Crimes of Empire ▶ Crimes of the Powerful

Crimes of Globalization David O. Friedrichs1 and Dawn L. Rothe2 1 Department of Sociology/Criminal Justice, University of Scranton, Scranton, PA, USA 2 Department of Sociology and Criminal Justice, Old Dominion University, Norfolk, VA, USA

769

C

traditional boundaries of criminological inquiry. Some forms of crime, broadly defined, have been visible to scholarly students of crime from the outset, with homicide as the classic example. But other forms of crime, broadly defined, have been far less visible – and in some cases virtually invisible – to criminological inquiry. This entry addresses one form of such crime: crimes of globalization and more specifically the crimes of international financial institutions. These “crimes of globalization” are situated within the context of the broad range of crimes that are linked to and driven by the expanding conditions of globalization. Some attention is given to two key concepts and phenomena: “crime” and “globalization”. The fact that the crimes of international financial institutions cannot be readily classified under either the heading of “corporate crime” or “crimes of states” is a core reason for the relative invisibility of this type of crime. Nonetheless, as “corporate crime” and “crimes of states” are both classified as organizational crimes, international financial institutions do fit within the realm of research on “white-collar crime” or “the crimes of the powerful.” Crimes of globalization are intimately interrelated with crimes of states and political white-collar crime (especially corruption), as well as the interests of powerful private sector entities such as corporations and investment banks. How is such crime best explained and understood? The basic parameters of an integrated theory of crimes of globalization are identified here. The overall lack of accountability and the profound challenges involved in controlling crimes of globalization are also addressed. Some concluding observations address the potential value of further criminological attention to crimes of globalization.

Overview The topic addressed in this entry is not one of the conventional topics within surveys of the terrain of criminology and criminal justice literature. But a criminology that aspires to remain relevant in a rapidly changing world during the course of the twenty-first century must transcend the

Crimes of Globalization: The Fundamentals Defining Crimes of Globalization Conventional crime and its control is principally a local phenomenon and somewhat less so a state and national (or federal) phenomenon.

C

C

770

Accordingly, most criminologists are not especially receptive to adopting a global framework for defining crime or for the study of crime and its control. In the conventional view, crime as defined by law on the state level – or locally or nationally – is the proper domain of criminology, along with its control, and it is through the study of such crime and its control that criminology is most likely to make its contribution to the advancement of human knowledge and to produce knowledge specifically useful for identifying optimal public policies in response to crime. Crimes of globalization are those demonstrably harmful policies and practices of institutions and entities that are specifically a product of the forces of globalization and that by their very nature operate within a global context. Although these crimes can involve violations of criminal laws on the state or international level, they may also incorporate harms not specifically addressed by statutory law. The vastly disproportionate influence of elite interests over the formal criminal law is accordingly taken into account in a definition that transcends the boundaries of such law. It is not typically the specific intent of those who engage in crimes of globalization to cause harm. Rather, the devastating harm to vulnerable people in developing countries is a consequence of the skewed priorities of institutions and entities which favor the interests of the powerful and the privileged. The concept of crimes of globalization is not synonymous with two formulations that have received significant recent attention: the “globalization of crime” and “globalization and crime” (e.g., Aas 2007; Friman 2009; Larsen and Smandych 2008). The first of these terms refers broadly to long-standing forms of crime now carried out in an increasingly global context, and the second term refers broadly to the influence of globalization on crime, conventionally defined. Simon Mackenzie (2006) has introduced the term “systematic crime” in his discussion of the broad forms of global harm emanating from the practices of international financial institutions, and their complicity in denying the link between supporting interests of advanced economies and harm in developing countries.

Crimes of Globalization

The concept of crimes of globalization, as originally formulated, was limited to the demonstrably harmful activities of international financial institutions, with a special focus on one of these institutions, the World Bank. However, these crimes intersect with a range of other forms of crime engaged in by powerful entities, including crimes of states, political white-collar crime, and state-corporate crime (Friedrichs 2010). Multiple complex interconnections exist between these different types of globalized harm. Some refinement of the definition of crimes of globalization seems warranted. In the interest of greater clarity, the notion of crimes of international financial institutions specifically is best classified as a subtype of the broader category of crimes of globalization. The two principal international financial institutions are the International Monetary Fund (IMF), which seeks to maximize financial stability, and the World Bank, primarily focused on promoting development (Woods 2006). The World Trade Organization (WTO) is often aligned with these international financial institutions and has many parallel attributes and issues but strictly speaking is an international regulatory entity, with its primary formal mission being to foster trade. In a rapidly changing global economy, the roles of the international financial institutions have been increasingly questioned. These institutions have many ties with each other, and the lines of demarcation between their activities can become quite blurred. The World Bank and the International Monetary Fund: International Financial Institutions The World Bank is here addressed in some detail, to exemplify the fundamental nature of an international financial institution (Goldman 2006; Marshall 2008; Weaver 2008). The World Bank, formally the International Bank for Reconstruction and Development (IBRD), was established at the Bretton Woods Conference in 1944 to help stabilize and rebuild economies ravaged by World War II. Eventually it shifted its focus to an emphasis on aiding developing nations. The


Bank makes low-interest loans to governments of its member nations and to private development projects backed by those governments with the stated aim to benefit the citizens of those countries. The World Bank claims to contribute to the reduction of poverty and to the improvement of living standards in developing countries. Today the Bank is a large, international operation with more than 10,000 employees, 180 member states, and annual loans of $30 billion. The World Bank was established, along with the International Monetary Fund, at the behest of dominant Western states, with little input from developing countries. They are disproportionately influenced by or manipulated by elite economic institutions and entities – e.g., transnational mining companies – and have been characterized as agents of global capital. In developing countries, they deal primarily with the political and economic elites of those countries with little direct attention to the perspectives and needs of indigenous peoples. They have loaned money to ruthless military dictatorships engaged in murder and torture and denied loans to democratic governments subsequently overthrown by the military. They have favored strong dictatorships over struggling democracies because they believe that the former are more able to introduce and see through the unpopular reforms their loans require. Borrowers typically are political elites of developing countries and their cronies, although repaying the debt becomes the responsibility of these countries’ citizens, most of who do not benefit from the loans. The Bank has lost $100 billion due to fraud and corruption, in one estimate, over a period of several decades. The privileged in developing countries have been the principal beneficiaries of World Bank and International Monetary Fund loans, not poor people in those countries. The World Bank and the International Monetary Fund have been the target of much criticism, especially in the recent era. They have been characterized as paternalistic, secretive, and counterproductive in terms of their claimed goals of improving people’s lives. They have been charged with complicity in policies with genocidal consequences, with exacerbating

771

C

ethnic conflict, with increasing the gap between rich and poor, with fostering immense ecological and environmental damage, with neglecting agriculture crucial to survival in developing countries, and with the callous displacement of vast numbers of indigenous people in these countries from their original homes and communities. Critics claim that many less developed countries that received World Bank and International Monetary Fund loans are worse off today in terms of poverty and that the severe austerity measures imposed on borrowing countries, deemed necessary to maximize the chances of loans being repaid, impact most heavily on the poorest and most vulnerable citizens. The “structural adjustment agreements” in developing countries have been shown to also impact negatively on human rights in those countries. The building of dams has been the single most favored World Bank project, but even its own experts concede that millions of people have been displaced as a result of these dams. In many of these dam projects, resettlement plans either have been nonexistent – in violation of the Bank’s own guidelines – or have been inadequately implemented. In one notorious case in the 1970s, anti-dam protesters in Guatemala were massacred by the military. At a World Bank meeting in Berlin in 1988, protesters called for the establishment of a Permanent People’s Tribunal to try the World Bank (and the International Monetary Fund) for “crimes against humanity.” An American anthropologist has characterized the forced resettlement of people in dam-related projects as the worst crime against them, short of killing them. An American biologist characterized the World Bank’s report on the environmental impact of a dam project in a developing country as “fraudulent” and “criminal.” Globalization in Relation to Crimes of Globalization Globalization clearly has many different dimensions. Those most pertinent within the realm of white-collar crime generally, and crimes of globalization specifically, include the following:

C

C

772

(1) the growing global dominance and reach of neoliberalism and a free market, capitalist system that disproportionately benefits wealthy and powerful organizations and individuals; (2) the increasing vulnerability of indigenous people with a traditional way of life to the forces of globalized capitalism; (3) the growing influence and impact of international financial institutions and the related, relative decline of power of local or state-based institutions; and (4) the nondemocratic operation of international financial institutions, taking the form of globalization from above instead of globalization from below. Crimes of Globalization, Transnational Crimes, and International Crimes The relationship of crimes of globalization to the familiar but sometimes unclearly invoked terms transnational crimes and international crimes requires some attention here. Transnational crimes are essentially forms of crime that are increasingly carried out across borders and via international or global networks. The forces of globalization are transforming and amplifying structures of opportunity for a wide range of different forms of criminal activity (Aas 2007). Some of these forms of crime (e.g., human, arms, and drug trafficking) are hardly new, but the transnational dimension of them has expanded. Some of these forms of crime are quite new (e.g., cybercrime), but their transnational dimension greatly enhances the challenges of controlling them. Organized crime has long had transnational dimensions. The processes of globalization have been transforming some dimensions of such crime, with these transnational dimensions increasingly central to the operation of emerging forms of organized crime. But Hazel Croall (2005) is surely correct in arguing that there has been a disproportionate amount of criminological attention to the globalization of organized crime, when the global activities of multinational corporations and financial institutions cause far greater harm. Certainly the threat of transnational or global terrorism is substantial, but the argument can be made that it also has received disproportionate attention relative to other forms of


transnational or global harm. Especially since 9/11, there has been a huge amount of attention to transnational or global terrorism. International crimes are best conceived of as violations of international law, which in their generic form (e.g., genocide, war crimes, crimes against humanity, and massive violations of human rights) have a long history. Such crimes have often been committed within national boundaries but are increasingly carried out globally. International crimes are most typically thought of as crimes of states but may also be committed by insurgencies, militias, and other parties. Corporations – and increasingly multinational corporations – are also complicit in international crimes. For example, some of the corporations operating in Nazi Germany and its occupied territories played a role in the Holocaust, serving as classic cases of such crimes. In the more recent era, such corporations as Blackwater (XE), Sandline, and Halliburton have been accused of violations of international law (Rothe 2009). The conditions of globalization produce expanding opportunities for such crime. The lines of demarcation between crimes of globalization, transnational crimes, and international crimes are sometimes fluid and complex. But the key actors typically involved, and the bodies of law violated, tend to be different. Any coherent discussion of “crimes of globalization” must also address this phenomenon historically and cross-culturally. Some activities carried out in ancient times can be described as early forms of globalized white-collar crime. Monumental historical crimes were committed in the name of imperialism and colonial expansion globally, over a period of many centuries. Obviously the global slave trade in conjunction with this, continuing into the nineteenth century, was a crime of epic proportions. Furthermore, imperialistic enterprises continued through the twentieth century into the twenty-first, and slavery and a thriving slave market have hardly become extinct in the contemporary world, despite the collapse of traditional colonial regimes and forms of slave trade. More broadly, “crimes of empire” are a feature of our world today.


Crimes of Globalization, State-Corporate Crime, and Crimes of States The crimes of international financial institutions (IFIs) have a generic relationship to statecorporate crimes insofar as they are cooperative ventures involving public sector and private sector entities and in some respects are hybrid public/private sector entities. The literature on statecorporate crime (e.g., Michalowski and Kramer 2006) has focused on crimes arising out of cooperative ventures involving states and corporations. In one sense, crimes of globalization could be characterized as a neglected, cognate form of such crime: i.e., state-international financial institution crime. While the above categories may appear to be separate phenomenon, we suggest the connections between them are not so easily separated. The intersection of business and government has led to increased cases of a “globalized criminality.” In the recent era, Western states as well as corporations have promoted neoliberalism or a supposed “free market” model for the global political economy. Within such an environment, the crimes of globalization of international financial institutions are intertwined with crimes of states. The policies and practices of the international financial institutions are largely driven by the global agenda of powerful developed states such as the United States. In many of the developing countries, corrupt political oligarchs facilitate the promotion of this agenda, despite it being largely at odds with the interest of their citizens. Cases of such corrupt practices have been especially pronounced and well documented in the case of sub-Saharan countries of Africa, such as the Democratic Republic of Congo, Uganda, Rwanda, and Senegal (Rothe 2010b). But altogether these corrupt practices are a global phenomenon. Origins of the Concept of Crimes of Globalization The concept of “crimes of globalization” was first put forth in a paper for the American Society of Criminology Annual Meeting in 2000 (subsequently published as an article in Social Justice in 2002) with the specific title “The World Bank

773

C

and Crimes of Globalization: A Case Study” by David O. Friedrichs and Jessica Friedrichs. The specific inspiration for this concept was the experience of the junior author, Jessica Friedrichs, living among poor fishing people in northeast Thailand whose way of life was being destroyed by a dam at least partially financed by the World Bank. Jessica Friedrichs returned to the United States in April 2000 from her junior year experience in Thailand, 2 months earlier than anticipated, to make a presentation on the Pak Mun Dam situation at an antiglobalization (or global justice) demonstration in Washington, D.C., that month. The senior author of the article, David Friedrichs, with a long-standing interest in crimes of the powerful, realized that while at least some of the policies and practices of international financial institutions such as the World Bank had demonstrably harmful consequences, this form of harm had been quite wholly neglected by criminologists. The Pak Mun Dam Case In the Pak Mun dam case, the World Bank helped finance the building of the dam in eastern Thailand in the early 1990s. The process of planning, constructing, and operating this dam was undertaken without obtaining input from the fishermen and villagers who lived along the river. The construction of the dam had a detrimental effect on the environment, flooding the adjacent forests. This effect violated the World Bank’s own policies on cultural property destruction. Many edible plants upon which locals were dependent for their sustenance and for income were lost. Villagers who used the river waters for drinking, bathing, and laundry developed skin rashes. Most importantly, a severe decline in the fish population occurred. As a consequence, the way of life of indigenous fishermen dependent upon abundant fish for food and income was annihilated. The resettlement of the fishermen and compensation for their losses were wholly inadequate. Traditional communities began to disintegrate. Many of those affected by these developments organized protest villages and engaged in other actions calling for the Thai government and the World Bank to take

C

C

774

responsibility for the devastation they caused by building the dam, which cost far more than expected and generated far less electricity than had been anticipated. Crimes of Globalization: Subsequent Studies In the wake of the original article focusing upon the World Bank and the case of the Pak Mun dam, a number of criminologists have applied the concept of crimes of globalization to other circumstances. For example, (Rothe et al. 2006) conducted research that explored the interrelations between the International Monetary Fund and the World Bank and legacies of colonialism along with foreign policies that set the stage for large-scale atrocities and crimes of states. Exploring the circumstances leading to the sinking of the ferry Le Joola, the authors demonstrated that the state of Senegal itself had core liability for this maritime tragedy, with its dramatic loss of lives. The government readily admitted its errors and several ministers either stepped down or were removed from their positions. However, despite unequivocal governmental responsibility, Rothe, Muzzatti, and Mullins advance the case that the sinking could not be characterized simply as a case of state crime. Rather, a thorough investigation and analysis of the reasons and forces behind the Le Joola sinking suggested that international financial institutions bore some clear culpability for the disaster. In response to structural adjustment programs (SAPs) imposed by the International Monetary Fund, the Senegalese government was forced to cut spending in many areas. These spending cuts extended to ferry programs central to transportation in Senegal, especially in relation to its geographic location. This had a direct impact on the upkeep and return of the Le Joola to open waters. The ferry capsized with only one of its two engines functioning, resulting in the deaths of 1863 passengers. This was the second largest maritime disaster in history. Most crucially, the authors of this study demonstrated why scholars need to examine the criminogenic effects of policies and practices of international financial institutions in developing countries such as Senegal. These policies and practices


privilege capitalistic profit over human lives and a better quality of life for people in developing countries. Accordingly, this is crime against vulnerable human beings. An article by (Rothe et al. 2009) took a parallel approach, exploring the role of international financial institution policies in the conditions leading to the Rwandan genocide in 1994. While the World Bank and the International Monetary Fund did not seek to instigate economic collapse or to promote genocide, their policies and their systematic inattention in Rwanda set the stage for political and economic disaster as well as the genocide itself. The authors suggested that these international financial institutions knowingly violated their own standards as well as international human rights principles. Through the imposition of harsh conditions tied to their financial aid, they facilitated criminal activities on a massive scale. In an article published in 2008, Ezeonu and Koku also adopted the crimes of globalization concept. They demonstrated the key contributing role played by the neoliberal policies of international financial institutions in sub-Saharan Africa, in expanding the vulnerability of people in this region to HIV infection. They called for more systematic criminological attention to the victimization of people in developing countries as a consequence of the promotion of neoliberal policies and practices in an increasingly globalized world (see also Ezeonu 2008). In a similar vein, Rothe (2010a, b) has provided an analysis of the complicity of international financial institutions in heightened levels of corruption and the suppression or violation of human rights in developing countries. Analyzing such complicity seems especially important given that these institutions claim to be engaged in combating corruption in developing countries, including those linked to transnational and multinational corporations. The anticorruption initiatives include threatening to withhold much needed economic aid and loans in the absence of action taken against corrupt activities in these countries. Rothe has illustrated the specific role of the international financial institutions in the illegal expropriation of the rich natural resources


of the Democratic Republic of Congo by the neighboring countries of Uganda and Rwanda. Beyond theft on a grand scale, Rwandan and Ugandan state forces and militias also engaged in especially atrocious human rights violations conducted against civilian populations, including forced labor, systematic rape, and widespread killing. Through their funding of African states engaged in crimes against both their own citizens and those of neighboring countries, the international financial institutions bear some responsibility for these crimes. Parallel circumstances have arisen in other parts of the world. Stanley (2009) has analyzed the role of the international financial institutions in Indonesia. They directed some $30 billion to the Suharto regime, despite its known record of massive corruption, false accounting, and a militaristic appropriation of aid funds. As the World Bank’s focus was on supporting Indonesia, the state was able to use funds supposedly intended to reduce poverty in its brutal campaign against civilians in the state of Timor-Leste. This campaign had as its purpose terrorizing people to deter them from voting for independence from Indonesia. One could identify many other cases in Asia and other parts of the world where the international financial institutions have been complicit in supporting corrupt, authoritarian regimes and facilitating their massive violations of human rights. The concept of crimes of globalization has also been adopted in relation to forms of crime that occur in the context of globalization but do not specifically involve the international financial institutions. Wright and Muzzatti (2007) have addressed the global restructuring of agriculture and food systems – agri-food globalization – with some specific attention to the victimization of huge numbers of animals: e.g., 58,000 sheep stranded at sea for almost 3 months in 2003, in violation of animal welfare law. Altogether, policies and practices relating to the global restructuring of agriculture and food systems were driving up food prices, pushing tens of millions of people toward hunger and starvation, and developed country farm subsidies were driving large numbers of farmers in developing countries

775

C

into desperate circumstances – to the advantage of corporate and high finance interests in the wealthy countries of the world. The notion of crimes of globalization has also been implicitly adopted by some authors who are not criminologists and who are addressing a broad public audience. A book by John Perkins (2005), entitled Confessions of an Economic Hit Man, became a best seller following its publication in 2005. He defined economic hit men (EHM) as “highly paid professionals who cheat countries around the globe out of trillions of dollars” (Perkins 2005, p. ix). In his case, as an employee of an international consulting firm, Perkins claims to have participated in a range of activities involved in funneling funds from international financial institutions and international aid organizations into the hands of major transnational corporations and a small number of wealthy and influential families in developing countries. Economic hit men are engaged in persuading developing country leaders to become part of a vast global network that ultimately serves the interest of US-based corporations and US businesses generally, at the enormous expense of the people of the developing countries. In a subsequent book, The Secret History of the American Empire, Perkins (2007) further explores some of these themes. Although these two books have been criticized as self-dramatizing, they may have succeeded in raising the consciousness of new audiences about crimes of globalization. In A Game as Old as Empire: The Secret World of Economic Hit Men and the Web of Global Corruption (Hiatt 2007), a book inspired by the Perkins’ best seller, various authors address aspects of “the corporatocracy” (“the powerful people who run the world’s biggest corporations, the most powerful governments, and history’s first truly global empire,” p. 20). These authors address such matters as the hundreds of billions of dollars that developing countries spend annually for servicing their debt, the world of offshore banking, the expropriation of Africa’s oil wealth, the role of export credit agencies in boosting overseas sales for multinational corporations, and the mirage of debt relief.

C

C

776

An Integrated Theory of Crimes of Globalization On the global level, any initiatives at explanation should begin with a sense of humility: i.e., the globalized world we live in is so endlessly complex, with countless different variables interacting on multiple different levels, that fullfledged explanation (and prediction) is tremendously difficult. Nonetheless, the simplistic and one-dimensional explanations of crime advanced by some criminologists – e.g., low self-control theory – are quite useless in relation to crimes of globalization. To the extent that we can hope to understand crimes of globalization, an integrated theoretical approach is necessary. There have been attempts to provide integrated theories to explain traditional street crime. Integrated theories have also been applied to organizational offending, including corporate crime, state-corporate crime and crimes of states, or violations of international criminal law (Rothe 2009). These integrated theories provide a foundation for the integrated theory of crimes of globalization outlined below. An integrated model of offending takes into account motivations, opportunities, controls, and constraints at four levels of analysis (interactional, organizational, structural, and international). In doing so, the integrated theory combines insights from criminological theories and other disciplines to explain the multiple levels at play within each specific case. Social learning (e.g., differential association), anomie, strain, rational choice, routine activities, techniques of neutralization, and control theories have been brought together within an integrated approach to address some specific components of organizational offending. An integrated theory of crimes of globalization also draws from sociological organizational analysis (e.g., network and system analysis) and from political economy models to explain policies and practices pursued at the state and international levels. This integrated approach recognizes that organizations are not monolithic entities and do not operate within a vacuum. Instead, it recognizes agency at the interactional level while simultaneously noting the impact of


organizational culture and the broader structure on the individual agency level of decision-making. Key components of an integrated theory of crimes of globalization include the motivating forces of pursuit of both legitimacy and profit as organizational goals, the internal structure and reward systems of organizations, and external pressures which are imposed on organizations. The specific political context within which international financial institutions operate is one in which the interests of the powerful countries that provide most of the funding for their programs are aligned with those of the political and business elites of the developing countries which are being aided. On a structural (or state-based) level, then, international financial institution programs are ultimately skewed to advance or protect the interests of elites while all-too-often inflicting harm on ordinary citizens and indigenous populations in developing countries. On the organizational level, a criminogenic environment is fostered within the international financial institutions. The officers and staff of these institutions are rewarded for getting out aid and loans and implementing programs in developing countries, not for ensuring that these programs do not cause long-term harm for the people of the countries being aided. The “success” of the programs is measured more in terms of their size and scope rather than their ultimate effect. International financial institutions reward their personnel for technical proficiency rather than for concerning themselves with the perspectives and needs of the ordinary people of developing countries. Despite professed noble objectives and the absence of a specific intent to do harm, the mode of operation of the international financial institutions is intrinsically criminogenic and undemocratic. Their key deliberations are carried out secretly and with an absence of accountability. Controlling and Responding to Crimes of Globalization Throughout the course of the twentieth century and into the early stages of the twenty-first century, it became widely recognized that local, state, and national (or federal) institutions of


social control were no longer sufficient for the challenges arising in an increasingly globalized world. The expanding adoption of transnational, international, and global institutions to address a broad spectrum of harmful activities – from cross-border trafficking to crimes of war – has been one of the defining attributes of the contemporary era. But the international financial institutions operate with a singular absence of effective accountability and oversight. There are a range of international declarations and accords that prohibit some of the harmful activities engaged in by the international financial institutions ranging from fostering corruption to complicity in violations of human rights. As specific examples, we have the United Nations Convention Against Corruption Article 2 (c) which defines officials of public international organizations in a way that could be applied to international financial institutions. Other such international accords define transnational criminal activities and human rights violations in a form that can also be interpreted as applying to the international financial institutions. However, no international institutions or tribunals specifically have jurisdiction over, take complaints on, or adjudicate the broad range of harmful activities engaged in by the international financial institutions. The powerful countries that dominate these institutions – notably, the United States and Western European countries – are highly unlikely to call them to account for their harmful activities, since their policies and practices are aligned with and advance the economic interests of these countries. The Global Justice Movement and Crimes of Globalization To date, global justice (or antiglobalization) activists and their protest rallies have been the principal entities holding international financial institutions accountable for their harmful policies and practices in developing countries. A World Bank Bonds Boycott is one manifestation of this, since the World Bank raises most of its funds by issuing bonds, and these bonds are purchased by ordinary citizens through pension funds, labor unions, churches, municipalities, universities, and other entities.

777

C

In a world where citizens of developing countries increasingly have access through the Internet to specific information about the exploitative activities of Western institutions in their own countries, demands for addressing crimes of globalization are likely to increase. Populist challenges to autocratic, corrupt regimes and the immense economic inequality that they sustain will surely be one of the defining themes of the twenty-first century. In developing countries, growing numbers of people are characterizing themselves as victims of monumental crimes carried out by these regimes, and Western governments and institutions are seen as complicit in fundamental ways in these crimes. During 20112012 this situation played out in Egypt, Tunisia, Libya, Syria, Bahrain, Yemen, and other Middle Eastern countries. The gross maldistribution in terms of consumption of natural resources that exists between the developed and developing world is also highly likely to be subjected to increasing challenge. Going forward, it seems highly likely that Western entities, including international financial institutions, will increasingly be “indicted” for their perceived crimes against the people of developing countries.

Concluding Observations The “crimes of globalization” concept has not generated significant controversy to date because it continues to lie outside the focus of almost all students of crime and criminal justice. But criminologists are in collective denial if they fail to recognize that the crimes of the powerful – including those of international financial institutions – will attract growing attention from ordinary people all over the globe. In an increasingly globalized world, international financial institutions are immensely important players and are likely in the future to assume even greater importance and influence globally. A core premise for this entry is that international financial institutions, whatever their stated missions, have in their policies and practices been complicit in very large-scale forms of social harm and that this harm should be recognized as a significant form

C

C

778

of crime, and accordingly it is both a legitimate and a useful project for criminologists to apply a criminological framework to the understanding of these crimes of globalization and the challenges arising in relation to the prevention and control of such crime.

Related Entries ▶ Crimes of the Powerful ▶ Human Trafficking ▶ International Crime Victimization Survey

Recommended Reading and References Aas KF (2007) Globalization and crime. Sage, Los Angeles Croall C (2005) Transnational white-collar crime. In: Sheptycki J, Wardak A (eds) Transnational and comparative criminology. Glasshouse Press, London, pp 247–268 Ezeonu I (2008) Crimes of globalization: health care, HIV and the poverty of neo-liberalism in Sub-Saharan Africa. Int J Soc Inquiry 1:113–134 Ezeonu I, Koku E (2008) Crimes of globalization: the feminization of HIV pandemic in Sub-Saharan Africa. The Global South 2 Friedrichs DO (2010) Trusted criminals: white collar crime in contemporary society, 4th edn. Cengage, Belmont Friedrichs DO, Friedrichs J (2002) The World Bank and crimes of globalization: a case study. Soc Justice 29:1–12 Friman RH (2009) Crime and the global political economy. Lynne Rienner, Boulder Goldman M (2006) Imperial nature: the World Bank and struggles for justice in the age of globalization. Yale University Press, New Haven Hiatt S (ed) (2007) A game as old as empire: the secret world of economic hit men and the web of global corruption. Berrett-Koehler, San Francisco Larsen N, Smandych R (eds) (2008) Global criminology and criminal justice. Broadview Press, Toronto Mackenzie S (2006) Systematic crimes of the powerful: criminal aspects of the global economy. Soc Justice 33:162–182 Marshall K (2008) The World Bank: from reconstruction to development to equity. Routledge, New York Michalowski RJ, Kramer RC (eds) (2006) State-corporate criminality: wrongdoing at the intersection of business and government. Rutgers University Press, New Brunswick Perkins J (2005) Confessions of an economic hit man. Ebury Press, London

Crimes of the Powerful Perkins J (2007) The secret history of the American empire: the truth about economic hit men, jackals, and how to change the world. Penguin, London Rothe DL (2009) State criminality: the crime of all crimes. Lexington Books, Lanham Rothe DL (2010a) Facilitating corruption and human rights violations: the role of international financial institutions. Crime Law Soc Change 53:397–412 Rothe DL (2010b) The dragon rising: international financial institutions and the emerging role of China. In: Chambliss WJ, Michalowski R, Kramer RC (eds) State crime in a globalized age. Willan Press, London, pp 152–169 Rothe DL, Muzzatti S, Mullins CW (2006) Crime on the high seas: crimes of globalization and the sinking of the Senegalese ferry Le Joola. Crit Criminol 14:159–180 Rothe DL, Mullins CW, Sandstrom K (2009) The Rwandan genocide: international finance policies and human rights. Soc Justice 35:66–86 Stanley E (2009) Torture, truth and justice, the case of Timor-Leste. Routledge, London Weaver C (2008) Hypocrisy trap: the World Bank and the poverty of reform. Princeton University Press, Princeton Woods N (2006) The globalizers: the IMF, the World Bank and their borrowers. Cornell University Press, Ithaca Wright W, Muzzatti S (2007) Not in my port: the “death ship” of sheep and crimes of agri-food globalization. Agr Hum Val 24:133–145

Crimes of the Powerful David Kauzlarich1 and Dawn L. Rothe2 1 Department of Sociology and Criminal Justice Studies, Southern Illinois University Edwardsville, Edwardsville, IL, USA 2 Department of Sociology and Criminal Justice, Old Dominion University, Norfolk, VA, USA

Synonyms Corporate crime; Crimes of empire; Crimes of globalization; Organizational crimes; State crime; State-corporate crime; White-collar crime

Overview Crimes of the powerful have occurred throughout humanity’s existence, from the ancient armies

Crimes of the Powerful

779

Crimes of the Powerful, Fig. 1 Crimes of the powerful by type

C

Crimes of the Powerful Or White-Collar Crimes

Occupational Crimes

Organizational Crimes

Corporate Crime State-Corporate Crime

and Roman leadership to the aristocratic criminal of the eighteenth Century to the modern corporate CEO and head of state. While these behaviors did not come under the scrutiny of criminologists until the last century, there is a growing body of critical theoretical frames that have been used to explain the etiological factors associated with crimes of the powerful.

Fundamentals Theoretical explanations of the crimes of the powerful are not straightforward or universally applicable. As such, it helps to first understand the typical forms by which these types of criminality are generally categorized as well as the levels of analysis with which theories can be applied. Crimes of the powerful, also referred to as white-collar crimes, can be broken down into two main categories: occupational and organizational crimes. Occupational crimes are those committed by individuals within the context of their occupation for their own self-interests whereas organizational crimes are those committed by individuals within the context of their occupation for the interests of or benefit of the organization. The latter does not imply that the individual may not benefit as well, however, a major component of organizational crime is the inclusion of organizational benefit (Fig. 1). Organizational crimes can be broken down into four major subtypes: (1) state crime,

State Crime


(2) corporate crime, (3) state-corporate crime, and (4) crimes of globalization. State crimes are illegal or socially injurious act of omission or commission by an individual or group of individuals in an institution of legitimate governance which is executed for the consummation of the operative goals of that institution of governance (Kauzlarich and Kramer 1998). Genocide, human rights violations, war crimes, illegal war, and crimes against humanity are actions that fall under the category of state crime. State crimes are historically and contemporarily ubiquitous and result in more injury and death than traditional street crimes such as robbery, theft, and assault. Consider that genocide during the twentieth century in Germany, Rwanda, Darfur, Albania, Turkey, Ukraine, Cambodia, Bosnia-Herzegovina and other regions claimed the lives of tens of millions and rendered many more homeless, imprisoned, and psychologically and physically damaged (Rothe and Kauzlarich 2010). Corporate crimes are committed on behalf of business interests, sometimes by individuals, sometimes by groups; they surface among the self employed and among executives of companies large and small. Corporate crime may include theft, fraud, and violence. Famous cases of financial fraud include those committed by Enron and Adelphia in which the manipulation of stocks and various forms of deceptive accounting resulted in the theft of billions of dollars from investors. Violations of worker

C

C

780

Individual


Organizational

Immediate/Structural

State/Structural

International

Crimes of the Powerful, Fig. 2 Continuum of theoretical levels of analysis

safety and environmental regulations are also major forms of corporate crime. State-corporate crimes are illegal or socially injurious actions that result from mutually reinforcing interaction between institutions of political governance and institutions of economic production and distribution (Kramer and Michalowski 1991: 5); also see Michalowski and Kramer (Michalowski and Kramer 2006) and Matthews and Kauzlarich (2000). The concept of state-corporate crime has been used to examine the space shuttle Challenger explosion (Kramer 1992; Michalowski and Kramer (2006), the environmental devastation caused by US nuclear weapons production (Kauzlarich and Kramer 1998; Michalowski and Kramer (2006), and the deadly fire at the Imperial Food Products chicken processing plant in Hamlet, North Carolina (Aulette and Michalowski 1993; Michalowski and Kramer (2006). Other examples of state-corporate crime include the I.G. Farben Company’s involvement with Nazi atrocities (Michalowski and Kramer (2006), the Wedtech case involving defense contractor fraud (Friedrichs 1996), and the violent and deadly crash of ValuJet flight 592 in May of 1996 (Matthews and Kauzlarich 2000). Crimes of globalization are activities that lead to a range of physical and economic injury committed by international agencies in the interest of facilitating global capitalism (Friedrichs and Friedrichs 2002). Most criminological literature has examined the harms resulting from the World Bank’s funding of capitalist expansion into less developed countries, which results in the marginalization of indigenous peoples, higher rates of income inequality, environmental disaster, health crises, and political corruption (Ezeonu and Koku, Ezeonu and Koku 2008; Rothe 2010;

Rothe et al. 2006; Rothe et al. 2009). Although most of the harms are not directly criminalized by political authorities as in the case of most traditional forms of corporate crime, expanded definitions of crime rooted in critical criminology are used to frame the activities as subject to criminological examination. Occupational crimes are committed against an employer are also understood by most critical criminologists as the outcome of the need for and fetish with money in capitalist societies. While not profit-seeking in the same way as organizations, both traditional street crime and occupational white-collar crime stem from economic factors, whether real or perceived. This distinction between the various forms of crimes of the powerful reviewed above is important for any discussion or application of theory. For example, a theory explaining the motivation of someone stealing 500.00 dollars from their employer for their own financial gain is unlikely to involve the same situational and motivational factors of a head of state committing genocide. Nor are the opportunity structures or controls going to be the same. Likewise, theories apply to different levels of analysis. Each of these can be thought of as operating along a continuum from the individual or interactional level of analysis to the supra-macro level or international level (Fig. 2). The distinction between these various levels of analysis is relevant to the type of crimes being analyzed as well as the theoretical model that can be applied. For example, a theoretical approach that explains individual level behavior may have limited explanatory power in the case of corporate crime, in which organizational culture or pressures facilitating or “causing” motivating forces. As such, theories that explain


occupational crimes for example differ from those that attempt to explain organizational crimes such as corporate, state-corporate, state crime or crimes of globalization. Critical Theories Given the range of theories that have been applied to the crimes of the powerful over the last century, it is impossible to review each approach. The emphasis here is on critical and integrated theories of crime, although their relationships to non-critical explanations will be briefly noted. What distinguishes critical theories of crime from other explanations is their opposition to – not just interest in – unequal fundamental political, economic, and social structures and relationships. The major forms of critical theories include Marxist, left realism, feminist, postmodernism, cultural, and peacemaking perspectives. Respectively, their critiques are centered on capitalism, stratification and inequality, patriarchy, modernity, positivist criminology, and warmaking. Critical theories of crime have roots in general sociological conflict theory, and those that focus on crimes of the powerful are almost exclusively inspired by Marxist sociological theory (now often referred to as political economy theory). While Karl Marx said little about crime, some criminologists, especially critical criminologists, recognize a substantial debt to this nineteenth century scholar. Marx believed that a society’s mode of economic production directly influenced the manner in which relations of production are organized and thus determines in large part the organization of social relations, the structure of individual and group interaction. Under a capitalist mode of production, there are those who own the means of production and those who do not. The former group is known as the bourgeoisie and the latter as the proletariat. The bourgeoisie, or ruling class, controls the formulation and implementation of moral and legal norms, and even ideas. Both classes are bound in relationship to one another, but this relationship is asymmetrical and exploitive. This relationship affects law and crime in fundamental ways. Laws are created by the elite to protect their

781

C

interests at the expense of the proletariat. Marxists might point out that even presumably simple and well supported laws may not work in the interests of the have-nots, though they may be perceived to be a representation of the collective will of a society. Over a century ago an intellectual follower of Marx, Willem Bonger, applied some of Marx’s arguments to crime in capitalistic societies. In Criminality and Economic Conditions (published in English in 1916), Bonger observed that capitalistic societies appear to have considerably more crime than do other societies. Furthermore, while capitalism developed, crime rates increased steadily. Under capitalism, Bonger argued, the characteristic trait of humans is self interest (egoism). Given the emphasis on profit maximization and competition, and the fact that social relations are class structured and geared to economic exchange, capitalistic societies spawn intraclass and interclass conflicts as individuals seek to survive and prosper. Interclass conflict is one sided, however, since those who own and control the means of production are in a position to coerce and exploit their less fortunate neighbors. Criminal law, as one instrument of coercion, is used by the ruling class to protect its position and interests. Since social relations are geared to competition, profit seeking, and the exercise of power, altruism is subordinated to egoistic tendencies. Applied to crimes of the powerful, Bonger’s argument, which is still a basic assumption of modern Marxists, is that those in positions of corporate power are almost always focused on making, sustaining, or increasing profit. Financial crimes by corporations are therefore seen as a logical extension of capitalist logic as are violations of worker and environmental regulations and laws. Quite simply, from a basic Marxist approach, corporate crime is caused by the overriding allegiance to profit no matter who might be hurt along the way. State and state-corporate crime can also be viewed from a Marxist perspective. Considering state-corporate crime, states are seen as primarily facilitators of capital accumulation for corporations. This is especially true in the case of state-facilitated state-corporate crime

C

C

782

in which “governmental regulatory institutions fail to restrain deviant business activities, because of direct collusion between business and government, or because they adhere to shared goals whose attainment would be hampered by aggressive regulation” (Kramer and Michalowski 1991: 6). As Barnett (1981: 7) and Chambliss and Zatz (1993) have noted, a major goal of the US state has been to promote capital accumulation, and the state’s regulatory function “must not be so severe as to diminish substantially the contribution of large corporations to growth in output and employment.” For example, while state regulatory agencies have been created to help protect workers (Occupational Safety and Health Administration), the environment (Environmental Protection Agency), and consumers (Consumer Product Safety Commission), these agencies will not undermine an industry’s fundamental contributions to the functional requirements of the economy (Matthews and Kauzlarich 2000). In sum, while laws and regulations governing corporate behavior are plentiful, Marxist scholars will point out that the laws lack serious enforcement and associated penalties which therefore do little in the way of deterring corporations from crime. This is not a coincidence since capitalist states are more interested in protecting capital than they are the environment, workers, and consumers. It is important to note that most Marxist scholars now see the state as having some relative autonomy in that it is not always directly working with and for capitalists, but rather the long term survival and expansion of business and capital in general. As Chambliss and Zatz (993: 10) note “Should the state represent only the interests of capitalists, the conflicts will increase in intensity, with workers pitted against the state.... Were the state to side with the workers. . .the system would likewise collapse and a new social order would have to be constructed. Faced with this dilemma, officials of the state attempt to resolve the conflict by passing laws, some which represent the interests of capitalists and some the interests of workers.” Similar logic is used by Marxist or political economy critical criminologists to explain the basic causal properties of crimes of globalization.


In the Marxist view, states will favor business in policy making, regulatory agency mandates, and international activities. The latter is especially important in recent decades as the process of globalization spreads capital and culture throughout the far reaches of the globe. International agencies such as the World Bank, the International Monetary Fund, and the World Trade Organization are clearly pro-capitalist, free market neoliberal agencies that see private ownership and corporate control, rather than socialist and democratic collectivist initiatives, as the main ways to link and grow economies. As such, many crimes of globalization are directly related to the economic interests of those in power rather than the welfare of indigenous populations. The most recent, cutting-edge critical theories of crimes of the powerful have been developed under the umbrella of “crimes of empire.” Michalowski (2009: 308) defines this as “the construction of development strategies, citizenship, sovereignty and culture in subordinate spaces through agreements with local elites if possible, and by force if necessary, in order to facilitate processes of capital accumulation and distributions of power that are disproportionately beneficial to a dominant cosmopolitan center.” Examples of crimes of empire include many illegal military invasions, most forms of imperialism and colonialism, training and support for totalitarian regimes by capitalist states, and the suppression of human rights across the globe. Major recent analyses of crimes of empire have occurred in the context of the United States invasion of Iraq in 2003 (Kramer and Michalowski 2005). Many powerful states, like powerful corporations, are in a constant search for growth, which may include expanded territory, expedited avenues for capitalist accumulation, and increased social and cultural power and resources. The history of state crimes is replete with these interests, whether we are considering historical or more contemporary empire building by European, North American, Middle East, or African countries. A constant thread in imperialism and “war for empire” is the use of military force not only against combatants but also civilians, which is


a grave crime against humanity, and one that is often normalized in the context of patriotism or self defense in communicative discourse (Kramer 2010). Very early on non-critical scholars of traditional street crime theorized about the manner in which deviance or crime can be neutralized (Sykes and Matza 1957). Modern critical scholars of state crime in particular have found that Sykes and Matza’s (1957) theory is very useful in understanding crimes of the powerful. Originally, Sykes and Matza (1957) proposed the concept of techniques of neutralization, which involves mental and social rationalizations both before and after crimes are committed. The key forms of neutralization and justification include (1) denial of responsibility, (2) denial of injury, (3) denial of victim, (4) condemnation of the condemner, and (5) appeal to higher authority. These techniques can best be understood in terms of the simple process of rationalizing one’s own behavior, whether in response to cognitive dissonance, as a precondition to acting, aiding a costbenefit analysis, or as a post-action to minimize or attempt to legitimize or neutralize the guilt of a person’s behaviors or after the fact accountability. While techniques of neutralization are often used to explain occupational crimes, this approach also aids in our understanding of specific individuals within organizations. Scholars of state crime and crime of the powerful in general have found evidence of the operationality of techniques of neutralization (Cohen 2001; Kauzlarich et al. 2000; Kramer 2010; Simon 1999; Vaughn 1996). Indeed, one of the most important things which separate state victimizers from their victims is their power to exert their will. Most often, victimizers do not acknowledge the degree to which their policies have caused harm while assessing the effectiveness of their policies to bring about desired change or to maintain their position of dominance. Unjust and injurious domestic and international policies can also be downplayed by neutralizing reasonable categorical imperatives (e.g. do no harm) by employing bankrupt utilitarianism, which may or may not be guided by ethnocentric paternalism. Often times harms are neutralized by denying responsibility, dehumanizing the powerless for

783

C

purposes of exploitation, and appealing to higher loyalties (i.e. the capitalist political economy and “national security”) (Kauzlarich et al. 2000; Simon 1999). With respect to the criminal historical treatment of Native Americans within the United States, “colonists quickly justified their violence by demonizing their enemies” Takaki (1993: 43). However, the transference of one’s own negative tendencies to another group is not something new. While Native Americans were seen as unruly, “God-less” savages, Takaki (1993) notes that the atrocities committed by the civilized whites against the Native Americans, were, in fact savage. It is in this light, then, that Native Americans became an enemy worthy of indiscriminate killing. In much the same manner, the indiscriminate killing of the “God-less” communists of Central America during the US imperialist project there were also justified in a similar manner. In sum, the most longstanding and powerful critical analysis of the crimes of the powerful in critical criminology hinges on the assumption that corporations pursue goals of profit and expanded market shares and along the way cause significant legal and illegal harm and injury. States do so as well, either for empire building or the facilitation of corporate capitalism. Some critical scholars have taken issue with this and suggest that not all crimes of the powerful can be delineated down to economic motivations as this ignores religious, political, ideological, and cultural factors. Further, classic Marxist theory ignores other levels of analysis as well as obvious factors such as social controls and opportunities for crime. This has led to the development of critical integrated theories of crime. While these models are designed to explain state crime, they are also applicable to many other forms of organizational crime. Critical Integrated Theoretical Perspectives The earliest attempts to generate a theoretical model of state crime date back to Ronald Kramer and Raymond Michalowski’s (1990) work on state-corporate crime (see Michalowski and

C

C

784

Kramer 2006 for a complete history). The theory was later expanded on and clarified by several other criminologists and remains the only comprehensive integrated theory for addressing organizational crimes, particularly state crimes, to date. For example, in 1998, David Kauzlarich and Ronald Kramer provided a detailed schematic of their proposed integrated theoretical frame which incorporated anomie and political economy theories at the state/structural level, organizational theory at the meso level and strain, rational choice, differential association, and routine activities at the interactional level of analysis. Kauzlarich and Kramer discuss how motivation is affected by one’s socialization within that environment, the social meaning given to his or her behavior, an individual’s goals, and issues of personality such as personal morality and obedience to authority. Borrowing from Sykes and Matza (1957), reviewed above, they include techniques of neutralization as a variable of control. At the organizational level, Kauzlarich and Kramer draw heavily from organizational theorists to include instrumental rationality, role specialization, and task segregation. At the structural or institutional level of analysis, the major social institutions and social structure are included, particularly the political and economic institutions and their interrelationship as well as anomic conditions. Kauzlarich and Kramer (1998: 146) suggest the primary assumption of that perspective is that the very structure of corporate capitalism provides the impetus toward organizational crime, thus becoming crimes of capital (Michalowski 1985). They further propose that the political economy perspective stresses the shaping and/or constraining influences of the broader historical structure of a society as a factor of organizational behavior. This includes factors such as the culture of competition, economic pressure, and performance emphasis. While this integrated model has proved useful in examining numerous cases of organizational crime (see Michalowski and Kramer 2006), the theory has been critiqued for its heavy emphasis on the criminogenic nature of capitalist social organization, which limits the theory to those


crimes associated primarily with the corporate culture or within the United States. Further, Dawn L. Rothe and Christopher Mullins have suggested that the theory fails to recognize several other key factors associated with crimes of the powerful, such as weakened and transitional states, the involvement of militias, ideological and religious motivating forces, and international relations. To strengthen and expand the explanatory power of Kauzlarich and Kramer’s (1998) theory, Rothe (2006, 2009) and Rothe and Mullins (2008, 2009) proposed an expanded version of the integrated model. This included expanding the levels of analysis to include the international to incorporate the increasingly international nature of organizational criminality (as recognized by global political economy theory) and by recognizing both formal controls and informal controls or constraints on organization as well as individuals. Another difference between Kauzlarich and Kramer’s model and that of Rothe and Rothe and Mullins is the inclusion of an often ignored component of some forms of crimes of the powerful – social disorganization. While much attention has been paid to organizational context and decision-making processes by scholars of organizational crime, there is a similarly rich criminological tradition which examines how social forces work within communities that are disorganized to produce criminal actions and actors. Coming out of the Chicago School of thought, social disorganization theory (Bursik and Grasmick 1993; Shaw and McKay 1942) suggests that when communities possess a diminished capacity to create and enact informal mechanisms of social control, crime rates increase. The expanded integrated model incorporates this to provide an explanatory tool for a host of crimes of the powerful including those of heads of state operating in specific conditions as well as other organizations including corporate endeavors. Others have followed this avenue by examining social disorganization in relation to private military contractors’ (PMC) criminality in fields of operation (Rothe and Ross 2010). It has been noted that PMCs often operate within a disorganized environment. This not only


includes the community level, but the state and organizational ones as well. After all, war-torn areas are by definition disorganized. In sum, Rothe and Mullins use many of the theoretical concepts advanced by Kramer, Michalowski, and Kauzlarich, but expand the theory to incorporate international variables, factors that are not associated with capitalistic endeavors of corporations and states, and components of social disorganization. In addition, they propose other factors and theoretical concepts that allow for the adaptation of catalysts that may be unique to specific cases (e.g., paramilitary groups, insurgencies, militias, postcolonial conditions, and weakened or illegitimate governments).

785

C

organizations and people within them come to learn and practice harmful or injurious ways in the courses of their personal and professional lives.

Related Entries ▶ Corporate Crime Decision-Making ▶ Crimes of Globalization ▶ Marxist Criminology ▶ Nuclear Weapons and State Crime ▶ State Crime ▶ State-Corporate Crime

Recommended Reading and References Summary Crimes of the powerful are complex. While there are numerous non-critical theories of crime (rational choice, routine activities theory, anomie theory, to name a few) that have been used to explain occupational and organizational crimes, the focus here has been on critical and integrated approaches. Critical theories focus on power, and in the study of white-collar crime, power is mostly considered to be synonymous with some form of capital accumulation or the practices designed to protect or expand economic resources. Marxist or political economy theorists have been criticized for being too simple in causal logic. Surely the desire for profit or money is a strong motivation for behavior, yet not all corporations and states commit the same amount or type of crime. Moreover, it has been noted that ideological and religious motivations can be as strong as economical drives. Further, contexts for crime are important for they help explain under what particular situations crime is more or less likely to occur. This is where integrated theories have helped enrich the study of the causes of occupational and organization crime. Crime is multi-dimensional, even if it stems from the same source, and using theories that attend to all levels of analysis help us understand how

Almond G (1990) A discipline divided: schools and sects in political science. Sage, Thousand Oaks Aulette JR, Michalowski RJ (1993) Fire in hamlet: a case study of state-corporate crime. In: Tunnell K (ed) Political crime in contemporary America. Garland, New York, pp 171–206 Barak G (ed) (1998) Integrating criminologies. Ashgate/ Dartmouth, Aldershot Barnett H (1981) Corporate crime, corporate capitalism. Crime Delinquency 27(1):4–23 Bonger W (1916) Criminality and economic conditions. Little, Brown, Boston Bonger W (1969) Criminality and economic conditions, abridged version. Indiana University Press, Bloomington Bursik RJ Jr, Grasmick HG (1993) Economic deprivation and neighborhood crime rates 1960–1980. Law Society Rev 27(2):276–280 Chambliss W, Zatz M (1993) Making law: the state, the law, and structural contradictions. Indiana University Press, Bloomington Cohen S (2001) States of denial: knowing about atrocities and suffering. Polity Press, Cambridge Ezeonu I, Koku E (2008) Crimes of globalization: the feminization of HIV pandemic in Sub-Saharan Africa. The Global South, 2, 2, Africa in a Global Age, 112–129 Friedrichs D (1996) Trusted criminals: white collar crime in contemporary society. Wadsworth, New York Friedrichs DO, Friedrichs J (2002) The World Bank and crimes of globalization: a case study. Soc Justice 29(1–2):1–12 Green P, Ward T (2004) State crime: governments, violence and corruption. Pluto Press, London Kauzlarich D, Kramer R (1998) Crimes of the American nuclear state: at home and abroad. Northeastern University Press, Boston

C

C

786

Kauzlarich D, Matthews R, Miller W (2000) Toward a victimology of state crime. Crit Criminol 10:173–194 Kramer R (1982) Corporate crime: an organizational perspective. In: Wickman P, Daily T (eds) White collar and economic crime. Lexington Books, Lexington, pp 75–94 Kramer RC (1992) The space shuttle challenger explosion: a case study of state-corporate crime. In: Schlegel K, Weisburd D (eds) White collar crime reconsidered. Northeastern University Press, Boston, pp 214–243 Kramer R (2010) Resisting the bombing of civilians: challenges from a public criminology of state crime. Soc Justice 36(3):78–97 Kramer R , Michalowski R (1990) Toward an integrated theory of state-corporate crime. Presented at the American Society of Criminology, Baltimore Kramer RC, Michalowski RJ (1991) State-corporate crime: case studies in organizational deviance. Unpublished manuscript Kramer R, Michalowski R (2005) War, aggression, and state crime: a criminological analysis of the invasion and occupation of Iraq. British J Criminol 45:446–469 Matthews R, Kauzlarich D (2000) The crash of Valujet flight 592: a case study in state-corporate crime. Sociol Focus 33(3):281–298 Michalowski RJ (1985) Order, law and crime. Random House, New York Michalowski RJ (2008) Power, crime and criminology in the new imperial age. Crime Law Soc Change 51:303–325 Michalowski R (2009) Power, crime and criminology in the new imperial age. Crime Law Social Change 51(3– 4):303–325(23) Michalowski RJ, Kramer R (1990) Toward an integrated theory of state-corporate crime. Paper presented at the American Society of Criminology, Baltimore Michalowski RJ, Kramer R (eds) (2006) State-corporate crime: wrongdoing at the intersection of business and government. Rutgers University Press, Piscataway Pearce F (1976) Crimes of the powerful. Pluto Press, London Rothe DL (2006) The Masquerade of Abu Ghraib: state crime, torture, and international law. PhD dissertation, Department of Sociology, Western Michigan University, Kalamazoo Rothe DL (2009) The crime of all crimes: an introduction to state criminality. Lexington/Roman and Littlefield, Lanham Rothe DL (2010) Facilitating corruption and human rights violations: the role of international financial institutions. Crime, Law Soc Change 53(5):457–476 Rothe DL, Kauzlarich D (2010) State-level crime: theory and policy. In: Barlow HD, Scott D (eds) Crime and public policy: putting theory to work, 2nd edn. Temple University Press, Philadelphia, pp 166–187 Rothe DL, Mullins CW (2008) Genocide, war crimes and crimes against humanity in Central Africa: a criminological exploration. In: Haveman R, Smeulers A (eds) Supranational criminology: towards

Crimes Without Victims a criminology of international crimes. Intersentia, Antwerp, pp 135–158 Rothe DL, Mullins CW (2009) Toward a criminology of international criminal law: an integrated theory of international criminal law violations. Int J Comp Appl Crim Justice 33(1):97–118 Rothe DL, Ross JI (2010) Private military contractors, crime, and the terrain of unaccountability. Justice Q 27(4):593–617 Rothe DL, Mullins CW, Muzzatti S (2006) Crime on the high seas: crimes of globalization and the sinking of the Senegalese Ferry Le Joola. Crit Criminol Int J 14(2):159–180 Rothe DL, Mullins CW, Sandstrom K (2009) The Rwandan Genocide: international finance policies and human rights. Soc Justice 35(3):66–86 Shaw C, McKay HD (1942) Juvenile delinquency and urban areas. University of Chicago Press, Chicago Simon D (1999) Elite deviance, 6th edn. Allyn and Bacon, Boston Sutherland E (1949) White collar crime. Holt, Rinehart and Winston, New York Sykes G, Matza D (1957) Techniques of neutralization: a theory of delinquency. Am Sociol Rev 22:664–670 Takaki R (1993) A different mirror: a history of multicultural America. Little Brown, Boston Vaughan D (1982) Toward understanding unlawful organizational behavior. Mich Law Rev 80:1377–1402 Vaughan D (1992) The macro–micro connection in whitecollar crime theory. In: Schlegel K, Weisburd D (eds) White collar crime reconsidered. Northeastern University Press, Boston, pp 124–145 Vaughn D (1996) The challenger launch decision: risky technology, culture, and deviance at NASA. University of Chicago Press, Chicago

Crimes Without Victims ▶ Moral Crimes

Criminal Activity of Sex Offenders ▶ Criminal Career of Sex Offenders

Criminal Career ▶ Desistance from Crime

Criminal Career of Sex Offenders

Criminal Career of Sex Offenders Patrick Lussier Laval University, Quebec City, QC, Canada

Synonyms Criminal activity of sex offenders; Development of sex offending

Overview The interest for the criminal career is not new and several commentaries and observations about sex offenders’ criminal activity have been made for quite some time. Most of these commentaries and observations were focused on the same underlying questions, that is, sex offenders’ dangerousness. The issue of dangerousness has been addressed by examining sex offenders’ likelihood of sexual recidivism using different methodologies. These descriptive studies of sex offenders’ criminal records were not supported by an organizing conceptual framework which led to the emergence of controversies among researchers about sex offenders’ nature and extent of their criminal behavior. These controversies certainly did not help to challenge common myths and false beliefs about sex offenders’ criminal behavior which have, in some instances, serve as the foundation to develop new criminal justice policies to tackle the problem of sexual violence and abuse. The current study aims to introduce the criminal career approach and, in doing so, aims to provide a common organizing framework for policy makers as well as researchers from various disciplines such as psychology, psychiatry, sociology, social work, criminal justice, and criminology. Although there is a long history of criminal career research with the publication of Criminal Careers and Career Criminals in 1986 by Dr. Al Blumstein and colleagues, such a framework was introduced to the field of criminal justice and criminology. The criminal career approach is concerned with

787

C

the description and explanation of the longitudinal sequence of offending. While the criminal career approach has been around for quite some time in criminological circles, it would take some time, however, before this framework would be introduced more explicitly to the field of sexual violence and abuse (Blokland and Lussier 2012; Lussier et al. 2005). Building on the criminal career approach proposed by Blumstein and colleagues, the current review examines the current state of knowledge regarding the criminal activity of sex offenders. While the criminal career approach should not be seen as a cure-all approach, it provides a conceptual framework to organize existing study findings, to guide future empirical research, as well as to help to think more clearly about sex offenders’ criminal behavior. Therefore, this chapter aims, first, to introduce researchers from the field of sexual violence and abuse to the criminal career approach; second, to organize the empirical knowledge on the criminal activity of sex offenders using a criminal career approach; and, third, to review the state of empirical knowledge on various dimensions of the criminal career of sex offenders.

The Criminal Career Approach Over the past five or six decades, no other offender type has been under more scrutiny from researchers than sex offenders have been. Researchers from the field of sexual violence and abuse have measured a wide range of factors to describe individuals having committed sex crimes, such as family background, victimization experiences, exposure to deviant models, attachment bonds, parental practices, childhood behaviors, psychiatric symptoms, personality traits and disorders, intelligence and cognitive skills, sexual arousal and interests, mood and temperament, cognitive distortions, sexual development and sexual behaviors, coping skills and coping strategies, and pornography use, to name just a few. One striking observation that can be made is that while sex offenders have been described in so many ways along so many dimensions and

C

C

788


Criminal Career of Sex Offenders, Table 1 Descriptive summary of criminal career parameters Parameters Prevalence Age of onset Frequency (volume) Lambda Continuity Career length Versatility (diversity) Seriousness Specialization Desistance

Definition The proportion of a given population committing a crime during a specific time period The age at first offense The number of crimes committed The number of crimes committed taking into account the time at risk (i.e., excluding periods during which the offender did not have the opportunity to offend (i.e., hospitalization, incarceration, death)) The passage from juvenile offending to adult offending The length of time between onset and termination of offending The number of different crime types committed The degree of gravity of the criminal activity The tendency to limit offending to one particular form of crime Termination of offending

factors, comparatively speaking, the very behavior that clinical researchers aimed to explain, sexual offending, has been largely neglected. This is not an overlooking but illustrates the fact that most theoretical views of sex offending is based on the assumptions that there is a stable propensity to commit sex crime and theoretical models should only be concerned by the description and the explanation of this propensity. These models, therefore, do not recognize the importance of distinguishing such aspects as prevalence, age of onset, persistence, frequency, seriousness, and desistence. This trait-like approach is not well suited to describe and explain why offenders start or stop offending and whether the same factors explain both. Some theoretical models do make distinction for various offending stages but have been limited to a few aspects of sex offending, such as onset and persistence (e.g., Marshall and Barbaree 1990). The criminal career approach provides a framework to think about how sexual offending starts, develops over time, and stops and whether such distinctions are theoretically, clinically, as well as policy relevant (Table 1). The Onset of Sex Offending The age of onset of sex offending refers to the age at which sex offending is initiated. The age of onset is particularly interesting because it marks the origins of the behavior and allows examining why and under what circumstances the criminal

behavior was initiated. The age of onset has been discussed in several empirical studies on sex offenders, but its operationalization has not always been clear and straightforward. In earlier investigations, clinical researchers have been concerned with the age of onset of sexual problems of adult offenders. Using the term sexual problems is problematic because it encompasses behaviors such as the onset of deviant sexual arousal, the onset of deviant sexual fantasizing, the onset of deviant sexual behaviors, as well as the onset of sexual offending. Early theoretical models put much great emphasis on the role of deviant sexual fantasies as a precursor to sexual offending. Clinical research, however, has shown that only a small proportion of adult offenders report deviant sexual fantasies and/or a paraphilia, and an even fewer proportion of them report having experienced such deviant fantasies prior their offending (Marshall et al. 1991). This reinforces the importance of distinguishing deviant fantasies, deviant sexual behaviors, and sex offending. Early studies looking at the onset of sex offending has described adult sex offenders as grown-up juvenile sex offenders. For example, in the Prentky and Knight study (1993), 49 % of their sample of adult rapists reported an onset prior age 18, while the rest of the sample reported an onset in adulthood. These results mirror those reported in the Groth et al. study (1982), which showed an average age of onset of 19 years old


for a sample of sexual aggressors against women, while in the Abel et al. (1993) study, it was 22 years old. The self-reported onset age for child molesters appears to be different than the one reported for rapists, but the findings are not stable across studies. In the Prentky et al. (1993) study, whereas 49 % of adult rapists were JSOs, that number increased to 62 % for child molesters. Therefore, given these results, one could expect that the average onset age for child molesters would be younger than the one reported for rapists. This is not the case and this could be attributable to sampling differences. More precisely, these earlier studies showed some discrepancies across child molester types. To illustrate, in the Marshall et al. (1991) study, the selfreported age of onset was 24 years old for extrafamilial offenders against boys, 25 years old for extrafamilial offenders against girls, and 33 years old for incestuous fathers. Similar numbers were reported by Smallbone and Wortley (2004) suggesting that the onset of extrafamilial child molestation starts sooner than the onset of intrafamilial child molestation. These differences may be explained by opportunity structure of the offense as one needs to have a biological child to offend against him or her. These studies used retrospective data to estimate the age at which adult sex offenders started their offending behavior. From these self-report studies (see, e.g., Abel et al. 1993), however, it is not always clear whether the onset refers specifically to sex offending or to some other behavior such as the onset of deviant sexual interests, the onset of deviant sexual fantasizing, and the onset of deviant sexual arousal. Not surprisingly, the age of onset based on self-report data is younger than those based on official data (e.g., Gebhard et al. 1965; Lussier et al. 2005; Smallbone and Wortley 2004). When looking at the official age of onset, results clearly indicate that it significantly varies across sex offender types. Reports suggest that sexual aggressors of women are typically charged for a first offense in their late twenties, while for sexual aggressors of children, it is typically in their late 30s. There is a gap between the age of onset reported in self-report studies with adult sex

789

C

offenders and those found in studies based on police data. That gap, however, is relatively unknown given that self-report and official data on the age of onset are not typically analyzed in the same study limiting the conclusions that can be drawn. Further, the utility of other sources of information, such as the police report and the victim statement, has not been examined in prior research. Lussier and Mathesius (2012) provided evidence that the official age of onset, as measured with criminal justice data, provides a distorted view on the actual onset of offending, at least for some sex offenders. Their claim was based on the observation that official data of offending does not take into consideration the offender’s ability to avoid and/or delay detection. In their study, official data, police data, and victim’s account were analyzed to compare and contrast the official and actual age of onset. On average, it was found that there is a gap of about 7 years between actual and official age of onset in sex offending. The findings showed that the gap between actual and official onset was much more important for child molesters, more specifically incestuous and pseudo-incestuous fathers. These findings may suggest that victims may take significantly longer to report the crime to the authorities (if they do) when the crime is committed by a parental figure. For the most part, while the actual age of onset does not vary across sex offender types, it does for the official age of onset, suggesting differential investment in detection avoidance across offenders. Further, findings show that close to 20 % of sex offenders have already desisted or are in the process of desisting by the time they are first charged for their sex crime. Frequency of Sex Offending Frequency of sex offending refers to the number of sex crimes committed. This refers to the offender’s volume of crimes. It is interesting to note that researchers have not spent much time describing this aspect of sex offender’s offending. Like many other crime types, measuring frequency of sex offending is not as straightforward as it may seem. This may be due to the nature of the offending behavior and various

C

C

790

offending strategies adopted by sex offenders (Lussier et al. 2011). As such, frequency of sex offending may refer to the number of victims an individual has offended against. Frequency can also refer to the number of sex crime event, or the number of times an individual has sexually offended against a person. Some offenders may adopt an offending strategy in which different victims are offended against on a very limited number of occasions (once or twice). This victim-oriented strategy may characterize offenders who offend against strangers. Other offenders may decide to limit the number of victims they offend against choosing instead the maximization of a single offending opportunity by reoffending on multiple occasions against the same person. This strategy may be referred to as the event-oriented strategy and characterizes incest and pseudo-incest offenders as well as offenders offending against their partner. It is difficult to estimate the frequency of offending using available data found in past research because too often criminal justice indicators are used, such as the number of arrests (or convictions) for a sex crime. Such indicators may not reflect well the actual behaviors that lead to the arrest. For example, a sex offender may have been convicted once for sex crime which involves the abuse of a child over a 5-year period in which the victim was molested on more than 500 occasions. Relatedly, an individual may also have only one conviction for a sex crime that involves the rape of three women, each raped on a single occasion. Frequency can also be influenced by the time at risk or the time an offender was at risk of perpetrating a crime (e.g., not incarcerated, not hospitalized, not dead). As such criminal career researchers have used the term lambda or annual frequency of offending that takes into account the time at risk. Empirical studies have shown much heterogeneity in the frequency of sex offending across sex offenders. In the Lussier et al. (2011) study using a sample of convicted adult male sex offenders, the average number of victim was 1.8, but the number of victims varied between one and 13. Similarly, Groth et al. (1982) reported that the number of self-reported sex crimes reported by


their sample of sex offenders varied between one and 30. Even greater variance was found in the Weinrott and Saylor (1991) study where the range of self-reported victims varied between one and 200. Heterogeneity in the volume of sex offending can also be found across sex offender type. Such discrepancies between aggressors of women and children using official data were not found in the Weinrott and Saylor (1991) study. They found that on average, sexual aggressors of women had 1.8 victims, while it was 2.0 for child molesters. Similarly, Groth et al. (1982) using self-reported data found that rapists and child molesters had a similar average number of sex crimes (about five). This may have something to do with sample composition. Pham et al. (1999) also reported that their group of intrafamilial offenders had a mean number of 1.6 victims, much lower than what was reported for extrafamilial child molesters. Indeed, intrafamilial child molesters tend to have a significantly lower number of victims (Abel et al. 1987). Intrafamilial offenders’ lower average number of victims may be a function of structuring opportunity factors. Indeed, the average number of victims corresponds to the number of children typically found in contemporary western-country families. This may suggest that these offenders are less likely to offend outside the family setting. It may well be also that incestuous fathers are seeking event-oriented opportunity where they can maximize the number of offending opportunities against the same victim. It could be, therefore, that the sample used in the Weinrott and Saylor (1991) study included a high proportion of intrafamilial child molesters. Weinrott and Saylor (1991) study findings are also insightful about the sex offenders’ volume of sex offending because their study also used selfreported data. In other words, they asked these men to report their number of victims. Hence, while sexual aggressors of women had, on average, about two official victims, these men reported, on average, having offended against close to 12 victims. In other words, the mean number of self-reported victims was more than five times the number of official victims. It should


be noted, however, that the median number of victims was 6, or three times the average number of official victims. The median is much lower than the mean, most probably because of the presence of small group offenders with a disproportionately high number of victims. The discrepancies between official and selfreport data were also observed for child molesters. If their sample of child molesters had, on average, offended against two victims, these men reported having offended, on average, against seven victims. Similar numbers were reported by (Marshall et al. 1991) as well as by Groth et al. (1982). Nonetheless, the number of self-reported victims reported in the Weinrott and Saylor study is a far cry from those reported by Abel et al. (1987). The Abel et al. study reported that extrafamilial offenders against girls had about 20 victims on average, while for extrafamilial offenders against males, it was 150. Note that the median numbers of victims for the two groups were 1 and 4, suggesting that 50 % of their sample of extrafamilial offenders against girls self-reported only one victim, while 50 % of the sample of extrafamilial offenders against boys reported no more than four victims. Perhaps, for some reasons, a small group of offenders included in the Abel et al. (1987) exaggerated the volume of their offending. Hence, while Abel et al.’s (1987) mean number of victims looks impressive, the inspection of the median suggests that most child molesters, to the exception of those offending against boys outside the family setting, typically offend against a single victim. This is not meant to say that there are no prolific sex offenders. Lussier et al. (2011) investigated the presence of prolific offenders in a sample of adult males convicted for a sex crime. Using several sources of information (self-report, police investigation, victim statement), the frequency of offending (and lambda) was estimated. The study was insightful about the presence of prolific sex offenders for several reasons. The authors examined frequency in terms of the number of sex crime events offenders had been involved in. The study findings highlighted that about 11 % of their sample had been involved in over 300 sex

791

C

crime events as opposed to about 40 % who had been involved in only one crime event. There were no differences in the number of different victims offenders of both groups had offended against, suggesting that maximizing the number of victims is independent of the offender’s decision to maximize the number of crime events. In other words, some offenders take advantage of low-risk short-term opportunities with different victims, while others opt to benefit from a single offending opportunity by repeatedly offending against the same victim over and over. The findings revealed that the most prolific sex offenders were older and had a more conventional background characterized by a stable relationship with an adult partner, a job at the time of the offense(s), no drug issues, and no prior record for a sex crime. The convention image of the prolific sex offender contrasts with the typical image of the chronic offender stemming from empirical research with prison populations of general offenders as someone who is young, single, unemployed, has significant alcohol and/or drug issues, and a lengthy criminal record. The findings of (Lussier et al. 2011) also contrast with the media portrayal of the “sexual predator.” Continuity in Sex Offending Continuity refers to the persistence of sex offending from adolescence to adulthood. Continuity, therefore, informs about the proportion of juvenile sex offenders who become adult sex offenders as well as the proportion of adult sex offenders who were juvenile sex offenders. The continuity hypothesis stipulates that today’s juvenile sex offenders are tomorrow’s adult sex offenders, and several policies are based on this assumption. The discontinuity however suggests that most juvenile sex offenders do not go on to become adult sex offenders and that, for the most part, juvenile sex offenders’ offending behavior is limited to the period of adolescence. The lesson learned from criminal career research is that retrospective data with samples of adult offenders tend to overestimate continuity of offending because such studies fail to include youths who have desisted from offending in adolescence. It is well accepted by criminologists that there is

C

C

792

much discontinuity in antisocial behavior, but antisocial personality disorder in adulthood virtually requires antisocial behavior in youth. Whether this conclusion applies also to sex offending is unclear. It has in fact been argued that JSOs do not continue their sexual offending in adulthood. While this is certainly true for most JSOs, empirical studies indicate that a small fraction (between 5 % and 10 %) may indeed continue their sexual offending in young adulthood (e.g., Lussier et al. 2012). The proportions of JSOs continuing in adulthood gradually increase with a longer follow-up period in adulthood (between 10 % and 15 %) (e.g., Hagan et al. 2001). Bremer (1992) reported a 6 % reconviction rate in a sample of serious JSO, but the recidivism rate rose to 11 % when based on self-reports. Therefore, while the use of official data underestimates recidivism rates, it is unlikely to be able to explain the fact that the vast majority are not rearrested or caught again for a sex crime. Taken together, there appears to be sizeable discontinuity in sexual offending from adolescence to adulthood superimposed on a little continuity. Hence, while aggregate data indicate that the overwhelming majority of juveniles do not persist their sexual offending in adulthood, it does not inform practitioners and policy makers about those few juvenile offenders who do persist. Versatility in Sex Offending Versatility refers to offenders’ tendency to commit a wide array of offenses. Criminal versatility is concerned with the crime-switching patterns of offenders as offending persists over time. Two types of crime-switching patterns can be distinguished. First, crime-switching can be analyzed in the context of sex offenders’ whole criminal activity. Hence, from that perspective, researchers are concerned with the number of different types of non-sex crimes sex offenders are involved in. An underlying theme would be to examine the crime mix, or the nature of the different types of criminal activities characterizing sex offenders’ criminal repertoire. The scientific literature on the criminal versatility of sex offenders has been described in details elsewhere


(e.g., Lussier et al. 2005). Secondly, crimeswitching can also be examined strictly in the context of sexual offending. This would refer to sex offenders’ tendency to limit themselves to one form of sex crimes. Within the context of sex offending, crime-switching patterns can occur along several dimensions such as victim’s age, gender, relationship to the offender, nature of sexual acts committed by the offender, and level and type of coercion used. Others have also used the term sexual polymorphism to describe a sexual criminal activity characterized by much versatility (see Guay et al. 2001; Lussier et al. 2008). Few studies have examined the level of sexual polymorphism and crime-switching patterns in the sexual criminal activity of persistent sex offenders. Based on the current state of knowledge, there are three broad conclusions that can be drawn in regard to the offending pattern of persistent sexual offenders. First, Soothill and colleagues came to the conclusion that while sex offenders are generalists in their criminal offending, they tend to specialize in their sexual offending, confining themselves to one victim type (Soothill et al. 2000). Similarly, (Radzinowicz 1957) also found specialization in victim choice in that only 7 % of his large sample of sex offenders had convictions for crimes against both male and female victims, a finding consistent with those of Gebhard et al. (1965). More recently, Cann et al. (2007) found that only about 25 % of their sample of incarcerated sex offenders was versatile when considering victim’s age and gender as well as the offendervictim relationship. On the other hand, crimeswitching patterns may vary as a function of the dimension of the sexual polymorphism considered. For instance, while they also found much stability as to the victim’s gender, Guay et al. (2001) reported considerable versatility for those targeting adolescents. While offenders targeting children and those targeting adults remained in the same category, those offending against adolescents were likely to switch either to adults or to children. (Guay et al. 2001) hypothesized that adolescents may be a sex surrogate choice when the preferred partner was not available.


Empirical studies conducted in clinical settings have shown a divergent picture of the sex offenders’ crime-switching pattern. Weinrott and Saylor (1991) as well as Heil et al. (2003) have argued that official data hide an enormous amount of sex crimes. Using official data only, Weinrott and Saylor (1991) found that only 15 % of their sample of offender was versatile considering only three categories: adult female, extrafamilial children, and intrafamilial children. Using a self-reported computerized questionnaire, however, that number rose to 53 %. Similarly, (Heil et al. 2003) reported that incarcerated offenders in treatment are not versatile as to victim’s age (7 %) and gender (8.5 %) when assessed with official data but are when interviewed using a polygraph (70 % and 36 %, respectively). Less dramatic numbers were reported for parolees which might be explained by sampling differences (i.e., incarcerated offenders were more serious offenders) and the fact that admitting a crime was a prerequisite to enter treatment. Abel’s well-known study conducted under strict conditions of confidentiality showed that 42 % of their sample targeted victims in more than one age group, 20 % targeted victims of both gender, and 26 % committed both hands-on and hands-off crimes (Abel and Rouleau 1990). Similar results have been reported elsewhere in a sample of sex offenders assessed in a forensic psychiatric institution (Bradford et al. 1992). The overlapping nature of different forms of sexually deviant acts found in the clinical studies is counterintuitive to current typological models of sex offenders based on the characteristics of the offense. The victim’s gender, the victim’s age, the offender-victim relationship, the level of sexual intrusiveness, and the level of force used during the commission of the crime are some examples of criteria that have been used over the years to classify sex offenders (e.g., Gebhard et al. 1965; Knight and Prentky 1990). Smallbone and Wortley (2004) came to the conclusion that diversity in paraphilic activities may be a function of general deviance. Indeed, looking at different activity paraphilia (e.g., voyeurism, frotteurism, sexual sadism) in a sample of child molesters, they found that a scale measuring

793

C

the versatility of sexual deviance correlated significantly and positively with nonsexual offending. In other words, as the frequency of offending increases, so does the versatility in paraphilic interests and behaviors. Similarly, (Lussier et al. 2005) found in a sample of adult sex offenders that versatility in sex offending was strongly related to versatility in nonsexual nonviolent offending as well as versatility in nonsexual violent crime. Furthermore, using structural equation modeling, they found that such pattern of general versatility was related to an early onset and persistence of antisocial behavior. In other words, sex offenders characterized by a life-course-persistent antisocial tendency were more likely to show much versatility in their sexual offending. In that regard, (Guay et al. 2001) hypothesized that crime-switching in sexual offending might be partly explained by low self-control. Lussier et al. (2008) analyzed crime-switching patterns of a sample of convicted adult sex offenders using transition matrices and diversity indexes. They observed that crime-switching in sex offending is multidimensional in that diversity index tends to be relatively independent from one another. In other words, if an offender offends against victims from different age groups, it does not imply that this person will offend against both males and females. Therefore, contrary to Abel and Rouleau (1990) conclusions, little evidence was found suggesting that sex offenders are sexual deviates offending against different types of victims in different contexts. Furthermore, the study findings highlighted that crime-switching patterns vary across dimensions of sex crimes. On the one end of the continuum, victim’s gender and level of physical force are relatively stable across crime transitions. The notion of preference is relevant and of importance in the understanding of persistence in sexual offending. The study findings were in line with those of Soothill et al. (2000) suggesting that some specialization can be found for certain aspects of sex offending and that some aspects of offending are far from being random upon certain situational contingencies. On the opposite end of the continuum, the victim’s age

C

C

794

and sexual intrusiveness involve more crimeswitching. In other words, sex offenders are not prone to switching from males to female victims (and vice versa) or to change their level of violence across offenses. It is the nature of sexual acts (hands-on, hands-off, oral sex, penetration) and victim’s age that tend to fluctuate the most across the longitudinal sequence of sex crimes committed by sex offenders. The concept of sex surrogate might also play a part in stimulating crime-switching (Guay et al. 2001). This appears to be especially true for those having offended against adolescent victims, who might represent the second best option in the absence of the preferred victim type (i.e., children or adults). This situation appears to be true for both child molesters and rapists. Of importance, and in keeping with the sex surrogate hypothesis, is that very few child molesters also offended against adults and vice versa. Finally, the study highlighted that crime versatility in sex offending tends to increase as a function of persistence of sex offending, especially for victim’s age, offender-victim relationship, and sexual intrusiveness. Hence, the more sex offenders offend against different victims, the more their sexual criminal repertoire will diversify along those dimensions. This might partly explain discrepancies reported in earlier studies as clinical samples including more serious and persistent offenders should report more evidence of crime-switching. Specialization in Sex Offending Crime specialization is another important aspect of the criminal career of sex offenders. Various definitions of crime specialization have been proposed over the years. Criminal career researchers have generally defined specialization as the probability of repeating the same type of crime (Blumstein et al. 1986). Crime specialization is important from a crime control perspective for obvious reasons. If offenders do specialize in a particular crime type, then it gives support to the implementation of crime prevention strategies targeting known offenders involved in such crime type. The sex offender registry is an example of assuming that sex offenders are sex crime specialists. Recording personal information


in a database helping to track down convicted sex offenders is considered useful from a law enforcement standpoint. It can be used as a tool to prioritize suspect by assisting in the criminal investigation of new cases of sexual assault and abuse. According to the specialization hypothesis, if the criminal activity of a sexual offender persists, it would be primarily in sexual crime. Crime specialization is understood as the tendency for some crime to involve a higher or lower level of repetition over time. In that context, researchers may be interested in comparing whether specialization in sex crime is similar or different than specialization in burglary, drugrelated offenses, driving under the influence, auto-theft, etc. The concept of crime specialist, on the other hand, refers to individuals with a higher probability of repeating the same crime over time. In that context, researchers may be interested in determining the proportion of sex crime specialists among sex offenders and their characteristics. Specialization can be studied in two contexts: (a) whether sex offenders tend to specialize in the type of sex crime they commit (e.g., Soothill et al. 2000; Lussier et al. 2008) and (b) whether sex offenders tend to specialize is sex crimes when considering their whole criminal activity. The latter has been the subject of closer empirical scrutiny by researchers. Lussier (2005) reviewed the scientific literature on crime specialization in sex offending and concluded that empirical studies have provided little empirical evidence supporting the specialization hypothesis (see also Simon 1997). More precisely, sex offenders do not limit themselves to sex crimes, quite the contrary. This is not to say that all sex offenders are criminally versatile or that sex offenders’ criminal record always includes other crime types. Sex offenders who have a persistent criminal activity tend to be involved in other crime types that are not sexual in nature. This conclusion, however, requires closer scrutiny. Recidivism studies do show that sex offenders have a greater likelihood of being rearrested for a sex crime than non-sex offenders are. For example, in the (Langan et al. 2003) study, using a 3-year follow-up of close to 270,000 prisoners, 5 % of ASOs were rearrested


for a sex crime compared to 1 % for non-sex offenders. These results do not take into account the fact that the recidivism rates significantly vary across sex offenders. For example, Quinsey et al. (1995) showed that for comparable followup periods, the sexual recidivism rate of incest offenders was 8 % as opposed to 18 % for child molesters offending against girls and 35 % for child molesters offending against boys. Recidivism studies are impacted measures of crime specialization because it does not consider the whole criminal activity of an offender but only two successive crimes. Others have used transition matrices and have reported on the probabilities of being rearrested for a sex crime while taking into account the offender’s entire criminal career. Results have shown that crime specialization is much lower for rape (e.g., Blumstein et al. 1988) than when using a broader definition of sex crimes that includes child molestation (e.g., Stander et al. 1989). This suggests that child molesters are more likely to specialize in sex crimes than rapists. Such conclusion seems to be reinforced by the analysis of their criminal record and the importance of sex crimes in their entire criminal activity. Indeed, studies have shown that sex crimes represent about 4–14% of the entire criminal activity of rapists, while it is about 40 % for child molesters (Gebhard et al. 1965; Lussier et al. 2005). These numbers should be seen as tentative given the small number of studies having examined specialization in sex crime using a ratio of sex crimes to all crimes committed by sex offenders. Desistance from Sex Offending The study of the criminal career of sex offenders is still in its infancy. Not surprisingly, therefore, several dimensions of the criminal career of sex offenders have not been subject of much empirical research. Of importance, patterns of escalation and de-escalation in sex offending have been largely overlooked. Another key criminal career dimension that has been overlooked until recently is desistance (e.g., Kruttschnitt et al. 2000; Laws and Ward 2011). Desistance refers to the termination of the criminal career and can be understood here as the termination of sex offending.

795

C

The concept of desistance is important to describe the age at which sex offenders stop their sex offending; it allows us to determine the length of sexual offending, whether desistance from sex offending is accompanied by desistance in other crime types. Criminal career researchers understand desistance as a discrete event (the period after which offending has stopped). Developmentalists, however, understand desistance as a dynamic process by which offending slows down and becomes more specialized until complete termination. Studies having discussed the issue of desistance in the context of sex offending have generally relied on sexual recidivism indicators to determine desistance, that is, desistance is implied for the absence of a new charge or conviction during the follow-up period. This approach is somewhat misleading because with a longer follow-up period, offender considered to be desistors may become sexual recidivists. Recidivism is also problematic in the context where desistance is seen as a process as the presence of a new conviction for a sex crime may not inform about whether or not offending is less frequent and less serious over time. In the field of sexual violence and abuse, desistance has generally been discussed in terms of the presence (or absence) of an age effect on the risk of sexual recidivism (Lussier and Healey 2009). In other words, it is possible that with age, aging, and the passage of time, sex offender’s propensity to commit a sex crime changes. Researchers generally agree on the recidivism rates of the younger adult offenders and older offenders, but there is controversy about the age effect occurring for other offenders. Three main points have been at the core of the debate about the link between aging and reoffending in adult offenders: (a) identification of the age at which the risk of reoffending peaks, (b) how to best represent the trend in risk of reoffending between the youngest and the oldest group, and (c) the possibility of differential age-crime curves of reoffending. One hypothesis states that, when excluding the youngest and oldest group of offenders, age at release and the risk of sexual recidivism might be best represented by a plateau. Thornton (2006) argued that the

C

C

796

inverse correlation revealed in previous studies may have been the result of the differential reoffending rates of the youngest and oldest age groups, rather than a steadily declining risk of reoffending. In this regard, one study presented sample statistics suggesting a plateau between the early 20s and the 60s+ age groups (Langan et al. 2003) No statistical analyses were reported between the groups, thus limiting possible conclusions for that hypothesis. Another hypothesis suggested there might be a curvilinear relationship between age at release and sexual recidivism, at least for a subgroup of offenders. (Hanson 2002) found evidence of a linear relationship for rapists and incest offenders, and a curvilinear relationship was found for extrafamilial child molesters (see also Prentky and Lee 2007). Whereas the former two groups showed higher recidivism rates in young adulthood (i.e., 18–24), the latter third group appeared to be at increased risk when released in the subsequent age bracket (i.e., 25–35). This led researchers to conclude that, although rapists are at highest risk in their 20s, the corresponding period for child molesters appears to be in their 30s. These results, however, have been criticized on methodological grounds, such as the use of small samples of offenders, the presence of a small base rate of sexual reoffending, the use of uneven width of age categories to describe the data, the failure to control for the time at risk after release, and the number of previous convictions for a sexual crime (Barbaree et al. 2003; (Thornton 2006). The controversy over the age effect led researchers to question whether risk assessors should consider the offender’s age at the time of prison release and, if so, how the adjustment should be done (Barbaree et al. 2007; Harris and Rice 2007). Subsequently, two prominent schools of thought emerged, and two main hypotheses have been used to describe and explain the roles of propensity, age, and reoffending in sexual offenders: (a) the static-maturational hypothesis and (b) the static-propensity hypothesis. The static-maturational hypothesis suggests that sex offenders’ risk of reoffending is subject to a maturation effect, as this risk typically follows


the age-crime curve (Barbaree et al. 2007; Lussier and Healey 2009). Importantly, the maturation hypothesis is based on the assumption of a stable propensity to reoffend, but the offending rate can change over life course. In other words, the rank ordering of individuals (between-individual differences) on a continuum of risk to reoffend remains stable, but the offending rate decreases (within-individual changes) in a similar fashion across individuals. It was determined that the offender’s age at release contributes significantly to the prediction of reoffending, over and above scores of various risk factors said to capture sex offenders’ propensity to reoffend. Multivariate analyses showed that when controlling for prior criminal history, the rate of sexual reoffending decreases by about 2 % for every 1-year increase of the offender’s age at release (Thornton 2006). Adjusting for sociodemographic and criminal history factors, (Meloy 2005) replicated this finding for probation failure and for nonsexual reoffending, but not for sexual reoffending. This could be explained by the low base rate of sexual reoffending for this sample (i.e., 4.5 %). Other studies indicated that age at release contributes significantly to the prediction of reoffending, even after adjusting for actuarial scores (e.g., Barbaree et al. 2003). Similar to (Thornton 2006; Hanson 2006) reported that after adjusting for the scores on Static-99, the risk of sexual reoffending decreased by 2 % for every 1-year increase in age after release. No interaction effects were found between scores of the Static99 and age at release. Though these preliminary results provide evidence in favor of the maturational hypothesis, many questions remain unanswered. The key question is whether sex offenders identified as high risk are also subject to an age effect. Because previous studies did not test the maturational hypothesis separately for high-risk offenders and considering that highrisk sex offenders constitute only a small minority of all convicted sex offenders, researchers might have been limited in finding a differential age effect. The static-propensity hypothesis suggests that, by using historical and relatively unchangeable


factors, adult sex offenders can be distinguished based on their likelihood of reoffending. The main assumption is that criminal propensity is stable over life course, and therefore, risk assessment tools should only be used for measuring the full spectrum of this propensity. An important point of contention for the static-propensity hypothesis is whether younger offenders at high risk to reoffend show the same or similar recidivism rates as older offenders with the same risk to reoffend. According to the staticpropensity hypothesis, older offenders with high scores on risk assessment tools represent the same risk of reoffending as younger offenders with similar scores (e.g., Harris and Rice 2007). For static-propensity theorists, the only age factor that risk assessors should include are those reflecting a high propensity to reoffend, such as the age of onset of the criminal activity. For example, Harris and Rice (2007) argued that the effect of aging on recidivism is small. In fact, they argued that age of onset is a better risk marker for reoffending than age at release. In other words, those who start their criminal career earlier in adulthood show an increased risk of reoffending. Their findings showed that the offender’s age at release did not provide significant incremental predictive validity over actuarial risk assessment scores (i.e., VRAG) and age of onset. This could be partly explained by the fact that age of onset and age at release were strongly related, that is, early-onset offenders are more likely to be released younger than late-onset offenders. The high covariance between these two age factors might have limited researchers in finding a statistical age at release effect in multivariate analyses. Furthermore, (Barbaree et al. 2007) found that, after correcting for age at release, the predictive accuracy of actuarial tools significantly decreased, suggesting that an age effect was embedded in the risk assessment score. Actuarial tools have developed by identifying risk factors that are empirically linked to sexual reoffending. If the risk of reoffending peaks when offenders are in their 20s, it stands to reason that characteristics of this age group are most likely to be captured and included in

797

C

actuarial tools. Consequently, scores of risk assessment tools might be more accurate with younger offenders but overestimate the risk of older offenders. The findings of Lussier and Healey (2009) were generally consistent with the age-crime curve. Most sex offenders do not reoffend sexually after being released from prison and the Lussier and Healey (2009) study provided additional evidence of this. Congruent with the findings of Kruttschnitt et al. (2000), their finding was further evidence against the argument that sex offenders respond to nothing but long-term imprisonment and intensive community supervision. All offenders eventually desist, albeit at a different rate. The factors or the mechanisms explaining desistance remain tentative (Laws and Ward 2011). To illustrate the importance of age on desistance, Lussier and Healey (2009) compared the predictive accuracy of the offender’s age at release to that of the scores of an actuarial tool (i.e., Static-99) that includes one item reflecting the offender’s age at release. The findings showed that, by itself, age at release was as good a predictor of reoffending as the score of the Static-99, an actuarial tool designed to determine the risk of reoffending in sexual offenders. These results suggest that the offender’s age at release should be an important component considered by risk assessors when considering cases for longterm incapacitation and intensive community supervision. It is plausible that even if the agecrime association is quite general, it is not necessarily invariant and some offenders might deviate from that pattern. Hence, it is possible that the age effect might not operate the same way for individuals characterized by different offending trajectories. Future studies should examine whether the age-crime curve is present for sex offenders characterized by different offending trajectories and whether the age effect has the same impact on sexual recidivism across these groups. The results of the Lussier and Healey (2009) study do not provide empirical evidence for a strategy of selective incapacitation aimed at sex offenders but rather highlight the limited understanding of the role of aging and the process of desistance in sex offenders.

C

C

798

Conclusion The study of sex offenders’ criminal activity using a criminal career approach is still in its infancy. Therefore, the conclusions drawn here should be interpreted accordingly. Longitudinal studies have shown that most juvenile sex offenders do not become adult sex offenders. In fact, studies suggest that about 10 % of convicted juvenile sex offenders become adult sex offenders. Similarly, retrospective longitudinal studies with adult sex offenders suggest that most adult sex offenders were not previously juvenile sex offenders. Said differently, there is not much continuity in sex offending from adolescence to adulthood. Such continuity appears to be more important for sample of adult sex offenders sampled in maximum-security psychiatric hospital suggesting that continuity in sex offending is associated with a mental health disorder. Empirical studies suggest, therefore, that the onset of sex offending for the majority of adult offenders occurs in adulthood. The investigation of the onset of offending shows a gap of about 7 years between the actual onset of sex offending and the age at first conviction for a sex crime. This gap varies across sex offender type. While there are not much age differences in terms of the actual onset of sex offending, there are differences in terms of age at first conviction for a sex crime. Child molesters are more likely to be first convicted for a sex crime later than sexual aggressors of women. Such differences may explain why child molesters are often found to be older than sexual aggressors of women and are most probably due to the fact that child molesters go undetected for longer time periods than sexual aggressors of women. The study of offending frequency reveals that most sex offenders will commit one crime against a single victim. Of those who persist, studies reveal two main offending strategies: one based on maximizing the number of victims (victim-oriented), while the other is based on maximizing the number of events (event-oriented). Those opting for a victim-oriented strategy tend not to reoffend against the same victim, while those following an event-oriented strategy will target few victims


that will be abused repeatedly, sometimes over long time periods that can span over a decade. The study of desistance suggests that all sex offenders eventually desist from sex offending but at a different rate. Empirical research also suggests that age and aging does impact sex offender’s risk of reoffending over time in a traditional fashion. Older sex offenders have a significantly lower probability of reoffending than younger offenders, yet the criminal justice system is often imposing the most stringent conditions and sentences to older offenders which may impact their ability for successful community reintegration. Such conclusions, however, are based on official data on recidivism and subject to known limitations about official statistics on crime. Using other sources of information in future research to measure crime represents an important challenge for researchers in the field of sexual violence and abuse. Greater understanding of the criminal career of sex offenders will inform policy makers about these offenders’ pattern of offending over time but also how to best tackle the problem of sexual violence and abuse.

Related Entries ▶ Sex Offender Treatment ▶ Sexual Recidivism ▶ Specialization and Sexual Offending

Recommended Reading and References Abel GG, Becker JV, Mittleman MS, Rouleau JL, Murphy W (1987) Self-reported sex crimes of non-incarcerated paraphiliacs. J Interpers Violence 2:3–25 Abel GG, Rouleau JL (1990) The nature and extent of sexual assault. In: Marshall WL, Laws DL, Barbaree HE (eds) Handbook of sexual assault: issues, theories, and treatment of the offender. Plenum Press, New York, pp 9–21 Abel G, Osborn CA, Twigg DA (1993) Sexual assault through the life span: adult offenders with juvenile histories. In: Barbaree HE, Marshall WL, Hudson SM (eds) The juvenile sex offender. Guilford, New York, pp 104–117 Barbaree HE, Blanchard R, Langton CM (2003) The development of sexual aggression through the life span. Ann N Y Acad Sci 989:59–71

Criminal Career of Sex Offenders Barbaree HE, Langton CM, Blanchard R (2007) Predicting recidivism in sex offenders using the VRAG and SORAG: the contribution of age-atrelease. Int J Forensic Ment Health 6:29–46 Blokland A, Lussier P (2012) Sex offenders: a criminal career approach. Wiley-Blackwell, Oxford Blumstein A, Cohen J, Roth JA, Visher CA (1986) Criminal careers and “career criminals”. National Academy Press, Washington, DC Blumstein A, Cohen J, Das S, Moitra SD (1988) Specialization and seriousness during adult criminal careers. J Quant Criminol 4:303–345 Bradford JM, Boulet J, Pawlak A (1992) The paraphilias: a multiplicity of deviant behaviours. Can J Psychiat 37:104–108 Bremer JF (1992) Serious juvenile sex offenders: treatment and long-term follow-up. Psychiatr Ann 22:326–332 Cann J, Friendship C, Gozna L (2007) Assessing crossover in a sample of sexual offenders with multiple victims. Leg Criminol Psychol 12:149–163 Gebhard PH, Gagnon JH, Pomeroy WB, Christenson CV (1965) Sex offenders: an analysis of types. Harper & Row, New York Groth AN (1982) The incest offender. In: Sgrol SM (ed) Handbook of clinical intervention in child sexual abuse. D.C. Heath and Co, Lexington, pp 215–239 Guay JP, Proulx J, Cusson M, Ouimet M (2001) Victimchoice polymorphia among serious sex offenders. Arch Sex Behav 30:521–533 Hagan MP, Gust-Brey KL, Cho ME, Dow E (2001) Eightyear comparative analyses of adolescent rapists, adolescent child molesters, other adolescent delinquents, and the general population. Int J Offender Ther Comp Criminol 45:314–324 Hanson RK (2002) Recidivism and age: follow-up data from 4673 sexual offenders. J Interpers Violence 17:1046–1062 Hanson RK (2006) Does static-99 predict recidivism among older sexual offenders? Sex Abuse-J Res Treat 18:343–355 Harris GT, Rice ME (2007) Adjusting actuarial violence risk assessments based on aging or the passage of time. Crim Justice Behav 34:297–313 Heil P, Ahlmeyer S, Simons D (2003) Crossover sexual offenses. Sex Abuse: J Res Treat 15:221–236 Knight RA, Prentky RA (1990) Classifying sexual offenders: the development and corroboration of taxonomic models. In: Marshall WL, Laws DR, Barbaree HE (eds) Handbook of sexual assault: issues theories and treatment of the offender. Plenum Press, New York, pp 23–54 Kruttschnitt C, Uggen C, Shelton K (2000) Predictors of desistance among sex offenders: the interaction of formal and informal social controls. Justice Q 17:61–87 Langan PA, Schmitt EL, Durose MR (2003) Recidivism of sex offenders released from prison in 1994. US Department of Justice, Washington, DC

799

C

Laws DR, Ward T (2011) Desistance from sex offending: alternatives to throwing away the keys. Guildford, New York Lussier P, Healey J (2009) Rediscovering Quetelet, again: the “aging” offender and the prediction of reoffending in a sample of adult sex offenders. Justice Q 26:827–856 Lussier P, Mathesius J (2012) Criminal achievement career initiation and cost avoidance: the onset of successful sex offending. J Crime Justice, in press Lussier P (2005) The criminal activity of sexual offenders in adulthood: Revisiting the specialization debate. Sexual Abuse: A Journal of Research and Treatment 17:269–292 Lussier P, LeBlanc M, Proulx J (2005) The generality of criminal behavior: a confirmatory factor analysis of the criminal activity of sexual offenders in adulthood. J Crim Justice 33:177–189 Lussier P, Leclerc B, Healey J, Proulx J (2008) Generality of deviance and predation: crime-switching and specialization patterns in persistent sexual offenders. In: Delisi M, Conis P (eds) Violent offenders: theory, public policy and practice. Jones and Bartlett Publishers, Boston, pp 97–140 Lussier P, Bouchard M, Beauregard E (2011) Patterns of criminal achievement in sexual offending: unravelling the “successful” sex offender. J Crim Justice 39:433–444 Lussier P, van den Berg C, Bijleveld C, Hendriks J (2012) A developmental taxonomy of juvenile sex offenders for theory research and prevention: the adolescentlimited and the high-rate slow desister. Crim Justice Behav, in press Lussier P, Mathesius J (2012) Criminal achievement, career initiation, and cost avoidance: The onset of successful sex offending. J Crim Justice 35: 376–394 Lussier P, van den Berg C, Bijleveld C, Hendriks J (2012) A developmental taxonomy of juvenile sex offenders for theory, research and prevention: The adolescentlimited and the high-rate slow desister. Crim Justice Behav 39:1559–1581 Marshall WL, Barbaree HE (1990) An integrated theory of the etiology of sexual offending. In: Marshall WL, Laws RD, Barbaree EH (eds) Handbook of sexual assault: issues, theories, and treatment of the offender, applied clinical psychology. Plenum Press, New York, pp 257–275 Marshall WL, Barbaree HE, Eccles A (1991) Early onset and deviant sexuality in child molesters. J Interpers Violence 6:323–335 Meloy ML (2005) Sex offender next door: an analysis of recidivism, risk factors, and deterrence of sex offenders on probation. Crim Justice Policy Rev 16:211–236 Pham TH, DeBruyne I, Kinappe A (1999) E´valuation statique des de´lits violents chez les de´linquants sexuels incarce´re´s en Belgique francophone. Criminologie 32:117–125

C

C

800

Prentky RA, Knight RA (1993) Age of onset of sexual assault: criminal and life history correlates. In: Hall GCN, Hirschman R, Graham J, Zaragoza MS (eds) Sexual aggression: issues in etiology assessment, and treatment. Taylor & Francis, Washington, DC, pp 43–62 Prentky RA, Lee AFS (2007) Effect of age-at-release on long term sexual re-offense rates in civilly committed sexual offenders. Sex Abuse: J Res Treat 19:43–59 Quinsey VL, Lalumie`re ML, Rice ME, Harris GT (1995) Predicting sexual offenses. In: Campbell JC (ed) Assessing dangerousness: violence by sexual offenders, batterers, and child abusers. Sage, Thousand Oaks, pp 114–137 Radzinowicz L (1957) Sexual offences: a report of the Cambridge department of criminal justice. Macmillan, London Smallbone SW, Wortley RK (2004) Onset, persistence, and versatility of offending among adult males convicted of sexual offenses against children. Sex Abuse: J Res Treat 16:285–298 Simon L (1997) The myth of sex offender specialization: An empirical analysis. New England Journal on Criminal and Civil Commitment 23:387–403 Soothill K, Francis B, Sanderson B, Ackerley E (2000) Sex offenders: specialists, generalists or both? Br J Criminol 40:56–67 Stander J, Farrington DP, Hill G, Altman PME (1989) Markov chain analysis and specialization in criminal careers. Br J Criminol 29:317–335 Thornton D (2006) Age and sexual recidivism: a variable connection. Sex Abuse: J Res Treat 18:123–135 Weinrott MR, Saylor M (1991) Self-report of crimes committed by sex offenders. J Interpers Violence 6:286–300

Criminal Careers

learn and further understand the dimensions of criminal careers as well as offending behavior across the life course. First articulated by Alfred Blumstein and colleagues in 1986, the criminal career approach to understanding patterns of offending behavior has provided a wealth of information that has fostered greater understanding of criminal behavior, greater knowledge to foster theoretical development, and realignment as well as important implications for the development of age-graded policy responses to crime and criminal behavior. While the strongest emphasis of the criminal career approach has been on generating a wealth of information about criminal careers, including factors related to their initiation, factors related to components of active offending activity, as well as factors related to career desistance or termination, the approach has important implications for informing policies and programs aimed at controlling or preventing crime and criminal behavior. This entry provides an opportunity to describe the criminal career approach as well as chart how the criminal career paradigm provides a useful vehicle for informing and facilitating the development of effective policies and practices aimed toward controlling and preventing crime and criminal behavior across the life course.

Criminal Careers Introduction ▶ Onset of Offending

Criminal Careers and Public Policy Responses Paul Mazerolle Arts, Education, and Law, Griffith University, Griffith, QLD, Australia

Overview The past three decades of research and theorizing in Criminology has provided an opportunity to

The criminal career paradigm has emerged over the past three decades and is one of the most dominant perspectives for facilitating empirical research about criminal offending behavior. The modern understanding of the criminal career approach dates to work by Alfred Blumstein and colleagues (1986); however, the precursors to this form of thinking had its origins in prior work examining offending birth cohorts (cf. Wolfgang et al. 1972) as well as research examining controversial policy responses to crime (Greenwood and Abrahamse 1982). The rise of the criminal career approach and the proliferation of research on aspects of criminal careers have coincided with the rise of

Criminal Careers and Public Policy Responses

the life course approach in Criminology. Life course Criminology, which incorporates Developmental Criminology (LeBlanc and Loeber 1998), represents an approach for examining offending behavior over time, through agegraded sequences as well as contexts that are both meaningful and subject to change (Piquero and Mazerolle 2001). According to this approach the life course incorporates age-graded roles, responsibilities, and institutional contexts and sequences that develop over time (Elder 1995). The life course approach provides a rich framework for understanding behavioral change and continuity over time and across the age-graded life span. It includes concepts such as trajectories over the life course as well as transitions and turning points (Sampson and Laub 1993). A related way of conceptualizing the life course approach involves embracing the concept of offending pathways which vary over time and are influenced by a range of social contexts. Overall, the life course perspective presents unique opportunities for developing a comprehensive understanding of criminal behavior. The life course approach is entirely consistent with the criminal career paradigm, and both approaches have implications for informing policy directions aimed at preventing and controlling criminal behavior. In the following sections, a presentation of the criminal career approach is provided. This is followed by an examination of policy responses to crime and criminal behavior which relate to key dimensions of criminal careers across the life course.

The Nature of Criminal Careers The criminal career approach examines dimensions of offending behavior and aims to categorize such meaningful dimensions into concepts that foster empirical assessment and understanding. The criminal career approach differentiates individual patterns into three broad phases, including the onset or initiation of offending, the active offending phase, as well as the desistance or termination phase. Criminal

801

C

careers differ with regard to when they begin (early or late onset), how they proceed (highly specialized vs. versatile, high or low frequency, etc.), and whether and when they finish or terminate (desist or persist). Over the past 20 years, a wealth of research has uncovered various aspects of criminal careers. For example, research supports the finding that offenders with an early age at onset to offending are at an enhanced risk for a serious and persistent criminal career (Moffitt 1993; Patterson and Yoerger 1997). Offenders with a later-onset age tend to have less prolonged offending careers and stronger prospects for completing school, securing employment, and leading a life with diminished offending risks in adulthood. Importantly, the precursors for early and late onset offending vary and consequently have implications for prevention efforts. The bulk of the criminal career can be described by an individual’s offending frequency, the nature of their offending with respect to offense type (e.g., violence, property crime), as well as whether offending is escalating in seriousness over time. Indeed, research consistently finds that serious offenders are those that begin their careers early, exhibit a high frequency of offending across diverse or versatile types of offending including violence, over a prolonged or persistent period of time (Moffitt 1993). Importantly, it is crucial to ascertain the nature, dimensions, and precursors for high-frequency offending, as well as factors that relate to versatile or specialist offending patterns because the opportunities for prevention are informed by the diverse nature of offending causation. In terms of the later stages of criminal careers, research illustrates factors related to desistance or career termination (Laub and Sampson 2003; Sampson and Laub 1993) as well as factors related to the process of desistance (Bushway et al. 2001). Increasingly, research reveals the value of employment skills and experiences, as well as the protective influences and consequent informal social control related to developing meaningful intimate relationships including marriage (Horney et al. 1995). Clearly, understanding when, how, and why active offenders

C

C

802

refrain from further offending is significant for informing a range of programs and practices to minimize offending persistence. The value of the criminal career approach rests in the meaningfulness of the concepts for capturing criminal career dimensions (e.g., onset, frequency), the ability to identity correlates for related career dimensions, the ability to inform extant theories about criminal behavior, as well as in utilizing the knowledge gained to inform policy and practice directions to foster greater crime prevention and control. It is to this issue we turn in the next section.

Criminal Careers and Crime and Justice Policy and Practice The past 20–30 years has borne witness to an increasing emphasis on understanding criminal careers as well as an emphasis on understanding crime across the life course. These developments have co-occurred during a period whereby an increasing emphasis on evidence-based policy and practice has occurred across the western world. Criminal justice policy making has been observed to be oftentimes a haphazard exercise fraught with ideological influences as well as misinformed conceptions about crime, its causes, as well as what constitutes effective crime control and prevention programs (Cullen and Gendreau 2001; Currie 1993). Over the past 15 years, there has been an increasing push toward evidenceinformed policy responses, and many criminologists have been instrumental in supporting such directions. These movements have also been reinforced by a public increasingly frustrated with ineffective and/or wasteful crime policy responses. It is within this context that the role of research knowledge about criminal careers across the life course is well placed to inform responses to criminal behavior. Indeed, if the precursors of criminal career dimensions vary, then the policy and practice responses need to be differentiated to address various aspects of criminal career dimensions. For example, the policy responses for reducing onset to criminality may


well need to differ from the policy responses that reduce violent offending, frequent offending, or chronic offending. A large body of research has indeed addressed these issues. We turn now to considerations of policy responses to crime across the life course.

Policy and Practice Focus Areas Across Criminal Career Dimensions and the Life Course: Preventing or Delaying Offending Onset Given that criminal careers are often established early in the life course, programs aimed at reducing or delaying offending onset represent important areas for investment. The past decade has seen a push toward early intervention programs as such efforts seek to provide effective supportive conditions and protective influences around young people at risk for embarking on an offending pathway (i.e., onset). Programs and policies designed to prevent onset to an offending pathway are generally targeted toward family or school settings because they represent important institutional contexts whereby effective interventions can occur. There are several examples of programs aimed at preventing offending onset. For example, a Nurse-Family Partnership program designed by David Olds has been aimed at supporting pregnant mothers through enhanced visits before and after the birth to assist them with supportive information and assistance in dealing with young children. Three randomized controlled trials have been conducted related to this program, and improvements in health and economic outcomes have been observed which have subsequent benefits for reducing crime. A study on the crime reduction benefits of this program demonstrates discernible reductions in arrests and convictions as well as reduced drug- and alcohol-related problems in comparisons to control subjects (Olds et al. 1998). Another area of focus for delaying or preventing onset involves parent training programs. Parent training programs are aimed at providing effective parenting models and skills


which subsequently translate into improved socialization of the child. These programs are especially aimed at fostering improved social and emotional competence in young people, readiness for school, as well as addressing conduct and behavioral challenges. One of the developers of an effective program, Dr. Carolyn Webster-Stratton has been at the forefront of research demonstrating the effectiveness of such approaches (Webster-Stratton and Taylor 2001). Numerous studies have demonstrated that early family/parent training programs are effective in reducing crime in adolescence and adulthood. A range of additional programs have shown good success in supporting children in their development and preparation for school, with the Perry Preschool Project a notable standout (Schweinhart 2007). Additionally, a number of school-based delinquency prevention programs have shown good promise in preventing delinquency initiation and programs which target cognitive processes and critical thinking and behavioral skills appear to be worthy of further investment (Gottfredson 2001).

Reducing High-Frequency and Chronic Offending Offenders who engage in serious, violent, frequent, and persistent criminal behavior generate a high degree of policy and program attention. Such chronic, high-rate offenders generate a high volume of the crime which is managed through the criminal justice system and which creates a high degree of harm for the community. In response, several policies and programs have been implemented with varying degrees of success. The 1980s, for example, included a push toward selective incapacitation approaches for chronic offenders (Greenwood and Abrahamse 1982). Selective incapacitation aimed to further prolong prison sentences for chronic, high-frequency offenders to ensure their offending risks could be removed from community exposure. Such approaches, while popular in some circles, failed to garner support due to concern over the effectiveness and fairness of such approaches.

803

C

A further example of a “get tough” policy approach relates to the three strikes and you’re in policy across many jurisdictions. Such policies, while popular among a subset of the community, tended not to persist because of cost concerns, concerns about their minimal impact of crime across the community, as well as concerns over sentencing equity and consistency. A further area of challenge in designing programs and responses for serious offenders relates to conceptions of specialized offending patterns which require program attention. For example, a range of programs exist for dealing with sexual offenders or domestic violence offenders but much of the evidence suggests that such programs are of mixed effectiveness. A major explanation for such equivocal results is that most high-frequency, serious offenders exhibit versatile offending patterns and therefore require responses that reflect their generalized criminal tendencies. Thus, more generalized policy and program responses have preventative implications for domestic violence as well as sexual offending. While there are encouraging results emerging from a range of prevention programs (e.g., Communities That Care, youth mentoring), the move toward multisystemic therapy (MST) programs among several program providers is especially encouraging. MST was developed by Dr. Scott Henggeler. The program provides intensive family- and community-focused treatment which aims to confront the institutional systems that influence chronic and violent juvenile offenders. MST views individuals embedded within various systems involving family, community, peers, etc. Thus, the intensive program response often includes teachers, friends, as well as family members. In general, the MST program is aimed at the high-frequency, high-rate, and serious offender group. Over a period of several years and studies, MST reveals consistently positive results across various domains including school, family, and crime. For example, evaluations of MST have revealed improvements in keeping youth in school, improvements in family relationships, reductions in drug and alcohol use, as well as large reductions in rearrests of up to 70 %.

C

C

804


Consequently, MST is one of the most impressive programs available for addressing serious, violent, and chronic offenders. Finally, important opportunities exist for correctional rehabilitation programs to target areas of intervention for persistent, high-risk offenders. While correctional rehabilitation has been out of favor at times, the weight of research evidence demonstrates the important prosocial outcomes which emanate from effective programs (Andrews et al. 1990). A series of studies consistently reveal the positive results that can be observed through investing in programs that target key areas of intervention need (antisocial attitudes, critical cognitive processing skills, etc.), especially for high-risk offenders, and in programs that are responsive or tailored to the learning styles of offenders (Andrews et al. 1990).

educational opportunities for active offenders, much like job skills, represent important opportunities to strengthen social control and social ties to prosocial institutions. With increased educational attainment, expanded employment opportunities arise, and such investments in conformity can lead to reductions in re-offending. Given the ongoing relationship between drugs and crime, it appears that desistance is not easily achieved among offenders with drug abuse problems. Thus, efforts at fostering drug abuse treatment among active offenders are an important investment toward reducing persistent offending over time. Moreover, there is clear and established evidence which shows that among serious offenders with drug abuse problems, participating in treatment is associated with reductions in drug use, higher employment rates, as well as reductions in crime.

Facilitating Desistance and Career Termination

Challenges and Implications

Fostering desistance from crime represents an ongoing public policy challenge. Research on criminal careers illustrates how offending changes across the life course and reveals multiple areas for prevention, in particular in the emerging adulthood phase of the life course. Key areas of focus for fostering desistance relate to the need for employment skills, educational attainment, substance abuse prevention, as well as opportunities for meaningful relationship development. Criminologists John Laub and Rob Sampson have been at the forefront in undertaking research on how informal social control in adulthood can foster desistance. Experiences in acquiring employment and meaningful intimate relationships (e.g., marriage) have been shown to support offending desistance (Sampson and Laub 1992) in adulthood. Thus, investing in the employment skills of established or active offenders represents an important opportunity for fostering desistance. A similar observation holds for education. Increased levels of education are consistently related to desistance and investments in

The explosion of research on criminal careers has shaped understanding about criminality across the life course. Beyond its direct contribution to shaping understanding as well as theories about criminal behavior, this knowledge is highly significant for informing practical responses to crime and ensuring effective crime policies are enacted, implemented, and supported. There are a range of implications regarding the importance of placing criminal career research findings into a policy and practice context. For example, while the evidence about effective policy responses to crime and criminal behavior evolves over time, the embrace of politicians and policy leaders to such knowledge is highly variable. At times, evolving knowledge is met with evolving acceptance. At other times, evidence-informed policy and practice appears to take a backward step. A case in point relates to the embrace of correctional boot camps for juvenile offenders. The research from the 1990s to 2000s consistently revealed problems associated with boot camps and that they were not effective in addressing juvenile offending (MacKenzie et al. 2001). Yet, a recently elected state


government in Australia has campaigned on a get tough on crime election platform and is implementing a “new style” boot camp to address serious juvenile crime. It is unsurprising that the leader in charge of this initiative has a background in the military and obviously knows what works, despite evidence to the contrary. In summary, gains have been hard fought and achieved in relation to the embrace of evidence to inform policy and practice around the western world, but such gains can be quickly lost. Thus, it is important to be ever mindful and vigilant about the fragility of such gains as new generations of politicians and political masters require exposure to evidence about the nature of criminal careers as well as evidence about what works to prevent and control criminal behavior across the life course. While there are strong connections between criminal career dimensions and areas for public policy focus and investment (e.g., delaying onset), it is important to understand the role of social context and related factors that can influence criminal career dimensions as well as the required policy responses to prevent and control crime. This observation simply calls attention to the point that context matters and rarely is it the case that patterns of offending are invariant across gender, racial groups, and social class differentiations among other considerations. Indeed prior research has recognized the significant variations in criminal career dimensions as well as various risk factors across gender, race, and social class (Piquero et al. 2007). Consider, for example, the role of gender. Although a number of common factors are found to increase risk for both males and females, some differences do appear in the overall sensitivity to certain risk and protective factors as well as in the developmental course and nature of antisocial behavior between the genders. For example, community-level factors such as socioeconomic disadvantage have been found to affect girls as young as 5 years old. There is also some evidence that females may be more responsive than males to the protective influence of factors such as prosocial role models, reduced

805

C

community violence, and enhanced extracurricular opportunities. Females appear to be more sensitive to the adverse effects of poor familial attachment and conflict but also more sensitive than males to the protective effect of strong attachment to family. Female juvenile offenders tend to present with multiple and chronic family-level risk factors and a family environment which is typically more adverse than that experienced by male juvenile offenders. Females associating with delinquent male peers, particularly females experiencing an early-onset menarche, are at significantly greater risk of adolescent-onset conduct problems (Moffitt et al. 2001). Moreover, puberty presents a time of enhanced risk for females, particularly those with preexisting externalizing behavior problems or otherwise salient risk factors such as academic difficulties, a history of abuse, or parental criminality/psychopathology (ChesneyLind and Sheldon 2004). At an individual level, the risk associated with neurocognitive difficulties seems to be more salient for males (Moffitt et al. 2001). Also, males typically exhibit higher trait levels of impulsivity and negative emotionality than females. Importantly, it is this difference which appears to primarily underlie the marked difference between males and females in the prevalence of adolescent antisocial behavior and conduct disorder (Moffitt et al. 2001). There is also some evidence that developmental trajectories to antisocial behavior and delinquency differ between the genders. Moffitt (1993) distinguished between two etiological trajectories to adolescent antisocial behavior problems: one which begins and typically ends in adolescence (adolescence-limited) and one with an early, childhood-onset which continues on into adulthood (life course persistent). However, more recently other researchers (e.g., Silverthorn et al. 2001) have argued that while Moffitt’s (1993) adolescent-limited and life course persistent taxonomy captures the timing of both male and female antisocial behavior/offending patterns, it may be failing to identify a subgroup of adolescent-limited females who exhibit a more serious pattern of

C

C

806

offending/antisocial behavior problems as well as a more problematic childhood risk profile. The above description illustrates that diversity exists in the criminogenic risk factors for males and females, and such diversity has implications for the flow on toward criminal career development and related dimensions. Such diverse contexts and influences on criminal career initiation and related dimensions obviously transcend gender and involve further contexts including race, age, and social class, among other areas. Prevention and crime control efforts developed to address certain aspects of criminal career dimensions (e.g., onset, frequency) require flexibility and adaptability to accommodate for contextual diversity. It is clearly the case that when policy simplicity tries to address behavioral complexity, very poor crime reduction results are observed. What communities and governments need to strive for are ensuring that policies and programs aimed at controlling and preventing crime are informed by the best knowledge around behavioral complexity of criminal career dimensions and diversity to support appropriate translation into comprehensive policy and programmatic development and delivery that moves importantly beyond the one size fits all approach so often observed around the world.

Conclusion The criminal career paradigm provides a useful approach for understanding unique dimensions of offending behavior. The approach enables researchers to examine correlates of unique career dimensions (e.g., onset, persistence, specialization) and uncover diversity in the nature of criminal careers. At the same time, the criminal career approach has several implications for fostering knowledge about policies and practices aimed toward crime control and prevention and in this way can contribute directly to effective crime reduction policies. This entry has demonstrated a range of policy and program approaches that emerge from considering the criminal career approach. In the


main, responses that are informed by key aspects of career dimensions are best placed to impact upon crime and criminal behavior. A range of policy and program options exist for addressing the initial stages of criminal careers. Examples include nurse home visitation programs, parent training, as well as early school transition programs. By contrast, more established criminal careers require specific approaches in response to career dimensions. For example, the frequent offender, the persistent, or the chronic offender requires more concentrated criminal justice related interventions. Examples of efforts that appear to be ineffective are evident (e.g., selective incapacitation, boot camps, three strikes), however, there are encouraging programs worthy of investment including multisystemic therapy (MST) as well as a range of cognitive behavioral treatment programs aimed at the prison-based population of offenders who will one day return to their communities. Responses aimed at fostering desistance and criminal career termination require attention to the criminogenic factors that prolong criminal offending and drug abuse treatment programs show good promise in this area. Moreover, greater attention to fostering informal social bonds to employment experiences as well as to intimate relationships represent good opportunities for turning points away from crime. Thus opportunities for acquiring job skills and education represent useful investments aimed toward fostering the process of desistance over time. Knowledge about criminal career dimensions and their understanding continues to evolve. While current understanding about effective, evidence-based policy responses to diverse crime problems is strengthening over time, a constant challenge is in ensuring that evidence about what works to control and prevent crime is embraced by the relevant authorities. Importantly, the criminal career approach is ideally placed for fostering knowledge about varying aspects of offending across the life course as well as illustrating a range of useful policy investments for preventing and controlling criminal activity.


Related Entries ▶ Age-Crime Curve ▶ Career Criminals and Criminological Theory ▶ Communities That Care ▶ Crime Specialization, Progression, and Sequencing ▶ Criminal Career of Sex Offenders ▶ Criminal Careers and Public Policy Responses ▶ Desistance and Supervision ▶ Desistance from Crime ▶ Desistance from Gangs ▶ Evidence-Based Policy in Crime and Justice ▶ Identification Issues in Life Course Criminology ▶ Measuring Crime Specializations and Concentrations ▶ Offense Specialization: Key Theories and Methods ▶ Officer Safety, Health, and Wellness ▶ Onset of Offending ▶ Pathways to Delinquency ▶ Risk Factors for Adolescent Sexual Offending ▶ School-Based Interventions for Aggressive and Disruptive Behavior: A Meta-Analysis ▶ Sex Offenders and Criminal Policy ▶ Specialization and Sexual Offending ▶ Specialization in Juvenile Offending

Recommended Reading and References Andrews DA, Zinger I, Hoge R, Bonta J, Gendreau P, Cullen F (1990) Does correctional treatment work? A clinically relevant and psychologically informed meta-analysis. Criminology 28:369–404 Blumstein A, Cohen J, Roth J, Visher C (1986) Criminal careers and ‘career criminals’, vol 1. National Academy Press, Washington, DC Bushway S, Piquero A, Broidy L, Cauffman B, Mazerolle PI (2001) A developmental framework for empirical research on desistance. Criminology 39:491–515 Chesney-Lind M, Sheldon R (2004) Girls, delinquency and juvenile justice, 3rd edn. Wadsworth, Belmont Cullen FT, Gendreau P (2001) From nothing works to what works: changing professional ideology in the 21st century. Prison J 81:313–338 Currie E (1993) Reckoning: drugs, the cities, and the American future. Hill and Wang, New York Elder G Jr (1995) The life course paradigm: social change, and individual development. In: Moen P, Elder G Jr, Luscher K (eds) Examining lives in context:

807

C

perspectives on the ecology of human development. American Psychological Association, Washington, DC, pp 101–135 Gottfredson DC (2001) Schools and delinquency. Cambridge University Press, New York Greenwood PW, Abrahamse A (1982) Selective incapacitation. Rand, Santa Monica Horney J, Osgood DW, Marshall IH (1995) Criminal careers in the short-term: intra-individual variability in crime and its relation to local life circumstances. Am Sociol Rev 60:655–673 Laub JH, Sampson R (2003) Shared beginnings, divergent lives: delinquent boys to age 70. Harvard University Press, Cambridge, MA LeBlanc M, Loeber R (1998) Developmental criminology updated. In: Tonry M (ed) Crime and justice: an annual review of research, vol 23, pp. University of Chicago Press, Chicago, pp 115–198 MacKenzie DL, Wilson DB, Kider S (2001) Effects of correctional boot camps on offending. The Ann Am Acad Polit Soc Sci Novemb 578:126–143 Moffitt TE (1993) Adolescence-limited and life-coursepersistent antisocial behavior: a developmental taxonomy. Psychol Rev 100:674–701 Moffitt TE, Caspi A, Rutter M, Silva PA (2001) Sex differences in antisocial behaviour: conduct disorder, delinquency, and violence in the Dunedin Longitudinal Study. Cambridge University Press, Cambridge Olds D, Henderson C Jr, Cole R, Eckenrode J, Kitzman H, Luckey D, Pettitt L, Sidora K, Morris P, Powers J (1998) Long-term effects of nurse home visitation on children’s criminal and antisocial behavior: 15 year follow-up of a randomized controlled trial. JAMA 280:1238–1244 Patterson GR, Yoerger KL (1997) A developmental model for later-onset delinquency. In: Deinstbeir R, Osgood DW (eds) Motivation and delinquency. University of Nebraska Press, Lincoln, pp 119–177 Piquero AR, Mazerolle P (2001) Life-course criminology: contemporary and classic readings. Wadsworth, Belmont Piquero AR, Farrington DP, Blumstein A (2007) Key issues in criminal career research: new analyses of the Cambridge Study in delinquent development. Cambridge University Press, Cambridge, UK Sampson R, Laub JH (1993) Crime in the making: pathways and turning points through life. Harvard University Press, Cambridge, MA Schweinhart LJ (2007) Crime prevention by the High/Scope Perry Preschool Program. Vict Offenders 2:141–160 Silverthorn P, Frick PJ, Reynolds R (2001) Timing of onset and correlates of severe conduct problems in adjudicated girls and boys. J Psychopathol Behav Assess 23:171–181 Webster-Stratton C, Taylor T (2001) Nipping early risk factors in the bud: preventing substance abuse, delinquency, and violence in adolescence through interventions targeted at young children (0–8 years). Prev Sci 2:165–192 Wolfgang M, Figlio R, Sellin T (1972) Delinquency in a birth cohort. University of Chicago Press, Chicago

C

C

808

Criminal Careers of Places Sue-Ming Yang1, David Weisburd2,3 and Elizabeth R. Groff4 1 Department of Criminology, National ChungCheng University, Min-Hsiung, Chia-Yi, Taiwan 2 Department of Criminology, Law and Society, George Mason University, Fairfax, VA, USA 3 Faculty of Law, The Hebrew University, Mt. Scopus, Jerusalem, Israel 4 Temple University, Philadelphia, PA, USA

Synonyms Criminology of places; Hot spots

Overview Though both individuals and places have long been the foci of criminological research, the majority of criminological research focuses on individual criminal involvement (Reiss 1986). As a result, criminologists often assume that places play a relatively minor role in explaining crime compared to an individual’s criminal propensity. In fact, there has been a long history of studying of crime at places dating back to the nineteenth century. In the recent decades, there has been a gradually increasing interest in academia to study crime at places. Particularly, research has focused on the distribution of crime across geographic places as well as the explanatory factors of crime at places (Weisburd et al. 2012; Sampson et al. 1997). To provide an overview for the recent development, this entry reviews literature related to crime concentration, longitudinal crime rates at places, and empirical findings as well as challenges faced when studying the criminal career of places. Among studies of crime distributions at places, one of the most well-known and widely agreed-upon findings is the concentration of crime at a small number of places. The earliest work known to demonstrate this empirically was

Criminal Careers of Places

from French scholar Michel-Andre´ Guerry and the Venetian cartographer Adriano Balbi. They published three maps on the distribution of crime in France in the years 1825–1827. In 1833 the influential Essai sur la statistique morale de la France was published, in which Guerry examined whether poverty and density of population might explain higher crime rates (Guerry 1833). His map shows clear patterns of differential concentration patterns of crime. The rich northern departements were confronted with higher property crime rates than the poor departements in the south of France. He concluded that the level of poverty was not the direct cause of crime. Similarly, his data suggested that population density was not a cause of crime. In 1836 his friend Alexandre Parent-Duchaˆtelet published an empirical study containing maps on the distribution of prostitution from 1400 until 1830 in Paris (Parent-Duchaˆtelet 1836). Because of the official control of brothels by the Paris authorities, systematic data were available on prostitutes, especially from the years 1817 to 1827. Even information regarding the departements where they came from was collected. Not surprisingly, the center of the city had the highest number of prostitutes. He used neighborhoods, as defined by administrative boundaries, as the unit of analysis. Other scholars also studied these issues in the 1830s. Most notably, the work of Adolphe Quetelet (1831[1984]) examined crime rates across provinces and countries in France and found that places with higher property crime rates tended to be wealthier. He pointed out on his maps that crimes, just like wealth, are not distributed equally across places. Guerry and Quetelet could be considered as pioneers of crime and place research, and their works set the stage for later theorizing linking crime and place. Almost 100 years after the work of Guerry and Quetelet, in a classic study of juvenile delinquency, Shaw and McKay examined the residential locations of juvenile offenders and then pinned the addresses on a big map of the city of Chicago (Shaw and McKay 1942). The geographic distributions of those addresses


showed clear clustering patterns in their residential locations. Through their close observations and explorations of the city, Shaw and McKay concluded that not all areas were plagued with the same amount of crime. The so-called transition zone outside of the center of the city had the highest amount of crime in the whole city. Specifically, places with higher juvenile crime rates tended to be closer to the center of the city, disorganized, occupied by the poor, and also possessed other social illnesses. Moreover, Shaw and McKay (1942) argued that as long as the structural factors of a place, such as poverty, racial heterogeneity, urban decay, and population turnover rates, remain stable, crime rates will as a result also remain stable over time. Despite the preponderant evidence found on the stability of crime places, others have found that crime rates do change under certain circumstances. For example, the argument that crime remains stable in Chicago was challenged by studies done in later years (Bursik and Webb 1982). Specifically, Bursik and Webb reanalyzed crime rates in Chicago neighborhoods with an extended dataset compared to the original one used by Shaw and McKay and found that the stability of crime in Chicago was purely a historical artifact. Crime rates in Chicago started to fluctuate after WWII. The lack of empirical credibility, the focus on individuals as the center of study, and other theoretical flaws contributed to the decline of the traditional social disorganization theory formulated by Shaw and McKay. However, in the 1980s a group of young scholars led by Albert Reiss started to take a new interest in the study of crime places, rather than individuals. Editing an early volume in the Crime and Justice series, Reiss and Tonry (1986) sought to bring Communities and Crime to the forefront of criminological interests. Reiss sought to raise a new set of questions about crime that had been ignored in earlier decades: “Recent work on communities and crime has turned to an observation that Shaw and McKay neglected: not only do communities change their structures over time but so often do their crime rates. . .a recognition that communities as well as individuals have crime careers” (Reiss 1986, p. 19). This volume

809

C

and other work developed in this period drew upon the identification of neighborhoods and communities to expand insights about the development of crime (Bursik and Webb 1982; Clarke 1983). Sampson (1993) also pointed out the importance of linking time and place in the study of crime. Criminology had been dominated by methodological individualism; as such, the understanding of crime had been decontextualized. Putting crime back to its place has then become salient for a more holistic understanding of crime. Another major problem was the lack of a temporal perspective, as well as the availability of longitudinal data, in crime and place research. Without taking into account the actual sequence of actions, it is not possible to understand the causal mechanisms behind the phenomenon (Sampson 1993, p. 430). To address this issue, Sampson argues that communities have “careers” in crime, and longitudinal study is needed to disentangle changes in crime rates in communities (see also, Reiss 1986). The initiative to refocus criminological inquiries at places has led to two different but tightly linked tracks of study examining crime at places. The first line of research focuses on the hot spots phenomenon that examines stability of crime across geographic units to identify where the concentration of crime locates. The second line of research moves one step further and focuses on tracing the trajectories of crime trends of places over time.

Crime Concentration at Places Crime concentration has been found in many studies across different places at different geographic levels. Early research done by Guerry, Quetelet, and Shaw and McKay clearly demonstrated that crime was concentrated in a few places. In more recent years, the research findings on crime concentration and the empirical findings about hot spots of crime enhanced development of theoretical ideas about crime distribution at places. The introduction of new theoretical ideas, such as situational crime prevention

C

C

810

(Clarke 1983), routine activities theory (Cohen and Felson 1979), environmental criminology, and crime pattern theory (Brantingham and Brantingham 1984), invited people to explore the linkage between crime and places. At the same time, the popularity of personal computer and the affordable mapping software allow for timelier and more in-depth crime analysis to visualize the concentration patterns of crime for both researchers and practitioners. This theoretical development and technological advancement jointly promoted the research of crime at place. Subsequent research continued to identify high-crime neighborhoods within cities (Bursik and Webb 1982; Sampson and Groves 1989). The concentration of crime has been found at various geographic levels, not just the neighborhood level. The law of concentration also applies to specific types of “places.” Spelman (1995) examined calls for service at public places such as schools, housing projects, subway stations, parks, and playgrounds, over a 4-year time period in Boston. Analysis results showed that there were strong place-to-place variations on crime and disorder rates. Specifically, the worst 10 % of public places were responsible for about 30 % of crime. In addition to the concentration of crime, Spelman also found that long-run risks at places remained quite stable and explained a substantial amount of variation in citizen’s calls for service. Concentration patterns have also been observed in “high-risk” places such as bars/taverns. For example, Sherman et al. (1992) examined tavern crimes in Milwaukee and found that as few as 15 % of taverns in the city accounted for more than 50 % of the crimes that occurred in all taverns. The extent of concentration is even more salient at small geographic areas like addresses and places, than in larger aggregations of geography such as neighborhoods or census tracts. In one of the pioneering studies in this area, Lawrence Sherman and colleagues found that the majority of citizens’ calls for service were generated by about 3 % of addresses in Minneapolis, Minnesota. Fifteen years later, using street segments as the unit of analysis, Weisburd et al. (2004)


demonstrated similar concentration rates in Seattle, WA. That is, about 4–5 % of street segments accounted for about 50 % of all crime each year. This percentage remained quite stable during their 14-year observation period. Using 16 years of data and adding refinement to the definition of street segments, Weisburd et al. (2012) analyzed extended data from Seattle and reconfirmed the concentration of crime at the street-segment level. In Seattle, over 9 % of the segments did not evidence a single crime incident over this 16year period. Consistent with the Pareto Principle (a.k.a. the 80–20 rule), they found that 80 % of crime incidents were found on between 19 % and 23 % of segments across their study period, while 100 % of incidents fell on between 60 % and 66 % of segments in a given year. Another study by Weisburd and Amram (forthcoming) found that 5 % of the street segments in Tel Aviv were responsible for 50 % of the crime incidents, a statistic remarkably similar to the results from Seattle. These studies and others (Brantingham and Brantingham 1984; Clarke 1983) have established crime places as an important focus of criminological inquiry and practical crime prevention. The concentration of crime at “hot spots” is also observed for specific crime types. Braga et al. (2010) examined gun crimes in Boston and concluded that street segments and intersections account for less than 5 % of the total streets in the city but had over 75 % of gun violence incidents within them. Additionally, Weisburd and Mazerolle (2000) found that approximately 20 % of all disorder crimes and 14 % of crimes against persons were concentrated in just 56 drug-crime hot spots in Jersey City, New Jersey, an area that comprised only 4.4 % of street segments and intersections in the city. Similarly, Eck et al. (2000) found that the most active 10 % of places (in terms of crime) in the Bronx and Baltimore accounted for approximately 32 % of a combination of robberies, assaults, burglaries, grand larcenies, and auto thefts. Finally, examining juvenile arrests over a 14-year period, Weisburd et al. (2009) also found an extreme concentration rate – 3–5 % of street segments were responsible for all juvenile arrests during


any given year. In short, the findings from studies provide strong support for the idea of a “law of concentrations” for crime at places. To date, the available empirical evidence clearly shows that crime does not randomly distribute across places. Rather, research has found that a small number of places contribute a disproportionate share of crime. This finding echoes the results from the famous Philadelphia Cohort Study, in which Wolfgang et al. (1972) found that less than 6 % of individuals were responsible for more than 50 % of the arrests. The comparison result led Sherman to argue that future crime is “six times more predictable by the address of the occurrence than by the identity of the offender” (1995, pp. 36–37). He then asked “why aren’t we doing more about it? Why aren’t we thinking more about wheredunit, rather than just whodunit?” As such, many scholars proposed the need to study of crime places using criminal career perspective.

Stabilities and Changes of Crime at Places Following the tradition of Chicago school, scholars like Robert Bursik and his colleagues started to focus on how crime develops in communities. Specifically, some scholars began to apply the concepts drawn from the criminal career paradigm to examine whether places, just like individuals, also have “criminal careers” (Reiss 1986, p. 19; Sherman 1995). For example, Schuerman and Kobrin (1986) applied a dynamic model to explaining neighborhood crime characteristics over time. Recognizing the importance of understanding developmental crime trends, Sherman (1995) pointed to the relevance of the criminal career paradigm to the study of longitudinal crime rates at micro places. Other important concepts have also been borrowed from the criminal career paradigm to study the onset, continuity, specialization, and desistance of crime at places (Spelman 1995). Among the research of crime at places, stability of crime trends is the most common

811

C

finding across different geography. For example, Griffiths and Chavez’s (2004) study of Chicago neighborhoods also found support for the argument that places might follow very different trajectories with regard to homicide trends between 1980 and 1995. Specifically, they found that different neighborhoods exhibited different homicide trends – some had stable trajectories, some had increasing trends, while others had experienced high number of homicides. But the majority of the neighborhoods demonstrated stable homicide trends, while a smaller percentage of neighborhoods showed high increasing homicide rates during the same time. The stability of crime at places is also found at smaller geographic units. Weisburd and colleagues applied the criminal career concepts to examine stability and changes in crime rates of street segments in Seattle, WA (Weisburd et al. 2004, 2012). Specifically, Weisburd et al. (2004) found chronic hot spots and chronic cold spots of crime over 14 years of analysis. With additional 2 years of data, (Weisburd et al. (2012)) reconfirmed the previous findings using group-based trajectory analysis (see Nagin and Land 1993) to classify street segments into different trajectory groups based on their levels and directions of crime trends. They found 22 distinct groups of street segments following different developmental trends. Out of the trajectory groups, almost half of the street segments (49.6 %) were classified into the “crime-free” trajectory group as they had crime counts at or near zero for every year of the study. The second biggest trajectory group was trajectories that featured with “low stable” crime trends. Totally, 32 % of street segments with relatively low crime rates that remained stable over the 16-year study period were classified into this category. The next category was the “moderate stable” trajectory that had a moderate crime rate that was flat over time and contained 1.2 % of the street segments in Seattle. Importantly, the stability of crime is not limited to low-crime places. In fact, they also found that about 1 % of street segments, the “chronic high” trajectory, evidenced a stable high crime rate over time. Despite the

C

C

812

nationwide crime drop observed in the early 1990s, those places continued to experience high crime problems over that period of time. In sum, Weisburd et al. (2012) found that over 80 % of streets in Seattle had extremely stable crime patterns over time. Though the hot spots phenomenon has been supported by numerous studies, this was the first time research was able to empirically demonstrate the long-term stability of hot spots (and cold spots) at such a micro level of geography. The extent of stability of crime trends is astonishing over such a long period of time. Nonetheless, they also found places with fluctuating crime rates. Consistent with the crime drop phenomenon across the US cities through the 1990s and early 2000s, they found places that experienced declines in crime. The “low-rate decreasing” category accounted for 9.3 % of segments that had relatively low crime rates which evidenced declines in crime. The “high-rate decreasing” group contained 2.4 % of segments that had high rates which decreased over time. While most segments remained stable or experienced crime drops over this period, there were some segments which experienced crime spikes over the study period. In the city of Seattle, about 3.8 % of street segments featured “low increasing” crime rates, while another 0.9 % of segments experienced “high increasing” crime rates over the 16-year period. The findings from Weisburd et al. (2012) produced two major conclusions about criminal career of places. First, crime is not only concentrated at small places, these “hot spots” remain stable across time. In fact, the majority of places feature extremely stable crime patterns longitudinally. This means that crime prevention practitioners can focus their resources on relatively few crime hot spots and deal with a large proportion of the crime problem. Importantly, places are not “moving targets.” Place-based crime prevention provides a target that “stays in the same place.” Thus, crime prevention policies targeting high-risk places could be beneficial. Secondly, despite the stability and continuity of crime rates at most places, there is still a great deal of variability across street segments over


time in Seattle. Thus, it is important to understand factors that determine the changes in crime at those places.

What Factors Predict Crime Change? While scholars have provided a strong empirical basis for the assumption that crime is strongly clustered at crime hot spots and that there are important developmental trends of crime at place, existing research provides little insight into the factors that underlie these patterns. For example, we could identify only three prior published studies that specifically examined developmental patterns of crime at micro places over time. One study conducted by Spelman (1995) looked at specific places such as high schools, public housing projects, subway stations, and parks in Boston, using 3 years of official crime information. Taylor (1999) examined crime and fear of crime at 90 street blocks in Baltimore, Maryland, using a panel design with data collected in 1981 and 1994. These studies are limited only to a small number of locations and to a few specific points in time. To provide a comprehensive examination of risk and protective factors that could affect the developmental patterns of crime at places, Weisburd et al. (2012) collected 25 data sources which represented key concepts from two major place-based theoretical perspectives – social disorganization theory and opportunity theory. They examined relationships between the developmental crime patterns derived from group-based trajectory analysis and social disorganization and opportunity variables. The explanatory power between variables was compared and contrasted to identify important risk and protective factors in explaining specific development processes of crime at place. Multinomial logistic regression was used to compare the two most distinct crime places, the crime-free places versus the chronic-crime segments, and see what factors best distinguish between the two places. The following sections report on findings from a comprehensive study done by Weisburd et al. (2012).


Opportunity Measures and Crime Hot Spots The importance of opportunity theories for understanding crime at place has a long history in criminology (Brantingham and Brantingham 1984; Clarke 1983; Cohen and Felson 1979). A focus on crime naturally leads scholars to specific places or situations and the opportunities that situations and places provide for crime. Based on opportunity theory, the key elements of the crime triangle are motivated offenders, potential targets, and capable guardians (Cohen and Felson 1979). To test the validity of opportunity theory in predicting crime trends at places, Weisburd and colleagues applied the theoretical framework to the Seattle data and found a general support for the theory. Firstly, an increase in motivated offenders, as represented by an increase in high-risk juveniles on a street segment, has a strong and significant impact on the likelihood of the street segment being classified into the high-rate chronic hot spots pattern. Indeed, for every additional high-risk juvenile found on a street segment, the likelihood of being in this group as opposed to the crime-free group more than doubles. A positive change over time in the number of high-risk juveniles on a street segment also increases the likelihood of being a chronic-crime pattern street segment. Of course, the fact that high-risk juveniles on a street increases crime risk does not mean that these juveniles are the culprits. It may be their friends who commit crimes on the street in the course of their routine activities (possibly visiting the high-risk juveniles who live on a particular street segment). Secondly, with all else being equal, it would be expected that as the number of suitable targets increases, the number of crimes would also increase. Weisburd and colleagues used four measures to capture the number and attractiveness of targets on a street segment: employment, public facilities, residential population, and retail business sales. The results show that employment is the single most important variable in explaining the likelihood of a street falling in the high chronic-crime group as opposed to the crime-free group. For every additional employee

813

C

on a street segment, the odds of falling in this pattern increase by 8 %. The change in number of employees over time also has a strong and significant impact in the same direction, though of a somewhat smaller magnitude. The presence of public facilities within a quarter mile of a street segment also significantly increases the probability of being in the chronic-crime-trajectory pattern. Having a public facility such as a community center, park, library, middle or high school, or a hospital within a quarter mile increases the likelihood of being in the chroniccrime group (as contrasted with the crime-free group) by almost 25 %. Residential population also has a very strong impact on the likelihood of being a crime hot spot in the data. As would be expected by opportunity theory, the larger the residential population, the more likely a street segment is to be in the chronic-crime group. In contrast, the measure of retail sales was not statistically significant in predicting street segments to be in the chronic high group. It may be that the amount of retail sales does not reflect the number of patrons or visitors to the street or that after accounting for employment, public facilities, and residential population, most of the variability has been captured on street segments. Nonetheless, the presence of suitable targets as indicated by employees, facilities, and residential population is a key factor in explaining crime hot spots. The findings for guardianship, however, are less clear. Guardianship was measured by the existence of a police or fire station within a quarter of a mile of a street segment. Another common measure used in the situational crime prevention research is street lighting, which was measured by wattage. However, both measures of guardianship reached statistical significance, but not in the expected direction. The presence of a police or fire station increases the likelihood of a street segment being in the chronic-crime pattern. Additionally, more wattage of lighting on a street is associated with a higher likelihood of being a crime hot spot. Weisburd et al. argue that these inconsistent findings could be a result of the fact that the presence of police/fire station and high wattage lighting are both more likely to

C

C

814

be found in areas with higher population density. Moreover, it is also possible based on situational crime prevention theory that having a police station, or increased street lighting, in location with more crime problems were responses to higher crime rates. Thus, the findings could be spurious and confounded with population density or prior crime problems. In addition to the key elements of opportunity theory, it is also found that accessibility and urban form, like types of street, are significant variables in predicting that a street segment will be a crime hot spot. For example, the number of bus stops and arterial roads both increase the likelihood of being in the chronic-crime group as contrasted with the crime-free group. Social Disorganization and Crime Hot Spots As mentioned earlier in the entry, social disorganization theory is another major theoretical approach for the study of places. A series of variables reflecting structural components of social disorganization have been tested in the past, including socioeconomic status, population heterogeneity, mobility, and female-headed households. A neighborhood that is poor, more disadvantaged, heterogeneous in terms of racial or ethnic composition, and more urbanized tends to be more vulnerable to crime and other social problems. Using the Seattle data, Weisburd et al. (2012) tested social disorganization theory at the street-segment level to see whether the core theoretical concepts help predict chronic-crime places. They found that socioeconomic status represented by the value of residential property on a street segment and the amount of housing assistance is both strongly and significantly related to a street segment falling in the chroniccrime group. The other two structural dimensions reflecting social disorganization were mixed land use and racial heterogeneity. However, both of these measures were not statistically significant in differentiating between the chronic-crime and crime-free groups. In terms of the extent of urbanization, it was captured by the distance of the street away from the city center. It had an overall impact on predicting crime, in that areas


closest to the center of the city had the highest crime rates. These were the areas where new immigrants and poorer residents were concentrated, where social control was weak, and juvenile delinquency and crime problems would accordingly be concentrated. In addition to those structural characteristics, Weisburd et al. also included description of physical conditions in the model. Physical disorder such as litter, trash, graffiti, and abandoned cars are the most direct indicators of social disorganization (Shaw and McKay 1942[1969]), and its relationship to developmental trajectories of crime at street segments is very strong. A street segment is much more likely to be in the chronic-trajectory pattern as opposed to the crime-free pattern if it has higher reports of physical disorder incidents. Additionally, increases in physical disorder are also positively associated with higher likelihoods of being in the chronic-crime trajectory. Recent conceptualizations of social disorganization theory draw distinctions between the structural characteristics of areas and the mediating factors that connect the structural factors and outcome variables like crime (see Bursik and Grasmick 1993; Sampson et al. 1997; Sampson and Groves 1989). The strength of social ties among residents, also called social capital, determines the extent to which social control functions in areas. Different measures identified in prior studies have been used to conceptualize the intermediating mechanism including participation in local organizations (Sampson and Groves 1989), willingness (or perception of responsibility) to intervene in public affairs (Sampson et al. 1997), local friendship networks (Sampson and Groves 1989), mutual trust (Sampson et al. 1997), and unsupervised teens wandering on the street (Sampson and Groves 1989; Sampson et al. 1997). These factors are believed to condition the effects of structural disadvantage on local crime problems. The empirical validity of these measures has been tested in many studies. However, Weisburd and colleagues (2012) were first to test the idea at the micro geographic level over a long period of time. Based on the data, they found that truant juveniles on a street segment significantly


increased the likelihood of it being in a crime hot spot, while high level of collective efficacy (as measured by percentage of active voters) decreased the chance of a place being a crime hot spot. It seems, accordingly, that the more involved the residents are in public affairs, the less likely the streets are to have chronic-crime problems. Thus, social control could be built at street level to reduce the crime problem.

Conclusion Overall, research has found that crimes are clustered in a small number of places, regardless of the geographic unit of analysis examined. The “hot spots” phenomenon suggests that we could identify and deal with a large proportion of crime problems by focusing on just a very small number of places. In addition to the concentration of crime, recent studies have found that crime hot spots remain very stable over time. However, despite the preponderant evidence for crime stability at places, other scholars have also identified a smaller proportion of places with changing crime rates (Griffiths and Chavez 2004; Weisburd et al. 2004, 2012). These studies find that changes in social characteristics and contextual factors can also lead to change in crime rates at places. Specifically, social disorganization theory and opportunity theory are both relevant in predicting crime at places. For example, the number of residents, unsupervised teens wandering the streets, the more (or less) committed citizens are to public affairs, and having more potential crime targets in an area all determine the occurrence of crime at any given place. Another recent development of place-based criminology is the emphasis on studying micro geographic areas like addresses or street segments, rather than larger geographic units such as communities, neighborhoods, or census tracts. Weisburd et al. (2012), Braga and colleagues (2010), Taylor (1999), and other scholars have demonstrated a stronger crime concentration effect at micro places and have thus argued that focusing on micro places could be promising and more efficient for crime prevention efforts.

815

C

Crime at place is very predictable, and therefore, it is possible not only to understand why crime is concentrated at place but also to develop effective crime prevention strategies to ameliorate crime problems at places. Criminologists and crime prevention practitioners can identify key characteristics of places that are correlated with crime. At a policy level, it is important to focus on initiatives like “hot spots policing” that address specific streets within relatively small areas. If police become better at recognizing the “good streets” in the bad areas, they can take a more holistic approach to addressing crime problems.

Related Entries ▶ Criminology of Place ▶ History of Geographic Criminology Part I: Nineteenth Century ▶ History of Geographic Criminology Part II: Twentieth Century ▶ Hot Spots and Place-Based Policing ▶ Law of Crime Concentrations at Places ▶ Longitudinal Crime Trends at Places

Recommended Reading and References Braga AA, Papachristos AV, Hureau DM (2010) The concentration and stability of gun violence at micro places in Boston, 1980–2008. J Quant Criminol 26(1):33–53 Brantingham PL, Brantingham PJ (1984) Patterns in crime. Macmillan, New York Bursik RJ Jr, Grasmick HD (1993) Economic deprivation and neighborhood crime rates, 1960–1980. Law Soc Rev 27(2):263–284 Bursik RJ Jr, Webb J (1982) Community change and patterns of delinquency. Am J Sociol 88:24–42 Clarke RV (1983) Situational crime prevention: its theoretical basis and practical scope. Crime Justice 4:225–256 Cohen LE, Felson M (1979) Social change and crime rate trends: a routine activity approach. Am Sociol Rev 44:588–608 Eck J, Gersh J, Taylor C (2000) Finding crime hot spots through repeat address mapping. In: Goldsmith V, McGuire P, Mollenkopf J, Ross T (eds) Analyzing crime patterns: frontiers of practice. Sage, Thousand Oaks

C

C

816

Griffiths E, Chavez JM (2004) Communities, street guns and homicide trajectories in Chicago, 1980–1995: merging methods for examining homicide trends across space and time. Criminology 42(4):941–978 Guerry A-M (1833) Essai sur la statistique morale de la France. Crochard, Paris Nagin DS, Land KC (1993) Age, criminal careers, and population heterogeneity: specification and estimation of a nonparametric, mixed Poisson model. Criminology 31(3):327–362 Parent-Duchatelet AJB (1836) De la prostitution dans la ville de Paris. Dumont, Bruxelles Quetelet A (1831) Recherches sur le penchant au crime aux différents âges. Hayez, Brussels Quetelet A (1833) La precision des r ´ esultats croit comme la racine ćarree du nombre des observations. Annales d’Hygiene Publique ‘et de Medecine L ´ egale ´ 9:308–336 Reiss Jr. AJ (1986) Why are communities important in understanding crime? In: Tonry M, Morris N (eds), Communities and crime. Crime and justice: a review of research, vol 8. University of Chicago Press, Chicago, pp 1–33 Reiss AJ Jr, Tonry M (1986) Communities and crime. University of Chicago Press, Chicago Sampson RJ (1993) Linking time and place: dynamic contextualism and the future of criminological inquiry. J Res Crime Delinq 30:426–444 Sampson RJ, Groves WB (1989) Community structure and crime: testing social disorganization. Am J Sociol 94(4):774–802 Sampson RJ, Raudenbush SW, Earls F (1997) Neighborhoods and violent crime: a multilevel study of collective efficacy. Science 277:918–924 Schuerman L, Kobrin S (1986) Community careers in crime. In: Reiss AJ Jr, Michael T (eds) Communities and crime, vol 8, Special refereed issue of crime and justice. University of Chicago Press, Chicago, pp 271–311 Shaw CR, McKay HD (1942) Juvenile delinquency and urban areas. University of Chicago Press, Chicago Sherman LW (1995) Hot spots of crime and criminal careers of places. In: Eck J, Weisburd D (eds), Crime and place. Crime prevention studies, vol 4. Willow Tree Press, Monsey, pp 35–52 Sherman LW, Weisburd D (1995) General deterrent effects of police patrol in crime ‘hot spots’: a randomized study. Justice Q 12(4):625–648 Sherman LW, Gartin PR, Buerger ME (1989) Hot spots of predatory crime: routine activities and the criminology of place. Criminology 27(1):27–55 Sherman LW, Schmidt JD, Rogan DP, Gartin PR, Cohn EG, Collins DJ, Bacich AR (1992) The variable effects of arrest on criminal careers: the Milwaukee domestic violence experiment. J Crim Law Criminol 83(1):137–169 Spelman W (1995) Criminal careers of public places. In: Eck JE, Weisburd D (eds) Crime and place: crime prevention studies, vol 4. Willow Tree Press, Monsey

Criminal Cases Review Commission Taylor R (1999) Crime, grime, fear, and decline: a longitudinal look. Research in brief. National Institute of Justice, Washington, DC Weisburd D, Amram S (forthcoming) The law of concentrations of crime at place: the case of Tel Aviv-Jaffa. Police Pract Res Weisburd D, Mazerolle LG (2000) Crime and disorder in drug hot spots: implications for theory and Practice in policing. Police Q 3(3):331–349 Weisburd D, Bushway S, Lum C, Yang S (2004) Trajectories of crime at place: a longitudinal study of street segments in the city of Seattle. Criminology 42(2):283–321 Weisburd D, Morris N, Groff E (2009) Hot spots of juvenile crime: a longitudinal study of arrest incidents at street segments in Seattle, Washington. J Quant Criminol 25:443–467 Weisburd D, Groff ER, Yang S (2012) The criminology of place: developmental patterns and risk and preventive factors. Oxford University Press, New York Wolfgang ME, Figlio R, Sellin T (1972) Delinquency in a birth cohort. University of Chicago Press, Chicago Yang S (2010) Assessing the spatial-temporal relationship between disorder and violence. J Quant Criminol 26(1):139–163

Criminal Cases Review Commission ▶ Innocence Commissions

Criminal Defense Profession Alissa Pollitz Worden and Andrew L. B. Davies School of Criminal Justice, University at Albany, State University of New York, Albany, NY, USA

Overview A defendant’s right to legal counsel in a criminal prosecution is one of the most familiar of the Constitution’s due process protections. In 1963, the Supreme Court declared in Gideon v. Wainwright that There are few defendants charged with crime, few indeed, who fail to hire the best lawyers they can get to prepare and present their defenses. That government hires lawyers to prosecute and

Criminal Defense Profession defendants who have the money hire lawyers to defend are the strongest indications of the widespread belief that lawyers in criminal courts are necessities, not luxuries.

This entry reviews the criminal defense profession. The first section reviews the historical origins of counsel in the English legal system, tracing its evolution into colonial courts. The history of the American legal profession is marked by the rapid changes of a nation that adjusted, over the course of a few generations, to political independence, frontier expansion, abolition of slavery, immigration, and the Industrial Revolution. The tiny legal profession of the 1700s grew rapidly during the 1800s and by 1900 had evolved into a stratified guild system that remains in place today. The second section describes the contemporary structures and characteristics of the criminal defense bar and examines how lawyers define their work and how they think about their clients. The third section considers means of measuring the quality of legal representation, comparing legal, occupational, performance-based, and client perspectives on effective (and ineffective) counsel. A substantial majority of criminal defendants are represented by publicly paid counsel, through a patchwork of state and local policies and programs; the fourth section describes the variety of these systems, maps the boundaries between public and privately retained representation, and reviews the empirical literature that compares the performance of lawyers working in these diverse settings. Two concluding sections explore the political context of criminal defense work and current and emergent challenges in defining the scope and appropriate roles for criminal lawyers in contemporary courts.

The Criminal Defense Profession in the United States The Evolution of Criminal Defense Most of United States legal structure and doctrines were borrowed and adapted from the common law and practices of England. Medieval adjudication processes were governed by clergy

817

C

(and attended largely to violations of religious laws), and the inquisitorial nature of those proceedings reflected faith in divine intervention to guide the hand of justice. But following the consolidation of English villages under William the Conqueror, law enforcement and prosecution increasingly targeted those who violated the King’s laws, not those of the church. A more secular and compartmentalized judicial system took root, staffed by appointed justices and eventually moderated by lay juries. While the origins of today’s adversarial and jury-based processes are not completely documented, it appears that the shift to formal adversarial procedures by the fifteenth century established the distinctive roles of prosecution (representing in either a private or public capacity the interests of the complainant) and defense lawyer (speaking on behalf of the accused). However, lawyers’ access to the courts was more limited than their access to clients (Donahue 1964). English lawyers were formally recognized as practitioners at the Inns of Courts by the 1400s. But even when retained for a fee, their presence in the courtroom was limited and sometimes prohibited, in part due to fears that their participation would obscure, not reveal, the truth. In subsequent centuries, English custom wavered on the issue: common law tended to treat lawyers as important if not essential, but legislators and judges often did not. In 1789, the authors of the US Constitution rejected British ambivalence about the right to counsel when they drafted the 6th Amendment, establishing an unambiguous right to retain an attorney: In all criminal prosecutions, the accused shall enjoy the right to a speedy and public trial, by an impartial jury of the State and district wherein the crime shall have been committed, which district shall have been previously ascertained by law, and to be informed of the nature and cause of the accusation; to be confronted with the witnesses against him; to have compulsory process for obtaining witnesses in his favor, and to have the Assistance of Counsel for his defence.

Originally, the guarantee of assistance of counsel meant that counsel could, and should, be present at all stages in criminal proceedings in federal prosecutions. Later, this right was

C

C

818

extended to state prosecutions, though many of the colonies had such a guarantee in place since the early 1700s. In principle, this history suggests the triumph of earlier English common law reasoning that counsel was an essential element of due process. Until the twentieth century, however, this principle coexisted uneasily with economic facts about American lawyers and defendants: the former were private businessmen, engaged in a for-profit profession, and the latter were often too poor to buy their services. There is scant historical evidence about the fates of defendants who went to court without paid counsel during the 1800s and early 1900s; one can only speculate that the legal services they received were perfunctory at best. Contemporary ideals about the right to counsel (regardless of ability to pay) did not take hold in the United States until the twentieth century (see, e.g., Smith 1919). A cascade of landmark cases established, at least in principle, the obligation of states to provide counsel in capital cases (Powell v. Alabama, 1932), felonies (Gideon v. Wainwright, 1963), many juvenile proceedings (In re Gault, 1967), and most misdemeanors (Argersinger v. Hamlin, 1972) (for a very useful review, see Drinan 2009). When the Constitution was ratified, the American legal profession was small, and its practitioners were known more for scholarship than for their legal practice. This changed quickly, as populations grew, moved westward, and established settlements that were swiftly incorporated into new states. The need for legal advice and advocacy exploded during this era of rapid social change: lawyers were needed to settle property disputes, manage business deals, and of course defend miscreants and troublemakers. New villages and new statehouses also bred new opportunities for aspiring politicians, and quickly lawyering became not only a profitable profession for adventurous young men but also a path to public office, insofar as it provided courtroom opportunities for showcasing oratorical skills (Friedman 2007). The barriers to entering the profession were low in the nineteenth century. In most states, after studying law with a local judge, a young man had

Criminal Defense Profession

only to earn the approval and certification of that judge to join the bar. This apprenticeship system began to fade in the mid-1800s, when attorneys began to recognize the value of more restrictive standards for admission to the bar. Within a generation, enterprising educators established law schools, advertising for students and institutionalizing the notion that formal education was requisite for practice. This development took a momentous turn in 1876 when Harvard reconstructed its law school curriculum, under the leadership of Christopher Columbus Langdell, to focus less on the practical issues of practice and common law understandings in favor of the “science” of precedent, jurisprudence, and legal logic. This educational model – still known as the Langdell method – quickly caught on in the curricula of new law schools. With a few adjustments, this model guides the standard 3-year degree program of today. This method did rather little to prepare lawyers for the reality of courthouse negotiations and client management, and even today most law graduates endure grueling postgraduation study programs to actually pass state bar exams. But Langdell left an important legacy: lawyers were thereafter honored with the academic status of Juris Doctor. During the three decades following Langdell’s revolution, state bars began to establish educational requirements for admission to the bar. This move paralleled those of other professions (such as medicine) that sought to increase legitimacy and prestige by establishing higher standards for professional certification. By the 1920s, most states required passing a standardized exam to practice law (Friedman 2007). By the early 1900s, the legal profession experienced another significant change that had implications for criminal defense practice: the field began to specialize and stratify in alignment with the social and economic status of clients. The solo general practitioner was gradually supplanted by firms that marketed their services to corporations, local businesses, and upper-middle-class families; their work involved handling contracts, taxes, probate, and other work organized around financial acquisition and


planning. At the lower end of the status continuum stood clients whose legal problems tended to be immediate, unexpected, and problematic: family crises, injuries and accidents, and, of course, arrests and criminal charges (Heinz and Laumann 1994; Heinz et al. 2005). This trend has continued into the twenty-first century: research suggests that, with some important exceptions, lawyers who take on criminal clients occupy low rungs on the professional ladder, are less likely than others to practice in large firms, and are most likely to combine criminal work with other “personal plight” areas of law, such as personal injury. Characteristics of Contemporary Criminal Defense Lawyers Americans are ambivalent about lawyers generally and criminal defense lawyers in particular. Shakespeare (Henry VI, Part 2) and, reportedly, Jesus had few kind words for lawyers (Book of Luke 11:46). One of the most prominent scholars in the field of legal studies recently published an academic treatise on the history of lawyer jokes (Galanter 2005). But Americans also idealize lawyers. Abraham Lincoln’s homespun oratorical skills were crafted in village courthouses and bought him enough political capital to become a presidential contender. In the twentieth century, fictional defense lawyers such as Atticus Finch and Perry Mason were heroic figures who resolutely confronted skeptical jurors and riled communities to represent clients who, though innocent, faced long odds. Today, law school has become increasingly expensive and entrylevel positions more scarce, and yet law school applications and enrollment have inclined steadily over the past 50 years. The most recent census data on employment reports that the nation has over one million practicing lawyers (excluding those who may hold law degrees but do not represent clients; Bureau of Labor Statistics 2011), a number that has doubled in three decades. But a contemporary census of the legal profession would provide only a partial picture of the criminal defense bar. This is in part because specializations are not uniformly recorded, in part because the mix of

819

C

cases that lawyers accept varies greatly, and in part because there are few sharp lines between public defenders – full-time salaried attorneys representing the poor on behalf of the state – and attorneys who represent criminal defendants under less-structured programs, in combinations of publicly paid and privately retained arrangements. Who are these lawyers, where do they work, and how do they view their clients? Private criminal defense work remains concentrated in small local firms. Over the past 50 years, the expansion of indigents’ right to counsel has meant that a large share of criminal defense cases that once went unrepresented or nominally represented by private practitioners is now in the hands of publicly paid attorneys. National data on practice settings indicate that lawyers who self-identify as legal aid or public defender attorneys constitute only 1 % of practicing attorneys, a small subgroup of the 6 % who begin their careers practicing any sort of public interest law (National Association for Legal Professionals 2006). Between 1980 and 2000, the percentage of all lawyers who worked in small private firms (of fewer than ten practitioners), where one might expect to find most private criminal defense work, has remained around 50 % (American Bar Association 2010). The profession remains predominantly white: today only 5 % of practicing lawyers are African American, and 4 % are Asian; independent of racial identity, 3 % identify as Latino. In 2010, women comprised 32 % of the licensed bar (Bureau of Labor Statistics 2011), a percentage that has risen steadily since the 1970s as law school cohorts have become nearly 50 % female. Some have argued that a largely white, middleclass profession lacks the perspective to effectively represent defendants who are disproportionately low-income people of color, and a small body of empirical research suggests that lawyers in general, but particularly white lawyers, have lower expectations about the outcomes of cases of hypothetical black clients (see, e.g., Eisenberg and Johnson 2003). Research also suggests that women and men differ slightly in their perspectives on the

C

C

820

criminal defense work: men more fully embrace a strong due process orientation, and women are more likely to perceive moral dilemmas working with criminal clients (Siemsen 2004). But generally, interviews with criminal lawyers reveal higher levels of commitment to due process and opposition to punitive sentencing policies (Worden 1998; Ehrhard 2008). Contradicting stereotypes of co-opted public defenders, empirical studies report zealous advocacy orientations in this group (McIntyre 1987; Emmelman 1996). Defining Effective Counsel and Performance The Supreme Court of the United States has held that the Sixth Amendment requires not only that counsel be present but also that it be “effective.” In the words of Justice O’Connor in Strickland v. Washington (1984), it is not enough that “a person who happens to be a lawyer is present at trial alongside the accused.” This question has been addressed not only by the courts but also by professional associations who have developed standards and guidelines on what it means to provide counsel effectively. Empirical work, meanwhile, has examined some of the assumptions which underlie these standards – in particular that effective representation will result in more favorable outcomes for the accused. Constitutional/Appellate Court Definitions

Writing for the majority in Strickland v. Washington, Justice O’Connor specified a twopronged test to determine whether the assistance of counsel the defendant received had been ineffective. First, the defendant must show that counsel’s performance was deficient. This requires showing that counsel made errors so serious that counsel was not functioning as the “counsel” guaranteed the defendant by the Sixth Amendment. Second, the defendant must show that the deficient performance prejudiced the defense. This requires showing that counsel’s errors were so serious as to deprive the defendant of a fair trial, a trial whose result is reliable.

The holding in Strickland was intended to set a high standard for claims of ineffectiveness. It stifled defendant claims by stipulating that courts


should generally defer to the judgment of trial attorneys’ strategic reasons for presenting (or not) certain defenses. In Washington’s case, the trial attorney had chosen not to enter mitigating evidence because of the danger that other prejudicial facts would be brought to light, and the Court held that this constituted a “strategic” exception. Moreover, the facts omitted were, in the opinion of the Court, insufficiently likely to have changed the verdict in the case. Accordingly, the appellant was held not to have met either prong of the test. The court returned to the question of when a decision might reasonably be considered “strategic” in Wiggins v. Smith (2003). The defense attorney was held to have erred by ceasing his search for mitigation evidence on the basis of a “strategic” focus on proving the client’s innocence. The court cited the capital defender standards promulgated by the American Bar Association, assessing the attorney’s conduct against the standard of “prevailing professional norms,” and ultimately concluded that his failure to investigate represented “inattention” rather than “strategy.” Strategic decisions to cease to pursue mitigation avenues must not precede, but should only follow full investigation. Professional Definitions

Professional groups such as the American Bar Association and many state bars have compiled lists of standards and performance guidelines that define effective representation. The National Study Commission on Defense Services, commissioned by the US Department of Justice, composed its Guidelines for Legal Defense Systems in the United States in 1978. These have since been followed by a range of other sets of standards and guidelines, some of which were compiled in 2000 as the Compendium of Standards for Indigent Defense Systems. Both the ABA and the NLADA now maintain online libraries of professional standards devoted not only to defender practice and system design but also to specific practice areas (such as juvenile or capital representation) and specific practice elements (such as the need to advise clients of collateral consequences of conviction such as


employment restrictions and placement on offender registries, and the need to sustain a healthy attorney-client relationship). Professional standards stipulate a variety of practice characteristics ranging from attorney qualifications, client interactions, case preparation methods, and contact with other court professionals. In addition, standards have been developed concerning the organization of defender systems themselves. These have focused less upon the work defenders do and more upon the systemic burdens under which they practice: standards of supervision and training, resources available for the investigation and preparation of cases, and, most prominently of all, caseloads. Performance-Based Definitions

Empirical studies of the quality of indigent defense tend to focus on the observable characteristics of cases which are understood to represent quality in defender services, as defined by the legal and professional standards aforementioned. Examples include the extent of motion practice, early intervention by counsel, the time spent by attorneys preparing cases, investigative work, client-attorney interactions (their number, duration, and quality), and the provision of ancillary services, such as assistance with civil legal matters or access to social work services (e.g., Flemming 1986). Researchers have also gone beyond the question of whether attorneys adhere to the procedures stipulated by professional standards and have attempted to find out if these better practices produce better case outcomes for clients. These outcomes are related directly to defendants’ fates: convictions, sentences, bail decisions, time to adjudication, days spent incarcerated, diversion into alternative programs, and success following reentry. Research findings on this question are inconsistent, likely reflecting the problems of sampling and measurement in court data. Nationally representative samples of defendants are hard to come by, and studies of indigent defense case outcomes are typically conducted on samples compromised by selection bias (such as incarcerated

821

C

populations) or limited to the operations of specific parts of the courts system (such as federal courts) or types of cases (such as homicides). With these caveats in mind, researchers have repeatedly investigated two questions which tap common assumptions about publicly funded defenders: first, whether defendants retaining attorneys privately do better in court than those represented by publicly funded lawyers and second, whether institutional providers employing salaried attorneys (such as public defender offices) obtain better outcomes than those that make ad hoc assignments of private attorneys paid in some other way. In contradiction to the stereotype that publicly funded defenders provide inferior services to privately retained ones, most studies have found little or no support for this thesis after differences between clients and their cases are controlled (Hartley et al. 2010). Publicly funded attorneys may obtain improved outcomes for clients in certain contexts, particularly when considering intermediate case outcomes such as pretrial release or charge bargaining in isolation from sentencing itself (Hartley et al. 2010). Hanson et al. (1992) found a lower likelihood of incarceration for defendants retaining private attorneys, but no difference in other outcomes. Nevertheless, the presumption that private attorneys, who may spend significantly more time on cases than publicly funded ones, obtain better outcomes for their clients has little basis in empirical evidence. The absence of findings of differences in the quality and effectiveness of defense attorneys has prompted speculation regarding other possible differences in the working environment or methods of publicly funded counsel in comparison with privately retained attorneys. It may be the case that publicly funded attorneys, because of their frequent appearance in court, develop working relationships with other court actors which allow them to work to their client’s advantage through informal means where traditional adversarial ones fail them. Flemming et al. (1992) observed the ways in which defenders necessarily formed and maintained working relationships with judges and even prosecutors. If this

C

C

822

is the case, “repeat players” in the private bar would demonstrate similar collegial behavior and positive results for their clients. Within public defense systems, comparisons of public defender offices with systems that rely on the private bar have also found few differences in case outcomes (Houlden and Balkin 1985). Some exceptions exist for specific demographic groups (Hartley et al. 2010), and several recent, well-designed analyses have established significant evidence of the detrimental effects for defendants of representation by assigned counsel in comparison to institutionalized providers in the context of the federal courts (Iyengar 2005) and among felony defendants in urban counties (Cohen 2011). The sources of such differences remain a point of controversy, though aside from the infrastructural advantages that public defender offices may provide, authors such as Iyengar (2005) also point to differences in the attorneys that each system attracts, and differences in the ways the operation of market forces change the composition of each. Though incipient, this research suggests that different systems may attract different defenders, who in turn obtain different outcomes for their clients. Clients’ Perceptions

Beyond examining case outcomes, other scholars, beginning with Casper (1971), began asking defendants themselves what they thought of their lawyers. These studies revealed that client perceptions are molded not only by the favorability of sentencing decisions but also by the interactions clients had with their attorneys. Predictably, defendants’ skepticism – their common assumption that appointed lawyers are at best incompetent and at worst indifferent state lackeys – presents particular obstacles in building the trusting relationship presumed to be required for effective defense ( Flemming 1986). Notwithstanding the well-intentioned nature of the advice defenders frequently offer their clients, the failure to couch that advice in a reassuring client-attorney relationship dulls the appreciation that defendants might otherwise be expected to feel for their attorneys.


Reassuring defendants that their attorneys are motivated, interested, and honorable is more than a matter of wanting to make clients happy, however. Researchers have also adduced evidence that the procedural fairness of court practices has other significant benefits. Defendants who are satisfied that their verdicts were arrived at via fair processes are more likely to accept their sentences, feel more confidence in criminal justice generally, and may be more likely to comply with its mandates (Casper et al. 1988). Attention to client satisfaction, therefore, is not merely a matter of pleasing consumers. It has been sustained by concern with assuring that defendants themselves – many of whom will eventually reenter society – regard the criminal justice system positively and are able to reintegrate successfully into society. Attorney-client interactions and their consequences for defendants’ evaluations have come under considerable scrutiny, and research suggests that attorney and client expectations about what constitutes effective service are at odds. Quite apart from gaining clients’ confidence, an obstacle which state-appointed attorneys find difficult (Flemming 1986; Casper 1971), attorneys vary in the style and degree of attention paid to the client and their willingness to involve them in decision-making. More controversially, some scholarship has also sought to demonstrate the ways in which attorney decisions may be guided by factors other than client interests, such as the fee structure by which they are reimbursed (Tata 2007). Accordingly, the extent to which attorneys are able to foster positive client experiences and the possibility that structural factors such as the source of their reimbursement act as potential impediments to effective representation remain issues of concern. Criminal Defense as Public(ly Funded) Function The immediate result of the 1963 Gideon decision was to create a massive demand for publicly provided representation. At times this demand has overwhelmed some states and communities. The retroactive nature of the decision meant that


many inmates were retried or, when this proved impractical, simply released (Eichman 1966). Quickly, however, a nationwide patchwork of defender providers was created. Though data on the immediate implementation of the Gideon mandate are scarce, the earliest national survey revealed that by 1972, 16 states had developed statewide programs with the remainder delegating the responsibility for defense provision to counties and localities (Benner and Lynch 1973). State financial support for these systems also varied, and in 1975, states funded just 41.5 % of defense costs, leaving the remainder for localities to supplement (Sourcebook of Criminal Justice Statistics 1977). In the decades since the 1970s, the general trend has been toward the centralization of both organizational and funding responsibility for defender services within states. Recent data show that these trajectories are far from complete, however, with a total of 22 states using statewide systems in 2007, funding an average of 66 % of system costs in 2008 (Langton and Farole 2010). National estimates on uptake of publicly funded defender services are scarce. Eligibility criteria themselves were left to the discretion of providers, with Gideon and subsequent decisions stipulating only that the defendant be “too poor to hire a lawyer.” As a result, estimates of indigence among the population of defendants are bedeviled by a lack of standardization, as well as a more fundamental lack of systematically collected data. Moreover, states vary in their definitions of the right to counsel: in some, the right has exceeded that mandated by the Supreme Court through state statutory revisions and case law. Most estimates suggest, however, that the proportion of defendants using free legal services averages in the range from 65 % to over 80 %, depending on the type of case. The freedom afforded states in determining the manner of delivery for defender services has resulted in a variety of structural approaches to service provision both among and within states. Historically, public defense was typically provided through informal ad hoc assignment

823

C

of counsel, frequently – though not always – with little attention to attorney qualifications or compensation. These traditional “assigned counsel” systems made no meaningful distinction between private and publicly funded lawyers, since most defenders mingled publicly funded assignments with cases on private retainer. Indeed, the term “public defender” did not always connote payment from the public purse, but also applied to lawyers working pro bono to “defend” the “public.” The institutionalization of states’ obligation to compensate counsel led to the necessity of systematized billing processes for attorneys as well as new bureaucratic structures with new responsibilities: assigning attorneys to cases, assuring they provided representation, and paying them for their services. This formalization brought the opportunity for greater oversight and systematization in the provision of defender services through assigned lawyers, though the extent to which these systems have expanded beyond simple billing functions to oversee quality in service provision is highly variable. Among these evolving assigned counsel programs, a now-familiar approach gained popularity: the Public defender offices are permanent offices staffed by salaried attorneys who provide legal representation in all cases assigned to them. Public defender offices were not uncommon in large cities by the early part of the twentieth century. Public defender offices seem to promise economies of scale, which probably explains why they are concentrated in urbanized areas (Worden and Worden 1989). Staff in such offices may be full or part-time and may or may not be permitted to retain private practices in addition to their work as state or local employees. Concern with efficiency has led to yet a third approach: contracts with attorneys or law firms to provide representation. The terms of such contracts may vary, but the approach is distinctive from the archetypal assigned counsel or public defender approaches because attorneys contract with local governments (typically counties) to handle all or a percentage of the jurisdiction’s caseload, during a set period of time. How they

C

C

824

manage that caseload is up to them; how local governments choose a bidder is up to those governments. Contracting offers mixed benefits. A bidding process may reduce costs for localities, though the consequence of that process may be that quality is compromised, particularly if the contract offers a flat fee for services regardless of case complexities. In State vs. Smith (1984), a low-bid contract for the provision of defense services in Mojave County, Arizona, was struck down on the basis that selecting a provider on a pure cost basis was insufficient to guarantee effective assistance of counsel. Moreover, as Worden (1993) showed, many counties are too small to generate an effective bidding process for contracts, so contracting can actually increase rather than decrease costs. Nevertheless, contracting has become increasingly popular, resulting in considerable interest in the processes of bidding and contract composition which avert the harm to quality that might result. A single provider of defender services is generally insufficient to provide all legal representation in a jurisdiction, however, because cases with multiple defendants present conflicts of interest. Except in the case of traditional assigned counsel systems (where each defendant can be assigned to a different lawyer from the pool), institutional defender systems need backup systems. Rules for the determination of conflicts differ, but in general a provider cannot provide representation to a defendant if it has previously represented either a codefendant or a witness in the case. Backup systems generally come either in the form of assigned counsel plans, additional institutionalized “conflict defender” offices, or firms or attorneys contracted for the purpose. The Political Context of Criminal Defense Crime and justice policy is, by definition, the product of political processes, and the organized bar, legal advocacy groups, and individual attorneys have been at the center of many reform debates. The American Bar Association (and many state and local associations as well) has advocated for changes in legal processes (including, but not limited to, higher standards for criminal defense of the poor). In many states,


public and nonprofit organizations provide assistance to defense attorneys, notably in capital cases and appeals. Organizations such as American Civil Liberties Union, The Innocence Project, the National Legal Aid & Defender Association, and the Death Penalty Information Center draw heavily on the contributions and experience of lawyers as well as legal scholars. Many such groups provide or support provision of legal representation in individual cases and also actively lobby for policy changes in substantive and procedural law. Most landmark changes in criminal procedural law have resulted not from lobbying and legislation, however, but from litigation. The due process revolution was waged by criminal defense lawyers representing convicted clients in high-level appeals, and these lawyers were motivated by the opportunity to change policy, typically in the direction of greater due process protections or equality of treatment for defendants. Scholars and professionals refer to this sort of work as “cause lawyering” or “impact litigation.” Clearly some of these initiatives have made lasting changes in operational definitions of justice and fairness, and indeed, some experts argue that litigation is a more promising strategy for effecting legal reforms than efforts in legislatures. However, others caution that while legal rulings can dramatically redefine defendants’ rights, courts have very limited enforcement powers, and litigants (and their lawyers) have little incentive to ensure that sweeping rulings are, in fact, transformed into routine practices that change how police, prosecutors, judges, and juries do their work (Scheingold 1974; but see also Scheingold 2004). More generally, the last four decades witnessed a sharp increase in adjudication and sentencing in the aftermath of a real increase in crime rates during the 1970s and 1980s and a much more prolonged increase in punishment into the current century. Scholarship on the causes of the latter (and on variability across states and communities as well) suggests not only increasing (and sometimes overheated) awareness of crime as a policy problem but also


social, economic, and cultural differences in how political entities responded to crime and justice issues (Davies and Worden 2009). The work of criminal defense lawyers is shaped by the legal contexts in which they practice. But while researchers have investigated the consequences of changing sentencing codes and caseloads for law enforcement and corrections, they know little about how prosecutors and defense lawyers have adapted to these shifts. Current Issues and Controversies Today’s lawyers face challenges and controversies that may redefine the scope and the standards for their work with criminal defendants. Some of these are structural: at what point is representation required? Others entail revisiting the standards for effectiveness (or ineffectiveness) of counsel. New (or expanding) clienteles, defined by charges and by personal characteristics, call for new expertise and skills. In a very different vein, some argue that recoding some minor crimes into violations or civil offenses might dramatically reduce the need for lawyers, privately paid or publicly provided. Finally, defense lawyers confront new ethical and professional questions as alternative courts recast their traditional identity as adversarial champion. The examples here illustrate some of these challenges. When Does the Right to Counsel Begin and End?

The question of when the 6th amendment right to counsel applies to cases is answered differently, in law and in practice, across jurisdictions. Counsel at trial is unquestioned, but there are legal and practical disputes about whether a defendant needs, and is entitled to, counsel in juvenile court, at arraignment and bail hearings, and in appeals (Worden et al. 2010–2011). Attorney General Eric Holder recently called for close scrutiny of court practices and customs that oblige defendants to make court appearances, possibly plead, and have bail set without the benefit of counsel. Policies on post-conviction review of cases that hinge on DNA evidence, and cases that involve sex offenders facing civil commitment proceedings, vary across states;

825

C

as such petitions become more numerous, expectations for legal assistance may increase. Revisiting “Effective Counsel” in the Context of Plea Negotiations

As noted above, “effective” counsel has historically been defined, simply, as the absence of egregiously “ineffective” counsel, and the bar for a ruling of ineffective counsel has been placed rather high. A pair of recent Supreme Court rulings signals a shift in this thinking, however. In Laffler v. Cooper and Missouri v. Frye, both decided by 5–4 votes, the court established that standards of effective representation should include judgments about the quality of lawyers’ advice about rejecting plea offers in cases that ended in trial convictions. New Clients with New Problems

Today’s defense lawyers face a more complex tableau of indictments, sanctions, and evidentiary questions than did their predecessors. While the last generation’s draconian drug laws are slowly being retracted, and street crime rates have fallen dramatically, today’s clients present a new mix of social, cultural, and legal problems. Law enforcement’s focus on illegal immigration adds a layer of complexity and anxiety to some defendants’ cases – an aspect of representation given special urgency by the Court’s recent Padilla v. Kentucky (2010) decision requiring that immigrant defendants be counseled on the immigration consequences of a guilty plea. The last decade’s construction of new sex offender laws carrying collateral sentencing consequences – registries, residency restrictions, and civil commitment – raise the stakes for lawyers negotiating for their clients’ futures. Lawyers (and prosecutors) face new questions about technology, evidence, and admissibility as it becomes easier to inadvertently leave an electronic trail of communications, travel, or financial transactions (and also, of course, easier for authorities to follow those trails; see, e.g., United States v. Jones (2012)). Reformed Legal Codes and Fewer Lawyers?

The cost of providing lawyers in minor cases is causing some states to revisit the question of

C

C

826

whether they are necessary at all. In reforms referred to collectively as “reclassification” or “decriminalization,” minor offenses such as sleeping in a public park, low-level harassment, or, most conspicuously, marijuana use have been reduced to the level of civil infractions without the possibility of jail time. (Absent a threat of incarceration, the Supreme Court does not compel states to provide legal counsel.) Reclassification is often accompanied by elimination of other legal consequences of conviction – for example, by providing assurances that the case will not create or add to one’s criminal record. Some believe that reclassification saves public tax dollars even as it reverses the criminal stigmatization of minor offenders. At its grandest, the move toward reclassification is seen as an attempt to begin to undo some of the overbuilt, “industrialized” aspects of the criminal justice system that sprang up in conjunction with the rise in crime and incarceration since the 1960s. This initiative also taps into the sentiment that lawyers, and indeed courts, may be unnecessary for the resolution of many minor infractions and disputes. New Models of Adjudication and Legal Representation

Specialized courts (sometimes called therapeutic courts) are designed as rehabilitative alternatives to the adversarial adjudicative process. Since the late 1980s, drug courts, domestic violence courts, and mental health courts have proliferated. While they vary in many ways, they have in common a commitment to addressing defendants’ social, psychological, medical, and economic problems as a route away from recidivism. A central tenet of most of these courts is the necessity of a collaborative effort on the part of defendants, judges, social services, prosecutors, and defense lawyers. These experiments offer the appealing promise of destigmatization, reform, and resource recovery (since rehabilitated offenders will commit fewer future crimes and require fewer criminal justice resources to process and punish). However, they also create new professional and ethical questions for defense lawyers. A spirited defense of a client’s innocence or an


attack on a prosecutor’s evidence is hardly the fitting prelude to this sort of diversion and treatment; indeed, a guilty plea and, typically, some evidence of remorse are prerequisites for these alternatives. These specialized courts have typically identified particular psychological and medical problems (addiction, mental illness) or specific offenses (domestic violence, prostitution). The newest experiment in this genre is the veterans’ court, which defines the experience of combat as a special condition meriting specialized treatment.

Conclusions From its beginnings as a modest and tradesmanlike craft, defense lawyering has become much more than only a profession. It has come to stand for both the worst and the best of the law: the morally questionable imperative to provide criminals with expert assistance in avoiding conviction and punishment, on the one hand; and, on the other, some of the Supreme Court’s finest prose on the need to equalize access to justice for rich and poor, state and citizen, alike. This ambivalence of idealism mixed with the appearance of disreputability continues to characterize the defense bar to this day. In order to pursue a case to the highest court and secure a significant policy victory, lawyers must be prepared to subject a single client to years of litigation – instead of simply encouraging them to plead guilty and dispose of their matter efficiently. Advocates argue that defense reform will not be complete until every criminal defendant is afforded the full criminal trial to which they are nominally entitled. But of course, lawyers stand to benefit a great deal from expansion of the right to counsel and guarantees of more generous compensation. What results, therefore, is a continuation of old arguments over how much defense counsel is really necessary to justice: whether it serves clients or causes and whether it protects the guilty or the innocent. These arguments conjure up incongruous images: on the one hand, courageous Atticus Finch in a hostile courtroom, and on the other, florid fast-talking lawyers


promising fast relief for drunk driving arrests on late-night advertisements. One reason that these old arguments have been so hard to resolve is that defense lawyering, unlike many other elements of the criminal justice system, has not been subjected to significant empirical scrutiny over its central claims. Police organizations, for example, have sought sophisticated approaches to proving their effectiveness; offender treatment programs are regularly subjected to performance evaluations. But the defense bar has engaged in far less selfexamination, and outsiders get few opportunities to evaluate lawyers’ work. Furthermore, the realities of criminal practice are more complex than most people, including lawyers themselves, may acknowledge. Lawyers’ place in adversarial adjudication is straightforward enough: zealous defense of clients’ interests, which usually means strategic pursuit of acquittal or minimal punishment. But lawyers usually operate in the more complex context of plea negotiations and in a court system that has become increasingly punitive, even as (especially recently) alternative adjudication systems, diversion programs, and treatment options have proliferated. Serving clients’ “best interests” has become more complicated. Defense lawyers have a unique interest in protecting their clients from the cycle of cumulative disadvantage – the hardships that start with the stigma of arrest, accrue with conviction and sanctions, and are cemented when reintegration and rehabilitation fail. Simple structural reforms – such as organizational changes which assure counsel reaches a defendant early or reclassification measures which assure defendants are not detained needlessly – can make a difference in whether a defendant is able to go home, keep a job, maintain custody of children, and even reintegrate into society in the long term. Broader awareness of these changes (and evaluations of their effectiveness) will not resolve historical, cultural, and economic dissension over defense lawyers’ roles. It might, however, emancipate lawyers from simplified ideological prescriptions about their work and lead to a more nuanced and realistic public understanding of their work.

827

C

Related Entries ▶ Comparative Legal Cultures ▶ History of the Death Penalty ▶ Innocence Commissions ▶ Juries, Lay Judges, and Trials ▶ Measuring Wrongful Convictions ▶ Mental Health Courts ▶ Plea Bargaining ▶ Problem-Solving Courts ▶ Public Defenders ▶ Scientific Evidence in Criminal Adjudication ▶ Scientific Evidence in Criminal Prosecutions ▶ Sentencing Research ▶ Sex Offenders and Criminal Policy ▶ Therapeutic Jurisprudence

Recommended Reading and References Benner L, Lynch B (1973) The other face of justice. National Legal Aid and Defenders Association, Washington, DC Bureau of Labor Statistics (2011) Occupational employment statistics. http://www.bls.gov/cps/cpsaat11.pdf Casper J (1971) Did you have a lawyer when you went to court? No, I had a public defender. Yale Rev Law Soc Action 1:4–9 Casper J, Tyler T, Fisher B (1988) Procedural justice in felony cases. Law Soc Rev 22:483–507 Cohen T (2011) Who’s better at defending criminals? Does type of defense attorney matter in terms of producing favorable case outcomes. US Bureau of Justice Statistics, Washington, DC, Working Paper Series. http://freealexander.org/wp-content/uploads/2011/12/ SSRN-id1876474.pdf Davies A, Worden A (2009) State politics and the right to counsel: a comparative analysis. Law Soc Rev 43:187–220 Donahue C Jr (1964) An historical argument for right to counsel during police interrogation. Yale Law J 73:1000–1057 Drinan C (2009) The third generation of indigent defense litigation. N Y Univ Rev Law Soc Change 33:33–427 Ehrhard S (2008) Plea bargaining and the death penalty: an exploratory study. Just Sys J 29:313 Eichman C (1966) Impact of the Gideon decision upon crime and sentencing in Florida. Florida Division of Corrections Research and Statistics Section, Tallahassee Eisenberg T, Johnson S (2003) Implicit racial attitudes of death penalty lawyers. DePaul Law Rev 53:1539–1556

C

C

828

Emmelman D (1996) Trial by plea bargain. Law Soc Rev 30:335 Flemming R (1988) Client games: defense attorney perspectives on their relations with clients. Am Bar Assoc Res J 1988:253–277 Flemming R, Nardulli P, Eisenstein J (1992) The craft of justice: politics and work in criminal court communities. University of Pennsylvania Press, Philadelphia Friedman L (2007) History of American law, 3rd edn. Simon & Schuster, New York Galanter M (2005) Lowering the bar: lawyer jokes and legal culture. University of Wisconsin Press, Madison Hanson R, Ostrom B, Hewitt W, Lomvardias C (1992) Indigent defenders get the job done and done well. National Center for State Courts, Williamsburg Hartley R, Miller H, Spohn C (2010) Did you get what you pay for? Type of counsel and its effect on criminal court outcomes. J Crim Justice 38:1063–1070 Heinz J, Laumann E (1994) Chicago lawyers: the social structure of the bar, revised. Russell Sage Foundation and Chicago: American Bar Foundation, New York Heinz J, Nelson R, Sandefur R, Laumann E (2005) Urban lawyers: the new social structure of the bar. University of Chicago Press, Chicago Houlden P, Balkin S (1985) Cost and quality of indigent defense: ad hoc vs coordinated assignment of the private bar within a mixed system. Justice Sys J 10:159–172 Iyengar R (2005) An analysis of the performance of Federal Indigent Defense Counsel. Princeton University, working paper Langton L, Farole D (2010) State public defender programs, 2007 NCJ 228229. Bureau of Justice Statistics, Washington, DC McIntyre L (1987) The public defender: the practice of law in the shadows of repute. University of Chicago Press, Chicago National Association for Legal Professionals (2006) Employment patterns – 1982–2004. NALP Bulletin, June 2006 Siemsen C (2004) Emotional trials: moral dilemmas of women criminal defense attorneys. Northeastern University Press, Boston Smith RH (1919) Justice and the poor. Carnegie Foundation, New York, Bulletin # 13 Sourcebook of Criminal Justice Statistics (1977) Washington DC: Bureau of Justice Statistics Spangenberg Group website: http://www. spangenberggroup.com/ Tata C (2007) In the interests of clients or commerce? Legal aid, supply, demand, and ‘ethical indeterminacy’ in criminal defense work. J Law Soc 34:489–519 Worden A (1993) Counsel for the poor: an evaluation of contracting for indigent criminal defense. Justice Q 10:613–637 Worden A (1998) Representing the accused: professional values and professional choices of small-town lawyers. Crim Just Rev 23:1–28

Criminal Interviewing Worden A, Worden R (1989) Local politics and the provision of indigent defense counsel. Law Policy 11:410–424 Worden A, Davies A, Brown E (2010–2011) A patchwork of policies: Justice, due process, and public defense across American states. Albany Law Rev 74:1423–1463

Criminal Interviewing ▶ Interview and Interrogation Methods Effects on Confession Accuracy

Criminal Investigation ▶ Investigative Psychology

Criminal Investigative Analysis Roger L. Depue FBI Retired

Synonyms Crime profiling; Crime scene analysis; Crime scene interpretation; Criminal personality profiling; Profiling; Psychological profiling

Overview This entry will discuss Criminal Investigative Analysis (CIA) from an historical perspective concentrating primarily on how it began and developed in the Federal Bureau of Investigation (FBI) Behavioral Science Unit (BSU), and continues into the present day (2013) in the FBI National Center for the Analysis of Violent Crime (NCAVC), the International Criminal Investigative Analysis Fellowship (ICIAF), and the Academy Group, Inc. (AGI).

Criminal Investigative Analysis

The idea of drawing personality characteristics of the criminal from an analysis of the crime scene and the behavior exhibited there had been around since before the days of Sir Arthur Conan Doyle and his fictional supersleuth Sherlock Holmes (1887). More recently (1968), Dr. James A. Brussel, a New York psychiatrist, prepared a profile of the Mad Bomber for the New York City Police Department. It turned out to be amazingly accurate in the description of perpetrator George Metesky. Other early attempts to describe a criminal by the analysis of the behavior exhibited during the commission of the crime were carried out by psychologists and psychiatrists, especially those connected with the judicial process. These professionals were usually engaged by the judicial system to ascertain the mental capacity of a defendant in order to determine if he was competent to stand trial. They were trained in diagnostic and treatment skills and their language was often replete with technical terms, affording little insight to investigators about how and why the crime was committed. The Early Days of Profiling The FBI’s contribution to behavioral interpretation of crime scenes began in the training division at the old post office building in the heart of Washington, D.C., in 1971. At that time, Special Agent (SA) criminologist Howard Teten, a former police officer from California, was working on the behavioral assessment of violent crime scenes in preparation for teaching classes to law enforcement officers of the FBI National Academy Program. His friend and colleague SA psychologist Patrick Mullany soon joined him in this work. The two instructors team-taught an applied criminology course to veteran law enforcement officers. In 1972, the FBI opened the new FBI Academy complex located in the center or the United States Marine Corps Base at Quantico, Virginia. A new and more comprehensive training and educational program was instituted with courses affording college credit due to an affiliation with the University of Virginia. There were five academic units (departments) at the academy: Legal

829

C

Instruction, Education and Communication Arts, Forensic Science, Management Science, and the Behavioral Science Unit (BSU). The BSU was a multidisciplinary training unit consisting of agents with educational backgrounds in psychology, sociology, criminology, and political science. Courses were taught in subjects such as Community Relations (SSA Jack Pfaff), Urban Police Problems (SSA Larry Monroe), Crisis Management (SSA Dick Harper), Hostage Negotiation (SSAs Conrad Hassel and Tom Strentz), and the refined course of Applied Criminology that would eventually evolve into CIA. During the next few years, other members of the BSU followed the lead set by Supervisory Special Agents (SSA) Teten and Mullany and began contributing to the further development of crime scene analysis and behavioral interpretation techniques. A Crime Epidemic in the United States Meanwhile in the mid-1970s, violent crime had begun rising steadily in the United States, until reaching epidemic proportions in the 1980s. For example, in 1962, there were 8,404 homicides; twenty years later, there were 21,012 homicides. Homicides cleared by arrest were 93 % in 1965 and 74 % in 1982 (FBI Uniform Crime Reports). The incidence of homicide was growing by leaps and bounds, and the clearance rate (as measured by arrest) was decreasing. A relatively new phenomenon emerged called “serial murder,” a term coined by BSU SSA’s John Douglas and Robert Ressler. Serial murder was defined as three or more homicides committed over a period of time with an emotional “cooling off” period between murders (as compared to mass murder where multiple victims are killed at one time). The alarming statistics for other crimes of violence were following the homicide trend. In an effort to deal with this epidemic of violence and decreasing clearance rates, the Attorney General of the United States, William French Smith (1981–85), ordered the directors of each federal law enforcement agency to submit to him ideas and plans for addressing this catastrophic state of affairs. Since the FBI BSU had been conducting training in violent criminal behavior

C

C

830

since its inception, Executive Assistant Director John Otto turned to the BSU for input to plan a response to the directive. As part of the BSU training programs, case consultation naturally evolved as law enforcement students asked questions about their unsolved cases. Officers would describe a crime, e.g., a brutal murder, and ask for the instructor’s opinion about what kind of person would commit such a crime. The instructor would speculate about the characteristics and traits of the unknown perpetrator. Instructors from different academic disciplines began to join together to contribute ideas about an unsolved violent crime from their respective disciplines and “street” experience as former field investigators. Their techniques were refined, and they soon gained a reputation for rendering valuable assistance in the analysis and solution of unsolved cases. An informal profiling program was begun. FBI Research on Violent Criminals and Their Crimes During classes at the FBI National Academy and during training seminars given around the country throughout the 1970s and 1980s, police students asked questions about practical aspects of cases that could not be answered by simply turning to the existing literature of criminology. Questions such as: Does the killer return to the scene of the crime and if so, why would he do that? Is the killer in the crowd when the body is found or when the crime scene is being processed? Does a killer ever insert himself into the investigation? BSU instructors recognized that answers to these and other similar questions would give investigators the ability to make predictive estimates of criminal behavior based on a fuller, deeper analysis of the crime scene and the probable mindset of the criminal perpetrator. These and other questions caused two members of the BSU, SSA’s Robert Ressler and John Douglas, to begin an informal effort to obtain some answers by going into prisons and conducting interviews of several incarcerated murderers. Convinced of the value of information obtained, they enlisted the aid of behavioral


scientists (especially Dr. Ann Burgess) from major universities to design research projects and interview protocols to be used for subsequent interviews of targeted perpetrators imprisoned for having committed multiple homicides. First, since empirical research of serial violent offenders and their crimes had never been conducted from the criminal investigator’s point of view, the initial research required the thorough examination of multiple, often horrific, crime scenes of a perpetrator in great detail prior to the prison interview. The crime scenes were carefully compared to one another, noting similarities and differences. Questions about how and why they were committed were listed. Once the detailed examination of the offender’s crimes was completed, other documentations (i.e., pre-sentence investigation reports, mental health records, arrest histories, etc.) were studied. Agents realized that conducting interviews without this comprehensive preliminary work would be the equivalent of trying to gain insight into the mind of a great artist (Vincent Van Gogh) by interviewing him without ever looking at his actual work. The quality of a work of art made the artist a famous person of interest to the connoisseur and likewise, the details of the crime scenes made the killer an infamous person of interest to the investigator. An additional important reason for a close examination of the crimes concerned the accuracy of the information obtained from the offender during the prison interviews. Skilled investigators know that criminals are notorious liars, so a secondary benefit from the careful preliminary examinations was that the offender had more difficulty lying to interviewers about his crimes. Aristotle noted that those who wish to succeed must ask the right preliminary questions (Metaphysics II). Secondly, researchers wanted to understand the developmental history of the perpetrator to try to determine how he became a violent offender. They wanted to gain insight into the behaviors and thought processes of the violent criminals. The protocols were developed to attain these ends. Interview protocols sought information about the subject in categories such as


developmental history, health, family, education, occupation, mental health, arrest and conviction, and the forms of mental aberration. Questions concerned things such as at what age did violent fantasies begin, and at what age did the violent behavior start? A subunit was formed in the BSU for conducting additional research and was initially led by SSA Ressler and later by SSA Richard Ault, Jr. Over a period of years, many BSU instructors went into the prisons with the comprehensive research protocols to conduct face-to-face interviews of violent criminals such as assassins, serial killers, rapists, child abductors, sexual sadists (and their wives), and, eventually, other criminals such as traitors and spies. The BSU agents sought to learn the “tradecraft,” fantasies, and habits of the perpetrators. The first interviews concerned assassins and serial killers and were initially conducted by SSAs Robert Ressler and John Douglas. SSAs Robert “Roy” Hazelwood and Kenneth Lanning conducted the initial research interviews of rapists and child abductors. Soon the research program expanded to include espionage and was supervised by SSA Richard Ault, Jr. The three approaches that were responsible for increasing the reservoir of new knowledge in the BSU were also the three primary tasks of the BSU itself: training, case consultation, and empirical research. These functions came to be referred to as “the three-legged stool” because each approach was interdependent with the others and simultaneously reinforced by the others. Training acquired investigative ideas and questions from course attendees that could be added to research protocols. Research interviews provided insight into the mind of the criminal regarding his thought processes and behaviors before, during, and after his crime. Casework allowed a venue for the application of new knowledge to ongoing cases. These functions, in turn, created “cuttingedge” materials for training programs. The three-legged stool became the foundation for sophisticated crime analysis and interpretation of criminal behavior that was to become known as “profiling.” The profiling theory stated that behavior could be identified in the crime scene:

831

C

extracted, examined, interpreted, and finally described in a profile of the unknown perpetrator. The Profiling Process During an investigation of an unsolved crime, the profiling process consisted of a detailed examination of all available case materials and was intended to answer four questions: (1) what happened? (a listing of all significant behaviors of offender and victim); (2) how it happened? (a reconstruction of the crime with supporting evidence for each step); (3) why it happened? (the reasons for each behavior and the formation of motive); and (4) who is responsible? (the type of person who would have committed this type of crime, in this manner, for these reasons). Characteristics of a perpetrator listed in the profile might include such descriptors as approximate age, gender, race, marital status, occupation, hobbies, vehicle information, appearance and grooming, arrest history, residential information, life style, psychological features and disorders, as well as intelligence level, emotional adjustment and ability to interact socially, etc. A fifth question was added to the above four questions: Where would such a person as described in the profile be found in the environment? This question allowed for the extrapolation of behavioral information drawn from the crime to be applied to lifestyle. Answering these questions provided insights and practical suggestions to investigators, leading to a more thorough and comprehensive investigation. Students were arriving at the FBI Academy with unsolved violent crime cases for analysis, and additional cases were being received from throughout the country and from several nations where students who had attended BSU classes resided. At this point, it became necessary to implement a procedure to systematically handle the influx of cases. Bureau case management policies were put in place. It was not long before the caseload continued to grow and the need for additional personnel became evident. But Bureau administrators wanted to evaluate the program before proceeding. In order to ascertain the value of the profiling program, a 1981 study was undertaken by the FBI

C

C

832

Institutional Research and Development Unit headed by SSA criminologist Howard Teten (the former BSU pioneer) who had been promoted to that specialized unit. Two hundred nine violent crime cases had been submitted by one hundred ninety-two law enforcement agencies to the BSU for analysis. They were followed upon by a survey to see if in the opinion of the case investigators, the crime analysis and subsequent profile furnished by the BSU had been helpful. These cases were primarily unsolved “old dog” cold cases where all traditional and logical investigative practices had been tried but failed to bring closure. It would have been remarkable if even 1 in a 100 of these cases were solved. As it was, the survey showed that 88 suspects had been identified largely because the profiling process had expanded and refocused the investigation. Even more remarkable was the fact that in 15 of the cases, profiling was reported to have identified a suspect outright and thereby led to the immediate solution of the cases. Needless to say, the funds necessary for expansion became available. The FBI Investigative Profiler Program Soon additional personnel were assigned to the unit. SSA John Douglas supervised a subunit of four “investigative profilers” brought into the BSU from the field and trained in the profiling process to work as permanent investigative profilers. The first four investigative profilers were SAs Jim Horn, Blaine “Mac” McIlwaine, Bill Hagmaier, and Ron Walker. In order to facilitate examination of cases from around the country, 55 field agents were brought back to the academy and trained in the process of case evaluation to facilitate submission of cases to the BSU. They were called “profile coordinators” and assisted in the examination and selection of cases in their divisions of the country to determine if they would likely benefit from the profiling process. By the mid-1980s the term “profiling” took on a negative connotation (i.e., racial profiling) due to the practice of some law enforcement officers using racial discrimination to unfairly target minority persons suspected of being involved in criminal activity. In order to more accurately


describe the investigative profiling process, it was renamed Criminal Investigative Analysis (CIA). CIA became the umbrella term that incorporated all of the tasks included in profiling (i.e., equivocal deaths analysis, cold case analysis, staged crimes, indirect personality assessment, planning interrogations, expert testimony in court, etc.). A Crime Classification Manual was developed by the FBI BSU practitioners and associates in order to standardize terminology so that more accurate communication could take place within the criminal justice community. Examples of several efforts to distinguish terms from one another as well as to clarify understanding of terms commonly used came about. For example the term “modus operandi (MO)” was distinguished from “signature.” MO was described as a learned set of behaviors the offender develops and maintains because they work operationally to facilitate the efficient commission of the crime. MOs are dynamic and malleable. “Signature” was described as the offender’s “calling card” in that it was an individualized set of indicators that usually point to an offender’s personality (i.e., behaviors that are not necessary to the efficient commission of the crime, but satisfy some idiosyncratic need of the offender such as taking a souvenir from the victim). They generally do not change. “Staging” was described as purposely altering of the crime scene prior to the arrival of police in order to redirect the investigation away from the most logical suspect, or to protect the reputation of the victim and the family (i.e., removal of pornographic materials from the death scene in an autoerotic death to make it look like suicide). FBI Police Fellowship Program Worldwide, law enforcement investigators expressed a growing interest in learning CIA skills and the Police Fellowship Program was born. The CIA skills were too complex to be taught in a short period of time. BSU SSA Roy Hazelwood, a veteran profiler, designed a program of instruction and case analysis for high-caliber, experienced investigators who would be invited to come to the FBI Academy


for 1 year to work as Fellows in the BSU. The philosophy of the program was to train officers coming from large law enforcement agencies to become CIA specialists so that they could return to their departments and establish similar CIA functions there. The program began on a trial basis with the selection of one officer (Detective Sam Bowerman of the Baltimore County, Maryland, Police Department), and was soon followed by a group of four more investigators from other large police departments (Ed Richards, Texas Department of Public Safety, Ray Pierce, New York City Police Department, Dennis Cremins, Los Angeles Police Department, and Eric Witzig, Washington D.C. Metropolitan Police). Thereafter, they were followed by groups of five to six carefully selected men and women investigators from large police agencies from the United States, Canada, Australia, and Europe. Sergeant Ron McKay of the Royal Canadian Mounted Police of Canada became the first international member of the Fellowship Program. The attendees of the Fellowship were given 3 months of classroom instruction with classes at the academy, the University of Virginia, and the Armed Forces Institute of Pathology. For the next 3 months of training, the attendees were assigned crime scenes from solved crimes of violence and asked to apply the CIA skills they had acquired. In the last 6 months, the attendees were assigned active unsolved cases to work with veteran profilers as regular CIA members of the BSU. Gradually, the yearlong program was shortened to 6 months. After most of the pioneering BSU CIA agents had retired, the program was discontinued despite its success and popularity. Beginning of the International Criminal Investigative Analysis Fellowship In 1985, the members of the first full Fellowship Program decided to form a group of investigators who had completed the Fellowship Program, a National Criminal Investigative Analysis Fellowship. In time, with the addition of attendees from other countries, the organization of diverse alumni became formalized and the International Criminal Investigative Analysis Fellowship (ICIAF) was

833

C

formed. Their goals were simply stated: to maintain the ethical standards of the FBI program and to assure that the quality of the CIA process would be continued. When the FBI Fellowship Program ended, ICIAF was well equipped to conduct their own CIA Fellowship training program, allowing for the acquisition of CIA knowledge and skills. Leadership in the organization, such as presidents Sam Bowerman and Ron McKay, developed a process for the careful selection of exceptionally qualified criminal investigators and affording them high-quality training. Conferences were held to bring together FBI and ICIAF specialists and other reputable forensic behavioral science speakers. Additionally, at regular regional and national meetings, ICIAF members invited law enforcement agencies to bring unsolved violent crime cases to the meetings where they were examined by ICIAF investigators for behavioral interpretation and profiling. ICIAF eventually expanded its purview to include a second division of CIA in order to deal with the developing field of geographic profiling. The National Center for the Analysis of Violent Crime By the mid-1980s, the BSU instructors had identified other projects to be joined to their CIA work. A Community Analysis Worksheet (a community profile) was developed to assist police executives to conduct a detailed examination of their communities for planning efficient delivery of police services (SSA Roger Depue). Since profiling had been successful for identifying and dealing with individual criminals, a group analysis protocol (a group profile) was developed to assess the characteristics and dynamics of criminal groups for purposes of successful penetration and exploitation (Depue). In 1984, following a meeting of law enforcement practitioners at Sam Houston State University, a Violent Criminal Apprehension Program (VICAP) created to track unsolved homicides and to link them to solved and unsolved cases from different jurisdictions (the idea of former LAPD homicide detective Pierce Brooks was added to the BSU). Among the first members of the VICAP team

C

C

834

joining Pierce Brooks were investigators from police departments around the country, including Terry Green of Oakland, California; Ken Hanflan of Salem, Oregon; Jim Howlett of Charlotte, North Carolina; and Winston Norman of the Washington Metropolitan Police Department (who was an FBI BSU instructor following his retirement from the police department). The computerized Arson Information Management (AIM) system developed by Dr. David Icove was brought to the BSU, and the first attempt to use artificial intelligence to augment profiling was made with use of a VAX 11-785 state-of-the-art computer. BSU CIA investigators observed that officers working extremely violent cases often developed personal problems themselves. Because of that fact, an innovative program eventually entitled Stress Management in Law Enforcement (SMILE) was developed by BSU SSA instructors (John Mindermann, James Reese, Robert Schaefer, and James Horn) to address these concerns. An effort to project crime trends into the future was developed by SSA Bill Tafoya. SSAs Terry Ethridge, Tom O’Maley, and Joe Harpold developed innovative crime prevention strategies and added them to the BSU repertoire. Assistant Director Jim McKenzie and his deputy, Dr. Jim O’Connor, cleared the way to bring all of the programs under one roof, as the BSU was gradually becoming a national center for the analysis of violent crime. The BSU was divided into two separate entities, one specializing in training and research (Depue) and the other in investigative support (SSA Alan “Smokey” Burgess). The original BSU had grown from 12 to 55 members. The national recognition of the innovative and important work of the BSU came in 1984 when President Ronald Reagan, speaking at the National Sheriff’s Association Annual Conference in Hartford, Connecticut, announced the institution of the FBI National Center for the Analysis of Violent Crime (NCAVC) to be located at the FBI Academy at Quantico, Virginia. Its mission was to address unusual, vicious, bizarre, or repetitive violent crimes. Roger Depue, long-time chief and member of the BSU, was made its first administrator.


Additional field agents from around the country were brought back to Quantico to work as CIA specialists in the NCAVC, and others were given a 2 week training in-service in CIA fundamentals so that they could function as CIA coordinators (previously called profile coordinators) between Quantico, FBI field offices, and law enforcement agencies within their areas. CIA work was separated into specialized units involving different crimes of violence. The United States was divided into geographic regions, with NCAVC CIA agents providing oversight and direction to the FBI CIA coordinators and law enforcement investigators in the regions. During the 1990s, the FBI NCAVC grew in size, became a field office support resource, and moved offsite from Quantico to neighboring Aquia, Virginia. Eventually, the NCAVC was joined to other operational functions from the FBI Academy (e.g., the Hostage Rescue Team, Special Operations, Conflict Management, and Hostage Negotiation) and the entire operation became a separate FBI entity placed under the mantle of the new investigative support group called the Critical Incident Response Group (CIRG). Expansion into the Private Sector In 1989, retired BSU chief Dr. Roger Depue consulted with Dr. Bertram Brown (former Director to the United States National Institute of Mental Health and White House psychiatrist) and Conrad Hassel, Esquire, (retired chief of FBI Special Operations and Research) about the feasibility of starting a center for CIA to assist private sector organizations in dealing with problems caused by aberrant and violent behavior. Shortly thereafter, The Academy Group, Inc. (AGI) was begun as the first forensic behavioral sciences services company to bring CIA expertise to the private sector. AGI practitioners began to apply profiling techniques to the problems of violence in workplaces and schools, authorial attribution of anonymous communications, expert behavioral analysis and testimony in a wide variety of civil cases, and the analysis of criminal cases when requested by security and law enforcement agencies. Many veteran agents retiring from the NCAVC joined their colleagues


at AGI such as CIA practitioners Peter Smerick (current Chairman of the Board) and R. Stephen Mardigian (current President), and Michael Napier, an interview and interrogation specialist. Fellowship alumni also joined AGI to continue doing CIA work such as Ken Baker, United States Secret Service, and Larry McCann, Virginia State Police. AGI located in Manassas, Virginia, allowed for continued contact with the FBI BSU and NCAVC. Work of the Critical Incident Response Group NCAVC Meanwhile back at the FBI during the early 2000s, the CIA units of the CIRG NCAVC were named Behavioral Analysis Units (BAU) and four of these units were designated each with separate responsibilities. BAU 1 contained SSA’s specializing in threat assessment, especially in the analysis of written documents. Linguist SSA Jim Fitzgerald (who joined AGI following his retirement) led the field in this important work. This unit dealt with the threatening communications that often preceded or accompanied dangerous acts. Writing is a form of behavior that can be analyzed to reveal personality characteristics of the writer. Communications that are anonymous can be examined as to writing style and content and often afford important clues as to the identity of the unknown authorship. BAU 2 specialized in violent crimes against adult victims. Most violent crimes fall into this category. Both the victims and perpetrators are adults. Also in this category there was a good deal of concentration on crimes against the other most vulnerable victims in society, the elderly (as developed by SSA Mark Safarik, Dr. John Jarvis and Dr. Kathleen Nussbaum). BAU 3 was exclusively dedicated to crimes against children. Both adult and child perpetrators prey on children. This unit works closely with the National Center for Missing and Exploited Children. It is noted that BAU’s 2 and 3 continue the work against serial murderers originally begun in the BSU of the 1980s. BAU 4 deals with counterterrorism. Since 2001 and the terrorist attack of the World Trade

835

C

Center in New York City, much emphasis and many resources have been directed toward this most serious problem. These NCAVC BAUs continue their CIA work until the present time (2013). Meanwhile at the FBI Academy BSU, SSA Andy Bringuel instituted the Terrorist Research Analysis Project (TRAP) expanding the “group profiling” concept to a more comprehensive Group Analysis Protocol that improved understanding of foreign and domestic terrorist groups. ICIAF Understudy Program Under the leadership of presidents such as Ron McKay, Kate Lines, Steven Conlon, and Keith Howard (2013), the ICIAF continued to identify, select, and train new investigators from around the world in the art and science of CIA. An Understudy Program has been created whereby applicants are vetted and trained by qualified ICIAF CIA practitioners. The process includes nomination and interview of a candidate, practicum of case assessments, reading requirements, video and audio reviews and analyses, academic and field practice exercises, examinations, review of analytical notes and reports, and final individual candidate screening board interviews. In 2001, the geographic profilers originally trained by the Vancouver Police Department (VPD), Canada, became part of the ICIAF (information provided by ICIAF Geographic Profiling Division members). They formed a separate Geographic Profiling Division with its own vice president, and eventually adopted the Geographic Understudy Program developed by VPD in 1997. The program was designed to provide comprehensive training to members of those agencies wishing to establish their own geographic profiling capability. Suitable candidates embark upon a year of study under the tutelage of a mentor who must be a fully qualified geographic profiler. The training program is divided into four blocks: (1) probability and the Rigel software; (2) serial crime and crime linkage; (3) violent sex offenders and criminal profiling; and (4) quantitative spatial techniques and geographic profiling.

C

C

836

The first three blocks are done through distance education, under the supervision of the understudy’s mentor. The fourth block (four months) is a residency at the mentor’s agency, involving both reviews of previous files and casework in active investigations. The understudy must pass a qualifying examination at the end of the training period in order to successfully conclude the program. Continuing education is important for the ongoing development of professional skills and expertise, and the ICIAF community, listserv, and conferences play a key role in this process. Geographic profilers from Canada, the United States, the United Kingdom, and The Netherlands are now members of the Fellowship, and they have worked on serial crime cases from around the world. As a result, ICIAF membership continues to grow in size and sophistication (2013). Response to Negative Critiques of Profiling Profiling may mean different things to different people. As reported above, the FBI had an inhouse research capability, the Institutional Research and Development Unit (IRDU), to look at programs to determine if they were cost effective and whether or not they should be continued. The definition of profiling as set forth by the IRDU was: “As used in this study psychological profiling is defined as the process of identifying the gross psychological characteristics of an individual based on an analysis of the crime(s) he or she committed and providing a general description of the person utilizing those traits.” In 1981 the FBI IRDU did “a cost benefit study to determine the extent to which the service has been of value to the users.” The service (profiling) was provided by special agent behavioral science investigators who were instructors at the FBI Academy (the profilers). The service was provided to the users (primarily police investigators) at their request because the crimes (primarily murder) apparently could not be solved by conventional investigative methods. From a practical standpoint FBI management was interested in whether or not the police investigators thought the service was valuable to them and should be continued to be provided by the FBI.


It should be pointed out at this time that the profiling process would have been discontinued if it was not demonstrated to have been of value to the users. Questionnaires were distributed to 192 law enforcement agencies regarding the 209 cases that had been analyzed and profiles of unknown offenders had been prepared for them. Results revealed that in 88 cases suspect(s) had been identified and the law enforcement investigators (users) reported that the profiles assisted in focusing the investigation properly and had helped in locating possible suspects in the majority of cases. In 15 cases they reported that the profile actually identified a suspect. Needless to say the profiling program was continued. In the FBI profiling was developed by the examination of numerous cases, analysis of the crime scenes and by identifying the behavior therein. Agents became adept at identifying the behaviors and constructing profiles of the unknown offenders. Research projects were conducted whereby agents interviewed offenders following a thorough review of crimes perpetrated by these offenders in order to gain insight into the thought processes operating in the offenders before the crime, during the crime, and after the crimes were committed. This empirical research contributed to the profiling process being refined and expanded. The FBI then brought outstanding investigators from federal, state and local law enforcement agencies into the FBI Academy for an intensive Fellowship Program where the process and techniques discovered by the agents were shared with them. Since that time many people have claimed to be “profilers” without working with law enforcement, and without doing the meticulous work of carefully examining numerous cases and crime scenes, and without getting feedback on the value of their efforts. Furthermore they have not analyzed large numbers of cases and crime scenes of known perpetrators and then gone into the prisons to interview these offenders in order to gain an indepth understanding of their criminal behavior. Now “research efforts” aimed at determining the value of “profiling” are being conducted by behavioral scientists and academics.


The results of this research often cast doubts on the “profiling” process studied. From the practitioner perspective, the one true method of determining the value of profiling is to ask the users of the profiling process whether or not it was helpful to them in solving crimes. If it is not, that profiling process should be discontinued. (The information quoted in the IRDU study is from an internal document of the FBI entitled “EVALUATION OF THE PSYCHOLOGICAL PROFILING PROGRAM” and reported by the Institutional Research and Development Unit of the FBI Academy, Quantico, Virginia in December of 1981).

837

C

and much of it must be attributed to CIA skills. It is largely due to a community of astute investigators with impressive CIA knowledge and skills who are conducting more proficient crime analysis, behavioral interpretation, and construction of accurate profiles of unknown dangerous offenders. These CIA skills have increased the sophistication and professionalism of investigators of violent crime in their struggle to assure that a higher level of security and justice is being achieved for the citizens of their respective communities.

Related Entries Conclusion It would be an understatement to say that CIA has caught the imagination of the media and entertainment industry. People are fascinated by the subject and crave information about how the process is done and examples of its effectiveness. The first fiction author to spend time at the FBI academy and the BSU researching cases and trying to gain insight into the profiling process and the mind of the serial killer was Thomas Harris (1982). His best-selling book Red Dragon created national interest in the BSU. His next best-selling book Silence of the Lambs and the subsequent movies about serial killer Hannibal Lector and FBI Special Agent Clarice Starling brought international attention to the BSU and the NCAVC. Media interests abounded. Additional books, magazines, television programs, movies, and games proliferated and have increasingly become available for public consumption. As with the box office success of the film Top Gun when there was a groundswell of young men and women eager to become the next generation of fighter pilots, young students of behavioral sciences and criminal justice aspire to become the next wave of CIA practitioners. Today worldwide, more and more perpetrators of violent crime are being identified, apprehended, prosecuted, and incarcerated earlier in their criminal careers. In the United States, the violent crime rate has dropped significantly

▶ Criminal Investigative Analysis ▶ Criminal Profiling ▶ FBI Influence on State and Local Police ▶ History of Criminal Investigation ▶ Investigative Psychology ▶ Modus Operandi of Sex Offenders ▶ Profiling Arson ▶ Psychological Autopsy of Equivocal Deaths ▶ Risk Assessment, Classification, and Prediction

Recommended Reading and References Depue RL, Schindehette S (2005) Between good and evil. Warner Books, New York Douglas J, Olshaker M (1995) Mindhunter. Simon & Schuster/Pocket Books, New York Douglas J, Olshaker M (1997) Journey into darkness. Simon & Schuster/Pocket Books, New York Douglas J, Olshaker M (1999) The anatomy of motive. Simon & Schuster, New York Douglas J, Ressler RK, Burgess AW, Burgess AG (1992) Crime classification manual. Lexington Books, New York Hazelwood R, Burgess A (1987) Practical aspects of rape investigation: a multidisciplinary approach. Elsevier, New York Hazelwood R, Michaud SG (2001) Dark dreams. St. Martin’s Press, New York Hazelwood R, Dietz PE, Burgess A (1983) Autoerotic fatalities. Heath Inc, Boston McCrary GO, Ramsland K (2003) The unknown darkness. New HarperCollins, New York Michaud SG, Hazelwood R (1998) The evil that men do. St. Martin’s Press, New York

C

C

838

Monahan J (1995) The clinical prediction of violent behavior. Jason Aronson Inc, Northvale Ressler RK, Shachtman T (1992) Whoever fights monsters. St. Martin’s Press, New York Ressler RK, Shachtman T (1997) I have lived in the monster. St. Martin’s Press, New York Ressler RK, Douglas J, Burgess A (1988) Sexual homicide: patterns and motives. Lexington Books, New York

Criminal Investigative Profiling

Criminal Punishment ▶ Institutional Theories of Punishment

Criminal Thinking Model ▶ Lifestyle Theory

Criminal Investigative Profiling ▶ Behavioral Investigative Advice

Criminal Justice System ▶ Drug Abuse and Alcohol Dependence Among Inmates

Criminal Law ▶ Identification Technologies in Policing and Proof

Criminalistics ▶ Forensic Science and Miscarriages of Justice ▶ History of Forensic Science in Policing

Criminalization and Occupational Health and Safety Fiona Haines1 and Paul Almond2 1 School of Social and Political Sciences, University of Melbourne, VIC, Australia 2 School of Law, University of Reading, UK

Synonyms

Criminal Personality Profiling

Health and safety law; Workplace safety crimes

▶ Behavioral Investigative Advice ▶ Criminal Investigative Analysis

Overview

Criminal Phenomenon: Crime, Criminal, Criminality ▶ Control Theory

Criminal Profiling ▶ Behavioral Investigative Advice

Work can be a dangerous place and provides the context for a multitude of deaths, injuries, and illnesses. These include those arising from major disasters, such as the chemical plant explosion at Bhopal, India, in 1984 or at Piper Alpha in the North Sea in 1988 and more recently in the Gulf of Mexico in 2010. There are the fatalities reported in local newspapers: deaths from a fall at a building site, a tractor rollover, the death of a truck driver on the road, and death from asphyxiation due to a trench collapse. Fatal and chronic illnesses are also important outcomes of work: heart attacks,

Criminalization and Occupational Health and Safety

cancers, lung disease, and the multitude of chronic ailments that arise from exposure to chemicals and other hazards both in the course of work and as a consequence of events like Bhopal and Chernobyl. Then, there are the legacies from particular occupations that lead to chronic disability: back pain from nursing, broken knees from carpet-laying, and repetitive strain injury from typing. Indeed, some ailments take their name specifically from work-related tasks: potter’s lung, policeman’s heel, and housemaid’s knee. The relationship between the harms above and criminalization as a means to reduce this toll of deaths, injuries, and illnesses is complex. Understanding criminalization in the context of occupational health and safety (OHS) requires analysis of the economic, social, and political context within which organizational and individual practices arise that lead to death, injury, and illness. Further, laws – including criminal prohibitions – are themselves a result of the interrelationship between economic, political, and social factors. Indeed, these elements often are evident in the contests around what should and should not be proscribed and prescribed by law. This entry begins with a brief overview of the nature of the problem of safety globally. Then, literature on the history of OHS law and enforcement is reviewed in order to tease apart the multiple layers of the connection between criminalization and safety at work. This history illustrates the enduring importance of key elements. These elements include the need to pay close attention to the demands of a capitalist economy and its impact on health and safety at work, the concern governments have to retain legitimacy through enacting (but not necessarily enforcing) law, the challenges of dealing with the letter of the law, and placing the variety of regulatory techniques that have been developed over time in an appropriately local and global context.

Occupational Health and Safety: A Global Problem Rates of OHS-related physical harm remain problematically high within modern industrial

839

C

societies. In 2007 (the last year for which full comparative figures are available), there were 5,785 deaths at work across the 27 EU member states; this translates into an adjusted rate of approximately 3.5 deaths per 100,000 workers (Eurostat 2010: 347). In the USA, the Occupational Safety and Health Administration’s (OSHA) Census of Fatal Occupational Injuries placed the number of fatal injuries to workers for 2008 at 5,214, or 3.7 deaths per 100,000 workers. In Canada, there were 1,036 work-related deaths in 2008, in Australia 444 in 2009 (286 “working” fatalities, 117 out-of-work fatalities, and 44 public or bystander fatalities), and in Japan 1,075. Death, illness, and injury due to poor working conditions extend well beyond the industrialized world. While the ILO recorded 18,067 fatal occupational “accidents” in 2003, the true rate is thought to be as high as 357,948, with a further 1.95 million deaths from work-related ill health (cited in H€am€al€ainen et al. 2009: 129). The vast majority of these deaths occur in the industrializing world, with China (97,248 accidents and 332,454 ill health), India (46,928 accidents and 355,863 ill health), and Indonesia (15,873 accidents and 61,572 ill health) featuring prominently. Africa is thought to account for over 470,000 work-related deaths each year. These statistics, though, should be treated with caution. The data are incomplete, as many injuries and deaths are not reported or recorded, even in industrialized nations with established administrative systems; the statistics available for many industrializing nations, which lack this infrastructure, are so sketchy as to be almost unusable. What counts as a death or injury at work also differs. Some jurisdictions (Australia being an example) have an inclusive definition, including workers, nonworkers, and deaths on the road (as a working driver and as a commuter), while most others have a much narrower definition (the UK excludes deaths on the road, and the USA excludes deaths to nonworkers). Incidents of occupational ill health (and resultant deaths) also tend to fall outside the statistics, primarily because they are harder to quantify accurately (see Tombs and Whyte 2007). For this reason, any comparison between countries and

C

C

840

jurisdictions needs to be treated carefully. Nonetheless, it is clear that the problem is substantial and requires criminological attention.

The Criminalization Phenomenon: The Legacy of History Criminologists can learn much from the history of OHS regulation – and from interpretations of that history. These interpretations include those who see primarily a progression towards greater civilization and higher standards. These authors point to the way the law has extended formal protections to an ever-widening portion of the working population, thereby reflecting broader democratizing trends within society (Thomas 1948). Others are more skeptical and see rather a complex deference to the demands of capital, contingent on local conditions (Carson 1979, 1980). Through a series of papers, Kit Carson argued that laws developed in the context of the industrial revolution in a manner that allowed industrialization and laissez-faire capitalism to thrive. Law reform served various purposes. The laws mediated competition within the emerging industrial class, as well as disciplining factory workers to obey the clock rather than the seasons (Carson 1979). But reform needed to take place in a way that ensured that the political legitimacy of the incumbent government was preserved. This was no easy task since critical voices in the Parliament argued that the toll on health that resulted from factory labor threatened the military capacity of the nation. Critically, both these interpretations of history, the civilizing and the critical, pointed to an important finding – namely that law reform did not necessarily mean adequate enforcement (Almond 2013; Carson 1979; Thomas 1948; Tombs and Whyte 2007). Law reform might be a powerful symbol of change, but implementation was another matter. Diverse interpretations of the industrial revolution in England, in short whether there was an identity of interest between workers and employers, were echoed in subsequent reforms that took place in the 1970s, when the debate centered on whether there was, or was


not, a consensus between workers and employers regarding the need for a safe workplace. This period has been identified as the “third wave” of safety regulation, which followed periods of regulation by contract (first wave) and by prescriptive rules (second wave) (Tucker 1995). A key figure of this period, Lord Robens, argued strongly that laws and regulations needed to be framed to enhance this common interest (Robens 1972). In particular, strict prescriptive rules determining what was, or was not, safe should give way to a focus on safe outcomes (through performance standards) and the processes by which they could be assured. The debates around the value of prescription or performance standards can be viewed in several ways. The first way is to evaluate the various arguments to identify which approach is more effective in reducing death, injury, and illness. From a perspective of evaluation prescription can play an important role, for example, in restricting access to confined spaces to prevent asphyxiation. But prescriptive standards can miss critical issues such as risky work processes (such as factory production lines). Further, prescriptive standards can quickly become obsolete as technology changes over time. Viewed in light of these limitations to prescription, the philosophical shift to performance standards begun in the 1970s was timely. However, it was also important in light of its economic and political context. The development of the Robens model needs to be understood in the context of 1970s Britain and in particular the centrality and subsequent decline of the coal industry. Robens, with his history in the union movement and his chairmanship of the National Coal Board was seen as “the voice of industrial common sense.” His status and biography, straddling both union and management, meant that his ideas could gain traction. Key among these ideas was the need for a flexible regulatory approach that had at its core a focus on the need for safety, the essence of a performance approach (Tucker 1995). Indeed, “Robens-style” legislation, albeit with some important differences in emphasis, was adopted broadly across anglophile countries. A key difference in emphasis between countries was the need for employee


representation and in the role of the unions. The state of Victoria, Australia, emphasized the need for clear worker representation in OHS at the same time as it moved to institutionalize performance standards. The level of control provided to workers within Robens-style legislation was seen as critical to its effectiveness. But an emphasis on worker representation also needs to take account of what kind of representation it is. An important influence on the development of OHS after Robens in the UK was the decline of union power during the Thatcher period in the 1980s. This is because performance standards emphasize the need to internalize safe systems of work and a safety management culture within corporate processes but, in doing so, they also “normalize” OHS risks by assuming that “the risks associated with increased efficiency and profitability can be responsibly managed and contained” (Silbey 2009: 343). This leaves unexamined the fundamental role played by the economic power of employers in determining exposure to risk in the first place. In this way, the neoliberal approach of contemporary governments is embedded via the movement of OHS away from the contested political sphere of prescription standards and industrial relations and towards the relatively uncontested arena of internal business procedures. The impact on the decline of union power on the development of performance standards, and indeed, levels of prescription, is complex. It is possible, for example, to understand prescriptive standards as the end point of negotiation about what it is and is not acceptable for an employer to ask an employee to do. Unions play a prominent role in this process. It is interesting to note that there remains a higher level of prescriptiveness within European health and safety regulations (where union representation in general is higher), a difference which causes friction within the EU (Walters 1996). In short, any claim of “consensus” around safety between employers and employees needs to be viewed as contingent and fragile. Without adequate worker participation, or where levels of industrial conflict are high, “consensus” quickly breaks down (Gunningham and Sinclair 2009). A vivid example of this is in the role and

841

C

effectiveness of health and safety representatives. Health and safety representatives, together with health and safety committees, can be an effective way for employees to voice their safety concerns to management and have them resolved quickly and effectively (James and Walters 2002). Yet, an increasing neoliberal emphasis on individual responsibility in the OHS regulatory framework can change the safety representative’s task, reducing their role to one of policing fellow workers’ “unsafe behavior,” such as nonuse of protective safety equipment, and fining them for “breaches” of locally generated safety procedures (Gray 2009). This is an inversion of the role as it was conceived within places such as Victoria, Australia, as the “consensus” shifts from tripartite (employer, employee, and government) responsibility for safety to an overriding emphasis on worker responsibility to ensure a safe workplace. Techniques and regulatory strategies (in this case the role of health and safety reps) must be understood and interrogated in context. Debates around the need for industrial manslaughter legislation to ensure a safe workplace are similarly complex. These debates are often couched in the need to emphasize adequate enforcement of OHS law (rather than just a focus on content) and the need for significant penalties to apply (Tombs and Whyte 2007). Many argue that an industrial manslaughter offense, a version of a homicide offense, is required to ensure safety at work. This would allow the communicative power of the “mainstream” criminal law to be applied to an individual or company following a workplace fatality. An offense of this sort was introduced into UK law via the Corporate Manslaughter and Corporate Homicide Act 2007 (Almond 2013) and was successfully used for the first time to convict a corporation of manslaughter in early 2011. Yet, for others, the demand for a “new” criminal offense concedes too much to history. That is, occupational health and safety legislation, properly conceived, always was and should remain criminal (Johnstone 2007). These authors argue that OHS law has been improperly conventionalized as quasi-criminal and that reform and enforcement should focus on using it

C

C

842

and reemphasizing its essentially criminal nature. However, over time, the persistent demands for separate industrial manslaughter offenses suggest that the quasi-criminal nature of OHS law remains prominent in the public imagination – at least in some jurisdictions. Recent surveys undertaken in the United Kingdom have shown that OHS breaches are not necessarily regarded as crimes per se, and commonly applied OHS laws are not regarded as having the status or symbolic effect associated with criminal law (Almond and Colover 2010). In general terms, safety laws are regarded rather ambiguously. Nonetheless, this ambiguity is not something that should be accepted unproblematically; it is a product of the struggles seen throughout the history of OHS law reform (Wells 2001). The problematic status of OHS law as criminal extends also to the consideration of just who - or what – should be made accountable. The intricacies of this debate are well outlined in several works including those by Fisse and Braithwaite (1993) and Celia Wells (2001). At issue here is the degree to which individuals and/or organizations should be held accountable. Wells (2001) explores the way criminal law itself has developed with a bias towards the identification of single individuals who need to be made criminally liable for their actions. Her account of history shows how this skews criminal liability onto the weak, while civil liability for physical harm (if any) is placed on the strong. In the OHS context, she points out the difficulties of individual liability within the workplace context. Often in worksites, it is difficult to identify a single individual who is responsible for a death in the workplace. Further, it is not easy simply to add up responsibilities of several individuals and apportion blame accordingly (the so-called aggregation model of responsibility). Rather, there is a system of work or an organizational culture that gives rise to corners being cut and pressures placed on individuals that can lead to disaster and death (Gray 2009; Silbey 2009). On the other hand, those who view the potential of criminal liability in instrumental terms, as a means of achieving a change in behavior via deterrence, advocate the pursuit of individual


rather than corporate liability strategies because individual directors and those whose decisions lead to safety violations are more directly deterred by the threat of imprisonment, fines, and reputational sanctions (Gobert 2005; Tombs and Whyte 2007). Teasing apart who should be held responsible, whether there should be corporate criminal liability or whether it should be the most senior manager or executive who should be accountable, forms a considerable part of the relevant literature here. What is clear is that industrial manslaughter offenses, which invariably derive corporate liability from the individual, have proved most successful against small business where lines of communication are clear and noncomplex (Wells 2001). A recognition of this fact was a key factor shaping the recent law reform process in the UK, where a specific “corporate manslaughter” offense, with liability based on a “management failure” model of liability, was introduced in 2007 to replace the existing common-law offense of manslaughter, thereby attempting to remove the requirement that a responsible individual be found as a basis for corporate liability. The history of reform, both that of targeted OHS law and that of developing a separate industrial manslaughter provision, reemphasizes the importance of taking account of the competing interests that find their way into the law. Certainly, there is no end in sight for these reforms. There is a helpful body of literature that explores whose interests end up written into law as a result of these struggles, as well as the necessary economic conditions for reform to be effective. Dan Curran’s (1993) history of mine safety legislation in the United States shows how reform that could improve safety required not only a strong union presence, but a union with a strong safety focus as well as a high demand for coal and a tight labor supply. And James Gobert’s (2005) work in the English context shows how potentially effective industrial manslaughter provisions were reduced or weakened under pressure from employers. In any case, whether “regulatory” or “criminal,” the laws that emerge in a field like health and safety tend towards compromise reflect the


competing interests and levels of influence of the groups that are party to debates in this field.

The Complex Division Between Criminal Behavior, Regulatory Breaches, and Desirable Behavior Drawing bright lines in law and regulation between what is criminal, what is harmful and what is desirable is difficult. Part of the difficulty here is the way that the benefits of a particular activity or industry are often intertwined with harm. The embedded nature of harm within benefit and the unequal distribution of those harms and benefits means that it is difficult (both politically and practically) to develop precise law or finely targeted regulation that can separate the two. Further, those with interests in the benefits often have unique leverage over policy-making within government. For example, a mine in a particular locale may provide necessary and welcome employment and inward investment to a town, and its continued presence supported by many residents as a result. It may well also provide significant profits for the mine’s owners. Yet, it is miners that are at most risk from death, illness, and injury. So, workplace safety breaches are categorized as side effects of an otherwise purposeful form of wealth-creating industrial activity (which is encouraged), and risks are managed through regulation rather than mining prohibited outright (Hawkins 2002; Kagan and Scholz 1984). Context-rich analyses of criminalization in the context of OHS illustrate well the complexity of the distinction between what is “criminal” behavior and prohibited and what is “merely” a regulatory breach of the law. While debates around the need for criminalization are important, they can distract attention away from the dynamic within which regulatory strategies (including criminalization) emerge. Again, it can be helpful to go back to Kit Carson’s work (1980), which shows how industrialists’ resistance to criminalization generated new forms of law in an attempt to bring them to account. Notions of mens rea and responsibility were developed in order to frame criminality in

843

C

terms of individual pathology rather than sociological factors such as poverty and inequality. But safety inspectors found that proving fault in relation to workplace safety breaches was exceedingly difficult and argued for the need for strict liability offenses that did not require intent or mens rea in order to prove a breach. But once in place, the industrialists’ argued that a truly criminal offense must require proof of intent, and so these safety “crimes” were different from “ordinary” deviance in legal form; this meant that they could then be classified as normatively different due to the lack of individual pathology involved (Carson 1979). This change of status was fundamental in legitimating the new laws in the eyes of the industrialists. Understanding enforcement in the OHS context means understanding why enforcement is responsive to the regulated – including strategies to deal with the level of political influence the industry is able to exert. Regulators find it difficult to rigorously enforce safety standards in the face of political ambivalence (Haines 2011). Hence, in the face of resistance, it is not surprising that enforcement officers seek legal provisions that are more easily proved, which gave rise to debates around ambiguity and criminality. This can be characterized as a problem of assimilation and differentiation; the problems of assimilating industrial offenders into existing concepts of fault proved insurmountable due to the fundamental individualism of the criminal law. From one perspective, strict liability offenses weakened the law, but from another, they at least allowed inspectors to prove an illegality against a corporate defendant. This “double shift” towards and then away from criminalization was born out of necessity and the strictures of a criminal system developed to deal with working-class “unruliness.” Hence, the “quasi-criminal” or regulatory character of the law developed from this point. The impact of this history of ambiguity on current government policy continues to be felt. For example, similar arguments attached to the “reasonable practicability” defense contained in the UK Health and Safety at Work Act 1974 (HWSA). The absolute duty for employers to provide a safe workplace

C

C

844

was mitigated by requiring that the costs of breaching the law outweigh the cost of prevention. This had the effect of legitimating the HSWA in the eyes of business without introducing any kind of “moralizing” mens rea element into play; at the same time, this requirement acts as a de facto fault element, in that it allows regulators to differentiate between “respectable” businesses and the morally culpable “bad apples” (Hawkins 2002). But such a requirement arguably also shields companies from rigorous enforcement by introducing a utilitarian calculus into the question of safety protection, allowing what should be an absolute moral obligation to be sidestepped where the costs of doing so would be detrimental to business competitiveness. There is another important result of this “responsiveness”; the diversity of law and enforcement techniques implemented in various jurisdictions. In the past 40 years or more, a proliferation of different techniques has developed in the OHS context. These techniques span different forms of penalties, such as “on the spot” fines and enforceable undertakings measures, and those that attempt to govern organizational behavior such as safety management systems or individual action in minute detail such as behavior-based safety training. Also important to understand are self-regulatory systems that include a wide array of accreditation schemes (often put in place by business defensively to preempt legislative reform) through to what Ayres and Braithwaite (1992) term enforced self-regulation: the safety case model in the regulation of major hazard facilities such as chemical plants and oil refineries. This latter case, when properly implemented, needs to be understood as far more than self-regulation. Under this model, each plant must come up with fine-grained and clearly articulated rules around safety at their facility. These are checked and, if satisfactory, approved by the regulator who then enforces this unique set of rules (the “safety case”) as it applies to each individual plant. Each of these initiatives above can, and should, command attention in terms of how far they provide an incentive for improved OHS standards on the ground.


Legal, organizational, and behaviorally focused techniques also need to be understood in context. This context includes contests over: (i) the definition of what laws should, or should not, pertain to the workplace; (ii) what is, and is not, acceptable in terms of both risk and behavior; and ultimately (iii) if death, injury, or illness occur, who should be held responsible and for what. There are a broad range of actors with potential influence here. There are the three major players: the government, the employers, and the employees. Further, inspection and enforcement officers need separate consideration to that of the government as their practices, formal and informal policies, and decision-making processes are significant in shaping regulatory outcomes. In particular, the work of scholars associated with the UK’s Oxford Centre for Socio-Legal Studies has shown how regulators utilize discretion, negotiation, and bargain and bluff, in order to maximize their influence while facing severe resource constraints (Hawkins 2002; Hutter 1997). In such cases, they need to find creative ways to implement the law and to generate higher standards. This can prove difficult, if not impossible.

Placing Regulation and Law in Context: The Direct Impact of Economic Conditions Important as law, regulation, and enforcement are to OHS standards, it is important to recognize that economic conditions also have a significant impact on safety. We noted above that small businesses were at most risk from prosecution under industrial manslaughter provisions in the UK. But small businesses in some sectors can have lower safety standards because of their weak position in the market place, both as discrete entities and as part of a contracting chain (Haines 1997). They have neither the financial nor human resources to invest in safety to the same degree as their large business counterparts. As a result, the safety afforded to employees of small businesses can suffer. This finding on small business also needs to be extended to include


the contracting relationships prevalent in some industries (e.g., construction) and increasing in others due to increasing levels of contract labor and the decision by some companies to contract out non-core parts of their business operations, such as cleaning or maintenance (James et al. 2007). This can create problems for safety. Firstly, these businesses may be small and so less able to ensure the safety of their employees. Secondly, the degree to which they can focus on safety may be dependent upon the contract price they are able to obtain; when stiff competition drives this price down, levels of safety can decline. Because of this, certain jurisdictions have enacted what is known as “chain of responsibility” legislation, making principal contractors responsible for the safety of subcontractors’ employees. This is a good example of law that is sensitive to economic conditions. But not all jurisdictions have these laws, or implement them effectively, and where they are absent, real problems in safety levels can arise (James et al. 2007). Finally, there has recently been an increased attention paid to workplace stress and its impact on ill health. For some, categorizing workplace stress as a health and safety issue is a step too far, but stress has been associated with significantly decreased life expectancy (La Montagne et al. 2007). In particular, work which is time pressured, but where individuals have little control over the pattern of their work, is particularly hazardous. Also, being at the bottom of the hierarchy in terms of power over the conditions of work within a workplace has been shown to have detrimental health effects, a key finding of the prominent Whitehall studies (see, e.g., Marmot and Smith 1991). The implication of this research is that a narrow intervention of the kind often associated with health and safety initiatives is unlikely to be effective. Rather, what is needed is a wholesale reorganization of work itself. Achieving such change seems at a considerable distance from the debates around the proper boundaries of the criminal law. But they share a similar feature in that reorganizing work in a manner consistent with the promotion of health, as with determining what is and is not criminal behavior, is politically fraught and likely to

845

C

engender strong opposition from those who benefit from current ways of working.

Globalization, Criminalization, and Workplace Safety

C Much current OHS research focuses on deaths, illnesses, and injuries that take place in the industrialized world, but (as stated at the outset of this entry) these are global problems. There are many high-profile examples of industrial disasters in industrializing countries, like Bhopal, mentioned at the beginning of this entry. Another example is provided by the Kader toy factory fire that occurred in Bangkok, Thailand, in 1993 (Haines 2005), which resulted in 183 deaths. Many lethal factory fires share similar features, including poor building design, oppressive working conditions, and substandard fire safety procedures. The role the criminal law can play in the aftermath of these disasters is also noteworthy and problematic, particularly in terms of who those laws seek to target. In the case of the Kader fire, it was a worker that was jailed for 10 years for causing the fire because he failed to extinguish the butt of his cigarette. His contribution was minor and pales into insignificance in light of inadequate building and fire safety standards at the Kader factory. Criminalization, in an international context, can act to scapegoat individuals at the expense of dealing with systemic problems. The international diffusion of corporate criminal liability has been driven in no small part by the prompting of international organizations like the OECD. This has the potential to “drive up” the value and effectiveness of regulatory processes in developing nations, but their implementation in very different jurisdictions can lead to problematic outcomes. The globalization of trade also has an appreciable impact on OHS. There are ever increasing numbers of global supply chains where the most risky jobs are located in rapidly industrializing contexts (James et al. 2007). China provides a salutary lesson in this regard; in the scramble for traction in the new world economy, the maximization of economic growth has taken

C

846

precedence over the enforcement of effective safety regulations. Chinese industry seeks to undercut production costs (and hence safety standards) elsewhere in the world and, in doing so, sets the floor for workplace conditions in competing developing nations. Governments of the Global South thus prioritize the need to attracting inward investment to spur economic growth over demands for improvement in safety levels. The challenge of OHS from a global perspective contains similar complexities to those reflected in the discussion earlier in this entry, in that harm is embedded within the benefits of international trade, including employment and income for individuals, extended families, and local communities. But this should not excuse the behavior of multinational corporations which scour the globe in order to find where production is cheapest and safety and environmental standards low. The strength and weaknesses of various industrializing states under global competition needs careful analysis, however. It should not be assumed that states such as Thailand, China, or India have no room to develop an effective OHS regime; indeed, local responses that are sensitive to local conditions can be more effective than those simply transplanted from the Global North (Haines 2005). Hence, a failure, or success, in improving safety needs to be evaluated in light of local political contests in addition to the impact of global economic conditions.

Conclusion This entry has argued that the connections between processes of criminalization and occupational health and safety need to be understood within the historical context of the recurrent struggles that have always accompanied law reform in this area. The ambiguous legal standing of OHS law is a result of the struggles in defining what is seen as acceptable within the workplace environment, as well as perceptions of the law’s capacity to effectively secure acceptable behavior on the part of the regulated. But at the same time, regulation here is about the value of human life and the capacity for one person to place another at risk.


It is about the pursuit of a fair and just system of economic citizenship, and there are important debates about how best to advance these values and the role that the notionally “moral” criminal law should play in this process. Is it more important to try and reinforce the normative value of health and safety regulation via the communicative power of the criminal law or to avoid doing so because of the disruptive, individualizing, and distracting impact that such action can have? The contest around criminalization should not simply be one around the need for industrial manslaughter but to understand the history of OHS law in different contexts and the nature of criminal law within the workplace itself. The struggles and contests also reflect broader economic, political, and social change. Just as prescription can be seen as a result of various contests, so too the Robens model’s emphasis on the consensus of interests needs to be understood within the context of the welfare state in Britain in the 1970s. It is not surprising that this expansive view has been challenged by the rise of neoliberalism and the regulatory state. But this is not simply a return to the past. Hence, the importance and effectiveness of health and safety representatives and committees depends critically upon local context. In one area, it can represent a very real opportunity for representation of those whose lives are most at risk. Yet, in another, it can signal the return of an emphasis on individual responsibility and victim blaming. These struggles and contests are reflected in attempts to introduce new criminal offenses and penalties to sanction egregious failing by employers for their negligence. The push to criminalize is met with resistance, which in turn is met on occasions with capitulation but also with creativity. It is in this political milieu that multiple techniques have been adopted to try and raise standards. These need careful scrutiny. In particular, the same initiative can work quite differently depending on economic and political context – including whether strong unions that priorities safety are, or are not, present. The constantly changing nature of production and consumption, too, places significant challenges in the path of improving standards. There is now


much greater complexity to relations between businesses and a real need for OHS to be included in relations between businesses, not just within the confines of one firm. Finally, economic globalization and the presence of supply chains for particular goods and services spanning the globe make law reform and adequate enforcement of standards difficult. It is a context, however, in which a criminological analysis that is cognizant of the economic and political dimensions of the problem has much to offer.

Related Entries ▶ Compliance and Corporate Crime Control ▶ Corporate Crimes and the Business Cycle ▶ Corporate Liability ▶ Environmental Regulation and Law Enforcement ▶ Victims of Corporate Crime

Recommended Reading and References Almond P (2013) Corporate manslaughter and regulatory reform. Palgrave Macmillan, London Almond P, Colover S (2010) Mediating punitiveness: understanding public attitudes towards work-related fatality cases. Eur J Criminol 7(5):1–16 Ayres I, Braithwaite J (1992) Responsive regulation: transcending the deregulation debate. Oxford University Press, New York Carson WG (1979) The conventionalization of early factory crime. Int J Sociol Law 7(1):37–60 Carson WG (1980) The institutionalization of ambiguity: early British factory acts. In: Stotland E, Geis G (eds) White collar crime: theory and research. Sage, Beverly Hills, pp 142–173 Curran DJ (1993) Dead laws for dead men: the politics of federal coal mine health and safety legislation. University of Pittsburgh Press, Pittsburgh EUROSTAT (2010) The social situation in the European Union 2009. Publications Office of the European Union, Luxembourg Fisse B, Braithwaite J (1993) Corporations, crime and accountability. Cambridge University Press, Cambridge Gobert J (2005) The politics of corporate manslaughter – the British experience. Flinders J Law Reform 8(1):1–38 Gray GC (2009) The responsibilization strategy of health and safety: neo-liberalism and the reconfiguration of individual responsibility for risk. Br J Criminol 49(3):326–343

847

C

Gunningham N, Sinclair D (2009) Organizational trust and the limits of management based regulation. Law Soc Rev 43(4):865–900 Haines F (1997) Corporate regulation: beyond ‘punish or persuade’. Clarendon, Oxford Haines F (2005) Globalization and regulatory character: regulatory reform after the Kader toy factory fire. Ashgate, Dartmouth Haines F (2011) The paradox of regulation: what regulation can achieve and what it cannot. Edward Elgar, Cheltenham H€am€al€ainen P, Saarela KL, Takala J (2009) Global trends according to estimated number of occupational accidents and fatal work-related diseases at region and country level. J Safety Res 40:125–139 Hawkins K (2002) Law as last resort: prosecution decision-making in a regulatory agency. Oxford University Press, Oxford Hutter B (1997) Compliance: regulation and environment. Clarendon, Oxford James P, Walters D (2002) Worker representation in health and safety: options for regulatory reform. Ind Relat J 33(2):141–156 James P, Johnstone R, Quinlan M, Walters D (2007) Regulating supply chains to improve health and safety. Ind Law J 36(2):163–187 Johnstone R (2007) Are occupational health and safety crimes a hostage to history? An Australian perspective. In: Brannigan A, Pavlich G (eds) Governance and regulation in social life (Essays in honour of W.G. (Kit) Carson). Cavendish-Routledge, London, pp 33–54 Kagan RA, Scholz JT (1984) The ‘criminology of the corporation’ and regulatory enforcement strategies. In: Hawkins K, Thomas JM (eds) Enforcing regulation. Kluwer-Nijhoff, Boston, pp 67–96 La Montagne A, Keegal T, Louie A, Ostry A, Landsbergis P (2007) A systematic review of the job stress intervention evaluation literature 1990–2005. Int J Occup Environ Health 13(3):268–280 Marmot MG, Smith GD (1991) Health inequalities amongst British civil servants: the Whitehall II study. Lancet 337(8754):1387–1401 Robens L (1972) Safety and health at work: report of the committee (‘The Robens Report’). HMSO, London Silbey S (2009) Taming Prometheus: talk about safety culture. Annu Rev Sociol 35:341–369 Thomas MW (1948) The early factory legislation: a study in legislative and administrative evolution. Thames Bank, Leigh-on-Sea Tombs S, Whyte D (2007) Safety crimes. Willan, Cullompton Tucker E (1995) And defeat goes on: an assessment of third-wave health and safety regulation. In: Pearce F, Snider L (eds) Corporate crime: contemporary debates. University of Toronto Press, Toronto Walters D (1996) Health and safety strategies in Europe. J Loss Prevent Process Industr 9(5):297–308 Wells C (2001) Corporations and criminal responsibility, 2nd edn. Oxford University Press, Oxford

C

C

848

Criminological Theory ▶ Juvenile Violence

Criminology of Place David Weisburd1,2, Elizabeth R. Groff3, Sue-Ming Yang4 and Cody W. Telep5 1 Department of Criminology, Law and Society, George Mason University, Fairfax, VA, USA 2 Faculty of Law, The Hebrew University, Mt. Scopus, Jerusalem, Israel 3 Temple University, Philadelphia, PA, USA 4 Department of Criminology, National ChungCheng University, Min-Hsiung, Chia-Yi, Taiwan 5 School of Criminology and Criminal Justice, Arizona State University, Phoenix, AZ, USA

Overview The primary focus of research in criminology has traditionally been on individuals and why they become involved in crime (Eck and Eck 2012; Sherman 1995). Weisburd and Piquero (2008), for example, found that studies that sought to predict crime were more likely to focus on individuals than any other units of analysis. Examining articles in the journal Criminology that tried to model crime, they found that 94 of 169 articles (55.6 %) published between 1968 and 2005 used the individual as the unit of analysis. Other units commonly found in the study included cities or counties (14 %) and neighborhoods (9 %). More recently, Eck and Eck (2012) reviewed all of the research papers in Criminology and Public Policy since its inception and found that none addressed crime at place. In recent decades, however, a growing number of scholars have begun to take a very different approach to the crime problem. This approach begins not with the people that commit crime but rather with the places where crime occurs. This is a radical reconceptualization of the crime

Criminological Theory

problem, but one that that is warranted by the reality of crime in the city. Why is it necessary to reconsider the “person-focused” crime and justice model of the last century? The main reason is simply that the yield of this approach has been questioned for more than three decades. Weisburd and Piquero’s (2008) review found that despite the focus primarily on individuals, prediction levels are fairly low and have remained low over time. Lawrence Sherman and colleagues (1989) categorized this new approach as the “criminology of place.” The focus here is on very small geographic areas within cities, often as small as addresses or street segments, and understanding their contribution to the crime problem. It focuses attention on “hot spots of crime,” or crime concentrations in such micro geographic areas and the factors that explain these concentrations over time, as well as the best ways to address these high crime locations. Drawing upon work by Weisburd, Groff, and Yang (2012), this entry will review the emergence of the criminology of place in recent decades and theoretical perspectives for understanding why places are such an important part of the crime equation. This entry will also review the main findings from a longitudinal study in Seattle, Washington, by Weisburd et al. (2012) examining the factors that contribute to why streets are high crime (or low crime) over time.

The Emergence of the Criminology of Place Criminologists have traditionally sought to understand why certain people as opposed to others become criminals (e.g., see Gottfredson and Hirschi 1990) or to explain why certain offenders become involved in criminal activity at different stages of the life course or cease involvement at other stages (e.g., see Laub and Sampson 2003). Criminologists have also been interested in places, but they have historically focused on larger geographic areas like neighborhoods or even larger areas like cities or states (see Weisburd et al. 2009). For example,

Criminology of Place

criminologists have often tried to explain why certain types of crime or different levels of criminality are found in some communities as contrasted with others (e.g., see Bursik and Grasmick 1993). While the individual and “macro” units of place, such as the community, have long been a focus of crime research and theory, only recently have criminologists begun to explore crime at very small “micro” units of geography. The roots of such approaches can be found in the efforts of scholars to identify the relationship between specific aspects of urban design (Jeffery 1971) or urban architecture (Newman 1972) and crime, but broadened to take into account a much larger set of characteristics of physical space and criminal opportunity (e.g., Brantingham and Brantingham 1981). These studies drew important distinctions between the site in question and the larger geographical area (such as the neighborhood, community, police beat, or city) that surrounds it. But the key to the origins of the criminology of place is a group of emerging theoretical perspectives that developed as a reaction to the limitations identified in offender-based criminology in the 1970s (e.g., see Martinson 1974). In a groundbreaking article on routine activities and crime, for example, Cohen and Felson (1979) suggested that a fuller understanding of crime must include a recognition that the availability of suitable crime targets and the presence or absence of capable guardians influence crime events. Cohen and Felson were to turn traditional conceptions of the crime problem on their head by suggesting that crime could be prevented without changing the supply or motivation of offenders in society. In the routine activities model, victims, offenders, and guardians were all essential parts of the crime equation. While traditional conceptions had focused on offenders, using data on crime rates in the United States in the postWorld War II period, Cohen and Felson (1979) illustrated that changes in other parts of the crime equation could impact the level of crime in society. For example, they found that changes in the value of property or the ease of which it could be

849

C

stolen affected burglary rates. Decreasing size and weight of electronic devices (e.g., televisions) and the increasing value of goods found in homes were found to be related to increases in victimization. They also found that changes in the presence of “capable guardians” influenced crime. In this case, the entrance of women into the work force and the resulting lack of “guardianship” at homes were associated with significant increases in house burglaries. The routine activities perspective pushed criminologists to extend their vision beyond the traditional concern with the causes of criminality and their focus on offenders. Indeed, Cohen and Felson (1979) took a pool of motivated offenders as a given and focused their attention more on the suitability of targets and the presence or absence of capable guardianship. Having introduced other elements of the crime equation, Cohen and Felson’s (1979) work naturally led to a focus on crime opportunities. Victims, offenders, and guardians are likely to intersect in physical space. The spatial component of crime thus became a key component of this perspective. As Cohen and Felson (1979, 589) note, “we take criminal inclination as given and examine the manner in which the spatiotemporal organization of social activities helps people to translate their criminal inclinations into action. Criminal violations are treated here as routine activities which share many attributes of, and are interdependent with, other routine activities.” Researchers at the British Home Office, in a series of studies examining “situational crime prevention,” also challenged the traditional focus on offenders and communities (Clarke 1983). Situational crime prevention moved the focus of the crime equation away from the people who commit crime and instead considered the context of criminal events to be critical. At the core of situational prevention is the concept of rational choice (Clarke and Cornish 2001). In contrast to offender-based approaches to crime prevention, which usually focus on the dispositions of criminals, situational crime prevention begins with the opportunity structure of the crime situation (Clarke 1995). Such opportunities affect whether

C

C

850

potential offenders will choose to commit crime. Importantly, such choices are seen to be the result of a decision employing “bounded rationality” (Clarke and Cornish 2001). That is, potential offenders use the limited knowledge about victims or guardians in specific situations that is easily available to consider relative costs and benefits. Situational crime prevention is concerned with the “opportunity structures” of specific contexts and places. By opportunity structure, advocates of this perspective are not referring to the broad societal structure of opportunities that underlie individual motivations for crime but to the immediate situational components of the context of crime. In this context, crime prevention may involve efforts as simple and straightforward as target hardening or access control and often follows a straightforward and commonsense notion of how to deal with crime problems (Clarke 1995). Importantly, place at a “micro” level is key to situational crime prevention theory since it focuses on the immediate opportunities for crime, which are generally structured within very small geographic areas. Around the same time as routine activities theory and situational crime prevention developed, Paul and Patricia Brantingham published their seminal book Patterns in Crime, which emphasized the role of place characteristics and human activity in shaping the type and frequency of human interaction (Brantingham and Brantingham 1984). Environmental criminology, also known as crime pattern theory, explores the distribution and interaction of targets, offenders, and opportunities across time and space (Brantingham and Brantingham 1981). Both rational choice and situational opportunities play a key role in crime pattern theory. The concept of place, in turn, is essential to crime pattern theory. Not only are places logically required (an offender must be in a place when an offense is committed) but also their characteristics are seen to influence the likelihood of a crime. Crime pattern theory links places with desirable targets and the context within which they are found by focusing on how places come to the attention of potential offenders.


The Importance of Examining Crime at a Micro Geographic Level One implication of these emerging perspectives was that places at a “micro” geographic level should be an important focus of criminological inquiry. While concern with the relationship between crime and place is not new and indeed goes back to the founding generations of modern criminology, the “micro” approach to places suggested by recent theories had just begun to be examined by criminologists (see Weisburd et al. 2009). Places in this “micro” context are specific locations within the larger social environments of communities and neighborhoods (Eck and Weisburd 1995). They are sometimes defined as buildings or addresses (e.g., see Sherman et al. 1989); sometimes as block faces, “hundred blocks,” or street segments (e.g., see Weisburd et al. 2004); and sometimes as clusters of addresses, block faces, or street segments (see Braga et al. in press). The criminology of place has from the outset fit its units of analysis to its theoretical interests. Routine activities theory (Cohen and Felson 1979), situational crime prevention (Clarke 1983), and environmental criminology (Brantingham and Brantingham 1981) naturally push the cone of geographic resolution to very low levels. Importantly, however, while pushing the geographic study of crime to much lower levels than traditional criminology, these theories did not specify a precise geographic unit for studying crime. At what level should crime at place be studied? There is an important trend over time toward the study of crime at place at smaller units of geography. But does that trend reflect a fact about the level of geography that is important for understanding crime, or is it simply a result of the specific data available and theoretical interests of scholars? Or is the best approach one that is eclectic in its understanding of crime at place? While the relevance of studying varying geographic units in coming to a more complete understanding of crime at place should not be discounted, it is important to recognize at the outset that studying crime at the “wrong”


geographic unit may lead to a very misleading portrait of how place and crime interact. In this context, examining crime patterns at larger geographic levels, even commonly used “smaller” units such as census tracts or census block groups, may mask significant lower order geographic variability. Importantly, beginning with a micro-level approach allows the researcher to examine the influences of larger geographic units, while starting at higher levels of geography may preclude examination of local variability. Collecting data at the lowest geographic level, or smallest units of analysis, allows one to examine how those trends relate to larger units of analysis at higher levels of geography. But data collection at larger units would not allow conversion to more micro units of analysis. For this reason, Weisburd et al. (2012) used a micro geographic unit – the street segment – as the basic unit of analysis in their study of Seattle. The street segment, however, is not the smallest unit of geography that can be used to study crime. Some scholars have argued for the importance of examining crime patterns across individual addresses or facilities (e.g., see Sherman et al. 1989). Policing data may include significant coding errors regarding a crime’s occurrence at the address level, and so one can be much more confident in the accuracy of crime incident data at the street-segment level, since getting the street of a crime’s occurrence wrong is much less likely than mistaking the specific address location for a crime. There is also strong theoretical justification for the identification of street segments as key behavior settings for crime (see below).

Social Disorganization and Opportunity Theories: Recognizing the Importance of Theoretical Integration Criminology of place scholars have looked primarily to what can be termed “opportunity theories” (Wilcox et al. 2003) as an explanation for why crime trends vary at places and as a basis for constructing practical crime prevention approaches (e.g., see Sherman et al. 1989;

851

C

Weisburd et al. 2004). Routine activities theory, situational crime prevention, and crime pattern theory all place great emphasis on the specific opportunities offered by specific places and situations. In contrast, study of crime at higher geographic levels has placed emphasis on the social characteristics of places, for example, the socioeconomic levels of people who live in certain areas (e.g., Bursik and Grasmick 1993) or the degree to which there is strong population heterogeneity (Shaw and McKay 1942). Such perspectives may be grouped more generally as social disorganization theories (see Sampson and Groves 1989). These themes are repeated again and again in traditional macro studies of place and crime. Scholars who study the criminology of place have virtually ignored social disorganization theories in empirical analysis and theoretical discussion. In one sense, this is understandable, since the impetus for study of micro crime places came from opportunity theories. Such theories justified examination of small geographic units because of their emphasis on the specific situations and contexts that make crime possible. In this context, the neglect of social disorganization in the study of the criminology of place can be traced to what some scholars have called “theoretical competition” (Bernard and Snipes 1996). All theories cannot be right in this perspective, and accordingly it is the job of criminologists to advance a single theoretical paradigm that can explain the phenomenon at hand. But a different perspective is interested in “theoretical integration.” As Bernard and Snipes (1996, 302) write, “(w)e argue that integration is the appropriate approach because the theories primarily make different but not contradictory predictions.” In the context of theoretical integration, one would draw from multiple theories that can increase overall understanding of the crime problem at micro levels of place. Theoretical integration of opportunity and social disorganization theories in a model for understanding crime at street segments would not make sense if social disorganization was a concept that is irrelevant to the criminology of place. This seems to be the position of many

C

C

852

scholars in this area. Sherman et al. (1989, 30), for example, argued in introducing the idea of a criminology of place, that “(t)raditional collectivity theories [termed here as social disorganization theories] may be appropriate for explaining community-level variation, but they seem inappropriate for small, publicly visible places with highly transient populations.” Often this position is taken indirectly, simply by ignoring social disorganization theories (e.g., Eck and Weisburd 1995). Sometimes scholars have specifically recognized the potential relevance of social disorganization to the understanding of variations in crime at micro geographic levels (e.g., Smith et al. 2000). However, such interest is the exception rather than the rule. There is reason to believe that social disorganization theories are relevant at both micro- and macro-level geographies. Street segments do not simply represent physical entities. They are also social settings, or following Wicker (1987, 614) “behavior settings,” which can be seen as “smallscale social systems.” In this context, street segments can be seen as examples of smallscale communities (see Taylor 1997). People who frequent a street segment get to know one another and become familiar with each other’s routines. Residents develop certain roles they play in the life of the street segment (e.g., the busybody, the organizer). Norms about acceptable behavior develop and are generally shared. Blocks have standing patterns of behavior, for example, people whose routines are regular like the mail carrier or the shop owner. In this context, we can see street segments as “micro communities” as well as “micro places.” They have many of the traits of communities that have been seen as crucial to social disorganization theory, in that these physical units function also as social units with specific routines. If the street segment can be seen as a type of “micro community,” then social disorganization theory has direct relevance to the criminology of place (see Rice and Smith 2002; Smith et al. 2000; Taylor 1997). For example, street segments, like communities, are often dynamic with people moving in and out as well as shops opening and closing. Such transitions have often


been seen to represent heightened social disorganization in studies of communities. Similarly, poverty and social disadvantage have been identified as strongly related to crime at higher levels of geography. Is variability in wealth or social class at the street-segment level also related to trends in crime? Do characteristics reflecting social disorganization vary at a streetsegment level of analysis, as they do across communities and neighborhoods? There are critical empirical questions addressed in the Weisburd et al. (2012) study discussed below.

Predicting Crime at Place: Findings from a Longitudinal Study Weisburd et al. (2012) recently focused on the criminology of place in an effort to advance both theory and practical crime prevention in this area. They examined crime incidents at the streetsegment level over a 16-year period in Seattle, Washington, assessing the concentration of crime over time and the geographic variability of crime across the city. They then combined data on a number of physical and social characteristics of the street segments, using variables from both opportunity and social disorganization theories to better understand what factors help explain the distribution of crime at the street level. Their study overall reached five major conclusions that are all discussed in more detail below: 1. Crime is tightly concentrated at crime hot spots, suggesting that one can identify and deal with a large proportion of crime problems by focusing on a very small number of places. 2. These crime hot spots evidence very strong stability over time and thus present a particularly promising focus for crime prevention efforts. 3. Crime at places evidences strong variability at micro levels of geography, suggesting that an exclusive focus on higher geographic units, like communities or neighborhoods, will lead to a loss of important information about crime. 4. It is not only crime that varies across very small units of geography, but it is also the social and contextual characteristics of places.


The criminology of place in this context identifies and emphasizes the importance of micro units of geography as social systems relevant to the crime problem. 5. Crime at place is very predictable, and therefore it is possible to not only understand why crime is concentrated at place but also to develop effective crime prevention strategies to ameliorate the crime problem at place. The Tight Coupling of Crime at Place Criminologists have traditionally assumed that crime is “loosely coupled” to place. The idea of loose and tight “coupling” has been used in many disciplines to identify the extent to which parts of systems are linked or dependent to one another. What is meant here by “loose coupling” of crime at places is that criminologists have traditionally not seen the bonds that tie crime to place as very strong, even though it has been clear from the outset that crime occurs in specific settings. Traditionally, criminologists have viewed crime opportunities provided by places as too numerous to make crime prevention strategies targeting specific places of utility for theory or policy. Recent studies point to the potential theoretical and practical benefits of focusing research on crime places and suggest a “tight coupling” of crime with the places where crime occurs. A number of studies, for example, suggest that there is a very significant clustering of crime at places, irrespective of the specific unit of analysis that is defined. The extent of the concentration of crime at place is dramatic. In one of the pioneering studies in this area, Sherman et al. (1989) found that only about 3 % of the addresses in Minneapolis, Minnesota, produced 50 % of all calls to the police. Fifteen years later, in a study in Seattle, Washington, Weisburd and colleagues (2004) reported that between 4 % and 5 % of street segments in the city accounted for 50 % of crime incidents for each year over 14 years. Weisburd et al. (2012) found that about 50 % of crime is found at just 5–6 % of street segments each year in Seattle over 16 years. More than 20 % of crime incidents were found at just 1 % of street segments.

853

C

Research also demonstrates that crime is not only concentrated at a small number of places but also that these concentrations remain stable over time, also reinforcing the idea of tight coupling. Weisburd et al. (2012) found similar levels of concentration for each of the 16 years they observed crime at street segments in Seattle. It is in some sense startling that over 16 years, about the same number of street segments produce about the same proportion of crime, especially since crime declined by more than 20 % in the city during this period. The 1 % of street segments that produce about 23 % of total crime in Seattle began the observation period as the most serious hot spots in Seattle and remained hot spots throughout the time period observed. The relative intensity of crime at these places and the stability over time of this developmental pattern suggest that there are specific characteristics of these places that generate or attract crime. Reflecting a similar underlying structure of tight coupling between crime and place was the more than 80 % of the street segments in the city which had very little or no crime throughout the 16-year period. Here, there would seem to be factors that help places resist or discourage crime. Street-by-Street Heterogeneity in Crime While a number of theoretical approaches in recent decades stress the importance of using small units of geography, it is important to demonstrate empirically that a focus on micro places is relevant to understanding the distribution of crime. That is, it could be the case that crime hot spots at a micro geographic level are just proxies for larger community-level hot spots. One could imagine a scenario, for example, where all the hot spots in a city are concentrated in a single neighborhood, which would suggest that neighborhood-level processes might be more relevant to consider. Weisburd et al. (2012) used a series of geographic analyses to demonstrate that the processes explaining crime patterns at street segments do not come primarily from higher geographic influences such as communities. There were indications of the influence of higher

C

C

854

order trends in the data, for example, in the fact that higher crime street segments were not distributed at random and were more likely to be closer to each other than would be predicted simply by chance. But these indications of macro geographic influences were much outweighed by evidence of the importance of looking at crime at the micro level (i.e., street segments). There is strong street-to-street variability in crime patterns in the data, and such variability emphasizes the importance of studying crime at place at a micro unit of analysis. Predicting Hot Spots of Crime Weisburd et al. (2012) were the first to examine the distribution of opportunity and social disorganization factors at the street-segment level over time. The key question here was whether or not variables in these two perspectives were concentrated at street segments. That is, are there hot spots of crime opportunities and hot spots of social disorganization at the street-segment level? And if so, are these hot spots found throughout the city, as crime hot spots are? They examined four main dimensions of opportunity theory: (1) motivated offenders, (2) suitable targets, (3) guardianship, and (4) accessibility/urban form. Opportunity measures were generally highly concentrated and evidenced strong variability across places. For example, 50 % of high risk juveniles (a proxy for “motivated offenders”) were consistently found on between 3 % and 4 % of the total number of Seattle street segments. In turn, half of all the employees (a proxy for “suitable targets”) in the city were located on less than 1 % of Seattle street segments. These concentrations of opportunity factors also remained fairly stable over time, suggesting one reason for the stability of crime patterns longitudinally. Weisburd et al. (2012) also found tremendous concentration and variation in most of the social disorganization measures they examined. While prior theory suggested that opportunity factors would be highly concentrated, the results for social disorganization were more surprising. Looking both at structural and mediating


variables of social disorganization, they found that there are hot spots of social disorganization at the street-segment level. For example, social disorder is one of the most direct indicators of the inability of a social system to control behavior. Fifty percent of reports of physical disorder in Seattle were found on between 1.5 % and 3 % of street segments. Fully 50 % of truant students (a mediating variable used as a proxy for “unsupervised teens”) were consistently found to live on between 2 % and 3.5 % of the total street segments during the study period. And these hot spots were not simply part of contiguous hot spots at larger geographic levels. They were not found only in specific neighborhoods; rather they were distributed across the city landscape. This suggests that a perspective that has generally been seen as relevant only higher levels of geography shows concentration and variability at the street-segment level. How do these findings relate to crime concentrations in the city? Weisburd and colleagues (2012) used a multinomial logistic regression to examine the association between opportunity and social disorganization factors and streetsegment crime trends. The overall model examined the influence of these factors on eight patterns of trajectories of crime. The focus here is on the comparison between the chronic street segments (that 1 % of street segments responsible for over 20 % of crime) and the approximately 80 % of street segments that were largely crime free over 16 years. How do opportunity and social disorganization variables help to differentiate whether a street segment is likely to be crime free or a chronic crime hot spot? The findings from the multinomial regression analysis overall suggest a high degree of statistical fit between theories of opportunity and social disorganization and crime at street segments. Using a pseudo R2 measure suggested by Nagelkerke, for example, Weisburd et al. (2012) found that the model explained about 68 % of the variance in pattern membership. This suggests that crime at place is very predictable, and even in a fairly early stage of theorizing and empirical research, predicting crime at place can be done with more


precision than most models predicting crime in individuals (see Weisburd and Piquero 2008). This also suggests the relevance of an “integrated theoretical approach” (Bernard and Snipes 1996) to the criminology of place. Both opportunity measures and social disorganization measures provide important information for understanding variation in developmental patterns of crime at place. The multinomial regression results suggested a number of opportunity and social disorganization variables significantly predict whether a street segment is likely to be a crime hot spot compared to being crime free. The two most important predictors of crime hot spots were drawn from the opportunity perspective. Both related to the suitability of places as crime targets. The larger the residential population of a street segment, the more likely it was to be a crime hot spot. Similarly, when there were more employees on a street segment, it was much more likely to become a hot spot of crime. These findings reinforce a more general conclusion of the opportunity perspective. The factors that increase the risks of crime relate directly to the situational opportunities that places present. When more people live on a block, there is more potential for crime because there are more potential victims (and perhaps higher numbers of motivated offenders). When more employees work on the block, they increase the amount of crime on that street, because they are likely to increase the volume of targets on the street. Another indicator of crime opportunities, arterial roads, also increased the risks of crime on street segments markedly. Arterial roads were much more likely to become crime hot spots. Arterial roads are more likely to bring together motivated offenders and suitable targets, because they are easily accessible. Three variables that reflect the social disorganization perspective also had very large impacts on the likelihood of a street segment being a crime hot spot. The higher the level of physical disorder on a street segment, the greater the likelihood of it being a high crime street segment. This is very much consistent with the idea that an

855

C

inability of the small social systems to control disorderly behavior will increase the risk of crime. High socioeconomic status, in contrast, acted as a protective factor for crime, with crime hot spots much less likely to be found at places with wealthier residents (as reflected in higher property values) who assumedly are able to bring into play both formal and informal social controls more effectively. Perhaps most interesting was the very strong impact of collective efficacy (see Sampson et al. 1997), as reflected by voter participation. Collective efficacy seems to act as a strong protective factor for crime at place. Accordingly, the direct situational opportunities that increase crime risk are one part of the crime equation at places, but social factors that act to insulate places from crime risks are also important. Implications If crime is strongly concentrated at place, and such concentrations are stable over long periods, then it is possible to assume that a focus on crime hot spots can have important crime prevention benefits. The concentration of crime suggests that police or other crime prevention agents can focus on a very small number of targets and have a large impact on crime problems overall. The fact that crime hot spots are stable across long periods reinforces the potential utility of such approaches, because it implies that if crime prevention agents did not intervene, the problem would persist. These are the implications of the tight coupling of crime and place that have been observed in prior studies. But the potential for the criminology of place in crime prevention is not just theoretical. These basic research findings have been reinforced by evaluations of practical crime prevention efforts. A series of randomized field trials demonstrate that police efforts focused on hot spots can result in meaningful reductions in crime and disorder (see Braga and Weisburd 2010). Braga et al. (in press) found in their Campbell Collaboration systematic review of the hot spots literature that 20 of 25 tests from 19 experimental or quasiexperimental evaluations of hot spots policing

C

C

856

reported noteworthy crime or disorder reductions. This strong body of rigorous evaluations led the National Research Council Committee to Review Research on Police Policy and Practices (2004, 35) to conclude “a strong body of evidence suggests that taking a focused geographic approach to crime problems can increase the effectiveness of policing.” Importantly, Braga et al. (in press) also reported little evidence of crime displacement as a result of hot spots programs. Crime did not simply move around the corner as a result of police intervention. But if important causal mechanisms underlying developmental patterns of crime at place can be found in factors such as economic deprivation or collective efficacy, then a much broader set of social interventions may also be required to change the trajectories of crime at crime hot spots. The focus on the specific places where crime problems are found provides an opportunity to “lower the scale” of social interventions and accordingly to make such interventions relevant to crime prevention practitioners. It is one thing to attempt change in the social conditions of an entire neighborhood or city. It is another to try to ameliorate problems on specific blocks. Perhaps the criminology of place provides a scale of intervention that can rekindle interest in the importance of social and structural interventions in doing something about crime. The work of Weisburd et al. (2012) suggests the importance of focusing crime prevention, whether it is at the level of local police agents or in terms of the development of social programs for hot spots of crime. Whatever the approach that is taken, it is time to recognize the need to focus crime prevention resources at micro places such as street segments. It is important to note that despite the wealth of data Weisburd and colleagues (2012) were able to gather on street-segment characteristics, they were limited because of retrospective data collection. These data availability issues suggest the importance of a prospective longitudinal study of crime at place that would capture at specific times both the characteristics of places and people (see Weisburd, Lawton, and Ready 2012). The National Institute on Drug Abuse within the National Institute of Health recently


provided funding for such a longitudinal study to take place. Weisburd and colleagues were awarded over $3 million to undertake a 5-year study of how living in drug or violent crime hot spots influences personal health, mental health, HIV and sexually transmitted disease rates, safe sex practices, drug use, crime, and other antisocial behaviors. The study will also develop knowledge on why places become drug or violent crime hot spots and how characteristics of street segments and their residents impact upon developmental trends of health, drug use, and crime. Results from this study should provide exciting new insights into the criminology of place.

Related Entries ▶ Geography of Crime and Disorder ▶ History of Geographic Criminology Part I: Nineteenth Century ▶ History of Geographic Criminology Part II: Twentieth Century ▶ Hot Spots and Place-Based Policing ▶ Law of Crime Concentrations at Places ▶ Longitudinal Crime Trends at Places

Recommended Reading and References Bernard TJ, Snipes JB (1996) Theoretical integration in criminology. In: Tonry M (ed) Crime and justice: a review of research, vol 20. University of Chicago Press, Chicago, pp 301–348 Braga AA, Weisburd DL (2010) Policing problem places: crime hot spots and effective prevention. Oxford University Press, New York Braga AA, Papachristos AV, Hureau DM (in press) The effects of hot spots policing on crime: an updated systematic review and meta-analysis. Justice Q Brantingham PJ, Brantingham PL (eds) (1981) Environmental criminology. Sage, Beverly Hills Brantingham PJ, Brantingham PL (1984) Patterns in crime. Macmillan, New York Bursik RJ Jr, Grasmick HG (1993) Neighborhoods and crime: the dimensions of effective community control. Lexington Books, New York Clarke RV (1983) Situational crime prevention: its theoretical basis and practical scope. In: Tonry M, Morris N (eds) Crime and justice: a review of research, vol 14. University of Chicago Press, Chicago, pp 225–256

Critical Criminology Clarke RV (1995) Situational crime prevention. In: Tonry M, Farrington DP (eds) Building a safer society: strategic approaches to crime prevention. Crime and justice: a review of research, vol 19. University of Chicago Press, Chicago, pp 91–150 Clarke RV, Cornish DB (2001) Rational choice. In: Paternoster R, Bachman R (eds) Explaining criminals and crime. Roxbury Publishing, Los Angeles Cohen LE, Felson M (1979) Social change and crime rate trends: a routine activity approach. Am Sociol Rev 44:588–608 Eck JE, Eck EB (2012) Crime place and pollution: expanding crime reduction options through a regulatory approach. Criminol Public Policy 11(2):281–316 Eck JE, Weisburd D (1995) Crime places in crime theory. In: Eck JE, Weisburd D (eds) Crime and place. Crime prevention studies, vol 4. Willow Tree Press, Monsey, pp 1–33 Gottfredson M, Hirschi T (1990) A general theory of crime. Stanford University Press, Stanford Jeffery CR (1971) Crime prevention through environmental design. Sage, Beverly Hills Laub JH, Sampson RJ (2003) Shared beginnings, divergent lives. Delinquent boys to age 70. Harvard University Press, Cambridge Martinson R (1974) What works? Questions and answers about prison reform. Publ Interest 35:22–54 National Research Council (2004) Fairness and effectiveness in policing: the evidence. Committee to review research on police policy and practices. In: Skogan W, Frydl K (eds) Committee on Law and Justice, Division of Behavioral and Social Sciences and Education. National Academies Press, Washington, DC Newman O (1972) Defensible space: crime prevention through environmental design. Macmillan, New York Rice KJ, Smith WR (2002) Sociological models of automotive theft: integrating routine activity and social disorganization approaches. J Res Crime Delinq 39:304–336 Sampson RJ, Groves WB (1989) Community structure and crime: testing social-disorganization theory. Am J Sociol 94:774–802 Sampson RJ, Raudenbush SW, Earls F (1997) Neighborhoods and violent crime: a multilevel study of collective efficacy. Science 277:918–924 Shaw CR, McKay HD (1942) Juvenile delinquency and urban areas. A study of rates of delinquency in relation to differential characteristics of local communities in American cities. University of Chicago Press, Chicago Sherman LW (1995) Hot spots of crime and criminal careers of places. In: Eck J, Weisburd D (eds) Crime and place. Crime prevention studies, vol 4. Willow Tree Press, Monsey, pp 35–52 Sherman LW, Gartin P, Buerger ME (1989) Hot spots of predatory crime: routine activities and the criminology of place. Criminology 27:27–55 Smith WR, Frazee SG, Davison EL (2000) Furthering the integration of routine activity and social

857

C

disorganization theories: small units of analysis and the study of street robbery as a diffusion process. Criminology 38:489–523 Taylor RB (1997) Social order and disorder of street blocks and neighborhoods: ecology, microecology, and the systemic model of social disorganization. J Res Crime Delinq 34:113–155 Weisburd D, Piquero AR (2008) How well do criminologists explain crime? Statistical modeling in published studies. In: Tonry M (ed) Crime and justice: a review of research, vol 37. University of Chicago Press, Chicago, pp 453–502 Weisburd D, Bushway S, Lum C, Yang S-M (2004) Trajectories of crime at places: a longitudinal study of street segments in the city of Seattle. Criminology 42:283–321 Weisburd D, Bruinsma GJN, Bernasco W (2009) Units of analysis in geographic criminology: historical development, critical issues, and open questions. In: Weisburd D, Bernasco W, Bruinsma GJN (eds) Putting crime in its place: units of analysis in geographic criminology. Springer, New York, pp 3–31 Weisburd D, Groff ER, Yang S-M (2012) The criminology of place: street segments and our understanding of the crime problem. Oxford University Press, New York Weisburd D, Lawton B, Ready J (2012) Staking out the next generation of studies of the criminology of place: collecting prospective longitudinal data at crime hot spots. In: Loeber R, Welsh BC (eds) The future of criminology. Oxford University Press, New York, pp 236–243 Wicker AW (1987) Behavior settings reconsidered: temporal stages, resources, internal dynamics, context. In: Stokels D, Altman I (eds) Handbook of environmental psychology. Wiley-Interscience, New York, pp 613–653 Wilcox P, Land KC, Hunt SC (2003) Criminal circumstance: a dynamic multicontextual criminal opportunity theory. Walter de Gruyster, New York

Criminology of Places ▶ Criminal Careers of Places

Critical Criminology ▶ International Responses to Victims in Criminal Justice ▶ Marxist Criminology ▶ So What Criminology?

C

C

858

Critical Incidents Tom Mijares1 and Ron McCarthy2 1 Texas State University-San Marcos, San Marcos, TX, USA 2 Director Emeritus, National Tactical Officers Association, Los Angeles Police Department (Retired), Doylestown, PA, USA

Overview This entry begins with a brief look at the traditional approach to critical incident management by individual law enforcement officers, placing particular emphasis on the recognition by the judiciary that special circumstances require special responses in the form of highly selective processes for selecting, instructing, and equipping the personnel tasked with these duties in an increasingly more violent and complex society. This entry then describes how social and technological changes led to the development of the current approach to incident control and problem resolution. The authors also summarize the controversies coming from this approach which features small, well-trained, and well-disciplined groups working as a cohesive and coordinated unit without depleting the law enforcement agency’s ability to respond to more routine calls for police service. The entry describes projections for the future of critical incident management based on the direction, frequency, and magnitude of current trends. The authors conclude with recommendations for future operations.

Historical Patterns Traditional methods of American law enforcement have centered on the beat cop making his rounds on foot and later in a motorized vehicle to take the appropriate action through his own initiative and limited resources. Backup assistance and radio calls to the station for advice were technically impossible until the relatively

Critical Incidents

recent times of the middle twentieth century. Specialized response capabilities were initially limited to horse-mounted police officers and were improved later by the advent of the automobile and the “flying squads” of major city detective bureaus. Labor unrest during the 1930s demonstrated the need for specialized police units possessing the means for immediate deployment, containment, and arrest capabilities to deal with unusual circumstances such as barricaded suspects, demonstrations, strikes, and other forms of unrest in a rapidly changing society. In 1931 the National Commission on Law Observation and Enforcement, commonly referred to as the Wickersham Commission, had been charged with the responsibility of reporting on the enforcement of the Eighteenth Amendment (Prohibition). The Commission’s 14-volume series of reports probed deeply into the American criminal justice system to offer the causes and effects of the situation and to suggest possible ameliorative efforts (Bopp and Schultz 1972). As police administrators recognized the need to maintain group integrity and cohesion for these critical incidents, specialized units were developed. Despite the assumption of elitism in the creation and maintenance of these units, the following shortcomings of specialized operations had already been identified by the Wickersham Commission: 1. The selection process for most police appointments and assignments was often based on nepotism and favoritism. 2. Training was irregular, inconsistent, and incomplete. As a result, a greater amount of force was used than what may have been necessary. 3. No continuous and concerted effort was made in large departments to integrate these units with other sections of the police departments. 4. Little research was conducted to find improvements to existing techniques and technology. 5. Specific organizational orders were not developed to define the deployment procedures to be followed by specialized units, and the lines of authority, responsibility, accountability, and communications were obscure.

Critical Incidents

An Increase in Violence in a More Complex Society From the middle and late 1960s and early 1970s to the present, social and technical factors worked to change the complexion of law enforcement in general and tactical policing in particular. First, the war in Vietnam and other conflicts resulted in a quantum increase in the number and variety of weapons available for law enforcement applications. Unfortunately, criminal elements throughout the world had (and continue to have) access to the same technology without the legal, social, and economic constraints endured by law enforcement agencies (Dobson and Payne 1982). Second, America’s space exploration program and its by-products have also had an impact on the technology of law enforcement and criminality. Limited space for orbiting payloads resulted in smaller, more efficient radio communications systems. Landline call boxes and one-way radios in police cars were expanded to full two-way portable communications. Where law enforcement had traditionally been dependent on the unplanned and unrelated efforts of patrol officers acting individually, the improvement in communications allowed all involved law enforcement officers the opportunity to coordinate their efforts more effectively in surveillance, pursuits, highrisk warrant service, tactical operations, and other activities that optimally require several individuals to act as a unit. Third, the frequency and magnitude of special threats increased beyond the imagination of the beat cop of a few decades ago. Regardless of the size of a law enforcement agency’s jurisdiction, it is extremely naive for a police administrator to ignore the possibility of extreme violence. History has already shown that these events are not limited by demographic and geographic factors. Finally, the effects of technical sophistication were accelerated by the development of criminal groups from all points on the politicalphilosophical spectrum and a willingness to collaborate against a perceived common enemy, i.e., the “establishment.” What is particularly disturbing for law enforcement has been the

859

C

realization that irrespective of their philosophies and ultimate goals, some of these groups share the commonalities of impressive armories of military weapons, the sophisticated training necessary to employ them, a proclivity for violence, and a total commitment to their causes. The multifaceted composition of the American populace suggests that even more dissident groups will develop. Events such as the Watts Riot in 1965, the Texas Tower incident in 1966, and the Howard Johnson shootout in New Orleans in 1973 held several separate but interwoven ramifications for the conduct of future law enforcement operations. First, they demonstrated that law enforcement technology was not yet sufficient to deal with many current problems. No longer was military surplus equipment adequate or appropriate for domestic applications. The Texas Tower incident in particular established an awareness of the need for a more sophisticated approach to communications. Second, these incidents also led to an awareness of the need for firm, effective crowd and traffic control. Considering how numerous people could have wandered into the field of fire in these commercial and residential areas during these critical incidents, a major factor in the successful resolution of the incident may be attributed to the fact that the responding officers were able to isolate the perpetrators and that they made considerable efforts to block traffic from the area. Third, these events illustrated how the complexion of many types of crimes was changing. The availability of large-capacity firearms and the sometimes direct involvement of a great number of people, both as perpetrators and as victims, were examples of how criminal activity was evolving and how law enforcement was forced into modifying its approaches accordingly. Fourth, alterations in criminal activity could also be seen by changes in overall strategies and specific tactics employed by the perpetrators. For the first time in the modern history of American law enforcement, responding police forces were confronted with perpetrators who were actually willing to die for their cause.

C

C

860

Fifth, where news coverage of these events had previously been limited to next-day reporting and analysis, technical advances in the news and entertainment media provided immediate and onscene telecast capability. Critics of law enforcement were given a means to record and scrutinize the substance of what had occurred and how the responding police forces responded to the incident. Improved news media coverage fostered another phenomenon: religiously, philosophically, and politically differing groups such as the Symbionese Liberation Army, the Branch Davidians, and The Order often achieved public martyrdom among some segments of society for their actions. Sixth, responding to critical incidents on an ad hoc basis led to a breakdown in the normal chain of command. Without this necessary element of organizational management, most of the officers who had been assigned respond to critical incidents simply were forced to standby until appropriate orders were received. However, during the Watts Riot a very small number of officers from the Metro Division of the Los Angeles Police Department immediately recognized the confused, overly cautious, and politically dictated approach displayed by the command staff. These officers had been assigned to the Division in part because of their long-established ability to exercise legitimate and purposeful initiative and because of their skill in employing aggressive patrol techniques. The officers recognized the threats and hazards of allowing the rioters to continue unchecked. Because these officers had often worked together as a cohesive unit, they were able to use an established principle of crowd control by identifying and removing the leaders of the riot, thus reducing the crowd to a collection of individuals without direction. Finally, these incidents pointed to the need for law enforcement specialists who are trained to plan for and deal with unusual events as a coordinated unit without depleting the law enforcement agency’s personnel or ability to respond to routine calls for police service. Special circumstances require specialized preparation in terms of personnel selection, technology, training, and tactics. Under the guidelines of Downs v.

Critical Incidents

United States (1975) and City of Winter Haven v. Allen (1991), a law enforcement agency could no longer assume that every member is prepared or equipped to perform any task on an ad hoc basis in response to this sort of increasingly complex and heinous criminal activity. This need has proved to be particularly true in large metropolitan areas where multiple calls for this type of police service have become commonplace. Specialized preparation revolves around an organization whereby all functions associated with these perplexing problems can be performed in an orderly fashion without disruption to the other elements of the overall law enforcement agency. The term organization implies more than a recognized position in an organizational chart. The term carries with it the implication that any specialized component possesses the appropriate authority and official means to perform any tasks for which it is responsible and that it receives the necessary administrative support to perform these tasks. Events such as the Munich massacre in 1972 demonstrated the same law enforcement shortcomings on an international basis (Mijares and Mijares 1994).

Present Trends in the Response to Critical Incidents Over the past 40 years, police administrators have progressively relied upon the concept of rigorously selected personnel who have been trained and equipped with special weapons and tactics (SWAT) to respond to these incidents while allowing the uninterrupted ability to respond to more routine calls for service. Since its formative years the SWAT concept has increased dramatically from a small number of large urban police departments to an almost universal acknowledgement of the guidelines defined in Downs v. United States and City of Winter Haven v. Allen recognized that special circumstances require special responses. When law enforcement agencies were financially unable to develop a full-time, organizationally dedicated tactical response capability, part-time SWAT units and mutual assistance programs were developed (Perkins and Mijares 1994).

Critical Incidents

Where officers were originally wearing military surplus fatigue uniforms and baseball caps, today’s operators don padded, fire- and ballisticresistant, and slash-impervious protective gear and fiberglass helmets. Where the original tactical personnel were armed with revolvers and pump shotguns, present SWAT personnel may select from a wide array of weapons and other equipment. Where yesterday’s officers were trained only on localized barricaded suspect and crowd control scenarios, the currently available training classes are many and varied through the state, regional, and national tactical training associations. Law enforcement managers and SWAT commanders have recognized that the future, as it relates to law enforcement’s obligation to be prepared to protect the citizens it serves, can be predicted. Economics and joblessness are commonly used predictors of crime in our society. Unpredictable calamities that are certain to take place, such as natural disasters, leave uncertainty as to the exact location, date, and time, but it is relatively certain that earthquakes will occur in California, tornados will occur in the Plains States, and floods will occur along the rivers. Accompanying these catastrophes and the damage they do is the predictable lawlessness on the part of some elements in society who see destruction of security, property, and chaos as opportunity to loot and assault. It is the mission of SWAT teams to provide a means for law enforcement to respond efficiently and effectively to all types of critical incidents in such a manner that routine calls for police service need not be ignored (Mijares and McCarthy 2008). The changing culture in all countries introduces crimes at a level and intensity not seen before. Active shooters have become a real and very dangerous problem throughout the world. The tragedies at Columbine High School and at the Texas Tower are two early examples. Societies that believe they are culturally immune from this phenomenon are not. More recent example took place in Oslo, Norway, Virginia Tech University, and Aurora, Colorado. These latest attacks demonstrated how a single suspect can easily kill dozens of people. Thus, it is highly

861

C

probable that mass murderers will continue to attack, especially where it is perceived that “It can’t happen here,” is a tenet of incompetent law enforcement managers and unrealistic politicians. Terrorism driven by both religious fanaticism and political extremism and the forecast of mass murder and civil disturbances created by extreme ideology are happening and will most likely continue. International terrorist groups such as Al Queda and its affiliates have planned and accomplished large-scale attacks in major cities throughout the world, e.g., New York in 2001, Madrid in 2004, Mumbai in 2008, and Moscow in 2009. Terrorist communications have been intercepted that tell us they intend to hit smaller targets in locations that are less prepared. The intelligence gathered on these groups suggests that large cities must continue to build and refine their tactical response capabilities and smaller jurisdictions must catch up. The Las Vegas Metro Police, Los Angeles Sheriff’s Department, Los Angeles Police Department, and Orange County California Sheriff’s Department have created a response known as Multiple Assault Counter-Terrorism Action Capabilities (MACTAC). These very large and capable agencies have joined together and combined their resources and training to be able to deploy their SWAT teams and other assets immediately in the jurisdiction that is under attack. This tactic allows the city or county team that is initially on scene to expect that as deployment is initiated, three additional teams from large agencies are already on their way with tactics and techniques have been practiced and will be immediately utilized. Some smaller agencies have begun to emulate this model and regionalize their training, logistics, tactics, and communications, so they have the same type of response capability. The MACTAC concept is being introduced nationally. A similar approach is being used in the area of gathering and analyzing intelligence in the law enforcement context. Usually described as “fusion centers,” this concept is the result of a joint project among elements of the Department of Homeland Security and the Department of

C

C

862

Justice and assigned personnel from state, county, regional, and local police departments. This approach involves the sharing of information on matters of mutual concern. Unlike the infamous “red squads” of a previous generation of law enforcement, fusion centers do not merely gather and store information but instead they analyze and disseminate it when verified and validated. Civil disturbance is a serious threat at all times in our free and open society with the myriad special interests, cultures, and human greed that exist in a heterogeneous society. By comparison, Japan had very little uncivil behavior and virtually no looting during the latest natural disasters. When there is an “excuse” for a riot in an open society, there is no shortage of willing participants. The victory of a sports team, either college or professional, can elicit mass civil disturbance. A hurricane can be the “opportunity” for the unprincipled elements in our society to assault, steal, burn, and murder. Law enforcement SWAT units must be ready and must train in joint exercises to protect those who cannot protect themselves and have a right to live in peace. Because of their high standards and advanced training, their organizational flexibility, and their disciplined approach to problem resolution, today’s SWAT teams have become much more than their original predecessors. They have become the operational first-response arm to thwart terrorism on the domestic front. Their ability to continue in this role is dependent on the perception of a threat by law enforcement leaders and elected officials. It is also dependent on a universal realization that the privacy rights and freedom of movement desired by all citizens in a free and open society also extend to those who would take them from us. Finally, it will depend on the resolve and ability of law enforcement personnel to balance the obligation to protect and serve with the constraints and privileges of the United States Constitution. The SWAT concept of small units has not been without its criticism. Much of this reproach has focused around claims that there has been an overuse of military metaphors and intimidation, intraorganizational elitism, and allegations that

Critical Incidents

the concept contradicts the community policing model, and that the overall approach is unnecessarily expensive and even counter-productive on the basis of cost-benefit analysis (Kraska and Kappeler 1997). However, these criticisms have been rejected on both methodological and substantive grounds that have led to inaccuracies in conclusions (Klein 2005).

Policing Critical Incidents in the Future Past events and current trends can provide a reasonable projection of likely scenarios and conditions under which tactical operations will be conducted in the future. Like all other aspects of modern society, police operations in general and tactical operations in particular will be greatly affected by advances in technology. Equally important, changes in society, fluctuations in the economy, variations in political demands, and continuous modifications of legislation and judicial interpretations will also make an impact on the conduct of law enforcement policies and procedures. Unfortunately these advances are likely to benefit the criminal perpetrator as well as the law enforcement responders to criminal actions. Consequently, there is no shortage of problems likely to be encountered in the future. Increased Encounters with Juveniles A review of the Uniform Crime Reports over the past several years has shown a decrease in the crime rate among the adult population. However, a similar review would produce evidence that the tendency among juvenile offenders is the opposite. Accordingly, likelihood is high that there will be an increase in encounters with heavily armed juveniles by SWAT units both in terms of frequency and intensity. A particularly disturbing aspect of this projection is the observed increase in acts of violence involving multiple victims perpetrated by juveniles against other juveniles. These acts have been manifested through schoolyard and theater shootings in various parts of the country. They have also been seen among the youthful

Critical Incidents

participants in the fiercely competitive subculture of illegal drugs. Although these events appear to have been committed independently, a disturbing set of common characteristics has developed which may have important consequences for law enforcement in general and tactical operations in particular. First, the economic and social backgrounds of the perpetrators vary considerably. In addition, the geography of these incidents suggests that similar incidents can take place at any location. Where most young criminals surrendered to the superiority of responding police units in previous incidents, today’s youthful offenders often refuse to lay down their weapons and show no reluctance to stand fast in an escalation of the situation into full-scale armed combat with the police. Whether this tendency is the product of misguided juvenile feelings of invincibility and immortality or simply a reflection of rising overall societal violence, the police face a difficult dilemma. If the officers are forced into using fatal force against juvenile offenders, they will be characterized as overzealous and trigger-happy. If they approach a critical incident with the deliberation shown by first responders to the Columbine High School incident of 1997 when they attempted to subdue and arrest the perpetrators, they will be criticized for being too slow to respond and uncaring about the perpetrator’s victims. Irrespective of the action taken, the responding law enforcement personnel will be criticized more severely and from more directions than ever before.

Suicide by Cop Suicide is hardly a new phenomenon. Various politicians, mental health professionals, and attorneys filing law suits against the police occasionally suggest that the police officers who confront suicidal suspects should use something besides deadly force to neutralize the situation. However, officers cannot be expected to accomplish immediate psychological evaluations and identify the mental and emotional problems leading to these suicidal incidents nor can they be expected to discover a solution to the

863

C

perpetrator’s problems in a few minutes when psychologists and psychiatrists who have given several hours of clinical sessions to their patients spread out over several months cannot do so. Some professionals such as psychologists, attorneys, and academicians have suggested that the police should have the skills to deal with the suicidal individual. At least they suggest that the police should be able to call a mental health professional to the scene to be utilized in some way to prevent a negative outcome. However, the availability of such a resource simply does not exist for most police departments. What mental health professional is available on Saturday at 2:20 a.m.? The law enforcement element of the criminal justice system, including well-trained tactical officers, may not be able to address the psychopathological needs of suicidal individuals in general or victim-precipitated shootings by police in particular. However, through training, education of the public, and continuous research, the impact of this phenomenon on law enforcement can be assuaged sufficiently to allow its etiology to identify appropriate remedial and preventative measures.

Terrorism and Police Involvement with Weapons of Mass Destruction In the absence of a formal declaration of war, most terrorist incidents that could be perpetrated within the borders of the United States are criminal acts. Although they may differ from the usual criminal acts in their gravity, the enormity of the criminal acts committed by these heinous offenders is so extraordinary that many citizens expect a military response or from a federal law enforcement agency. However, the response, particularly at the initial stage of the incident, will inevitably be performed by local law enforcement officers. Because of the magnitude of these events, a tactical unit will be directed to resolve the situation. Whether it is a mass homicide with firearms or a suicide bombing, the accelerated violence characteristic of a terrorist incident will require a response beyond

C

C

864

the capability of a single officer or even group of individual officers. It will require the attention of a cohesive and coordinated unit. In addition, the extreme violence of terrorism is likely to be imitated and displayed by members of the illegal drug trade as well as by delusional individuals who have acquired the tools of the terrorists. Even without a political, social, or religious agenda, these individuals cannot be taken lightly by law enforcement. Thus, it will become increasingly important for police officers to stay abreast of world events as to develop and hone their tactical skills. Where police officers may have relied solely on their respective intelligence units for information on criminals, tactical officers will need to stay more active in their pursuit of information. Terrorist events could be catastrophic. Only with a coordinated and cooperative response can the situation be managed. If such an incident were actually to take place, direct involvement by SWAT personnel alone may be unlikely. Representatives from fire departments, public health agencies, hazardous material units, and explosive ordnance disposal squads would be called on to neutralize the threat and to relieve any effects. However, Hillman (1999) suggested that SWAT personnel will have specific roles to fulfill in order to allow these other units to complete their missions safely and expeditiously: 1. They will accompany threat neutralization and medical treatment personnel into contaminated areas, particularly when there is a need to confront suspects and to control injured victims. 2. They will provide security for these specialized personnel. 3. They will provide much of the labor-intensive tasks of identifying, seizing, marking, and preserving physical evidence in such a manner that the custody and chain of evidence can be legally maintained for later criminal prosecution. 4. They will assist with the evacuation of intended but unaffected victims. These individuals, though not necessarily symptomatic, may require quarantine procedures and may be in a highly agitated and irrational state. The

Critical Incidents

demeanor, expertise, and methods employed by police personnel will be a major factor in maintaining relative calm and avoiding an intensification of the problem. Changing Environments During the formative years of the SWAT concept, most of the violent encounters were against barricaded suspects in homes or businesses. These encounters were usually the result of escalated family disputes or aborted robberies. While these trends continue today, there is a need for a tactical response in locations that historically were not just unthinkable, but were often regarded even by most criminals as sacrosanct. The shooting of 90 young people at a youth camp in Norway is a case in point. Whatever the motivation, first responders must become familiar with operating in environments and under conditions that are very different from the traditional residence and business place. This familiarization will involve not only differences in physical environments but also diversity in the motivations of perpetrators as well as the reactions by the victims. In most cases, the use of violence by criminals was primarily intended to be an aid for escape with an occasional outburst by an extremist crusader with an agenda. Based on statements made by captured perpetrators, each copycat use of violence appears to be an attempt to be even more grandiose than its predecessors and to leave a mark that will not be soon forgotten. Crime as a Diversion It is reasonable to assume that criminals and terrorists know and understand at least some of the basic requirements of a police tactical response. With this information it would also be reasonable to assume that terrorists or criminals would precede a major incident with a smaller but extraordinary event that would divert police assets to a geographical location within a jurisdiction that would make response to the primary event limited and very difficult. It is also prudent to train and familiarize other officers with at least a rudimentary ability to respond to further incidents when the primary tactical unit

Critical Incidents

is already engaged. Officers assigned to the function of providing a holding action need not be permanently assigned SWAT personnel but could be taken from a list of officers who may have been selected for training and later permanent assignment to the unit.

Recommended Changes in Approaches to Critical Incident Resolution Maintaining Standards for the Profession A certain measure of creativity and modification are often expected and occasionally necessary for the resolution of the myriad of tactical problems expected to be encountered by a modern SWAT team. However, the actual execution of plans must stay within the parameters of the United States Constitution, existing legislation, current case law, department policy, and standards of the profession. Standards must be met in a variety of ways through organizational readiness (City of Winter Haven v. Allen 1991), personnel procedures (Moon v. Winfield 1973), equipment renovation and replacement, and tactical applications (Downs v. United States 1975), in order to be consistent with standards of the profession. It is incumbent upon all law enforcement personnel to stay abreast of these standards and in any changes in the state of the art. Such information is readily available through several different media. Attending the conferences of the various national, regional, and state tactical officers associations provides current information on the development of new techniques as well as an opportunity to view demonstrations of innovations in the technology. Equally important, these conferences provide association members an opportunity to exchange case histories and opinions regarding approaches to be followed under similar circumstances. Printed articles in professional magazines such as Law Enforcement Technology, the Tactical Edge, and the FBI Law Enforcement Bulletin furnish information on the newest developments in equipment and its most effective use. Academic journals such as Police Liability Review and Journal of Contemporary Criminal

865

C

Justice provide commentary on relevant and directly applicable court decisions and legislation. In short a large volume of substantive information is available for the active learner in this area. Modifications in Approaches to Training The anticipated problems plus unforeseen others that will certainly occur will require tactical units to incorporate a variety of innovative approaches to situational preparation. The types of problems anticipated to be encountered in the future by SWAT teams suggest that the resolution will frequently require more than one organization. The trends toward incidents of greater scale and which often cross jurisdictional lines will require the cooperative effort of several law enforcement agencies and support organizations. Some police departments automatically mobilize assistance from the local fire department and emergency medical services. However, the likelihood of incidents involving weapons of mass destruction will involve the responses from state and federal personnel. Whether the support comes from within the organization or from external sources, the possession of individual skills and knowledge is insufficient without coordination of effort. Without the coordination coming from supervised joint training and participation in mock disaster drills, the effectiveness of any attempts to resolve these situations is very limited. Conducting Continuous Research Scientific research is an organized search for the truth involving problem identification, parameter definition, data collection and analysis, and the realization of a research decision. In the field of criminal justice in general and tactical operations in particular, applied research could address issues such as techniques, technology, and legal issues. While there may not be a legal requirement specifically mandating any form of research, the ability to display and document a continuous effort to find solutions for improving SWAT responses is certainly a helpful approach in promoting the image of any tactical unit and its management.

C

C

866

As an example, a tactical unit may wish to improve its ability to respond to barricaded suspect situations by examining the efficacy of various pieces of intelligence-gathering equipment or sensory-enhancing technology such as a thermal imaging surveillance system. The research decision would very likely involve a recommendation to conduct follow-up studies on the most effective techniques for using the equipment during a tactical scenario. It would likely also include a recommendation to conduct further research on the issues associated with the legal use of the equipment.

Conclusion Extreme violence requiring a coordinated police response is a phenomenon with no foreseeable end. Most of the critical incidents requiring a tactical response are still committed by individual perpetrators as a result of either aborted criminal activity or personal psychological disturbances. Consequently tactical preparations in the form of training, equipment, and tactical responses continue to focus on the individual criminal. However, a disturbing trend has been a growth in the number and strength of various wellorganized domestic terrorist groups. Often using freedom of religion as a justification for their crimes and a self-serving brand of patriotism as a rallying cry, these groups seem to be particularly adept at attracting popular support by concentrating on their philosophy and ignoring their criminal activity. Law enforcement management will continue to be responsible for protecting the peaceful majority of society from those who would use the available opportunity and technology to further their criminal motivations. Whether the criminal is an individual or a group, motivated by economic gain or political and religious philosophy, rational and calculating, or psychologically maladjusted is irrelevant. Law enforcement will continue to deal with these problems as they arise. The managers of police agencies will be required to anticipate these problems,

Critical Incidents

develop proactive solutions, stay within a limited budget, and still provide routine services with minimal disruption. Of vital importance is the requirement that all of these tasks be done while staying within the guidelines of existing case law and legislation and protecting the agency and its employees from avoidable litigation. The success of future police tactical operations during critical incidents will require active and supportive management. Success or failure will be highly dependent on how well management can assess agency and community needs through scientific research, provide relevant training for police personnel, utilize current and developing technology from many different fields, coordinate both internal and external activities toward the achievement of agency goals, and account for the entire organization with both objectivity and sensitivity.

Related Entries ▶ History of Technology in Policing ▶ Hot Spots and Place-Based Policing ▶ Juvenile Violence ▶ Police and the Military Nexus ▶ Police Use of Firearms ▶ Policing in Developing Democracies ▶ Policing of Peacekeeping ▶ Policing Terrorism and Legitimacy ▶ Risk Management in Policing ▶ Strategies of Policing Terrorism ▶ Terrorist Organizations

Recommended Reading and References Bopp WJ, Schultz DO (1972) A short history of American law enforcement. Charles C. Thomas, Publisher, Ltd, Springfield City of Winter Haven v. Allen, 541 So.2d 128 (Fla. App. Dist. 1991) and 689 So.2d 968 (Fla. App. Dist.1991) Dobson C, Payne R (1982) The terrorists: their weapons, leaders, and tactics. Facts on File, New York Downs v. United States, 522 F.2d 990 (6th Cir.1975) Hillman M R (1999) Biological/chemical terrorism and SWAT response. The Tactical Edge 17(3)

Critical Report on Forensic Science Klein G (2005) The militarization of the police? Or ten ways Dr. Kraska got it wrong. Police Forum (14, 5) Kraska P, Kappeler V (1997) Militarizing the American police: the rise and normalization of paramilitary units. Social Problems (44,1) Mijares T, McCarthy R (2008) The management of police specialized tactical units, 2nd edn. Charles C. Thomas, Publisher, Ltd, Springfield Mijares T, Mijares K (1994) The Munich massacre: its effects on law enforcement. Command, Summer Moon v. Winfield, 306 F. Supp. 843 (1973) and 383 F. Supp. 31 (1974) Perkins D, Mijares T (1994) Police liability issues associated with inter-agency mutual assistance pacts. Police Liability Review 8

Critical Report on Forensic Science Robert E. Gaensslen and A. Karl Larsen Forensic Science Group, Department of Biopharmaceutical Sciences, College of Pharmacy, University of Illinois at Chicago, Chicago, IL, USA

Overview In the USA, the National Research Council (NRC) Committee on Identifying the Needs of the Forensic Sciences Community issued a lengthy report based on several years of hearings and study. The report contained 13 major recommendations covering a wide spectrum of issues and disciplines within the forensic sciences. The NRC is an arm of the National Academies, which are often regarded as the most prestigious and authoritative bodies on matters of science, medicine, and engineering. The NRC has previously issued reports on forensic DNA technology (NRC 1992, 1996), but had never undertaken a comprehensive evaluation of forensic sciences before. The NRC report was directed by the US Congress, prompted by a number of factors and developments. Two important ones were the advent of DNA profiling of biological evidence and a major US Supreme Court decision in 1993 (Daubert et ux. etc. 1993) setting down

867

C

the criteria for the admissibility of scientific evidence in the courts. Daubert was the first case on admissibility of scientific evidence ever considered by the Supreme Court. The decision emphasized that admissibility was to depend on a sound scientific basis for the proffered evidence. DNA typing was always considered to have a sound scientific basis, and so became a sort of “paradigm” for the rest of the forensic sciences to follow. Almost everyone, critics and defenders alike, agreed that the forensic sciences were under-resourced. Thus, was born the idea that a comprehensive study was needed and that NRC was the body to conduct it. Authorized in 2005, the study was completed and the report issued in 2009. This entry will mainly discuss the NRC report on forensic science, highlighting the main recommendations, some of the background leading up to it, and some of the reaction to it.

Background and Introduction For almost 200 years, forensic laboratory and medicolegal services have been provided by government agencies. In the nineteenth century, most of these services were provided by medicolegal institutes in continental Europe and in a few of the larger cities in the US Forensic laboratories, as we know them today, developed during the twentieth century. In the years before the Second World War, several forensic laboratories were established in the USA. They were housed within the Federal Bureau of Investigation (FBI) and in some larger municipal police departments, such as the Los Angeles Police Department. With time, state and county forensic-science labs were established. The Omnibus Crime Control and Safe Streets Act of 1968 (P.L. 90-351) provided a considerable amount of funding for forensic laboratory expansion, mainly through block grants to the various states. More recently, considerable federal funding for the support and expansion of DNA technology in forensic labs has been available. This entry will mainly discuss the NRC report on forensic science, some of the background leading up to it, and some of the reaction

C

C

868

to it. Another report, on the forensic science services (FSS) in the UK, will be addressed in another contribution. Unlike the US report, the emphasis on the British report (from a Committee of the House of Commons) is on the implications of the government’s plan to shut down the forensic science service (FSS), essentially ending government-supported forensic-science services in the UK (see ▶ Organizing and Supplying Forensic Science Services). Recently, Ross commented on the NRC report from the standpoint of the forensic sciences in Australia and New Zealand (Ross 2011). He was looking at the forensic sciences in those countries through the lens of the relevant issues considered by the NRC committee that produced the report in the USA. Of special interest in this treatment is the fact that Australia has a National Institute of the kind recommended by the NRC report for the USA. The National Research Council (NRC) of the National Academies is a highly prestigious body that assembles expert panels from the scientific, medical, and technological world to look at difficult or controversial questions and make recommendations. This is the first time the NRC has comprehensively examined forensic sciences and services, and its findings and recommendations have been widely cited and quoted. The US Congress, various executive branch agencies, and the courts are likely to rely on them. As noted, a number of factors coalesced to prompt Congress to commission the NRC report. In the Daubert decision, the US Supreme Court emphasized the need for the “scientific” results proffered as evidence in a courtroom to be based on solid scientific principles gathered through hypothesis testing. The court emphasized the concept of empirical testing of a potential falsifiable hypothesis, a notion set forth by Popper (1989). In subsequent decisions (General Electric v Joiner 1997; Kumho Tire v Carmichael 1999), the court extended these principles to “technological” evidence elicited by way of expert testimony. In effect, the “Daubert trilogy,” as the three cases are sometimes called, modified and sharpened the focus of Rule 702 of the

Critical Report on Forensic Science

Federal Rules of Evidence. The Federal Rules of Evidence are established by the federal judiciary and govern the federal courts. There are several of these rules that pertain to forensic science and expert witnesses, but 702 is probably the most critical one. Rule 702, Testimony by Expert Witnesses, states: A witness who is qualified as an expert by knowledge, skill, experience, training, or education may testify in the form of an opinion or otherwise if: (a) the expert’s scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue; (b) the testimony is based on sufficient facts or data; (c) the testimony is the product of reliable principles and methods; and (d) the expert has reliably applied the principles and methods to the facts of the case. Until Daubert, an older 1923 District of Columbia appellate court case called Frye v. US was widely used as the standard for scientific evidence admissibility (Frye 1923). Under Frye, the determinant was “general acceptance in the scientific community” to which the subject area belonged. There were legal debates through the years over what constituted “general acceptance” and over the definition of the relevant “scientific community.” Daubert set down six criteria that a judge could use in deciding on admissibility: (1) Is the proposition testable? (2) Has it been tested? (3) Are there accepted standards? (4) Has there been peer review and/or publication? (5) Are the procedures generally accepted? (6) Does the procedure have a known error rate, and if so, what is it? The decision made the judge the “gatekeeper.” He/she has considerable latitude in using the stated criteria to decide on admissibility. The Court has made clear these criteria do not constitute a “checklist.” Strictly speaking, the US Supreme Court decision only applied to the federal courts, but many states (around 30 as of 2010) have adopted it (or its basic principles) as well. Those that have not adopted it continue to use the Frye precedent, or a modified version of it. In the wake of Daubert, there was considerable commentary that DNA typing and profiling constituted a sort of paradigm of proper scientific method, and other forensic


disciplines should be held to similar scientific standards. Criticism of the pattern evidence specialties, handwriting comparison, fingerprint comparison, toolmark and firearms identification, and evidence such as tire or footwear impressions and bite marks, has been especially pointed in this regard. There was scant discussion of chemistrybased specialties, such as controlled substance identification, toxicology, and forensic chemistry. Concerns were also raised about measurement of error rates, transparency concerning the accuracy and precision of tests, and validation of procedures. In the 1980s, the FBI, in collaboration with the forensic labs doing DNA testing at the time, had assembled a peer standard-setting body called the Technical Working Group on DNA Analysis Methods (TWGDAM). The goals of TWGDAM included standardizing and validating methods and choosing consensus protocols that would enable smooth operation of the DNA database (called CODIS, for Combine DNA Indexing System). Most people would agree that this TWG worked well and realized its objectives. For a time, the TWGDAM was supplanted by a committee known as the DNA Advisory Board (DAB), established by the DNA Identification Act of 1994 (P.L. 103-322). This Board was disbanded by law about 5 years later, and the Technical Working Group on DNA came back into existence, now as the Scientific Working Group on DNA Analysis Methods (SWGDAM), to establish and monitor consensus standards in DNA. Building on the TWGDAM model, the National Institute of Justice assisted in forming a number of other TWGs, and the FBI formed additional SWGs in this period. The purpose was to establish consensus standards in various forensic-science disciplines. There are now quite a number of these committees. The exoneration of over 250 convicted felons in 34 states prompted by investigations carried out by the Innocence Project through DNA analysis (Innocence Project visited 9/12/11) shed some harsh light on practices in the criminal justice system that had led to the original convictions. In some cases, these included poor

869

C

or unacceptable forensic lab practices (Garrett 2008; Collins and Jarvis 2008). Note that the extent to which unacceptable forensic practices contributed to the wrongful convictions in these cases has been challenged (Collins and Jarvis 2008). Reports of incompetent scientists and even fraud by people working in forensic laboratories and on forensic cases, and of almost complete meltdowns in a few laboratories, also came to the fore (www.corpus-delicti.com/ forensic_fraud.html, 2011). In this climate, there was also concern about the limited funding available for forensicscience services, the limited funds available for forensic-science-related research, and how forensic-science academic programs and education fit into the overall picture. A census of the public forensic science laboratories in 2002 (Peterson and Hickman 2005) indicated that the labs were encountering serious backlogs due to increased workloads and decreased resources. Looking at the calendar year 2002, forensic labs had about 289,900 cases in backlog as of January 1, received nearly 2,707,000 new cases during the year, completed over 2,495,000 of them, and ended the year with a backlog of over 501,000 cases. A follow-up study using almost identical methodology for 2009 (Durose et al. 2012) showed that labs started the year with over 1,131,000 cases backlogged and ended it with almost 1,154,000 despite having worked 3,632,000 cases. All these forces and developments persuaded Congress that forensic science needed some review and attention from the federal government, and thus, in late 2005, the US Congress authorized the National Academy of Sciences to conduct the inquiry that led to the NRC report. The report was published in 2009.

The NRC Report In charging the National Academy to conduct its inquiry, Congress stated that While a great deal of analysis exists of the requirements in the discipline of DNA, there exists little to no analysis of the remaining needs of the

C

C

870 community outside of the area of DNA. Therefore... the Committee directs the Attorney General to provide [funds] to the National Academy of Sciences to create an independent Forensic Science Committee. This Committee shall include members of the forensics community representing operational crime laboratories, medical examiners, and coroners; legal experts; and other scientists as determined appropriate.

This Forensic Science Committee was to: 1. Assess the present and future resource needs of the forensic science community, to include State and local crime labs, medical examiners, and coroners; 2. Make recommendations for maximizing the use of forensic technologies and techniques to solve crimes, investigate deaths, and protect the public; 3. Identify potential scientific advances that may assist law enforcement in using forensic technologies and techniques to protect the public; 4. Make recommendations for programs that will increase the number of qualified forensic scientists and medical examiners available to work in public crime laboratories; 5. Disseminate best practices and guidelines concerning the collection and analysis of forensic evidence to help ensure quality and consistency in the use of forensic technologies and techniques to solve crimes, investigate deaths, and protect the public; 6. Examine the role of the forensic community in the homeland security mission; 7. [Examine] interoperability of Automated Fingerprint Information Systems (AFIS) and 8. Examine additional issues pertaining to forensic science as determined by the Committee There were 17 members of the Committee. Two were forensic pathologists, four were academic or practitioner representatives of forensic science, three were attorneys (one cochairperson was a federal judge), and the remainder were science and engineering professors. They held a number of meetings and received both written and verbal comments from a wide variety of interested parties and representatives of interested organizations.


The report, issued in February of 2009, has 11 chapters. The first four concern defining forensic science, making a case for the integration of different services, the legal environment around admissibility of forensic-science evidence, and a discussion of scientific method and data interpretation. Next, some of the forensic-science subdisciplines are described. For purposes of this inquiry and report, forensic science was construed to include toxicology, firearms and toolmarks, questioned documents, trace evidence, controlled substances, biological and serology screening (including DNA analysis), fire debris and arson analysis, impression evidence, blood pattern analysis, crime scene investigation, medicolegal death investigation, and digital evidence, following NIJ 2006. The report is summarized and discussed below under headings corresponding to the chapter subjects. Some of these chapters contained recommendations, while others did not. The major recommendations are stated in their entirety and are integrated into the chapter headings in which they occur in the original report.

The Forensic-Science Community and the Need for Integrated Governance One of the problems identified is the breadth of scientific, technical, and practice specialties that fall under “forensic science” as broadly construed. From molecular biology, chemistry, and physics, to medicine, nursing, pharmacology, and toxicology, to fingerprints and firearms, to crime scene investigation, the umbrella covers a variety of practitioners from Ph.D.-level scientists to police personnel trained in crime scene processing and investigation (see ▶ Forensic Science Culture). Any proposal for uniformity covering such a wide spectrum of areas and interests presents a challenge, to say the least. Public-sector lab and medicolegal services are also delivered at different levels of government, from federal to municipal, with correspondingly different resources. Some state laboratories primarily serve local law enforcement agencies, because that is where the majority of criminal


cases are investigated. There are private, forprofit laboratories in the USA. It is fair to say that more private testing labs have grown up for DNA analysis because the need was there. A number of public laboratories have contracted with private labs to conduct the DNA profiling for data banking; to work old, cold, and backlogged cases; and in some cases even to work current cases as a way of catching up. In addition, there are many “identification” units or “fingerprint” units in law enforcement agencies that are not part of larger forensic labs. Lack of adequate funding is a theme running throughout the forensic lab and medicolegal services community. This has led to chronic backlogs in the laboratories as noted above. The National Institute of Justice (NIJ), which funds virtually all forensic-science research, has managed the infusion of millions of dollars to the public-sector labs and medical examiner offices over the past several years to improve their DNA programs, to help investigate and hopefully solve cold cases, and for general service improvements. The Coverdell program is the main vehicle for laboratory improvement funding, part of which is competitive. The program has helped materially, but has not solved the underlying resources problem. Much more research is needed in forensic science than is able to be funded with NIJ’s current resources. NIJ’s budget is minimal if compared with that of the National Institutes of Health (NIH) or the National Science Foundation (NSF), and it is generally able to award 20 or 30 grants a year for forensic-science research, not including Coverdell or President’s DNA Initiative funds. In fiscal year 2011 (the year from September 1, 2010, to August 31, 2011), for example, NIJ awarded about 385 grants and cooperative agreements totaling over $200 million. Only a small part of these funds was given for forensic-science research as such. NIJ also manages the activities of several funded centers through its Forensic Resource Network. The research funds administered by the Office of Science and Technology are competitive, and expert panels review the applications. NIJ recently announced it will develop review panels whose members serve multi-year terms, similar

871

C

to NIH’s study sections, instead of constituting new panels for each review cycle. It is to be noted that the FBI maintains an active research and development section of its Laboratory Division that involves in-house scientists and collaborations with university and other government agency scientists. The National Institute of Standards and Technology (NIST) is also involved in some forensic-science activities that fall under its overall mission, such as devising standard reference materials. To try and address both coordination and oversight of this myriad of activities, the Committee recommended that a National Institute of Forensic Science be created. Recommendation 1. To promote the development of forensic science into a mature field of multidisciplinary research and practice, founded on the systematic collection and analysis of relevant data, Congress should establish and appropriate funds for an independent federal entity, the National Institute of Forensic Science (NIFS). NIFS should have a full-time administrator and an advisory board with expertise in research and education, the forensic science disciplines, physical and life sciences, forensic pathology, engineering, information technology, measurements and standards, testing and evaluation, law, national security, and public policy. NIFS should focus on: (a) Establishing and enforcing best practices for forensic science professionals and laboratories; (b) Establishing standards for the mandatory accreditation of forensic science laboratories and the mandatory certification of forensic scientists and medical examiners/forensic pathologists—and identifying the entity/entities that will develop and implement accreditation and certification; (c) Promoting scholarly, competitive peerreviewed research and technical development in the forensic science disciplines and forensic medicine; (d) Developing a strategy to improve forensic science research and educational programs, including forensic pathology;

C

C

872

(e) Establishing a strategy, based on accurate data on the forensic science community, for the efficient allocation of available funds to give strong support to forensic methodologies and practices in addition to DNA analysis; (f) Funding state and local forensic science agencies, independent research projects, and educational programs as recommended in this report, with conditions that aim to advance the credibility and reliability of the forensic science disciplines; (g) Overseeing education standards and the accreditation of forensic science programs in colleges and universities; (h) Developing programs to improve understanding of the forensic science disciplines and their limitations within legal systems; and (i) Assessing the development and introduction of new technologies in forensic investigations, including a comparison of new technologies with former ones. This recommendation sits at the core of the report, and many of the other recommendations are conditioned upon its implementation. This NIFS would be a new breed of federal agency, having research and educational support as well as oversight and enforcement powers and responsibilities. The Committee elaborated what it believed to be minimal criteria for a federal coordinating agency. They noted they had considered NSF and NIST, but did not think either met the criteria. NIJ was rejected because it is part of the US Dept. of Justice whose primary mission is law enforcement and prosecution. The Committee recognized that budgetary and other constraints could delay or even prevent the implementation of this core recommendation. NIJ argued in its response to the report that it was structured to handle the tasks described and that many of its initiatives followed the recommendations (NIJ 2011). There have been two legislative initiatives in the US Congress that directly responded to this recommendation (THOMAS – Library of Congress). One, a bill in the US Senate (S132 introduced 1/25/11 by Sen. Patrick Leahy of


Vermont), the Criminal Justice and Forensic Science Reform Act, contains provisions to accomplish many of the recommendations in the NRC report, though not using the same language and terminology. Senate Bill S3378 (and an identical bill in the House, HR 6106) set out to accomplish many of the same goals but differ from S132 in the administrative structure. Sen. Leahy’s bill would create an Office of Forensic Science within the US Attorney General’s Office and an appointed Board to oversee all the proposed activities. Sen. Rockefeller’s bill would create a National Forensic Science Coordinating Office at NSF. There would be substantial involvement from NIST in the creation of standards. Neither of these bills has progressed out of committee, and there does not appear to be much chance that either of them will be enacted.

The Admission of Forensic-Science Evidence in Litigation The report has a lengthy, detailed, and informative review of the transition from Frye to Daubert criteria for admissibility and the evolution of Federal Rule 702 (which many state courts follow, along with the federal courts). This subject was sufficiently discussed above for purposes of this summary. A number of decisions on the admissibility of different categories of evidence are discussed. An important point recognized by the Committee is that courts and admissibility rulings are not the appropriate mechanism for establishing solid scientific underpinnings for the forensic sciences; that must be done by carefully controlled research. The admissibility discussion leads naturally to the next chapter of the report, concerning scientific principles and data interpretation.

The Principles of Science and Interpreting Scientific Data This entry discusses scientific method and the principles of method development and validation that characterize the traditional sciences. In the


view of the Committee, these must be followed by forensic sciences. There is a review of the scientific method and of the subsequent validation of scientific findings and results. Emphasis is placed on the validation of methods and of the methods used by forensic sciences to classify/ identify substances, to associate/dissociate evidentiary items with potential sources, and sometimes to attempt to reconstruct aspects of past events based on the physical evidence record. Method validation should be, in the view of the committee, published in peer-reviewed literature. It is suggested that the principles of ISO 17025 for testing laboratories be used as the basis for forensic method validation. These principles include calibration of reference standards and materials, comparison of results on one method with another, interlaboratory comparisons, systematic assessment of factors that influence results, and assessment of uncertainty. Many forensic labs are now using ISO 17025 principles as the basis for accreditation. The QA (quality assurance) guidelines developed for DNA analysis by TWGDAM/SWGDAM (discussed above) are enumerated as a model. There are now many TWGs/SWGs for different categories of evidence that have established consensus principles for the analysis and interpretation of the particular type of evidence (e.g., SWGDRUG, drug and controlled substance identification and quantitation analysis; SWGFAST, fingerprints and other friction ridge patterns; SWGGUN, toolmark and firearms identification; SWGMAT, materials such as fibers). It is noted that forensic labs should be reporting measurement error with results, and keeping track of error rates to the extent possible. Techniques and comparisons need to be subjected to blind testing to ferret out the rate of false positives and false negatives. Bias and sources of bias, particularly as applicable to comparisons where “source/not source” conclusions are reached, are discussed. Context bias and observer bias have been discussed more recently in the context of forensic-science examinations (see ▶ Cognitive Forensics: Human Cognition, Contextual Information, and Bias, Saks et al. 2003; Crime Lab 2009b; Thornton 2010).

873

C

The larger problem outlined by those critical of forensic sciences contains the “law-enforcement context bias” as well. It has to do with the effects of unconscious influences of bias on an observer seeing, recording, and interpreting data.

C Descriptions of Some Forensic-Science Disciplines This entry discusses the forensic analysis of several categories of evidence, including biological, controlled substances, friction ridge pattern analysis—usually fingerprints, other patterns such as footwear, toolmarks and firearms, hairs, fibers, questioned documents, paint, explosives residues, fire debris, odontology, and bloodstain patterns—and digital media analysis. Each one is described with respect to its sample data and collection, the methods of analysis employed, the scientific interpretation and reporting of results, and the committee’s conclusions. There is broad variety across the disciplines regarding techniques, methods, reliability, error rate assessments, reporting language and guidelines, underlying research, and the educational background of practitioners.

Improving Methods, Practice, and Performance in Forensic Science This entry focuses on improvements in practice across disciplines, with a view toward bringing more scientific approaches and methods to bear, particularly but not exclusively, in the pattern evidence comparison areas (see ▶ Forensic Science and the Paradigm of Quality). Recommendation 2. The National Institute of Forensic Science (NIFS), after reviewing established standards such as ISO 17025, and in consultation with its advisory board, should establish standard terminology to be used in reporting on and testifying about the results of forensic science investigations. Similarly, it should establish model laboratory reports for different forensic science

C

874

disciplines and specify the minimum information that should be included. As part of the accreditation and certification processes, laboratories and forensic scientists should be required to utilize model laboratory reports when summarizing the results of their analyses. Although it may at first sound trivial, terminology is a very critical issue in forensic sciences, especially the words and phrases forensic practitioners use as “terms of art.” Criminalists, for example, use the word “identification” to mean placing something into its proper class or category, such as “the stain represented as item # 6 was identified as human blood” or “item # 9 was identified as cat hair.” Fingerprint and firearms identification specialists, on the other hand, use “identification” to mean “individualization,” such as “the developed latent print represented as item # 8 was identified as the left middle fingerprint of John Doe” or “the bullet represented as item # 4 was fired by the 9 mm semiautomatic pistol, item #2.” We also talk about “identifying the human remains as that of Jane Doe.” In cases where there are comparisons made between a known specimen and an evidentiary one, the terminology issue can become even more confusing. Suppose a 13-locus DNA profile of the male fraction from a vaginal swab is identical in every respect to the profile of the suspect. One could write or testify that “the profiles match,” or that “the suspect could not be excluded as the source of the semen,” or that “the suspect and evidence profiles are alike in every respect that was tested and compared.” In a DNA comparison like this, these statements would be accompanied by an estimate of the probability of chance duplication. One could construct similar examples from paint comparisons, hair or fiber comparisons, etc. The issue here is whether these statements mean the same thing to a trier of fact. They probably do not. It was thus recommended that a standardized terminology be adopted and used in laboratory reports and in testimony, so that given the same results, the wording is always the same. It is understood that there might have to be a certain amount of within-discipline terminology standardization,


that is, trace analysts might not be able to use exactly the same terminology and vocabulary as a fingerprint or firearms examiner. The Committee noted that even if some or all of its core recommendations were ignored or delayed because of funding or political considerations, the terminology issue was one that should receive immediate attention. Another one was conformity of lab reports with external standards such as ISO 17025 (2005), where methods, materials, procedures, results, and conclusions are recorded, along with sources and measurements of uncertainty and levels of confidence where appropriate. Recommendation 3. Research is needed to address issues of accuracy, reliability, and validity in the forensic science disciplines. The National Institute of Forensic Science (NIFS) should competitively fund peerreviewed research in the following areas: (a) Studies establishing the scientific bases demonstrating the validity of forensic methods. (b) The development and establishment of quantifiable measures of the reliability and accuracy of forensic analyses. Studies of the reliability and accuracy of forensic techniques should reflect actual practice on realistic case scenarios, averaged across a representative sample of forensic scientists and laboratories. Studies also should establish the limits of reliability and accuracy that analytic methods can be expected to achieve as the conditions of forensic evidence vary. The research by which measures of reliability and accuracy are determined should be peer reviewed and published in respected scientific journals. (c) The development of quantifiable measures of uncertainty in the conclusions of forensic analyses. (d) Automated techniques capable of enhancing forensic technologies. The wide range of different disciplines comprising “forensic science” comes with an equally wide range of bases for accuracy, reliability, requirements for validation of a technique, etc.


The recommendation calls for research to quantitate reliability measures. It also calls for examiners to provide quantitative statements of measurement uncertainty where possible. This recommendation also implies that the research should try to get at error rates for different techniques and procedures. It is designed primarily to abstract best practices from the research data. The committee were especially concerned about disciplines like fingerprint comparison, where there have been claims of “100 % accuracy” or “zero error rate” (see ▶ Probability and Inference in Forensic Science). The Brandon Mayfield case discussed just below was a stark reminder of the possibility of misidentifications, even by highly experienced examiners. There has been lengthy discussion by lawyers, law professors, and fingerprint examiners around these questions (Koehler 2008; Haber and Haber 2008; Mnookin 2008; Dror et al. 2006; Dror and Charlton 2006). Recommendation 4. To improve the scientific bases of forensic science examinations and to maximize independence from or autonomy within the law enforcement community, Congress should authorize and appropriate incentive funds to the National Institute of Forensic Science (NIFS) for allocation to state and local jurisdictions for the purpose of removing all public forensic laboratories and facilities from the administrative control of law enforcement agencies or prosecutors’ offices. Bias on the part of forensic examiners as an intrinsic result of being administratively located within law enforcement agencies or prosecutors’ offices has been discussed for years. The problem may be made even more acute if the examiners are themselves agents of law enforcement (see ▶ Cognitive Forensics: Human Cognition, Contextual Information, and Bias). The poster-child case for this viewpoint is Brandon Mayfield (U.S. DOJ Inspector General 2006). The details are complicated, but briefly, FBI examiners mistakenly identified latent fingerprints from the Madrid, Spain, train bombing incident of March 11, 2004, as those of Brandon Mayfield, a Muslim US citizen living in Oregon. Critics point to this case as a clear example of context bias on the part

875

C

of the FBI analysts. But critics also argue that this sort of bias can be more subtle, that analysts working in a law enforcement environment are bound to be influenced toward that viewpoint, and that objectivity is not possible. Many forensic scientists and lab administrators do not agree, and there are published discussions on both sides of the question. Recommendation 5. The National Institute of Forensic Science (NIFS) should encourage research programs on human observer bias and sources of human error in forensic examinations. Such programs might include studies to determine the effects of contextual bias in forensic practice (e.g., studies to determine whether and to what extent the results of forensic analyses are influenced by knowledge regarding the background of the suspect and the investigator’s theory of the case). In addition, research on sources of human error should be closely linked with research conducted to quantify and characterize the amount of error. Based on the results of these studies, and in consultation with its advisory board, NIFS should develop standard operating procedures (that will lay the foundation for model protocols) to minimize, to the greatest extent reasonably possible, potential bias and sources of human error in forensic practice. These standard operating procedures should apply to all forensic analyses that may be used in litigation.

Strengthening Oversight of ForensicScience Practice The four recommendations growing out of the “oversight” have to do with quality programs, accreditation, certification, proficiency testing, code of ethics, and further research into the quantitation of measurement variability and error. Recommendation 6. Congress should authorize and appropriate funds to NIFS to work with the National Institute of Standards and Technology (NIST), in conjunction with government laboratories, universities, and private laboratories, and in consultation with

C

C

876

Scientific Working Groups, to develop tools for advancing measurement, validation, reliability, information sharing, and proficiency testing in forensic science and to establish protocols for forensic examinations, methods, and practices. Standards should reflect best practices and serve as accreditation tools for laboratories and as guides for the education, training, and certification of professionals. Upon completion of its work, NIST and its partners should report findings and recommendations to NIFS for further dissemination and implementation. Recommendation 7. Laboratory accreditation and individual certification of forensic science professionals should be mandatory, and all forensic science professionals should have access to a certification process. In determining appropriate standards for accreditation and certification, the National Institute of Forensic Science (NIFS) should take into account established and recognized international standards, such as those published by the International Organization for Standardization (ISO). No person (public or private) should be allowed to practice in a forensic science discipline or testify as a forensic science professional without certification. Certification requirements should include, at a minimum, written examinations, supervised practice, proficiency testing, continuing education, recertification procedures, adherence to a code of ethics, and effective disciplinary procedures. All laboratories and facilities (public or private) should be accredited, and all forensic science professionals should be certified, when eligible, within a time period established by NIFS. Recommendation 8. Forensic laboratories should establish routine quality assurance and quality control procedures to ensure the accuracy of forensic analyses and the work of forensic practitioners. Quality control procedures should be designed to identify mistakes, fraud, and bias; confirm the continued validity and reliability of standard operating procedures and protocols; ensure that best practices are being followed; and correct


procedures and protocols that are found to need improvement. Recommendation 9. The National Institute of Forensic Science (NIFS), in consultation with its advisory board, should establish a national code of ethics for all forensic science disciplines and encourage individual societies to incorporate this national code as part of their professional code of ethics. Additionally, NIFS should explore mechanisms of enforcement for those forensic scientists who commit serious ethical violations. Such a code could be enforced through a certification process for forensic scientists.

Education and Training in Forensic Science The committee noted that major strides had been made recently in bringing some standardization to forensic-science academic programs through the Forensic Science Education Program Accreditation Commission (FEPAC). The FEPAC accreditation program and guidelines grew out of a NIJ-supported Technical Working Group for Education and Training in Forensic Science (NIJ 2004). There are still many programs offered, however, that are not FEPAC accredited (FEPAC accreditation is voluntary), and they vary in their uniformity and science content. For graduate programs in particular, more funding should be available if there is to be a proper research base developed. Another issue is forensic-science education for attorneys and judges. More interactions and information exchange are encouraged. Law schools are in a position to enhance this interaction. In disciplines that require training as well as education, a firm foundation in scientific principles is recommended for both the educational and the training components (see ▶ Forensic Science Culture). Recommendation 10. To attract students in the physical and life sciences to pursue graduate studies in multidisciplinary fields critical to forensic science practice, Congress should authorize and appropriate funds to the


National Institute of Forensic Science (NIFS) to work with appropriate organizations and educational institutions to improve and develop graduate education programs designed to cut across organizational, programmatic, and disciplinary boundaries. To make these programs appealing to potential students, they must include attractive scholarship and fellowship offerings. Emphasis should be placed on developing and improving research methods and methodologies applicable to forensic science practice and on funding research programs to attract research universities and students in fields relevant to forensic science. NIFS should also support law school administrators and judicial education organizations in establishing continuing legal education programs for law students, practitioners, and judges.

Medical Examiner and Coroner Systems: Current and Future Needs Various problems exist in the USA with respect to death investigation, and this entry addressed them. There are two systems: medical examiner and coroner. Medical examiners are forensic pathologists, but there are few medical examiner jurisdictions compared with coroner jurisdictions. Coroners generally have little or no formal medical or forensic training. This situation creates a patchwork quilt, and there is considerable lack of uniformity across jurisdictions. The situation also inhibits the collection of accurate death statistics for deaths that come under forensic investigation. There is now a Scientific Working Group for Medicolegal Death Investigation (SWGMDI), and it has already issued several reports. A recent draft report shows that there are not nearly enough forensic pathologists in the USA to meet current needs. Recommendation 11. To improve medico-legal death investigation: (a) Congress should authorize and appropriate incentive funds to the National Institute of Forensic Science (NIFS) for allocation to states and jurisdictions to

877

C

establish medical examiner systems, with the goal of replacing and eventually eliminating existing coroner systems. Funds are needed to build regional medical examiner offices, secure necessary equipment, improve administration, and ensure the education, training, and staffing of medical examiner offices. Funding could also be used to help current medical examiner systems modernize their facilities to meet current Centers for Disease Control and Prevention-recommended autopsy safety requirements. (b) Congress should appropriate resources to the National Institutes of Health (NIH) and NIFS, jointly, to support research, education, and training in forensic pathology. NIH, with NIFS participation, or NIFS in collaboration with content experts, should establish a study section to establish goals, to review and evaluate proposals in these areas, and to allocate funding for collaborative research to be conducted by medical examiner offices and medical universities. In addition, funding, in the form of medical student loan forgiveness and/or fellowship support, should be made available to pathology residents who choose forensic pathology as their specialty. (c) NIFS, in collaboration with NIH, the National Association of Medical Examiners, the American Board of Medico-legal Death Investigators, and other appropriate professional organizations, should establish a Scientific Working Group (SWG) for forensic pathology and medico-legal death investigation. The SWG should develop and promote standards for best practices, administration, staffing, education, training, and continuing education for competent death scene investigation and postmortem examinations. Best practices should include the utilization of new technologies such as laboratory testing for the molecular basis of diseases and the implementation of specialized imaging techniques.

C

C

878

(d) All medical examiner offices should be accredited pursuant to NIFS endorsed standards within a timeframe to be established by NIFS. (e) All federal funding should be restricted to accredited offices that meet NIFSendorsed standards or that demonstrate significant and measurable progress in achieving accreditation within prescribed deadlines. (f) All medico-legal autopsies should be performed or supervised by a board certified forensic pathologist. This requirement should take effect within a timeframe to be established by NIFS, following consultation with governing state institutions.

Automated Fingerprint Identification Systems The primary issue under this heading is automated fingerprint identification system (AFIS) interoperability. There are several commercial AFIS system vendors, and the software and search algorithms are proprietary. Different jurisdictions have different systems, making it difficult to have a fully nationally integrated database and search system. Recommendation 12. Congress should authorize and appropriate funds for the National Institute of Forensic Science (NIFS) to launch a new broad-based effort to achieve nationwide fingerprint data interoperability. To that end, NIFS should convene a task force comprising relevant experts from the National Institute of Standards and Technology and the major law enforcement agencies (including representatives from the local, state, federal, and, perhaps, international levels) and industry, as appropriate, to develop: (a) Standards for representing and communicating image and minutiae data among Automated Fingerprint Identification Systems. Common data standards would facilitate the sharing of fingerprint data among law enforcement agencies at the local, state, federal, and even international


levels, which could result in more solved crimes, fewer wrongful identifications, and greater efficiency with respect to fingerprint searches; and (b) Baseline standards—to be used with computer algorithms—to map, record, and recognize features in fingerprint images, and a research agenda for the continued improvement, refinement, and characterization of the accuracy of these algorithms (including quantification of error rates).

Homeland Security and the ForensicScience Disciplines The Committee briefly explored potential roles of forensic sciences in homeland security, as that was part of its charge (see ▶ Forensic Science and Criminal Investigation and ▶ Forensic Science Culture). In doing so, it noted that these suggestions were preliminary and could comprise a separate study. Forensic-science involvement in homeland security situations differs from its involvement in criminal cases. There needs to be federal agency—local laboratory coordination, earlier involvement in the investigations to help direct them, and specialized technical expertise having to do with chemical, biological, or nuclear threats. Digital forensic sciences are expected to have a major role in antiterrorism efforts. Recommendation 13. Congress should provide funding to the National Institute of Forensic Science (NIFS) to prepare, in conjunction with the Centers for Disease Control and Prevention and the Federal Bureau of Investigation, forensic scientists and crime scene investigators for their potential roles in managing and analyzing evidence from events that affect homeland security, so that maximum evidentiary value is preserved from these unusual circumstances and the safety of these personnel is guarded. This preparation also should include planning and preparedness (to include exercises) for the interoperability of local forensic personnel with federal counterterrorism organizations.


Discussion The Committee’s report and recommendations have gotten a mixed reception in the forensic science and legal worlds. Some organizations have issued official or public responses or reactions to the report (ASCLD 2009; SWGFAST 2009; AAFS 2009; SWGDE 2009; White House OSTP 2010; Suresh 2012). So have various commentators and groups, some from forensic science and others from the legal profession (Crime Lab Report 2009a; Budowle et al. 2009; Kaye 2010; Chatman 2009; Koehler et al. 2011). Some of the organizations generally supported the principles on which the NRC recommendations were based and stated that they, in effect, stood ready to be part of the forward progress toward better forensic science. Most supported the call for more resources. Some objected to certain recommendations and criticisms. SWGFAST, representing one of the disciplines that was most criticized in the report, disagreed with many of the conclusions about the insufficiency of a scientific basis for friction ridge comparisons. Some disagreed with the recommendation of separating forensic laboratories from law enforcement administrations, suggesting that the apparent intrinsic bias it represented was not in fact necessarily present and that the relationship did not automatically cause bias. Not everyone agreed with the recommendation to form the NIFS. Some thought the necessary resources could be assembled from and within existing agencies that already have involvement in forensic-science research and support. NIJ, which was explicitly excluded by the NRC report as an appropriate setting for NIFS, responded with a lengthy justification for why it was actually the most appropriate place. NSF has stated, through the testimony of its director on the Rockefeller bill, that it stands ready to act as the primary federal agency for forensic science if the bill should be enacted. The harshest critics of forensic science from academic settings and law schools take the report

879

C

as a vindication of their long-held beliefs about and criticisms of the sorry state of forensic science. Some have seen the report as a blueprint for progress and positive change, even if all the recommendations cannot be implemented (especially see Kaye 2010; Koehler et al. 2011). Given the current state of the US economy and the concern with federal spending, it does not seem at all likely that the NIFS will come to exist. The failure of two congressional bills aimed at the NRC Committee’s primary goals to emerge from Committee is an indication that big, new, comprehensive legislation focused on improving forensic science is not likely. The majority of forensic-science laboratories are operated by local or state government. It should also be understood that there are significant limitations on what the federal government can do to regulate their activities. Another unfortunate consequence of the economic recession is that state and local governments find themselves under extreme pressure to streamline services and reduce expenses. This situation cannot be seen as favorable for any increase in most forensicscience lab resources and could even mean that resources will decline. Notwithstanding the different points of view and parochial interests of various parties and organizations, the report must be seen as a milestone in the history of forensic sciences in the USA. Never before has there been such a large, comprehensive look at the forensic sciences by anyone, and this one was done by the prestigious NRC. There is validation of and support for the TWG/SWG model, and those organizations will continue to make progress. There may be increased federal funding for forensic-science research and for operational improvements. Some of the recommendations will be implemented incrementally in and by the profession, even without a centrally focused federal forensic-science entity. The report will likely stand as the primary reference document for any discussion of forensic science laboratory improvement or funding for years to come.

C

C

880

Related Entries ▶ DNA Profiling ▶ DNA Technology and Police Investigations ▶ Drug Enforcement ▶ Expert Witnesses: Role, Ethics, and Accountability ▶ Fingerprint Identification ▶ Forensic Anthropology ▶ Forensic Environmental Evidence ▶ Forensic Facial Analysis ▶ Forensic Linguistics ▶ Forensic Palynology ▶ Forensic Science ▶ Forensic Science and Criminal Investigation ▶ Forensic Science and Miscarriages of Justice ▶ Forensic Science and the Paradigm of Quality ▶ Forensic Science Culture ▶ Forensic Science Effectiveness ▶ Legal Rules, Forensic Science and Wrongful Convictions

Recommended Reading and References AAFS [American Academy of Forensic Sciences] (2009) Position statement in response to the NAS “forensic needs” report, 4 Sept 2009 ASCLD [American Society of Crime Laboratory Directors] (2009) Comments on the release of the NAS report on forensic science, 19 Feb. 2009 Budowle B, Bottrell MC, Bunch SG et al (2009) A perspective on errors, bias, and interpretation in the forensic sciences and direction for continuing advancement. J Forensic Sci 54(4):798–809 Chatman Q (2009) How scientific is forensic science? Champion, Aug 2009 Collins J, Jarvis J (2008) The wrongful conviction of forensic science. Crime Lab Report, 16 July 2008 Crime Lab Report (2009a) A response to the testimony of Judge Harry T. Edwards before the Senate Judiciary Committee April 1, 2009 by Crime Lab Report. [Crime Lab Report is an independently edited and produced report on crime laboratory and forensic science issues; its views often reflect those of the forensic laboratory community] Crime Lab Report (2009b) Post-conviction activists ‘contaminate’ evidence in Texas [discussing contextbias on the part of fire investigators in a highprofile death-penalty case in Texas called “Willingham” in which the defendant was ultimately executed]

Critical Report on Forensic Science Daubert et ux. etc. et al. v. Merrell Dow Pharmaceuticals, Inc., No. 92-102 (1993), 509 U.S. 579, 113 S.Ct. 2786 (often referred to as “Daubert” or “Daubert vs Merrell Dow”) Dror IE, Charlton D (2006) Why experts make errors. J Forensic Ident 56(4):600 Dror IE, Charlton D, Pe´ron AE (2006) Contextual information renders experts vulnerable to making erroneous identifications. Forensic Sci Int 156:74 Durose MR, Walsh KA, Burch AM (2012) Census of publicly funded crime laboratories 2009, U.S. Department of Justice, Bureau of Justice Statistics, 2012 August, NCJ 238252 Frye v. United States (1923) Frye v. United States, 54 App. D.C. 46, 293 F. 1013, No 3968, Court of Appeals of District of Columbia (often referred to simply as ‘Frye”) Garrett BL (2008) Judging innocence. 108 Columbia L.R. 55 General Electric Co et al. vs Robert K. Joiner et ux., No. 96-188 (1997) 522 U.S. 136, 118 S. Ct. 512 (often referred to as “Joiner”) Haber L, Haber RN (2008) Scientific validation of fingerprint evidence under Daubert. Law Probab Risk 7(2):87 http://www.corpus-delicti.com/forensic_fraud.html, visited 9/12/2011 Innocence Project: http://www.innocenceproject.org/ visited 9/12/11 ISO 17025 (2005) ISO [International Organization for Standardization]/ IEC [International Electrotechnical Commission], General requirements for the competence of testing and calibration laboratories, 2nd edn. ISO, Geneva Kaye D (2010) The good, the bad, the ugly: the NAS report on strengthening forensic science in America. Sci Justice 50:8–11 Koehler JJ (2008) Fingerprint error rates and proficiency tests: what they are and why they matter. 59 Hastings Law J 59:1077 Koehler J, Mnookin JL, Cole SA, et al (2011) The Need for a research culture in the forensic sciences, UCLA LR 58:725 (and available from the Northwestern University School of Law Faculty Working Papers) Kumho Tire Co. Ltd., et al. v. Patrick Carmichael, etc. et al., No. 97-1709 (1999) 526 U.S. 137; 119 S. Ct. 1167 (often referred to as “Kumho Tire”). The Daubert case along with the Joiner and Kumho Tire cases are sometimes referred to as the “Daubert trilogy” Mnookin JL (2008) The validity of latent fingerprint identification: confessions of a fingerprinting moderate. Law Probab Risk 7(2):127 NIJ [National Institute of Justice] (2004) Education and training in forensic science: a guide for forensic science laboratories, educational institutions, and students, office of justice programs, NCJ 203099

Cross-National Performance in Policing NIJ [National Institute of Justice] (2006) Status and needs of forensic science service providers: a report to congress, NCJ 213420; this report followed a 180-Day Study Report on the Status and Needs of U.S Crime Laboratories by the American Society of Crime Lab Directors [ASCLD] in May 2004 NIJ [National Institute of Justice] (2011) The national institute of justice response to the report of the national research council: strengthening the national institute of justice, NCJ 234630 NRC [National Research Council] (1992) Committee on DNA technology in forensic science, DNA technology in forensic science. National Academies Press, Washington, DC NRC [National Research Council] (1996) Committee on DNA in forensic science, the evaluation of forensic DNA evidence. National Academies Press, Washington, DC NRC [National Research Council], National Academy of Sciences, Committee on Applied and Theoretical Statistics, Committee on Identifying the Needs of the Forensic Sciences Community (2009) Strengthening forensic science in the United States: a path forward. National Academies Press, Washington, DC, p 254 Peterson JL, Hickman MJ (2005) Census of publicly funded forensic crime laboratories, Bureau of Justice Statistics Bulletin, NCJ 207205 ‘Popper K (1989) Conjectures and refutations: the growth of scientific knowledge, 5th edn. Routledge, London, Karl Popper and Conjectures and Refutations in Wikipedia, retrieved 9/9/2011 from http://en. wikipedia.org/wiki/Karl_Popper and http://en. wikipedia.org/wiki/Conjectures_and_Refutations Ross A (2011) Forensic science in Australia: can we learn from international reports? Austr J Forensic Sci 43(2–3):135–145 Saks MJ, Risinger DM, Rosenthal R, Thompson WC (2003) Context effects in forensic science: a review and application of the science of science to crime laboratory practice in the United States. Sci Justice 43(2):77–90 Suresh S (2012) Testimony before the U.S. senate commerce, science, and transportation committee on the science and standards of forensics, 28 March 2012 SWGDE [Scientific Working Group on Digital Evidence] Position on the National Research Council Report to Congress, 17 Sept. 2009 SWGFAST [Scientific Working Group on Friction Ridge Analysis, Study, and Technology] (2009) Position Statement on the NRC Report, posted 3 Aug. 2009 THOMAS—Library of Congress—http://thomas.loc.gov/ home/thomas.php—visited 9/5/12, U.S. Congressional legislation can be queried and tracked from this site Thornton JI (2010) A rejection of ‘Working Blind’ as a cure for contextual bias. J Forensic Sci 55(6):1163 U.S. DOJ [Department of Justice] Inspector General (2006) The Brandon Mayfield case, see at http:// www.usdoj.gov/oig/special/s0601/PDF_list.htm; also

881

C

see FBI Responds to the Office of Inspector General’s Report on the fingerprint misidentification of Brandon Mayfield, National Press Release, 6 January 2006. More information and viewpoints: Crime Lab Report, Crime Labs under Police—Unresolved Issues, 15 October 2008; and S. Weinberg, Keystone Cops at the Police Lab, Miller-McCune.com, 18 June 2009 White House OSTP [Office of Science and Technology Programs] Response to NRC Report, October 2010

Cross-National ▶ Comparative Incarceration

Cross-National Crime Rates ▶ Institutional Anomie Theory

Cross-National Performance in Policing Mark Brunger1 and Graham Ellison2 1 Department of Law and Criminal Justice, Canterbury Christ Church University, Canterbury, Kent, UK 2 Institute of Criminology and Criminal Justice, School of Law, Queen’s University Belfast, Belfast, UK

Overview This entry considers some of the conceptual and methodological issues researchers might face when conducting cross-national performance comparisons of policing systems. In doing so, it asks whether any similarities can be identified across jurisdictions that would allow national police forces to be compared. Many jurisdictions now have some mechanism in place by which to assess police performance at the national level. Clearly, debates around performance,

C

C

882

effectiveness, and accountability are important when modern police organizations command multimillion dollar budgets, but they are also important in terms of public legitimacy: how the police perform within the context of their mandate and the degree to which they inspire public trust and confidence. This entry suggests that considered in cross-national terms, there is considerable variation in the mechanisms used to assess and measure police performance. However, the trend towards internationalization and diversification in public policing has made the desire for cross-national analysis of policing systems more pressing. In the European Union (EU), for example, there have been attempts to both harmonize the Home Affairs policies of its 27 member states and standardize policing policies (particularly around drugs and antiterrorism). However, as we outline in the discussion, these efforts remain limited and are hindered by the differences among member states regarding how police organizations are structured, individual national political priorities, different legal systems, languages and political cultures, rivalries between national police organizations, and not to mention difficulties in compiling agreed performance criteria and targets that have a cross-national relevance. Quantitative cross-national crime data do exist, however, from which comparisons can be made, but such undertakings are also hampered by a multitude of problems, with reliability and validity issues particularly in abundance. These difficulties are also compounded by a relative lack of clarity about what exactly needs to be compared. The entry concludes by suggesting that while some correlations and similarities can be deduced, the existence of local cultural, political, and legal differences ensures that the act of comparison remains a difficult proposition.

Assessing the Key Data Sources Academic Research When considering cross-national analysis, the main question that arises is to whether any

Cross-National Performance in Policing

similarities and differences can be made measurable to any relevant extent. It is the case that some significant attempts have been made at cross-cultural comparisons and are, therefore, worthy of consideration. Bayley (1990), for one, has provided a comparative analysis of policing developments across different jurisdictions. Surprisingly, given their prominent role in maintaining social order across societies, he notes that the police have often been ignored in comparative analysis (Bayley 1990, p. 4). The main problem he identifies is that despite the fact that all policing systems have similar basic concepts, definitional elements, and organizational similarities, the actual practice of policing varies widely across jurisdictions, therefore, making any effort at cross-national comparison a difficult one (see Bayley 1990, Chapter 1). Nevertheless, there have been some attempts at measuring police performance internationally. For example, Dennis and Erdos (2005) in Cultures and Crimes: Policing in Four Nations compare policing methods in the UK, France, Germany, and the USA. They point out that what these nations have in common is that they have witnessed steep rises in crime and disorder from the late 1960s onwards. In order to make comparisons, they concentrate on measuring trends in easily comparable offense categories, such as robbery and burglary, for example. The authors come to the general conclusion that there is a clear link with between maintaining increases in police numbers and reducing crime. To this effect, they argue that continued rising crime rates in the UK can, in some part, be attributed to the lack of visibility of police “on the beat,” which has made crime, in their terms, a low-risk activity. Furthermore, they argue that in France, Germany, and the USA, there has been a more “effective” attempt to deal with crime through a larger and sustained resourcing of the criminal justice system. However, like other analyses that make use of crime statistics to measure police (and policy makers) activity, it raises all sorts of problematic questions around the reliance on “officially recorded” crime data. The differential patterns of development in community and


public/private policing practices and the diversity in legal systems have produced divergent and inconsistent ways of recording offenses and what offenses are actually deemed recordable in the first place. Other studies have mapped the police use of force in a variety of international contexts (e.g., see Stenning 2003). Researchers assessed both the extent to which force, including deadly force, was used and the normative justifications for it. Again there are considerable variations in the use of force by police in the jurisdictions chosen for the study with established democracies much more likely to have a tighter regulatory framework in place, and within the EU, a statutory requirement to abide by supranational legal directives such as the European Convention on Human Rights. The researchers noted that given the variations in levels of political and economic development and the pressures facing the police in the jurisdictions studied, it was difficult to draw any definitive cross-national comparison about the circumstances under which force would be deployed and its character. The International Centre for the Prevention of Crime (Lelandais 2007) has recently engaged in an attempt to compare police performance in eight jurisdictions: the USA (New York City & Chicago), the UK, Canada (province of Quebec), Chile, Belgium, France, Australia, and New Zealand. What the research demonstrates is that the instruments used to assess police performance vary wildly within and between jurisdictions: Some national governments use the same managerialist performance indicators that are applied across the public sector as a whole (e.g., the UK); some assess performance via existing financial or public administration legislation (e.g., Quebec, France), while others adapt varieties of Computer Statistics packages (or COMPSTAT) (e.g., the USA – New York, New Zealand, Australia). Chile and Belgium adopt aspirational guidelines and action plans, while within the USA there are significant differences with the city of Chicago, for example, preferring the Alternative Policing Strategy (APS) to COMPSTAT. Even the nomenclature

883

C

of what is being assessed varies from “police excellence” in Belgium, “high performance” in New Zealand, “best value” in the UK, and “quality of service” in Chile (Lelandais 2007). In some jurisdictions “performance” is taken to mean effectiveness in dealing with crime; in others it relates to public satisfaction and confidence in the police (which may have nothing to do with crime) or to fiscal and budgetary probity and managerial dexterity. Even if it was accepted, based on this analysis, that it just might be possible to construct cross-national performance indicators, there remain significant problems insofar as it equates policing with the police. This position, which is discussed below, is becoming untenable given the increasing blurring of the boundary between highly variant modes of public and private policing provision. There are therefore huge difficulties in conducting cross-national research into policing organizations which hinge on the differences that exist between police organizations nationally and also differences in the way that the police are organized within the same jurisdiction. While nations in the west and north may subscribe to a notion of democratic policing, this takes on a rather different form in those countries that have a British common-law heritage (the USA, the UK – excluding Scotland, Australia, Canada, Republic of Ireland, New Zealand) compared to a Napoleonic or Roman civil law tradition (Scotland, all of continental Europe and Latin America, the province of Quebec, and so on). Most notably, in the UK, for example, public policing is based on the doctrine of “public consent,” while the “office of constable” which is rooted in common law ascribes all police officers irrespective of rank, a set of original, delegated powers which exist independently of their statutory authority or position on the bureaucratic hierarchy. Furthermore, even within the regions of the United Kingdom (Northern Ireland, Wales, Scotland, and England), there are considerable differences in how the police are organized. Scotland, for example, has a Roman law heritage, which impacts on the substantive nature of police powers and judicial procedure compared to

C

C

884

England, Wales, and Northern Ireland, which share a common-law legal tradition. In these jurisdictions, until relatively recently (1985 in England and Wales and 2000 in Northern Ireland), the police retained prosecutorial powers that were largely absent in the system of Procurator Fiscal in Scotland and even more so in the Napoleonic/Roman law tradition of independent institution between accuser and accused. However, other common-law jurisdictions (e.g., the USA and Canada), recognizing the anomaly in the British system, typically have had district attorneys (or their equivalent) to direct investigations for a considerably longer period of time. In addition, it is difficult to compare the style and organization of policing within the UK historically. Until the recent reforms associated with the peace process, Northern Ireland had a highly militarized form of public policing undertaken by the Royal Ulster Constabulary (RUC) that owed more to the colonial form adopted across the British Empire in the nineteenth century and the continental gendarmerie than to the archetypal “British Bobby.” The United States has a traditionally decentralized model of police organization with an amalgam of town, city, municipal, state, and federal forces. This system of tiering is also found in some European nations (Spain, Czech Republic, France, Italy, Germany, etc.) but not in others, for example, Norway, Sweden, the Republic of Ireland, and New Zealand who all maintain a single national police, albeit one that is demarcated into functional specialisms. Unlike many European police forces, the UK does not have a national police that is directly answerable to a central government department. Rather law enforcement is organized territorially in each legal system (England & Wales, Scotland, and Northern Ireland) with the police under the guidance of the relevant devolved administrations (Scotland and Northern Ireland) or the Home Office (England and Wales) but are operationally independent (in theory) from political direction. In those European countries governed by a Napoleonic/Roman civil law tradition, the lines governing the responsibilities of the police


and military authorities are also rather blurred with many countries (Austria, France, Portugal, Italy, Belgium, Switzerland, etc.) maintaining a gendarmerie that is administratively under the control of the civil authorities in some cases, but under military control in others. For example, gendarmes (individual officers) can be dispatched overseas as part of humanitarian and peacekeeping missions alongside the regular army, but in the majority of circumstances, they are used for regular policing duties. In some cases, the roles and responsibilities of the gendarmerie in relation to the civil police are unclear, and in Italy, for instance, you can report a crime either to the Carabinieri (gendarmerie), the Polizia di Stato (national civil police), or the Vigili urbani (municipal or local police). It is not unusual to find both the Polizia di Stato and the Vigili urbani turning up to the scene of a road traffic accident, and determining what organization will deal with the situation is often just a case of negotiation or more likely “who got there first.” Furthermore, Italy also maintains a corps of the regular army that is dedicated exclusively to “police” duties – the Guardia di Finanza – that deals with financial crime as well as illegal immigration and border controls. Somewhat unusually, Guardia di Finanza officers have the right to demand and check customer receipts for purchases made in shops and restaurants. To complicate matters further, some European countries, such as Belgium, have disposed of their gendarmerie and moved towards decentralized and regionally autonomous police services, while others such as the Netherlands and Scotland (as a semiautonomous region of the UK) are seeking to nationalize and centralize their police operations. There are also considerable variations in the degree of functional specialization found among national police organizations. In the UK, all police officers can find themselves performing crowd and riot control duties, but in France, this is the almost exclusive responsibility of a reserve unit of the Police Nationale – the Compagnies Re´publicaines de Sećurite´ (CRS). Again, however, the situation is muddied by the existence of a specialized unit of the gendarmerie


(the gendarmerie mobile) that performs similar duties. Even within Europe then, assessing cross-national police performance begs the following question: performance of what and of whom? There are simply too many organizations performing policing type functions and where the lines of authority and responsibility are unclear to make any kind of clear comparison. Fundamental differences are also evident when we start to go beyond the core European Union nations. Postcommunist policing systems within those countries that comprised the former Warsaw Pact have moved (or are moving) to less authoritarian and repressive dispensations, but even here there is considerable variations, with those states that have negotiated access to the EU being further along the developmental trajectory than those that have not. Compare the situation in the Czech Republic with that of Belarus, for example. Similarly, in Asia there is considerable differentiation in the organization and structure of the police, not to mention adherence to democratic norms and human rights standards. In South Korea and Japan policing is generally “rights based” and in accordance with democratic norms, but in other states, India, Sri Lanka, Pakistan, and Bangladesh, the situation is described by the Commonwealth Human Rights Initiative (2007) as “grim,” with key concerns in and around human rights, accountability, and governance abound. Policing in Hong Kong represents a rather curious hybrid case, having structural similarities with both the British “colonial” and “metropolitan” models given the legacy of its history, but which also appears to be influenced by policing developments in mainland China (Lau 2004). Importantly, however, we know next to nothing about the structure and organization of policing in the People’s Republic of China, which has received almost no attention in the comparative research literature (though see Tanner 2005). In the Middle East, policing structures reflect particular religious, cultural, and political sensibilities that make comparative analysis difficult, if not futile. In Islamic theocracies such as Iran and Saudi Arabia (and Afghanistan

885

C

historically, though increasingly so again), the Mutawwa’in (religious police, though the religious police are referred to as Basij-e Mostaz’afin in Iran.) rigorously enforce Sharia law which applies to everyone irrespective of whether they are Muslim or not. Since there is no formal separation of church and state in Iran and Saudi Arabia, the Mutawwa’in have powers, duties, and responsibilities that rival, if not exceed, those of the regular police, not to mention a degree of influence that has simply no equivalent in Western democracies. The argument being made here is that due to the huge variation in the organization of policing between and within nations, any kind of cross-national comparison is difficult if not nigh on impossible. This is further complicated by the variability of their core mandate (citizen safety, public order, or religious/cultural concerns) and huge disparities in adherence to democratic norms and international human rights standards. Academic research is further hampered by the rivalries between national police organizations or between the police and the gendarmerie that makes such comparative assessments difficult. One of us (Ellison) has had personal involvement with a reform endeavor involving the Turkish police which was seriously (though not fatally) hampered by the unwillingness of the Jandarma (gendarmerie) to participate in the study because of a legacy of hostility and mistrust of the Turkish National Police (TNP). This might not have mattered but for the fact that the Jandarma (gendarmerie) are the de facto police in much of eastern Turkey and perform all the tasks and functions that the TNP perform in the rest of the country. Therefore, the research was only able to glimpse a partial picture of the organization of public policing in Turkey and as such was limited in terms of the overall conclusions and recommendations that could be made. Questions also arise in relation to the context of modern “policing” and what it should be taken to mean in the twenty-first century. Against a background of advancing economic liberalism, the European Union (EU), North American, and Australasian nations have shifted to forms of

C

C

886

partnership policing, which has overseen a flattening out of policing across civil society to harness more private enterprise in its provision and delivery. Who or what “the police” can be taken to mean under these circumstances is becoming increasingly muddled. In the UK, for example, many police forces employ Police Community Support Officers (PCSOs) who are uniformed personnel, but not sworn officers, and who have a range of limited powers in relation to antisocial behavior, traffic, and so on. PCSOs have been described by the previous Commissioner of the London Metropolitan Police, Sir Ian Blair, as part of what he termed “the extended police family,” but have been criticized by the police employees association, the Police Federation, for attempting to police on the cheap. The public meanwhile are unsure of what the role of PCSOs is supposed to be, since their legal powers are rather limited and their status somewhat vague. Indeed, in England and Wales the distinction between public and private is likely to be further distorted by the current government’s reform agenda via the Police Reform and Social Responsibility Bill, which will likely result in a fundamental restructuring of the British police. Proposals are currently underway to outsource a number of key services to the private sector such as forensics, emergency call logging, crime information systems, as well as the widespread establishment of strategic partnerships with private firms being encouraged to engage in crime reduction initiatives. Already, in many UK towns and cities, parking and minor traffic enforcement duties are performed almost exclusively by private companies who are contracted by local authorities. However, such public-private hybrid activities are not just confined to crime prevention and reduction. There is considerable evidence that national security policing (or high policing) is increasingly contracted to the corporate sector, and in the context of global insecurity, we are seeing something of a state-corporate symbiosis emerging whereby public and private actors engage in “a mutually beneficial application of game theory to achieve proximate objectives”


(O’Reilly and Ellison 2006, p. 14). Increased privatization, marketization, commodification, and hybridization have all problematized the very nature of policing and its delivery across a myriad of domains. One cannot, for example, analyze the counterterrorism “performance” of the US police separately from the huge corporate infrastructure (security consultancies, security contractors, risk advisors, and so forth) that has grown up around homeland security. To put the point more simply, it is becoming much harder to track and measure who does what, when, and how, making efforts at constructing crossnational baselines that are empirically verifiable, reliable, and generalizable difficult, if not again, nigh on impossible. Furthermore, in those jurisdictions that are characterized by high levels of violent crime (South Africa and Brazil are cases in point), private security arguably eclipses public policing – at least for those that can afford it. In both jurisdictions thousands of heavily armed private security firms with alluring names such as “Blue Angels” provide security services in response to a near total lack of trust and confidence in the public police. Furthermore, high levels of crime in the Brazilian favelas have spurred grassroots forms of civil policing, with hundreds of authoritarian paramilitary style, Autodefesas Comunita´rias (citizen defense committees), emerging to effectively take the law into their own hands and to deal with transgressors through a variety of extralegal punishments, such as shootings and beatings. This blurring of the boundaries between public and private policing/security further complicates any attempt at comparative analysis, unless we know in detail what these relationships are, and are able to control for the activities of public and private actors in a way that makes analytical sense. Even focusing on specific models of policing does not lead to any productive conclusions. A general strategy of policing that we find being implemented across the globe is community orientated policing, which, therefore, might provide us with a base from which to make comparative analyses. Irrespective of whether it “works” or not, community policing allows


police forces across the world to use a common language and as a philosophy offers a meeting point for “stakeholders” that becomes a forum for discussion and international collaboration via bi- and multilateral exchanges, reform efforts, conferences, seminars, and the movement of policing personnel within what Otwin Marenin has termed the “transnational policing policy community”(Marenin 2007). So at least in theory, it may provide a unified vision of global policing by establishing a series of international benchmarks. However, much of this remains aspirational, and as Brogden and Nijhar (2005) assert, owing to the different historical and political development of policing systems, the relationships between state institutions such as the police and the community are greatly differentiated between jurisdictions. This results in community policing becoming quite differently understood across and even within nations (2005, pp. 104–107). Brogden and Nijhar (2005, p. 110) expose such stark differences in the way community policing is deployed, particularly in regard to whether policing functions are centralized or decentralized that it is almost impossible to generalize from jurisdiction to jurisdiction. In some jurisdictions, such as Belgium, Northern Ireland, and Chicago, for example, attempts have been made to place community orientated policing strategies at the core of all police activities. But, in the most part, such strategies remain on the periphery of policing priorities, often playing second fiddle to national security priorities and the maintenance of social order. Certainly, the extent to which policing systems are autonomous from the state authority has a significant impact on the relationship between citizens and criminal justice system, skewing some police systems more towards maintaining social order and the political status quo than others. Quantitative Databases If we assume (at a stretch) that the core mandate of many police organizations is crime reduction, then sources of international crime data might be useful when attempting to make cross-national comparisons between individual forces. These

887

C

include data collated by INTERPOL, the United Nations Crime Surveys, and the European Sourcebook (see Stamatel 2006). In the INTERPOL data, police representatives of the member nations complete quantitative returns on police records in 11 offense categories: murder, sex offenses (including rape), serious assault, all kinds of theft, robbery, breaking and entering, motor offenses, fraud, counterfeiting, and drug offenses. However, as Stamatel reports (2006, p. 16), although the definitions for each offense are rounded as closely as possible within clear definitional frameworks, the data process is a voluntary one which provides little in terms of quality control measures over offense categorization. Even INTERPOL itself notes that “the figures must be interpreted with caution” (Interpol 1999). With regard to the United Nations Crime Survey, the UN has been collecting crime data since 1970 – with “a view to improving the analysis and dissemination of that information globally” (United Nations Office on Drugs and Crime 2007). The data is based upon multilingual questionnaires given to UN crime and police coordinators in member countries (Stamatel 2006, p. 17). They record aggregate national level crime figures within four areas: police, prosecution, courts, and corrections. Again, data is based upon offenses reported to the police and is collected within 10 offense categories: homicide, rape, assault, thefts, fraud, embezzlement, drug-related crime, bribery, corruption, and kidnapping. However, not all member states report back evenly nor are the data easily fixed within the same categories. For the data that compare prosecution and courts systems, it is not at all clear how the effects of different prosecution systems are controlled for, such as whether they are adversarial or inquisitorial, nor do they control for the effects of plea bargaining where some cases do not go to trial, which is used much more extensively in the USA and the UK than in other jurisdictions. Such data must be approached with caution, and it is difficult to make any cross-national assessments of the police or indeed of criminal justice agencies more generally.

C

C

888

Thirdly, the European Sourcebook of Crime and Criminal Justice Statistics started collecting data in 1990. It collates data within the categories of homicide, assaults, rape, armed robbery, motor vehicle theft, burglaries, drug-related offenses, and people trafficking. Unlike the other two sources, the European Sourcebook also collates information about the number of offenders by crime type and the percentage of offenders who are female, minors, and nonnationals (Stamatel 2006, p. 18). The sourcebook also claims to offer a higher level of quality control in the data collection effort (Stamatel 2006, p. 19). This includes standard classification schemes that are implemented over the course of the data collation process and the fact that data collection are easier to undertake within the European Union since most states have a closer affinity with regard to their crime classification. However, the sourcebook suggests that even within the European Union, there are multiple dimensions to crime classification and data collection that vary considerably between Western, Central, and Eastern European nations (Krajewski 2011). To take just one example, there is considerable cross-national variation in the statutory definition, reporting, recording, and prosecution of rape within European jurisdictions (Lovett and Kelly 2009). International Victimization Survey By its very nature, policing is a subjective exercise, located in huge amounts of discretion and partiality in what crimes are prosecuted and recorded. Thus, police statistics provide a skewed picture of offending and are limited to illustrating, at best, broader trends in crime (Maguire 2002; Skogan 1975). Moreover, it has been extensively documented that there are validity and reliability issues with official crime statistics as a means for assessing levels of crime and police performance in any given jurisdiction (see Maguire 2002; Skogan 1975). So, in addition to police records that are the mainstay of the three sources indentified above, the International Crime Victimization Survey (ICVS) collects data on victimization


rates (see Van Dijk et al. 1990). The ICVS provides researchers with some detailed points of comparison with officially recorded data and, therefore, is worth considering further. Most industrialized democracies have a long tradition of publishing victimization surveys (e.g., see British Crime Survey or the National Crime Victimization Survey). The ICVS surveys cross-national household experiences of crime, perceptions of policing, as well as general feelings of safety and security. The ICVS was initiated in 1987 by a group of European criminologists with expertise in national crime surveys (Van Dijk et al. 1990), and initially it was operationalized in 14 industrialized countries. The survey seeks to advance international comparative criminological research beyond the constraints of officially recorded crime data to produce estimates of victimization that can be used for international comparison. Surveys have taken place in 1992, 1996, 2000, 2005, and 2008, with 140 surveys now having been deployed in over 78 different countries – sampling over 320,000 citizens. The present database covers 325,454 individual respondents. Crime categories are broken down into vehicle-related crimes (theft of a car, theft from a car, theft of a motorcycle or moped, theft of a bicycle), burglary, attempted burglary, theft of personal property, and contact crimes (robbery, sexual offenses, and assault and threat). Arguably, the ICVS is the most comprehensive instrument yet developed to undertake an international comparative study of volume crimes, victimization, and perceptions of the police and criminal justice agencies. Significantly, the data is derived from surveys among the general public and, therefore, not influenced by political or institutional prerogatives. For the IVCS, validity and reliability is better maintained through the standardization of questionnaires, and results are presented within statistical confidence margins (see Lynch 2006). But full standardization, in all aspects of design, has still to be proven, especially for the surveys deployed in developing countries. Even if the ICVS provides data that are fit for international comparison,


we have to be wary of country-specific circumstances that will not always allow for fully standardized methodology to be applied. Moreover, although the ICVS provides a measure of common crimes, the comparatively small samples sizes precludes estimations of less prevalent crimes such as rapes or aggravated assaults. The ICVS also ignores victimization by complex crimes such as largesse forms of corruption or organized crime, which victimizes collective populations rather than individuals. Victim surveys and recorded crime statistic invariably produce two quite different pictures of the volume and distribution of crime. Yet, it is not clear whether this is because victim-based statistics are more accurate than recoded crime statistics or vice versa (Skogan 1975). Although some might argue that levels of recorded crimes cannot be reliably used for comparing levels of common crime across countries, measurement procedures in both formats illustrate characteristic errors. Nevertheless, victim surveys seem a slightly more productive source from which to undertake cross-national comparative analysis, although this should not be exaggerated and we still need to be cognizant of a whole range of conceptual and methodological issues.

Dilemmas and Problems Difficulties with Cross-National Analysis The entry has so far described three main sources of data that might be used to help in the production of cross-national comparisons of policing systems. Besides some of the specific issues already observed within each method, a number of general problems with undertaking such analyses can be identified. For instance, there are no unilinear patterns of development that all nations follow, and likewise, in many developing nations the structures of public policing and criminal justice generally diverge considerably from those found in advanced liberal democracies. Critics have observed that “western” ethnocentric concepts and categories are often superimposed on less industrially

889

C

developed countries and their translation varies enormously (Shichor 1990, p. 65; Nelken 2009). Cultural and geopolitical differences, therefore, make direct comparisons difficult (Nelken 2009). Still, it might be argued that crossnational analyses could be made across nations that have evolved into similar advanced democratic civic systems – such as the EU or North America – where measurable, complex, bureaucratic, and recording infrastructures such as performance indicators are deployed (see Collier 2001). For example, an inordinate number of indicators have been used to measure success and failure in crime reduction. Some indicators measure public satisfaction rates of police organizations, others crime clear up rates, while others track victimization rates. Certainly, patterns can be found on a number of key measures. Nevertheless, although liberal democracies are similar and more transparent, difficulties still arise, as performance indicators often track spurious clearance rates and measure arrests and detections rather than what has been achieved through say crime prevention measures. Performance indicators, however, usually coalesce around what have been termed “volume crimes” – due to their often uniformity in definition and large volumes in occurrence – and, therefore, are perhaps the easiest to measure. Again, differences arise in the way nation states not only record crimes but also in the administrative infrastructure available for recording to take place (see Krajewski 2011). On another level, offense categories themselves often prove problematic, as they do not easily subscribe to exactly the same definitional boundaries. Dealing with antisocial behavior is a key issue for the UK police, frequently ranking as the number one priority concern among the public, but it proves problematic when trying to fit it into a singular definitional or legal framework. Reducing analysis to one variable that is common and uniform in most societies, such as “homicide,” for example, might provide a good indicator, not just of rates between countries but also police performance in investigative terms. Indeed, various studies

C

C

890

have used homicide to make cross-national/ cross-cultural comparisons (e.g., see Avison and Loring 1986; Krahn et al. 1986). Still, what might look superficially similar is often framed within local, cultural, political, and legal contexts that limit our ability to make generalizations (Nelken 2009). Even within the UK, for instance, Scotland defines homicide as including the offenses of “murder” and “culpable homicide,” whereas in England and Wales, it is defined as including “murder,” “manslaughter,” and “infanticide” (Richards 1999). Elsewhere in the European Union, causing death by dangerous driving qualifies as manslaughter in some jurisdictions, but not all. In the UK suicide was only decriminalized in 1961 and prior to that time was regarded in law as a form of self-homicide. Currently, many jurisdictions regard death as the result of assisted suicide second-degree murder or manslaughter. However, in Australia’s Northern Territory, the use of the so-called Deliverance Machine to perform an assisted suicide was legal between 1995 and 1997 with the death not recorded as homicide. Similarly, in Switzerland assisted suicide remains legal with critics claiming that current legislation contributes to “suicide tourism” from other jurisdictions. Even for those offenses deemed to be homicides, initially around 15 % are reclassified as “other” following police or court action (Richards 1999). In a major study of homicide statistics in 36 European nations (Anamort European Project 2008), the researchers found that the misclassification of homicides was a significant problem in 14 of the countries surveyed. The combined effects of such misclassification were felt to “lead to underestimation of the magnitude of the deaths due to homicides in all of these countries” (Anamort European Project 2008, p. 3). The main problem appeared to relate to the lack of accurate information provided to coders on the cause of death which had not been forwarded from the police or coroner’s office. This appeared to be less of a problem among the established democracies of northern and Western Europe owing mainly to higher standards in


forensic investigation/pathology but also, as noted above, to more efficient transmission of medicolegal data to national statistical offices (Anamort European Project 2008, p. 3). Furthermore, as Richards (1999, p. 30) points out, in a number of jurisdictions, homicide statistics are compiled from when a death was registered as a homicide rather than when the death occurred. In the UK, all cases of suspected homicide are referred to the coroner’s court. However, in complex cases it is not unusual for the determination of cause of death to be made up to several years and in some cases decades later. The Police Service of Northern Ireland’s (PSNI) Historical Enquiries Team (HET) which was established to investigate over 1,800 cold-case deaths arising from the conflict is currently reinvestigating deaths that occurred 40 years ago. So far, a number of these cases have been referred to the coroner’s court for a determination on the actual cause of death. Our point then is that even attempting to compare police investigative performance between jurisdictions for something as ostensibly straightforward as homicide that seems to be measurable across a number of jurisdictions is fraught with difficulties around definition, classification, and ultimately interpretation.

Conclusion To reiterate the starting points, clearly, as policing has become a multimillion dollar and a more globally transnational exercise, the pursuit for efficiency has produced an expected desire to measure police performance on a cross-national basis. This entry asked whether any similarities could actually be identified across different nations in order to undertake such measurements. Three data sources have been considered: (a) academic comparative research, (b) official crime data, and (c) international victimization surveys. In these contexts, the entry has directed the reader to some of the difficulties that encompass carrying out any adequate and robust cross-national comparisons of policing systems. Comparison


both requires understanding and interpreting what those in other jurisdictions are trying to achieve. Whether it is comparing a particular police organization’s performance data or using crime statistics as a barometer of police performance, the cultural and administrative differences that characterize the organization of policing at the cross-national level have to be considered. Surveys such as the European Sourcebook can provide rich comparative data, which given the continued harmonization among states within the EU can lead us to perceive that international comparisons are becoming easier. However, within the EU there are inconsistent levels of technology and robust administration systems to accurately record crime which tends to give rather anomalous results in comparative terms. For example, some of the new accession states (Latvia, Lithuania, Estonia, Poland) appear on some indices to have even crime rates lower than that of Sweden, which traditionally within Western Europe has had low crime rates coupled with high levels of social welfare provision (Krajewski 2011). In addition, as noted above, some EU nations are shifting to decentralize their police (Belgium), some are tending towards enhanced centralization (the Netherlands, Scotland), and some are maintaining traditional organizational structures (Germany, France, Italy, Spain), while others are further privatizing aspects of the police role and function (England and Wales). These organizational differences are coupled with variant cultural understandings of “crime” and “justice,” variations in recording practices, the lack of comprehensive national datasets, and different legal cultures and political systems that lead us towards the conclusion that, to say the least, an objective benchmark comparative analysis is a tough proposition. Yet, despite the problems that have been identified, aggregate crime statistics may provide the most useful indicator of cross-national police performance, although these need to be qualified heavily and interpreted with some caution. Overall, however, the international vagaries of policing mean that a definitive cross-national comparative analysis remains a rather elusive goal.

891

C

Related Entries ▶ BCS ▶ British Police ▶ Comparing Police Systems Across the World ▶ Crime Mapping ▶ Development of the UCR and the NCVS ▶ History of the Dutch Crime Victimization Survey(s) ▶ Methodological Issues in Evaluating Police Performance ▶ National Victimization Surveys ▶ NCVS

Recommended Reading and References Anamort European Project (2008) Homicide-related deaths in an enlarged European Union, Anamort European Project, June. http://ec.europa.eu/health/ph_projects/ 2004/action1/docs/action1_2004_frep_22_a10_en.pdf Avison WR, Loring PL (1986) Population diversity and cross national homicide: the effects of inequality and heterogeneity. Criminology 24:733–749 Bayley DH (1990) Patterns of policing: a comparative international analysis. Rutgers University Press, New Brunswick Brogden M, Nijhar P (2005) Community policing: international models and perspectives. Willan Publishing Collier P (2001) Police performance measurement and human rights. Public Money Manag 21(3):2001 Commonwealth Human Rights Initiative (2007) Feudal forces: democratic nations: police accountability in commonwealth South Asia. Available: http:// www.humanrightsinitiative.org/publications/police/ feudal_forces_democratic_nations_police_acctability_in_cw_south_asia.pdf Dennis N, Erdos G (2005) Cultures and crimes: policing in four nations. Institute for the Study of Civil Society, London, Civitas European Sourcebook of Crime and Criminal Justice Statistics (2007) http://www.europeansourcebook.org/ Interpol (1999) International crime statistics. ICPOInterpol General Secretariat, Lyon Krahn H, Hartnagel TF, Gartrell JW (1986) Income inequality and homicide rates: cross national data and criminological theories. Criminology 24:269–295 Krajewski K (2011) Crime and punishment in central and Eastern Europe during the last 20 years – is the region different from rest of Europe? Plenary paper given at the European Society of Criminology annual conference, Vilnius, Sep 2011 Lau RWK (2004) Community policing in Hong Kong: transplanting a questionable model. Crim Justice 4(1):61–80

C

C

892

Lelandais I (2007) Measuring police performance: international experiences. International Centre for the Prevention of Crime, October Lovett J, Kelly L (2009) Different systems, similar outcomes? Tracking attrition in reported rape cases across Europe. Child & Woman Abuse Studies Unit, London Metropolitan University. Available: http://www.sccjr.ac.uk/projects/Different-Systems-Similar-Outcomes-Tracking-Attrition-in-Report ed-Rape-Cases/63 Lynch JP (2006) Problems and promise of victimization surveys for cross-national research. Crime Justice 34:2 Lynch JP, Addington LA (2007) Understanding crime statistics: revisiting the divergence of the NCVS and UCR, Cambridge studies in criminology. Cambridge University Press, New York Maguire M (2002) Crime statistics: the data explosion and its implications. In: Maguire M (ed) The Oxford handbook of criminology, 3rd edn. Oxford University Press, Oxford Marenin O (2007) Implementing police reforms: the role of the transnational policy community. In: Goldsmith A, Sheptycki JW (eds) Crafting transnational policing: police capacity-building and global policing reform. Hart Publishing, Oxford, pp 177–203 Mayhew P (1987) Residential Burglary: a comparison of the US, Canada and England & Wales. National Institute of Justice/Government Printing Office, Washington, DC Nelken D (2009) Comparative criminal justice beyond ethnocentrism and relativism. Eur J Criminol 6(4):291–311 Newman WL, Berger RJ (1988) Competing perspectives on cross-national crime: an evaluation of theory and evidence. Sociol Q 27:27–41 O’Reilly C, Ellison G (2006) Eye spy private high: reconceptualising high policing theory. Br J Criminol 46(4):641–660 Richards P (1999) Homicide statistics, House of commons research paper 99/56, House of Commons Library. Available: http://www.parliament.uk/documents/ commons/lib/research/rp99/rp99-056.pdf Shichor D (1990) Crime patterns and socioeconomic development: a cross-national analysis. Crim Justice Rev 15:64 Skogan WG (1975) Measurement problems in official and survey crime rates. J Crim Justice 3(1):17–31 Stamatel JP (2006) An overview of publicly available quantitative cross-national crime data. IASSIST Q Stenning PC (ed) (2003) Police use of force and human rights. Simon Fraser University, Criminology Research Centre, Occasional Paper 2003–02. Available: http://www.sfu.ca/crc/fulltext/pufrep.pdf Tanner MS (2005) Campaign-style policing in China and its critics. In: Bakken B (ed) Crime, punishment and policing in China. Rowman & Littlefield, Lanham, pp 171–188

Cryptology and Security United Nations Office on Drugs and Crime (2007) United Nations Surveys on crime trends and the operations of criminal justice systems. Retrieved 3 July 2007 from http://www.unodc.org/unodc/en/ crime_cicp_surveys.html Van Dijk JJM, Mayhew P, Killias M (1990) Experiences of crime across the world. Key findings from the 1989 international crime survey. Kluwer Law and Taxation Publishers, Deventer

Cryptology and Security Jonathan C. Hoskin Clemson Computing and Information Technology, Clemson University, Clemson, SC, USA

Overview Commerce in the modern world is based on virtually instantaneous electronic transmission of financial data including movement of money. The Internet has allowed us to move away from physical transactions via checks and currency. However, it is dependent on an unprecedented level of trust with an equally unprecedented potential of catastrophic financial loss. Illegal movement of money in the electronic form requires no physical access nor leaves physical evidence and can be accomplished from the other side of the world. Thus, the need for unquestionable security has arguably equaled or eclipsed that of military needs from just a few decades ago. With so much to gain and so little to lose, the criminal element has been transitioning to take advantage of every opportunity afforded by less than adequate security measures. Security must be well planned, implemented, monitored, enforced, and reviewed. Use of encryption and decryption is an absolute necessity and thus must be more dependable than any time in the history of the planet. The cryptographic algorithms used ideally must not simply be rock-solid, but beyond question, thus our dilemma. At best we know that cryptography is a work in progress and we unfortunately cannot

Cryptology and Security

prove it is perfect, – we can only prove it is imperfect. We use computer systems and all the rest of the infrastructure knowing that it is flawed by virtue of being made by humans. This short introduction notes some of the early methods which often translate into present-day concepts and vulnerabilities. Use of letter substitutions and transpositions defines early cryptology but was nevertheless a dependable method when used correctly. It has always been the misapplication of techniques that rendered cryptographic systems vulnerable. Topics include stream and block ciphers, passwords, and symmetric versus Asymmetric cryptography.

Fundamentals and Future Directions One of the more dependable themes in modern entertainment is the thriller in which the supposedly impossible was done. Security of a bank, museum, or fortified building is breached, and something is taken defying incredible odds. Invariably many types of security systems are circumvented, and by luck, guile, and intelligence, the deed is done. Its comparable to an airplane crash in that not one, two, or even three things went wrong but a combination and other factors that caused the accident, any of which if recognized would have changed the outcome. Computer security can be viewed in much the same way. Human intelligence is a marvelous thing, but considering every possibility requires that people are perfect and humans can at best overcompensate for their inadequacy. Recognizing that advances in technology may swiftly undermine security, overcompensation should not be considered a luxury. This short survey of cryptology begins with the above statement because it reflects reality. Every week it seems that another incredible security weakness has been exploited, usually resulting in or expecting to cause significant financial loss. Computer magazine, the flagship publication of the IEEE (the Institute of Electrical and Electronic Engineers) Computer Society noted in its October 2011 issue that “Hackers steal $13 million in 1 day via online security

893

C

breach (Garber).” History has shown that criminals constantly get better; they find and attack the weak parts of physical and computer/networked systems where they exist and circumvent systems when they do not. Contributing on the positive side are white-hat computer hackers that test and evaluate systems, often discovering weaknesses and security holes that plague software written with too little attention paid to security. The following document addresses the topic believing it to be fundamentally sound and capable of providing the security needed. However, like everything else in life, it is invariably the implementation and correct use of the system that complete the process. A simple password renders the best encryption system useless, and readers should understand any limitations will be exploited. At best cryptography and related security is a work in progress. In the modern world, encryption and decryption via use of credit cards, security cards, passwords, and other tokens are part of everyday life and becoming ever more important. No one would ever think to leave a $100 note bill in clear view in their automobile or on their desk at work, but somehow they are not as concerned about a password which may represent significantly more financial risk. This disconnect with reality is the target of social engineering, attempting to obtain such information via sociological means. Although social engineering is not the specific topic of this document, it should be considered at each point of a security audit. There are many approaches to address the topic of cryptology. A mathematical explanation is appropriate for those with the required mathematical background; however, this document will minimize formulaic expressions. Mathematics will be limited to numbers of possibilities and understanding the significance of large numbers (for a mathematical view, see Menezes et al. (1996)). Several years ago a story was circulated around the Internet that began: “Aoccdrnig to a rscheearch at Cmabrigde Uinervtisy, it deosn’t mttaer in waht oredr the ltteers in a wrod are, the olny iprmoetnt tihng is taht the frist and lsat ltteer be at the rghit pclae.” While the attribution was

C

C

894

wrong, the scrambled message makes a very important point. Breaking an encrypted message does not necessarily require getting every “t” crossed and “i” dotted. As a consequence, linguists and puzzle solvers were critically important in this field, and it is only the recent development of the computer-based cryptology that has reduced their need. Keeping messages private has always been an obvious requirement for military or diplomatic endeavors. The general study of such matters is called cryptology. It can be divided into two disciplines: obscuring or inhibiting messages from being understood by some manner of encryption, the study of which is called cryptography, and the breaking of those coded messages called cryptanalysis. In general use, cryptology is often referred to as cryptography which may lead to some confusion. This may be further complicated by steganography, the concept of hiding a message in some fashion and which itself may be encrypted. Often, the modern study of cryptology gives less recognition to early methods and origins, progressing quickly to complex computational methods which when used correctly can surpass the capability of being broken by even the most sophisticated computing system. While a general understanding of computer-based models is appropriate, its utility is beyond many criminals although those with a technical understanding or ability to employ those technically savvy should recognize that in fact it can be unbreakable. This document will therefore note various aspects of the historical development because the many possibilities remain useful in everyday life (Callery 2008). The classic document of cryptology leading into the computer age is The Code-Breakers (Kahn 1967) although some may find The Code Book a smaller and more digestible volume (Singh 2000). A shorter review can be found in the journal Cryptologia (Al-Kadi 1992.) Perhaps the earliest and most obvious method of obscuring information was the written form when few people could read. Related to that was use of a different language. The royal courts in Europe invariably spoke a different language than the common mostly illiterate person of


a given country, thus protecting conversations from workers and limiting interaction to all but the educated. Changing languages throughout the conversation or in written messages was almost guaranteed to protect information when the unintentional listener/reader is only marginally literate. We continue to employ this technique today to keep information about birthday parties or similar concerns from our children. This may also be accomplished by using more complex wording without necessarily needing to change languages, but children (and criminals) learn quickly. During World War Two, the United States used the Native American dialect of the Navaho to communicate securely between marines. This technique could only work if the opposing side had no expertise in the language chosen which had been carefully predetermined. Navaho was selected from among a wide variety of American languages because the population of speakers was high enough to meet military needs in terms of number of recruits. Traditional Navaho words also had to be reinterpreted to describe modern military hardware and other issues which were not part of the language. In a similar way, the “language of the street” incorporates words that replace others such as drug names or various criminal acts. Another simple method of passing such messages is to physically hide it such that the other side cannot find it. In early Roman times, a slave could have his hair cut, have a message tattooed on his head, and after hair regrowth, the message was hidden. Recutting the hair revealed the message after traveling to an appropriate location. Clearly not a mechanism for timely disseminating of information, it can have the advantage of hiding the message from the slave. Unfortunately for him, he need not be alive to convey the message. Using a locked strong box was a viable option for sending messages and other more tangible items, but rather than a single lock, a system was devised to use two locks. The owner placed items needing protection in the box and secured it with their lock. It was sent to another person, and instead of opening the box, something he or she


could not do without the appropriate key, they added their own lock via a second hasp and sent it back unopened to the sender. Upon the return of the box, the original lock was removed and resent to the previous sender. It could then be opened because only the original receiver’s lock remained on the box. However, this technique could be exploited by what is known as a “man in the middle attack.” This required the box be diverted to another person before reaching its original target, a second lock was added by the interloper, and it was returned to the original sender. If the owner had no verification that the box reached its original destination, its contents could be obtained by that intervening person, and the sender might never know what happened. This was a concern hundreds of years ago and remains an issue today with regard to digital information. There are numerous other ways to physically hide information. A message can be scratched into a material such as wood then covered with a layer of wax to hide it. A heat source will reveal the message after the wax drips away. In early roman times, the use of pigmented beeswax was an artistic technique now known as encaustic; thus, a message could masquerade as an artistic work. A slight twist to the above process would include the use of invisible ink. Typically, this has involved the use of “ink” which is clear when initially used. The ink may be as simple such as lemon juice or a specific chemical compound or mixture. Revealing the message involves treatment of heat or exposure to a chemical spray. Today, we can add notes to computer documents, and they remain hidden in plain sight because most people do not know about or consider their existence. Although the above methods work, eventually a need developed for a more capable system to convey complicated or lengthy information in a routine fashion. Usually, individuals encrypt and decrypt information themselves which implies another security concern – more than one party will know the mechanism. The ultimate goal of a cryptography system is to protect information even if adversaries understand how it works.

895

C

A fundamental concept was the substitution cipher where letters are substituted to obscure the message. The Atbash cipher used in the bible was perhaps not strictly for security purposes. Its name comes from its use, substituting the last for the first letter, the second from the last for the second letter, continuing as defined by the Hebrew alphabet. Another simpler cipher is ROT13 which rotates back on M the thirteenth letter. If you cut the alphabet in half between M and N and then slide the second half under the first half were A aligns with N, B with O, etc., you have a cipher that is easier to use. The message “hello” becomes URYYB and there is no confusion to complicate decryption. Letters can be transposed rather than substituted within words somewhat as noted above referencing Cambridge, but the anagrams generated are perhaps even less secure. Another transposition example will show the concept with a very short message. For the message ABCDEFGHI, the first nine letters of the alphabet, simply write the first three on the first line, the second three on the second line, and the remaining three on the third. Reading down the first column and then the second and finally the third, we get ADGBEHCFI for the encrypted message. The mechanism or cipher needs to allow for uneven rows and is not limited to only three lines. Obviously, there are many other possible ways of manipulating the message, reading down one column and then up the following column or going backwards. The one limitation is that all letters in the original message are present and there are a limited number of words and messages that can be made from them. It is easy to understand that compared to transposition, substituting letters can make the encrypted message much more difficult to decrypt. However, adding simple transposition to other building blocks of encryption can dramatically increase complexity to the process. An early device allowed for the simple sliding of one alphabet against another. This became known as the Caesar cipher because Julius Caesar used it with his generals. It consisted of two disks, one smaller rotating on the top of the other and both having the alphabet imprinted there on.

C

C

896


Cryptology and Security, Table 1 Plain and encrypted text created using simple methods Encrypting data is one way to keep it secure GPETARVKPI FCVC KU QPG YCA VQ MGGR KV UGEWTG

Plain text Encrypted text created by shifting two letters right Encrypted text using atypical characters (with underlines indicating spaces for clarity)

This simple substitution is known as a cipher which is a series of steps followed to encrypt or decrypt a message. The original information is known as plaintext (often shown in lowercase letters) with the enciphered version (shown in small uppercase letters) known as the ciphertext. We can envision they can be lined up so that in English, the “a” on the top disk is aligned with the “a” on the bottom disk. Turning the smaller disk two letters to the left (Caesar used three letters), we can then use this cipher disk to encrypt a message. This movement of letters by two is the auxiliary information usually called the key or sometimes the cryptovariable (shown in uppercase letters). This is easier than Atbash or similar techniques because the letters remain in alphabetic order. Using this substitution cipher above, the word “ate” becomes “CVG” and the line of text “encrypting data is one way to keep it secure” becomes “GPETARVKPI FCVC KU QPG YCA VQ MGGR KV UGEWTG,” which initially seems imposing (Table 1). This would likely be unreadable to someone with no cryptography skills but is nevertheless very weak. Removing the spaces will make the message look even more difficult but is no match for the cryptanalysis methods below. Difficulty of code breaking may also appear to be increased by using alien or nontraditional characters such as those taken from other languages, cultures, or additional fonts. This is deceptive only for those not familiar with encryption. Each language has their own peculiarities, there are a limited number of letters which are repeated, and some letters occur in one order but never reversed. In English a double “h” is unlikely compared to a double “e,” and a “u” will follow a “q” but never in the reverse order.

More important in deciphering messages are the frequency of the letters which cannot be hidden in the above examples. The standard frequency distribution of letters is different for a given language. Generally, in English the letter frequency order is E, T, A, O, I, N, S, H, R, D, L, C, U, M, W, F, G, Y, P, B, V, K, J, X, Q, and Z. A message of sufficient length or multiple simple substitution messages combined allow you to break the message. Simply replace the most frequent letter with E and the second most frequent with T and continue with all letters, and even if not perfect, the result is probably decipherable. Another way of representing letters is to use not one but two or three characters. The benefit of this technique is a single letter can be represented by multiple combinations. When each letter is represented by two alphabetic characters, the number of possibilities is 262 or 676, thus providing 26 combinations for each letter if distributed evenly. Using three characters, the number is 263 or 17,576 with even greater possibilities. Thus, the letter “a” might be “AEL,” “MLT,” and “ZYE,” for example. When two or three characters are used to represent one letter, not only can each letter have multiple representations but they might also key for words. Thus, “AML” might be code for “bank,” and “LTL” could translate to “dollars.” To decrypt a mixture of substituted letters and code words requires its’ nomenclator. Certainly one of the most historically important encrypted messages was known as the Zimmermann telegram and used a numeral-only nomenclator with numbers up to 99,999. The purpose of that telegram was to entice Mexico into World War One on the German side and keep the USA out of Europe long enough to subdue England. Having used the same nomenclator previously, it had


already been partially broken, and its decryption was a significant point in the war. The usefulness of this possibly more secure technique using a nomenclator is sometimes diminished by the necessity of agreeing on all likely code words beforehand and requiring a code book. If security is the purpose, the nomenclator must be stored and used securely to not fall into the wrong hands. But as noted above, increasing use results in decreasing security. To expand the possibilities of the Caesar cipher and increase its complexity and thus security simply requires not keeping the disk or substitution the same. If, for example, the first letter is moved by two and the second by four, “ee” now becomes “GI,” and if repeated, the result is not so obvious. That initially makes the cryptanalysis more difficult, but using frequency analysis it would eventually be recognized that there are two different groups of letters – the simplest polyalphabetic cipher. The natural end to this technique is to use all the possibilities and was ultimately done by Blaise de Vigene`re (see Table 2). Continuing on the same road, what if a word was used for the cipher key? If the word “CIPHER” was the cipher, that would mean lining up the “a” with the “c” and using the wheel to determine the first letter and then lining up the “a” with the “i” for the second letter. After the sixth letter a lined up with “r,” the same code is repeated until the message is complete. As the code word cipher becomes longer, cryptanalysis becomes ever more difficult but not impossible. Up to this point in the development of cryptology, the cryptanalysts were winning but that was about to change. Consider what happens if the code keyword is 100 random letters from a to z, now the encrypted message becomes extremely difficult to break. True randomization makes the encrypted message random characters and thus unbreakable when used correctly. One important requirement becomes access to truly random numbers. Before computers became available, the process of generating random numbers might be similar to the numbered balls used for lottery drawings. However, because encryption has become ubiquitous and the need for random numbers dramatically

897

C

Cryptology and Security, Table 2 Vigene`re square Key 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

abcdefghijklmnopqrstuvwxyz BCDEFGHIJKLMNOPQRSTUVWXYZA CDEFGHIJKLMNOPQRSTUVWXYZAB DEFGHIJKLMNOPQRSTUVWXYZABC EFGHIJKLMNOPQRSTUVWXYZABCD FGHIJKLMNOPQRSTUVWXYZABCDE GHIJKLMNOPQRSTUVWXYZABCDEF HIJKLMNOPQRSTUVWXYZABCDEFG IJKLMNOPQRSTUVWXYZABCDEFGH JKLMNOPQRSTUVWXYZABCDEFGHI KLMNOPQRSTUVWXYZABCDEFGHIJ LMNOPQRSTUVWXYZABCDEFGHIJK MNOPQRSTUVWXYZABCDEFGHIJKL NOPQRSTUVWXYZABCDEFGHIJKLM OPQRSTUVWXYZABCDEFGHIJKLMN PQRSTUVWXYZABCDEFGHIJKLMNO QRSTUVWXYZABCDEFGHIJKLMNOP RSTUVWXYZABCDEFGHIJKLMNOPQ STUVWXYZABCDEFGHIJKLMNOPQR TUVWXYZABCDEFGHIJKLMNOPQRS UVWXYZABCDEFGHIJKLMNOPQRST VWXYZABCDEFGHIJKLMNOPQRSTU WXYZABCDEFGHIJKLMNOPQRSTUV XYZABCDEFGHIJKLMNOPQRSTUVW YZABCDEFGHIJKLMNOPQRSTUVWX ZABCDEFGHIJKLMNOPQRSTUVWXY ABCDEFGHIJKLMNOPQRSTUVWXYZ

increased, we are limited to computer-generated or pseudorandom numbers. Such numbers may not be perfectly random but are sufficiently random to not reduce the security of the encryption. For cryptographic purposes, the process of random number generation must be verified to ensure quality, or unbiased random numbers are generated. During World War Two, making a cipher unbreakable involved using a one-time pad, a piece of paper with key codes. Critically important is that it is only used once. Codes repeatedly used create patterns cryptanalysts use to break the code. Judicious use of a sufficiently long and truly random one-time pad may fulfill the cryptographic needs, but it creates other problems – it is only as secure as the one-time pad, and transmission of such information creates its own problems. When the adversary finds 30 days or more

C

C

898

of one-time pads, because they may already be disseminated to ships at sea, for example, you may have lost the ability to keep messages secret. Likewise, given the length of the one-time pad should be at least as long as the message itself, this unbreakable method is unfortunately of limited utility in the current world. By the early 1900s, there had been sufficient interest in cryptography to entice inventors to attempt making devices to accomplish this task. The best documented is the Enigma machine used by the Germans in World War Two which was an electromechanical device to encrypt/ decrypt messages. It might be simply described as an electrical–mechanical version of multiple Caesar cipher disks called rotors, chained together such that when a typewriter key is pushed, the result is an illuminated letter. It had the possibility of being nearly unbreakable, but the allies had one so they knew how it worked and it was misused. The Germans essentially repeated a portion of the same message each day because the daily weather report began with the same words. Thus, if settings on the Enigma could be manipulated to decrypt the ciphertext back to those words, they could decrypt the entire message. Although this was not an easy task given the state of computing devices, it allowed messages to be broken with hard work. Not every transmitted message would be decrypted, and those that were might not have been decrypted in time to be useful. The above examples are based on traditional characters or ones that replace those characters. With the advent of computers, all characters were replaced by a binary code, strings of ones and zeros. Seemingly elementary this enabled a vast and wide-ranging new world of unprecedented complexity. For the uninitiated, a simple example is no further than a typical garage door remote. To set or change the code, you open it and move the 8, 12, or more switches in both the sender and receiver. (Leaving the original settings unchanged is the same as keeping the default password on a computer – it is only marginally better than not having a password or code.) Eight similar switches grouped together are known as a computer byte.

Cryptology and Security Cryptology and Security, Table 3 Number of possibilities for selected numbers of bits Bits 1 2 3 4 5 6 7 8 12 16 24 32 64 128 256 512 1,024 2,048 4,096 8,192 16,384

Bytes 1/8 1/4 3/8 1/2 5/8 3/4 7/8 1 3/2 2 3 4 8 16 32 64 128 256 512 1.024 2.048

Math 21 22 23 24 25 26 27 28 212 216 224 232 264 2128 2256 2512 21,024 22,048 24,096 28,192 216,384

Number of possibilities 2 4 8 16 32 64 128 256 4,096 65,536 16,777,216 4,294,967,296 1.84467 1019 3.40282 1038 1.15792 1077 1.34078 10154 1.79769 10308 3.23170 10616 1.04439 101,233 1.09075 102,466 1.18973 104,932

A byte consists of 2 2 2 2 2 2 2 2 possibilities also represented as 28 or mathematically as 256 possibilities. This respectively represents a doubling of possibilities with each additional character or 2, 4, 8, 16, 32, 64, 128, and 256. This is a key concept because the code possibilities increase geometrically, while the ability to test a code or pass-phrase is linear (Table 3). This has a dramatic effect as passwords get longer, and the difficulty of breaking passwords becomes increasingly difficult depending on the current state of the art of cryptanalysis. If we spend twice as long inputting pass-phrases, we have only attempted twice as many codes. Ultimately, the trade-off is defined by the cost of breaking the code (the computational expense) vs. the worth of the protected information. By choosing the option of 12 bits on a garage door opener, we effectively have a byte and a half or 2 2 2 2 2 2 2 2 2 2 2 2 possibilities also represented as 212 or 4,096 possibilities. This is not an overwhelming number,


and a device that produces all possible signals simulating all garage door-opening possibilities is easy to envision. Another byte (220 or 1,048,574 possibilities) of additional complexity would make it increasingly unlikely that a code generator could produce all the possible code signals in a short span of time. The garage door opener is a security device meant to keep honest people honest and cannot be considered sufficient to protect someone that has been targeted by those with criminal intent. Although computer technology may have been pressed into the cryptology arena very early because of world wars, passwords were not the initial concern. Nevertheless, we use them every day and they contain information which may be built upon to understand others’ aspects of cryptology. Portrayal of the mathematical complexity of password encryption is difficult and most obviously incorrect in popular media. Breaking a password is often presented as identifying each character, one at a time in sequence, which might accurately represent a safe cracker listening to the tumblers fall in place using stethoscope on an old-style safe but has nothing to do with passwords. This is particularly disconcerting because, as any computer or ATM (automated teller machine) user inherently understands, you either get the password right or you do not. Unfortunately, the underlying message conveyed to the public is that passwords cannot be safe and are easily broken, thus unintentionally undermining the need for using good passwords. Believing you know all but the first or last character of a password is the same as not knowing the password. At that point, it seems only a brute force attempt at breaking the password or portions believed to be unknown will allow it to be determined. A brute force attack means trying every possibility until the correct one is discovered. The goal of the cryptographer is to limit access to only the correct password or a brute force attempt because the latter is defined by the state of computing technology and the implementation of the system. But this only works dependably if the user has a good password, hereafter known as a pass-phrase, and keeps it protected.

899

C

Use of pass-phrase instead of password is an important change meant to entice and empower the user into choosing more secure passwords. The user need not have to remember a passphrase; it can be recorded on a piece of paper and subsequently stored in a wallet or purse. Thus, it is considered as important as money, credit cards, and other information which the user intentionally keeps out of reach by others. Regardless of complexity, the user ultimately remembers the pass-phrase even if it takes several days. However, it should therefore be obvious that requiring frequent pass-phrase changes discourages use of strong pass-phrases and conveys the entirely wrong message. With regard to pass-phrase parameters, two important factors are complexity, known as entropy, and length. Entropy is dependent on the possibilities of the pass-phrase. If a user is limited to only numbers or 0 to 9, the difficulty of breaking it is much reduced. Normally, systems should allow pass-phrases consisting of the alphabet containing 26 upper- and 26 lowercase letters along with the ten digits and a variety of special chapters. The standard for keyboard input is the ASCII (American Standard Code for Information Interchange) table and consists of 128 possibilities. Key combinations with the Ctrl, Alt, and other keys can dramatically increase the number of possibilities. But because every input device does not have every character or modifying keys, even the standard ASCII list of 128 may be impractical. The much more important factor is passphrase length. If the pass-phrase is limited to 26 letters (case being irrelevant), ten numbers, and four special characters (e.g., @, #, $, and %), that would be 40 characters. Therefore, using only one character, the number of possibilities is 40, and with two characters (402), the number of possibilities increases to 1600. Moreover, using six characters (406) dramatically increases the possibilities to over four billion. (Note that using repeating characters such as 111111 for a pass-phrase is more likely to be observed, and thus, systems will generally limit the number of repeated characters to two.) Similarly, if upper- and lowercase letters can be used

C

C

900

(52) and numbers (10) and eight special characters, one additional character increases the possibilities by a factor or 70 and even more substantially with additional characters. Assume you have 70 possible characters and you are required to have an eight-character pass-phrase. That would be 708 or 576,480,100,000,000 possibilities. You could correctly guess the pass-phrase the first time or the last time, but on average it would require trying half of the possibilities or take 288,240,050,000,000 attempts – a brute force attack. If you try one every 10 s, it would take 48,040,008,333,333 min or 91,400,302 years to guess the pass-phrase on average for someone using a good pass-phrase. That seems like reasonable security risk but assumes the system will let you continuously attempt to input the pass-phrase without slowing down or stopping, an obvious implementation error. But more importantly, it is a good idea because we do not know how technology might change in the future. Knowledge of a person can sometimes be used to guess a pass-phrase: the name of an owner’s dog or other known or easily obtainable information is a poor choice, and their use perpetually hounds the lives of celebrities. Generic bad pass-phrase lists can be found on the Internet and invariably include the word password. The next and most likely mechanism to correctly guess the pass-phrase is by what is known as a dictionary attack. They involve trying all words and probably some simple derivatives also including foreign language words. That files exist purposely for dictionary attacks should be considered from a positive point of view; they emphasize the need for a good passphrase. Although many types of specific cryptanalysis attacks exist, they are relegated to longer texts, and the only remaining possibility for a pass-phrase is a brute force attack. Guidelines for choosing pass-phrases exist, but only systems which require complexity (at least one letter, number, and special character), define minimum length, and limit character repetition can ensure sufficiently secure pass-phrases are used. Often they allow pass-phrases to be reused


after some number of changes or allow changing to very similar pass-phrases, another potential risk. Secure systems do not allow continually trying pass-phrases which is a necessity but also problematic in itself. If a system only allows three attempts before blocking an account, a given account can be rendered useless if anyone can try three one-character pass-phrases to block it. The price for this type system must include the cost of an administrator required to frequently reset pass-phrases and both time and effort diverted from the user’s efforts. Another method to deal with such problems requires the user to wait 5 min before he or she can again attempt to login, and only after a greater number of attempts is the account blocked requiring an administrator reset. Giving the user a method for self-resetting of pass-phrases is another option but is itself a security consideration. The mechanism used by the system to process pass-phrases is often not known by users, they just assume it is recorded somewhere, and when the correct pass-phrase is input, the user gains the appropriate access. It should be obvious that if the user’s pass-phrase is compared to a recorded version, then the pass-phrase must be available in some way which is necessarily a security issue. Thus, for security reasons, such a process is no longer used, and in fact generally pass-phrases should not be known by anyone except the user. Instead of keeping the pass-phrase on the computer to compare it with input, a hash of the passphrase is saved. As the word hash implies, the pass-phrase is essentially chopped like the meat and potatoes combination with the same name. A computer hashing algorithm (used interchangeably with cipher) converts the input to a string of characters that cannot be reconstituted. It is a secure one-way function. The hashing process is similar to the CRC or cyclic redundancy check that quickly checks to determine if a file is believed to be the one associated with the file name. However, CRC is not intended to be a secure check as with the passphrase hashes noted above. Nevertheless, it is extremely difficult to create an illegitimate file


901

C

Cryptology and Security, Table 4 Examples of text hashed with CRC-16 and SHA-1 algorithms Text (underline indicates differences) Encrypting data is one way to keep it secure Encrypting data is one way to keep it secure Encrypting data is one way to keep it secure Encrypting data is one way to keep it Secure Encrypting data is one way to keep it secure

Hashing Algorithm result CRC-16 SHA-1 AA1C C799E6694ACA06AB82A390D5746645E1A2FE74F3 5225 ACC68E63A2986E8979E05FCEDB87E174A48D8AA7 6838 6284A2EC6C476A99C42C0AC838E443299D76A9CF CA1B 7D4D20D0272C2F3C359272EEA92076FE71FE23F8 E3AB F3DF22162D6DD7FE589E0B996B349B9FC6DE5C9F

with the same CRC value also having the same name as what it is attempting to impersonate. This is what makes the CRC value such a useful tool to monitor file changes. For the hashing algorithm (examples include MD-5, Whirlpool, and CRC-16) to work as a test of pass-phrase accuracy, the same pass-phrase must always result in the same hash value. Fortunately, system administrators need not know a pass-phrase for a specific user; they merely need access to change a pass-phrase. Changing the pass-phrase to something different and subsequently requiring the user to change it again within a few uses helps to ensure the system limits what system administrators can do when all such activity is recorded or logged. However, hackers routinely erase or modify logs in an attempt to hide their tracks. Several examples of text that has been hashed give the reader an appreciation for the technique shown in Table 4. The examples are derivatives of the text “Encrypting data is one way to keep it secure.” Each line is hashed using two algorithms. First is the shorter CRC-16 (cyclic redundancy check) of 16-bit or 2-byte length and designed for quickly checking to see if a file has changed. Converting the larger text to only four characters of the resulting hash shows the power of a simple hashing algorithm. (When the results do not include letters above F, it indicates that characters are hexadecimal or base 16 which includes numbers 0–9 and letters A, B, C, D, E, and F.) For this simple example above, I used the free program Easy Hash v. 1.6 by Tomasz Kapusta to generate the hashes, but similar programs can be found on the Internet. The more

complex SHA-1 is a widely used hash algorithm and has a 20-byte or 160-bit length equivalent to 40 characters. It was designed for the US National Security Agency in 1995. Although each line in the example only differs from the first line by an extra space or the case of one letter, the result of the hash is dramatically different. SHA-1 received extensive testing before being selected by the NSA, but, like all widely used algorithms, eventually attacks will reveal weaknesses, and this will soon be replaced by a more secure standard. Unfortunately, users can create situations that require a correct pass-phrase to be used without the ability to simply change it. An example of this issue is when a drive, folder, or file is securely encrypted and does not come under the control of an administrator and the person in question has lost/forgotten the pass-phrase or is no longer available. If the pass-phrase is cryptographically sound, then the protected entity essentially no longer exists because breaking that pass-phrase maybe virtually impossible. As encryption of data becomes more popular with individuals, important family information is put into jeopardy if that person has not given the pass-phrase or key to their lawyer or other family member. A more secure way of protecting critically important information is to use a technique whereby multiple users are required to find the correct pass-phrase. This can be envisaged using an X, Y grid. Assume the correct pass-phrase is located at zero on the Y axis and is equivalent to some number X. For simplicity we will use the pass-phrase at 10, 0 (X, Y) on the grid. A straight

C

C

902

line can be drawn through that point and through points 11, 1 and 8, 2. Knowing any two points anywhere along the line will define the line and thus define the pass-phrase as 10 because that is where the line passes through the Y axis. Ten different people could be given different X–Y coordinates, and any two can determine the pass-phrase, but one person alone could not define the Y axis location. In a similar fashion, use of other mathematical functions can require three or more sets of coordinates to define where the line crosses the Y axis. Thus, collusion or compromising one point does not immediately compromise the pass-phrase, but its complexity does not warrant use for anything except critical data. However, whoever controls the system is therefore another security consideration. Given the limitation of some number of characters for a pass-phrase, it is possible to generate all the possibilities for a given hashing algorithm, saved as a “rainbow table.” These are available on the Internet but are large, in the range of a terabyte or more. Thus, if one can obtain the hashed pass-phrase, locating it in the table will produce the originating pass-phrase. This works because a comparison of results of hashing programs should generate the same hash although in use, and unbeknownst to users, the result can purposely be different. Just as adding salt to food hash will change it, adding a salt (an additional factor) to a hashing algorithm changes its result as well. For example, simply appending an additional single character to the end of each pass-phrase will change the hashed result dramatically, as implied above and thus rendering a hash table useless. Because the same character is added each time to each pass-phrase, the resulting hash will be the same for each specific passphrase but will be different from the non-salted pass-phrase. Probably the most simplistic and yet elegant operation in computer-based cryptology is the Exclusive Or Logical Operator also known as XOR (pronounced X-Or). It is symmetrical, meaning it works the same for both encryption and decryption. Remembering that computers deal with 0 s and 1 s, it is obvious that they work well with logic table based on true and

Cryptology and Security Cryptology and Security, Table 5 Encryption and decryption using the exclusive or logical operator 1 2 3 4 5 6 7

Message Key XOR Encrypted Key XOR Unencrypted

0 1

1 1

0 0

0 1

0 0

0 1

0 1

1 1

0 1

0 0

1 1

0 0

1 1

1 1

0

1

0

0

0

0

0

false (Table 5). Given the result, this is a deceptively simple operation. The message (line 1), a string of 0 s and 1 s, is XORed with the key (line 2) also a string of 0 and 1 s. The key and the encrypted message are lined up, and for each bit, if both are 0, or both are 1, the resulting encrypted bit (line 4) is 0. Conversely, if they are different, the unencrypted bit is a 1. Un-encrypting the encrypted message works the same way. The result is the original message (line 7). This is a very fast operation with security dependent on the key. A short simple key of eight bits (256 possibilities) can be broken easily, while a non-repetitive random key is essentially the one-time pad and cannot be broken when used correctly. Therefore, we continually increase the key length for secure functions with 256 bits (1.15 1077 possibilities) on the horizon and even longer keys available as that eventuality is expected. Most of the above consists of cryptographic basics and how they may be used. Many operations are chained together to increase the difficulty of cryptanalysis. Much of what is used consists of the two major types of modern ciphers which include stream and block types. The stream cipher may be considered similar to the XOR system above and the one-time pad. It encrypts and decrypts the message bit by bit but is very fast. Unlike the one-time pad which requires a key as long as the actual message, the stream cipher uses a key stream generated with a much smaller key. An example of a stream cipher is RC4. In contrast to the stream cipher, the block cipher, as the name implies, processes the message blockwise or equally sized chunks. Over


time the block size has been increasing to stay ahead of more capable cryptanalysis techniques and computer advances. The minimum length has moved to 128 bits but will continue to increase. Actual use may be similar to the polyalphabetic substitution noted above, but as expected it is not that simple. Block ciphers might be compared to a chain knot in that each subsequent knot is dependent on what goes before it and is therefore unlike the stream cipher in that a single missed bit might not obfuscate the whole message. Security is enhanced by repeating the simple function known as a round. Multiple rounds and other manipulations the security. Recognizing the need for such a cipher for nonmilitary or official uses led the US government to release a standard, the Digital Encryption Standard in 1976. With only a 65-bit key length, it was considered obsolete no later than 1998 when it was broken, but a variant – Triple DES – was used in its place. A competition resulted in the Advanced Encryption Standard or AES being released in 2001 after several years of testing. Other well-known block ciphers include Blowfish, Serpent, and Twofish. Both stream and block ciphers are symmetrical, meaning they are used for both encryption and decryption. Cryptography had only been symmetrical until the 1970s when asymmetrical ciphers were first created. Conceptually, this would be more secure because previously the key was the same in both directions, and therefore, more than one person would need to know the key. One inherent security problem that existed was the inability to protect a symmetric key. This was finally solved using a key pair or what is known as public key cryptography (Diffie and Hellman 1976). Initially, its strength relied on the difficulty of factoring prime numbers of very large numbers and more recently elliptic curves. A key pair is generated with one used as a private key and the other a public key. The significance of the keys is simply that at least one is kept secret and which is irrelevant. The most important consideration is that a cryptographically secure key needs to be much larger than others mentioned above, 1,024 bits or 128 bytes

903

C

(representing 21024 or 1.8 10308 possibilities) which makes it a noticeably slower, on the order of a thousand times slower. But the additional time and key length is used to encrypt a key which is subsequently used in the much faster stream cipher above to encrypt or decrypt the message. Consider a simple scenario with email; someone sends you a message which is encrypted with your public encryption key which might be posted on the Internet for anyone to use. Note the similarities between this example and the locked box scenario above. You in turn decrypt it with your personal key. Because a different key is needed to decrypt the message, anyone could send a message to you, but only you, by virtue of the fact that you have the private key, can decrypt it. Although this sounds of little use by itself, consider that the original message is encrypted using the sender’s personal key and also encrypted using your public key. The result is a message that could only be sent by the originator because it is encrypted with their private key (you decrypt it with their public key) and subsequently only you can decrypt it with your private key because it was also encrypted with your public key. While governments and corporations may question its use, consider that the vast majority of email is unwanted spam and that public/private key encryption could dramatically reduce unwanted email. It could significantly impact both network bandwidth concerns and infrastructure costs but is unlikely to become a standard. The above represents a very limited introduction to cryptology without considering possible future developments. A new technology like quantum computing having the ability to much more quickly break codes makes constant vigilance a necessity for cryptographic security professionals. Ramifications of these or other new technologies make cryptographic security always a work in progress. But no matter how much we focus on the technology, that cannot be the end or as Bruce Schneier (2004) put it “I’ve realized that the fundamental problems in security are no longer about the technology; they’re about how to use the technology.”

C

C

904

CTC

Related Entries

Cultural Criminology ▶ Automated and Manual Forensic Examinations ▶ Computer Forensics

Jeff Ferrell Texas Christian University, Fort Worth, TX, USA University of Kent, Canterbury, UK

Recommended Reading and References Al-Kadi IA (1992) The origins of cryptology. The Arab contributions. Cryptologia 16(2):97–126 Callery S (2008) Codes and Ciphers: all the tips and tricks you need to make them and break them. Harper Collins Publishers, NY, NY. Diffie W, Hellman ME (1976) New directions in cryptography. IEEE Trans Inf Theory 22(6):644–654 Garber L (2011,October) News briefs. Hackers steal $13 million in one day via online security breach. Computer p 15. IEEE Computer Society Kahn D (1996) The code-breakers: the comprehensive history of secret communications from ancient times to the internet. Revised edn. Scribner, NY, NY. Menezes A, von Oorschot P, Vanstone S (1996) Electronic format 1997 (www.cacr.math.uwaterloo.cc/hac). Handbook of applied cryptography. CRC Press Schneier B (1996) Applied cryptography: protocols, algorithms, and source code in C, 2nd edn. Wiley, New York Schneier B (2004) Secrets & lies: digital security in a networked world. Wiley, New York, p. 398 Singh S (2000) The code book: the science of secrecy from ancient Egypt to quantum cryptography. Anchor Books, Random House

CTC ▶ Communities That Care

CTC Prevention Coalition ▶ Communities That Care

CTC Prevention System ▶ Communities That Care

Synonyms Labeling and Deviance; Labeling: History and Concept; Modern Marxist and Radical Theory; New Media and Crime Images; Postmodern Criminology; Reality Based Television Shows

Overview Cultural criminology is designed to place issues of meaning, symbolism, and interpretation at the forefront of the criminological enterprise. Cultural criminologists argue that both crime and crime control operate as cultural endeavors, with their personal and social consequences constructed out of contested processes of collective representation and interpretation. Given this, criminological inquiry must encompass more than crime and criminal justice as conventionally understood; it must also focus on transgressive subcultural practices, media representations of crime and victimization, public displays of anger and elation, and public performances of justice and injustice. In order to fully theorize such phenomena, cultural criminologists contend, criminology must in turn resurrect older traditions of cultural and interactionist analysis while also drawing on newer perspectives from visual theory, cultural geography, and related fields. In this way cultural criminology offers a critique of conventional criminology and its circumscribed range of theories and methods and at the same time attempts to reconstruct criminology as a more expansive, interdisciplinary, and intercultural undertaking. Cultural criminology likewise attempts to reorient criminological analysis to the distinctive social and cultural dynamics of the present late modern

Cultural Criminology

period and so to enhance criminology’s ability to penetrate the contested global politics of crime and crime control as they are now developing.

Fundamentals of Cultural Criminology While cultural criminology has emerged as a distinct theoretical and methodological orientation within criminology only during the past two decades, the intellectual roots of cultural criminology can be found amidst earlier perspectives within sociology and criminology. For cultural criminologists, the goal has been to reclaim these perspectives as essential components of criminological inquiry, to reimagine and reinvigorate them in the context of contemporary crime and crime control trajectories, and to enliven them with new perspectives from other disciplines. Intellectual History At its core, cultural criminology constitutes a synthesis of two theoretical orientations: the first North American and the second British. The North American orientation dates to the mid-twentieth century and the emergence of symbolic interactionism and labeling theory in the analysis of deviance and crime. For interactionist and labeling theorists, the appropriate subject matter of criminology was not crime as an individual act, but rather the web of social interactions and interpretations by which any act might or might not be constructed as crime. At issue was the meaning of crime and the exercises of power and discrimination by which that meaning was determined and redetermined within an ongoing social process. As regards crime and crime control, the “symbolic” dimensions of symbolic interaction suggested that a transgressive act could be made to symbolize heroism or horror, self-defense, or assault; likewise, the practice of crime control could be presented as a matter of good policing, patriotic participation, or political repression. For these theorists, the reality of crime was inherently a social and cultural reality – and a political reality as well. By understanding the symbolic construction of crime and crime control,

905

C

criminologists could begin to penetrate the proselytizing claims of moral entrepreneurs, the discriminatory dynamics of the criminal justice system, and they lived experiences of those labeled criminal. These lived experiences could in turn be documented through “naturalistic” case studies and careful ethnographies that revealed patterns of shared meaning and symbolic communication within marginalized subcultures. Here, then, was one foundation of cultural criminology: an attentiveness to symbolism and meaning in the construction of crime and a critical analysis of crime, culture, and power. Influenced by the work of the North American interactionists, sociologists and criminologists associated with the National Deviancy Conference, the Birmingham School of Cultural Studies and the “new criminology” in Great Britain began in the 1970s to develop what would become cultural criminology’s second intellectual foundation. Sharpening the critical potential of interactionist and labeling approaches, they carefully investigated the cultural avenues through which power was by turns enforced and resisted. Situating crime and crime control in historical and political context, they likewise analyzed the shifting ideological utility of crime, exposing the ways in which crime concerns and anticrime campaigns were made to fit within larger political agendas and historical trajectories. These researchers also engaged in the sort of case study and ethnographic research developed among interactionist criminologists, finding amidst the leisure worlds and illicit subcultures of their subjects moments of stylized defiance and collective resistance to legal authority. Taken as a whole, this work not only continued the interactionists’ reconceptualization of crime as a social and cultural construction; it also began to reconceptualize the very nature of power and social control in contemporary society. Power in many ways now denoted the power to control the meaning of public events and social control the ability to control the terms by which the activities of marginal groups came to be understood and interpreted. In the 1990s these two orientations were synthesized for the first time into a distinct “cultural

C

C

906

criminology” that explicitly drew on both the North American and British traditions in an attempt to theorize contemporary intersections of crime and culture. To supplement and expand these earlier approaches, cultural criminologists now included a variety of additional disciplinary and interdisciplinary perspectives: anarchist criminology and other critical criminologies, postmodern theory, cultural geography, and others. Utilizing this synthetic approach, they focused especially on the components of style, language, and symbolic meaning that animate illicit social worlds and on the ways in which legal and political authorities in turn utilize mediated representation to criminalize such worlds (Ferrell and Sanders 1995). Cultural criminologists also sought to revitalize the tradition of ethnographic research into illicit subcultures, both by theorizing the epistemic and emotional dynamics of this approach in distinction to quantitative research methods and by engaging in indepth ethnographic explorations of graffiti writers, neo-Nazi skinheads, street-level sex workers, and other groups. Current Theoretical Orientations Contemporary cultural criminology has spawned a variety of theoretical models that build from earlier work on the situated meanings and symbolic politics of crime and crime control. A number of these models explore both the immediate, sensual experience of crime and the dynamics that link such illicit experience to larger patterns of cultural meaning and historical change. The concept of edgework, for example, is utilized by cultural criminologists to analyze incidents of extreme, risky behavior that are generally seen as deviant or illegal (Lyng 2005); cultural criminologists have employed this model in investigating graffiti writing, street racing, BASE jumping, anorexia, sadomasochistic sexual practices, and other contemporary phenomena. In contrast to conventional understandings of such practices as simply destructive or individually out of control, cultural criminologists argue that these practices often involve a conscious exploration of the “edge” between chaos and order, and danger and self-determination. Essential to this exploration is


the way in which such practices blend high levels of risk with the practiced skills necessary to negotiate such risk; for participants, great risk requires great skill, and the greater the skill, the greater the risk that can be taken. This dynamic of risk and skill in turn appeals to more and more people precisely because both risk and skill are increasingly expunged from contemporary life; low-skill, service sector jobs predominate in the emerging consumer economies of late capitalism, and whether at work or play, “risk management” strategies now regulate most every aspect of social life. If in this way the concept of edgework links illicit experiences with larger social and economic changes, it also helps to explain a particular crime control problem. Cultural criminologists have found repeatedly that edgework produces for participants an alluring “adrenalin rush” out of its mixture of extreme risk and practiced skill – but because of this, crime control strategies meant to stop such activities often serve only to heighten the risks and sharpen the skills involved and so to enhance the appeal of edgework activities for those engaged in them. This edgework model resonates with the broader notion of the seductions of crime (Katz 1988). From the view of cultural criminology, much traditional criminological theory has been constituted back to front; that is, such work has theorized the background factors that lead up to the criminal event but has had little to say about the immediate dynamics of the criminal event itself. Cultural criminologists argue that this immediate “foreground” of crime, animated as it is with interactional exchanges and powerful emotions, is indeed worthy of analytic attention. Within it, participants can come to be “seduced” by the momentary meanings and emotions that emerge; within the criminal event, issues of stigma, honor, and respect can become a powerful, if fleeting, impetus for violent, exploitative, or courageous behavior. As with the model of edgework, the notion of crime’s seductions suggests that crime cannot be reduced to simple, individualistic explanations; instead, moments of criminality often harbor a host of complex, negotiated meanings that both reflect and


reconstruct larger cultural forces. Moreover, these meanings may drive the very dynamics by which a criminal event is resolved; more than physical violence or monetary gain, they are often what matter most to those involved. A similar intertwining of meaning, emotion, and culture emerges with the cultural criminological theory of the carnival and crime; here the frame of reference is also historical and comparative (Presdee 2000). Cultural criminologists note that, historically, carnival has served in many societies as a well-defined time of vulgarity, ridicule, and ritualized excess. Because it was predefined and regulated, though, carnival functioned both to celebrate and encourage misbehavior and to contain it within social and cultural boundaries; in this sense, carnival maintained the sort of delicate balance between dangerous human desires and defined social roles essential to the vitality of the societies in which it flourished. In contemporary Western societies, though, the delicate balance of carnival has largely been upended, in some cases through the criminalization of carnivalesque activities and in other cases through the legal and commercial cooptation of carnival into commodified experience and abstract spectacle. As a result, some aspects of carnival’s historical practice can now be bought and consumed, in the form of explicit pornography or degrading “reality television” programming. Other aspects are today enacted as crime – drug taking, sexual predation, and “joyriding” in stolen automobiles – but now all the more dangerously because freed from their containment within community ritual. As with the models of edgework and the seductions of crime, the model of carnival in this way embodies another key concept in cultural criminology: the idea of transgression. From the view of cultural criminology, there is often more at stake in crime and crime control than simple law-breaking and the legal response to it. Instead, crime seems often to embody a breaching of social and cultural boundaries, a dynamic of crossing over or breaking through – and if those boundaries to be breached encode dominant economic arrangements or longstanding cultural norms, then there is indeed

907

C

much more at issue than simple law-breaking. Yet, as the model of carnival shows, transgression also involves a moving dialectic between boundaries made and boundaries breached; in this sense, transgression constitutes an ongoing social and cultural process, a series of negotiated actions and reactions, more than a single act. In addition, as the edgework model suggests, those engaged in transgressive activities sometimes cross the boundaries of safety or respectability with an emerging intentionality; they begin to understand the world and their place in it differently precisely because of the boundaries they breach and so acquire from transgression a critical lens for seeing anew those arrangements that engender it. Two final theoretical approaches incorporate the historical perspective of the carnival of crime model but with an eye in particular toward the distinctive dynamics of the present, late modern period. The theory of exclusion/inclusion posits that contemporary society suffers from a particularly problematic contradiction (Young 1999). On the one hand, the social order is defined by increasing economic and political inequality and with this the increasing economic and legal exclusion of larger and larger segments of the population from full membership in mainstream society. Ongoing financial and housing crises, high levels of unemployment and homelessness, the prevalence of low-wage and temporary work among those who remain employed, and mass incarceration and systematic legal disenfranchisement in the United States and elsewhere all serve to exclude many among the poor, ethnic minorities and the formerly middle class from full participation in conventional social worlds. At the same time, though, these and other groups are, by design, culturally included. The power of the mass media, and especially mass advertising, is such that these groups are systematically taught to desire the same consumer goods as others. Moreover, like others in late capitalist society, they are encouraged to see these goods as essential markers of lifestyle success and to define their own status and identity by goods and services consumed. The result of this contradiction is a culturally induced epidemic of

C

C

908

resentment, insecurity, and humiliated desire and with this a parallel explosion in economic crimes of acquisition and expressive crimes of retaliation and frustration. With this model of exclusion and inclusion, then, cultural criminologists attempt to build from Merton’s foundational formulation of socially induced strain and adaptations to it while at the same time reimagining it within the contours of late capitalist culture and economy. The model of media loops and spirals likewise examines the distinctive cultural dynamics of late capitalist/late modern society, with a focus especially on the emerging interplay of media, crime, and criminal justice (Ferrell et al. 2008). According to this model, the links between crime, criminal justice, and the media in contemporary society are neither linear nor straightforward; for criminologists, therefore, it is no longer a matter of simply inquiring into how accurately the media reports on a particular type of crime, for example, or to what degree individuals imitate crimes that they see portrayed in the media. Instead, the saturation of everyday life with media-producing technology, and the nonstop consumption of mediated images and information, is such that distinctions between an event and its mediated representation often collapse. Criminologists seeking to understand crime and media, then, are increasingly confronted by a self-perpetuating interplay of criminal acts, mediated representations, and mediated responses – by situations, that is, where crime and its image circle back on one another – and so must consider the nature of media loops and spirals if they are to understand the contemporary construction of crime and criminal justice. As recent research in cultural criminology has shown, graffiti practitioners blend illegal street graffiti with sophisticated, legal media productions; gang members stage violent assaults so as to record and sell copies of them on the Internet; “reality television” shows set up violent confrontations and at times entrap their participants in assaults and arrest; and police officers utilize Facebook postings to track suspects, alter their street enforcement strategies due to their own police car cameras, and employ televised policing shows as points of reference in their own


street policing. Moreover, these looping entanglements of crime, criminal justice, and media regularly reproduce themselves over time and so become ongoing spirals of image and activity. An initial cell phone image of a criminal act, for example, may over time become an Internet posting, a piece of legal evidence, a segment in a video compilation to be viewed or purchased, and an image within a crime segment on a televised news program; similarly, a police car video may later be subpoenaed as part of legal proceedings, utilized for police training, or compiled into a “world’s wildest police chases” video. In such a world, cultural criminologists argue, criminological analysis must appreciate the power of mediated representation and must be sensitive to the ways in which such representation feeds back into the practice of crime and criminal justice. Methods of Cultural Criminology The research methods utilized by cultural criminologists embody the theoretical orientations that define cultural criminology itself. Given their analytic focus on shared meaning and emotion, cultural criminologists require methods that can take them inside the illicit situations where meaning and emotion are negotiated and that can attune them to the subtleties of symbolic interaction among people and groups. To understand the allure of edgework or the seductions of criminal moments, cultural criminologists argue, they must find ways to be present when such seduction is underway and to share as best possible in this seductive emotional process. Likewise, the focus on symbolism and mediated representation means that cultural criminologists must find ways to decode the complex symbolic universes of marginal groups and to penetrate the processes by which mediated representations are made and circulated. In all of this, cultural criminologists argue for a general “methodology of attentiveness” – that is, a methodological orientation that enables close attention to the nuances of human culture and human transgression. Broadly speaking, cultural criminologists in this sense prefer methods that embody something of the phenomenological imperative – the imperative to


encounter “the thing itself,” on its own terms and in situ – rather than methods that spawn artificial circumstances, broad generalizations, and abstract conclusions. Because of this, cultural criminology’s methodological orientation has become also a critique of conventional methods within mainstream criminology; in arguing for and employing their own attentive methods, cultural criminologists have also emerged as perhaps the most vocal critics of contemporary methodological practices in criminology (Ferrell et al. 2008). From the view of cultural criminology, the most widely used methods in criminology – survey research, statistical analysis of survey results, statistical analysis of governmental data, and the massive data sets that result – primarily serve to distance criminology from its subject of study. By their own design and internal logic, such approaches preclude on-the-ground engagement with situations of crime and crime control and with the meanings and emotions that emerge within these situations. Likewise, such methods systematically miss the subtleties of symbolism and style that circulate within illicit worlds, as they reduce these subtleties to a set of simplistic choices among survey answers or governmental record-keeping categories. As such, these methods in turn reduce the human identities and human interactions that construct the reality of crime and justice to statistical abstractions, all while removing researchers and their own identities from creative engagement with the research process. Most damningly, cultural criminologists suspect that such methods are perhaps so widely and popularly employed precisely because they do keep criminological inquiry at a distance, forfeiting critical engagement with the subject matter in the interest of political leverage and agency funding and thereby reproducing stereotypes and assumptions more than investigating them. Building from the pioneering field research of the earlier North American and British theorists whose work set the course for cultural criminology, cultural criminologists turn to ethnography – long-term, in-depth field research with subjects of study – in place of more conventional methods

909

C

of survey research and statistical analysis. Through deep immersion in the lives and interactions of criminals, crime victims, and crime control officials, cultural criminologists seek to become part of the process by which such people and groups construct the various meanings of crime. Likewise, cultural criminologists seek through ethnographic research to attune themselves to the shared languages, symbolic codes, and public performances that emerge within and between groups as they work to make sense of crime and crime control. This sort of ethnographic research embodies another, more controversial dimension as well. As already glimpsed in the models of edgework and the seductions of crime, cultural criminologists argue that the emotional dynamics of crime are often essential to the meaning and experience of it – and because of this, they argue, ethnographic researchers should also seek a degree of emotional intimacy with subjects and situations of study. This notion of subjective or appreciative understanding is embodied in Max Weber’s classic sociological concept of verstehen and specifically in what cultural criminologists call criminological verstehen (Ferrell and Hamm 1998). With this concept in mind, cultural criminologists have in recent ethnographies analyzed, for example, their own emotional participation in the “adrenalin rush” of graffiti writing, the macho pleasures of high-powered weapons use among off-duty police officers, the stark alienation of right-wing terrorist groups, and the dismissive stigmatization of phone sex workers and street prostitutes. Significantly, the controversial nature of this research stems not only from its deep immersion in the lives and emotions of illicit groups but from its reversal of the “objective” and “value-free” underpinnings of more conventional forms of social scientific research. Here it is not the “objectivity” of surveys and statistical analysis that produces research insights, but the opposite: the emotional subjectivity achieved through ethnographic research. Cultural criminologists have in addition sought to reimagine and reinvent ethnographic research in light of current theoretical models and in the context of contemporary social and

C

C

910

cultural developments. If the situated seductions of crime can emerge within a criminal event and evaporate with its termination, for example, then perhaps ethnography must be organized not only around long-term researcher participation but researcher attentiveness to immediate situational dynamics. Cultural criminologists refer to this approach as instant ethnography – ethnographic research that delves deeply into moments and situations, deconstructing them in search of their larger implications and understanding them as instances in which crime and its meaning are performed. Likewise, given the fluid, dislocated nature of many social interactions and identities in the late modern world, cultural criminologists argue that ethnography’s traditional focus on single, unitary social groups must now be supplemented by forms of liquid ethnography. This sort of ethnography pays special attention to issues of migration and movement, to the ways in which seemingly solid social identities are made and remade, and to the interplay of mediated images in the ongoing construction of individual and group perceptions. Finally, as suggested by the notion of verstehen and by cultural criminologists’ analyses of their own emotional involvement with crime issues, autoethnography has also emerged as a viable research method within cultural criminology. With autoethnography, the ethnographer’s own self and identity become a focus of study – but not as form of isolated self-examination. Instead, autoethnography demands that researchers account for the ways in which their participation in illicit groups or dangerous situations has altered their own sense of self or status or otherwise forced them to confront previously unexamined privileges and assumptions. In this way, autoethnography is designed to create a fuller ethnographic account of crime and crime control, one in which the ethnographer’s own experiences become yet another window into larger social and cultural patterns. Paralleling the distinction between survey research/statistical analysis and ethnographic inquiry is the distinction between conventional approaches to research on media and crime and those approaches taken by cultural criminologists.


Traditionally, criminologists have utilized content analysis – the measuring of static, discrete content categories within media texts – in studying, for example, the amount of media coverage given to a particular type of crime or trends in the amount of violence in popular films. In the context of contemporary media saturation and in light of the model of media loops and spirals already seen, though, cultural criminologists find that this approach is often unable to capture the ongoing, fluid interplay of contemporary crime and media. Calculations of media content and its amount miss the broader cultural and aesthetic frames of reference within which such content takes on meaning. In addition, utilizing content analysis in an attempt to measure the degree of divergence between the “real” nature of crime and its “biased” media representation ignores the spiraling process by which the “real” and the representational increasingly come to constitute one another. To supplement conventional content analysis, cultural criminologists utilize two types of alternative methods. Ethnographic content analysis conceptualizes mediated representation as an unfolding process of decision making, action, and reaction among a variety of groups and so undertakes the study of media and crime from the view of researcher immersion in this ongoing process. The approach likewise advocates an ongoing interplay between researchers and the media texts they study; the method is meant to provoke deep involvement with media texts, such that researchers over time develop thoroughgoing accounts of texts and their emergent meanings. In this way ethnographic content analysis allows researchers to understand media texts not as isolated embodiments of content but as emergent cultural process incorporating a variety of political and cultural dynamics. While this method enables researchers to identify and analyze textual patterns, as with content analysis, it also accounts for the liquid, looping process that emerges between crime, media, and justice. A second alternative approach links crime and media analysis even more closely to the overall practice of ethnography. Here, in the spirit of liquid ethnography, researchers engage in


fieldwork with criminals, criminal justice workers, and crime victims as these groups engage with the mass media, investigating the ways in which such groups attempt to manage images and accounts of their lives and experiences or, increasingly, undertake to produce their own mediated accounts of crime and justice. Similarly, cultural criminologists at times conduct ethnographic research inside media organizations, so as to be present as decisions are made and accounts developed regarding the representation of crime and justice issues. Substantive Themes and Research Trajectories As cultural criminology has matured over the past two decades, its theories and methods have been applied to a wider and wider range of phenomena. Certainly research is still being conducted on subcultural interactions and aesthetics, popular culture representations of crime and violence, and the seductive dynamics of edgework and other forms of situated transgression. In addition, though, cultural criminologists are both refining their focus on particular forms of crime and crime control and widening their analysis of contemporary social and cultural changes as regards crime and social control. One emerging focus within cultural criminology is the relationship between crime, crime control, and everyday life (Ferrell et al. 2008). Cultural criminologists note that the nature of law and the criminalization process is such that the magnitude and meaning of a criminal activity remains under construction – and that because of this, seemingly inconsequential or invisible crimes often merit as much investigation as those defined at any one time as consequential. In addition, cultural criminologists point out that contemporary legal and social control is increasingly encoded in the small situations of everyday life. Pervasive surveillance/CCTV cameras in urban areas, CPTED (crime prevention through environmental design) programs in parks and train stations, corporate risk-prevention strategies, computerized data and web monitoring, cell phone tracking devices – these and other phenomena build often unnoticed ideologies of

911

C

legal and cultural control into the practices of daily life. Likewise, popular contemporary policing approaches like the “broken windows” model focus police attention on everyday situations and low-order criminality, though generally with little reflection on the assumptions underlying such approaches. In light of all this, cultural criminologists argue that if we are to understand the meanings of crime and the ideologies of crime control, we must investigate their presence in everyday life; that is, we must investigate the dynamics of routine traffic stops, the layout of city centers, the daily policing of the homeless, and the prevalence of everyday crimes like shoplifting or loitering. Such research provides added benefits as well; it serves as a corrective to the media focus on sensational and unusual crime, and it encourages scholars and students alike to engage in accessible criminological inquiry. This mention of city centers and urban CCTV cameras highlights a second substantive focus within cultural criminology: the city and its spatial arrangements (Hayward 2004). Cultural criminologists note that, with the dramatic and increasing urbanization of the world over the past few decades, many conventional criminological analyses of crime and crime control may now need to be reconceptualized as analyses of distinctly urban patterns of crime and control. Drawing especially on work in cultural geography, cultural criminologists have, for example, investigated the ways in which new urban economies organized around high-end consumption reorganize the city’s spatial arrangements while also fostering new models for policing these consumerist spaces. Cultural criminologists have also explored these spaces as sites of legal and cultural conflicts that pit recent arrivals against long-time residents, corporate developers against public space activists, and automotive traffic patterns against advocates of alternative transportation – with these conflicts in turn entangled with ongoing mediated representations, arrests, and court proceedings. Recent ethnographic work in cultural criminology has likewise explored the cultural practices of illicit urban subcultures like those of skateboarders, graffiti writers, and

C

C

912

homeless trash scroungers, with special attention to the process by which such practices remake urban space, imbuing it with illicit meaning and remapping it along the lines of subcultural status and values. Two additional trajectories of research attempt to engage cultural criminology directly with the distinctive political and economic arrangements, and political and economic crises, that define the contemporary late modern/late capitalist period. The first of these is a developing cultural criminology of the state and of global political conflict. Recent work in cultural criminology has explored, for example, the increasingly mediated nature of military recruitment, military training, and military warfare itself; it has also shown that the boundary separating such mediated militarism from the worlds of popular culture and entertainment is becoming blurred at best. Similarly, cultural criminologists have documented the activities of militant or terrorist groups in self-producing sophisticated media and have shown how these and other images of war and violence loop and spiral through the global mediasphere. Linking these sorts of insights to the more traditional concerns of criminology, other recent research has explored the culture of punishment in the United States and other countries – a culture which flows in and out of the prison and into popular media, public perceptions of crime and justice, and even the tourism and entertainment industries (Brown 2009). Similarly, research into gangs and gang culture has documented the role of law enforcement agencies in defining perceptions of gang threat and in moving gang members and their culture back and forth across national boundaries; it has in turn documented gang members’ attempts at self-empowerment and resistance to these state practices. A second trajectory has cast cultural criminology into the broader cultural currents of late modernity. Building in part from the theory of exclusion/inclusion, cultural criminologists have found a complex of values, practices, and identities that coalesce around drift, dislocation, and the particular forms of transgression that accompany them. In this shifting world, status and


identity are increasingly cut loose from traditional anchors of steady career or home-based stability; indeed, the notion of “home” itself loses its conventional comfort as more and more people migrate from city to city or country to country in search of something other than parttime work and political insecurity. Especially in light of the ongoing global economic crisis, aspirations of upward achievement are often replaced by a desperate sense of surviving with what is at hand or with what might come to be at hand – yet advertisers continue to aggressively market identities that, while made desirable, are neither affordable nor achievable. In such a world, cultural criminologists argue, insecurity and uncertainty become a way of life, and in response, some people seek to find identities in acts of individual risk-taking and transgression; others engage in expressive crimes fueled by resentment, fear, and failure. In such a world crime control also changes, becoming increasingly a matter of risk management and the containment of transitory populations. Here is an emerging global milieu permeated not only by inequality and injustice, exclusion and inclusion, but by a seemingly intractable culture of ambiguity and uncertainty; to make sense of it, cultural criminologists argue, criminology may itself have to become comfortable with ambiguous findings, liquid analysis, and alternative forms of inquiry (Young 2007; Ferrell et al. 2008). In this context cultural criminology has continued to build from its original cross-cultural roots in North American and British criminology, with significant bodies of theoretical and substantive work now coming out of the Netherlands, Australia, Brazil, and other countries. In keeping with the early intellectual interplay between North American and British scholars, this contemporary work is not only geographically dispersed in its origins and substantive concerns but intentionally conversational, with cultural criminologists from around the world sharing ideas and working toward new intellectual syntheses (Carvalho et al. 2011) that can account for emerging social circumstances. Likewise, cultural criminologists continue to challenge orthodox approaches within criminology and to develop

Customary Justice

new means of engaging in criminological analysis. Drawing on film and visual theory and revisiting long-standing traditions of documentary photography, much contemporary work in cultural criminology is oriented toward both analyzing existing images of crime and producing an alternative body of informed visual representation as regards crime and justice (Hayward and Presdee 2010). Other cultural criminologists continue to experiment with the narrative forms of criminology, producing in place of conventional criminological discourse a range of manifestos, vignettes, poems, and short stories in both print and digital media. In all of this, the intent is to bring cultural criminology solidly into the present and the future and to position it for ongoing, critical engagement with the cultural dynamics of crime and crime control.

913

Ferrell J (1999) Cultural criminology. Ann Rev Sociol 25:395–418 Ferrell J, Hamm M (eds) (1998) Ethnography at the edge. Northeastern University Press, Boston Ferrell J, Hayward K (eds) (2011) Cultural criminology: theories of crime. Ashgate, Farnham Ferrell J, Hayward K, Morrison W, Presdee M (eds) (2004) Cultural criminology unleashed. Glasshouse, London Ferrell J, Hayward K, Young J (2008) Cultural criminology: an invitation. Sage, London Ferrell J, Sanders CR (eds) (1995) Cultural criminology. Northeastern University Press, Boston Hayward K (2004) City limits: crime, consumer culture, and the urban experience. Glasshouse, London Hayward K, Presdee M (eds) (2010) Framing crime: cultural criminology and the image. Routledge, London Katz J (1988) Seductions of crime. Basic Books, New York Lyng S (ed) (2005) Edgework. Routledge, New York Presdee M (2000) Cultural criminology and the carnival of crime. Routledge, London Young J (1999) The exclusive society. Sage, London Young J (2007) The vertigo of late modernity. Sage, London

Recommended Reading and References Brown M (2009) The culture of punishment. New York University Press, New York Carvalho S, Neto MP, Mayora M, Linck JAG (2011) Criminologia cultural e rock. Sao Paulo, Lumen Juris

C

Customary Justice ▶ Post-Conflict Traditional Justice

C

Calibration of Simulation Models

Calibration of Simulation Models

Suggest Documents

Calibration of Whole Building Energy Simulation Models

Calibration of Simulation Models - Technische Universiteit Eindhoven

Efficient simulation and calibration of general HJM models by splitting ...

Improved calibration of simulation models in railway ...

Bottom-up simulation calibration of zone and system level models ...

Calibration of Microscopic Traffic Simulation Models with Aggregate ...

Innovative Calibration Method for System Level Simulation Models of ...

Calibration of Microscopic Traffic Simulation Models ... - Tomer Toledo

Calibration – Manual Calibration of GSSHA models - Aquaveo

Validation of calibration models

simulation models

SIMULATION FOR AIRBAG SENSOR CALIBRATION

CALIBRATION OF SCHEDULES IN BUILDING SIMULATION USING ...

Calibration of Time History Simulation Methods - GeoScienceWorld

Simulation and Calibration Between Parameters of ... - SciELO

Assessment of Traffic Simulation Calibration Criteria ...

Assessment of Traffic Simulation Calibration Criteria ...

Bayesian calibration of building energy models - ScienceDirect

Objective Calibration of Regional Climate Models - American ...

PARAMETER cALIBRATION OF TRAFFIc MODELS IN ... - Core

Calibration of numerical models based on advanced

calibration of water distribution hydraulic models ...

ARTICLE: Calibration of Smooth Camera Models

calibration of steady-state car-following models