can enhancement be distinguished from prevention in ... - CiteSeerX

ERIC T. JUENGST

CAN ENHANCEMENT BE DISTINGUISHED FROM PREVENTION IN GENETIC MEDICINE? ABSTRACT. In discussions of the ethics of human gene therapy, it has become standard to draw a distinction between the use of human gene transfer techniques to treat health problems and their use to enhance or improve normal human traits. Some dispute the normative force of this distinction by arguing that it is undercut by the legitimate medical use of human gene transfer techniques to prevent disease - such as genetic engineering to bolster immune function, improve the efficiency of DNA repair, or add cellular receptors to capture and process cholesterol. If disease prevention is a proper goal of medicine, these critics argue, and the use of gene transfer techniques to enhance human health maintenance capacities will help achieve that goal, then the "treatment/enhancement" distinction cannot define the limits of legitimate gene therapy. In this paper, I argue that a line can be drawn between prevention and enhancement for gene therapy (and thus between properly medical and norunedical uses of gene therapy), but only if one is willing to accept two rather old-fashioned claims: 1) Some health problems are best understood as if they were entities in their own right, reifiable as processes or parts in a biological system, with at least as much ontological objectivity and theoretical significance as the functions that they inhibit. 2) Legitimate preventive genetic health care should be limited to efforts to defend people from attack by these more robust pathological entities, rather than changing their bodies to evade social injustices. Key Words: enhancement, ethics, genetic medicine, prevention

I. THE PROBLEM

In discussions of the ethics of human gene transfer, two distinctions have become standard rhetorical tools. The first distinction separates interventions that will only affect somatic cells from interventions aimed at modifying human germ-line cells. The second distinction contrasts the use of human gene transfer techruques to treat health problems with their use to erJiance or improve normal human traits (Walters, 1986). Both distinctions are Eric T. Juengst, Ph.D., Center for Biomedical Ethics, School of Medicine, Case Western Reserve University, Cleveland, Ohio, U.S.A.

The Journal of Medicine and Philosophy 22:125-142,1997. © 1997 Kluwer Academic Publishers. Printed in the Netherlands.

126

Eric T. Juengst

widely debated, but in practice they constitute the most infiuential moral matrix to date for evaluating new human gene transfer iriterventions. In this essay, I am interested in pursuing the debate over the second distincfion. .While the soniatic/germ-line distingtion is accused of lacking a.dequ¥te ethical force (Moseley, 1991), the conceptual line between those two'classes of intervention is at least clear. The treatment/enhancement distinction, however, often seems in danger of evaporating entirely under its conceptual critiques, and to that extent seems to pose the larger risk to our efforts at assessing human gene transfer technology. The treatment/enhancement distinction is usually used to argue that curative or therapeutic uses of genetic engineering fall within (and are protected by) the boundaries of medicine's traditional domain, while enhancement uses do not, and to that extent are more problematic as a professional medical practice or a legitimate health care need (Anderson, 1989; Baird, 1994).^ There are several interesting rejoinders to this argument. Some argue that medicine has no essential domain of practice, so that a coherent distinction between medical and non-medical services can never be drawn in the first place (Engelhardt, 1986). Others accept the distinction between treating and enhancing, but take on the traditional values of medicine, by arguing that privileging treatment over enhancement is itself wrong (Silvers, 1994). Others argue that, in any case, for psychological and economic reasons, the line between treatment and enhancement will be impossible to hold in practice (Gardner, 1995). The rejoinder I want to discuss, however, takes another tack. This response criticizes the distinction by showing how it dissolves in the case of using human gene transfer techniques to prevent disease when such interventions involve the enhancement of the body's health maintenance capacifies. The argument is that to the extent that disease prevention is a proper goal of medicine, and the use of gene transfer techniques to strengthen or enhance human health maintenance capacities will help achieve that goal, then the treatment/enhancement distinction cannot confine or define the limits of the properly medical use of gene transfer techniques. This argument is bolstered by the fact that the technical prospects for such preventive-enhancement interventions already look good, given gene therapy protocols now underway to treat ill patients in just those ways (Wilson, 1992; Rosenberg, 1990). One gene therapist summarizes this current biomedical work by saying:

The Enhancement/Prevention Distinction

127

Over the next few years, it appears that the greatest application will be in the treatment of cancer, where a number of genes that have been isolated have the potential to empower the immune system to eliminate cancer cells Human gene therapy cancer trials have also been initiated for insertion of the tumor necrosis factor (TNF) gene into T-lymphocytes in an effort to enhance the ability of T-lymphocytes to kill tumors. Another approach has been to insert the TNF gene into tumOr cells in an effort to induce a more vigorous immune response against the

tumor (Culver, 1993, emphasis added).

If human gene therapy protocols like these are acceptable as forms of preventive medicine, the critics ask, how can we claim that we should be "drawing the line" at enhancement? Once accepted, this rejoinder to the treatment/enhancement distinction can be used to cut both ways in the normative discussion of human gene transfer. Critics of human gene transfer com.bine it with a slippery slope argument to warn against the whole enterprise of gene therapy (Krimsky, 1990). Defenders use it to put a foot in the door of the distinction, to keep open the possibility of some legitimate enhancement uses of human genetic engineering (Walters and Palmer, 1997). In this paper, I follow neither of these branches in the argument through to normative conclusions about the moral merits of gene therapy or genetic enhancement. Instead, I want to focus on their common rejoinder to the treatment/erihancement distinction, and address the problem they pose of distinguishing between enhancement and prevention in genetic medicine. I will argue that a line can be drawn between prevention and enhancement for gene therapy (and thus, perhaps, between properly medical and nonmedical uses of gene therapy), but only at a conceptual price that most comtnentators seem unaware of, even when they are already paying it. The price of maintaining the prevention/enhancement distinction is that one must be willing to supplement the accounts of health and disease that (tharJcs to Boorse (1977) and Daniels (1986)) currently dominate discussions of the boundaries of medical practice with an older, and somewhat old-fashioned way of looking at health problems. In short, in order to preserve a distinction between preventive and enhancement uses of genetic engineering, one must be willing to accept that some health problems (or "maladies," to borrow this useful generic term from Clouser et al. (1981)), in addition to being "reductions in the organizational readiness of an individual to perform functions within

128

Eric T. Juengst

the efficiency range typical of members of its reference class (age, gender, species) in the absence of external sustaining causes," also need to be understood to be entities in their own right, reiHable as processes or parts in a biological system, with at least as much ontological objectivity and theoretical significance as the functions that they inhibit. On this view, legitimate preventive genetic health care needs, in addition to being limited to those that involve maintaining individuals within their personal (genetically constrained) subsets of the range of fimctional capabilities tjîcal for members of their reference class, limited to efforts to defend people from attack by these more robust pathological entities. Or so I will try to argue. First, though, I should point out a few more limits on my anibitions in this paper: 1. I am not concerned here with arguing whether distinguishing genetic prevention from genetic enhancement is the best way, in the final analysis, to settle the question of the moral limits of human genetic engineering. As I suggested above, there may be other considerations that undermine the whole line-drawing enterprise as well. 2. I will only point to, rather than pursue, the major implications of my thesis. If my thesis is correct, it looks as if it would raise considerations relevant to a number of larger issues. In short: a) My thesis seems to help resolve some "hard cases" for the currently dominant approach, suggesting that Daniels and company should pay its price and embrace it. b) It does improve the boundary defenses of medical practice, allowing health professionals to carry their commitment to prevention as well as cure into genetic medicine without opening the door to illegitimate enhancements. This suggests progress in the professional ethics of gene therapy, at least, if not in either "private" or social policy. c) On the other hand, it also increases the possibility of medical hegemony with respect to genetic engineering practices in our society, and preserves a traditional opening through which cultural values and prejudices can influence medical reasoning. This suggests that the slope remains slippery, and the story is not over yet. d) It also illuminates a theoretical weakness in the dominant model of health problems, and suggests that, far from being


129

ready to dispense with the concept of disease, philosophers of medicine should pay more attention to the theoretical role that pathological entities play in the ways in which medicine frames its domain.

II. DRAWING A LINE BETWEEN PREVENTION AND ENHANCEMENT

A. The Treatment/Enhancement Distinction and "Normal Function"

Theories of Health Needs. To understand the prevention/enhancement problem, it is useful to begin by looking back at the discussion of the treatment/erJiancement distinction from whence it springs. Its most useful exposition comes through the writings of Norman Daniels and his co-authors, who attempt to use the distinction in the course of defining the limits of "legitimate health care needs" for health policy purposes. They argue that conceiving health problems and healih care needs in terms of deviations from species-t5îcal normal functioning and an individual's capability range - what Daniels has recently called the "Normal Function" model of health needs (Sabin and Daniels, 1995) - is useful in helping to draw an objective distinction between curative treatments and erihancement uses of ordinary medical services like psychotherapy and prescriptions of human growth hormone. Others have used similar understandings of human malady to help explicate a distinction between "negative" (e.g., therapeutic) and "positive" (e.g., enhancing) human genetic engineering (Berger and Gert, 1991). On these accounts, to be healthy is to fvmction, under t)îcal circumstances, with the tjîcal efficiency of members of one's age, gender and species. Legitimate health care needs or "health problems" or "diseases" or "maladies" are all characterized by a fall from that level of functional readiness. All proper health care services, therefore, should be aimed at getting people back to "normal," e.g., restoring an individual's functional capability to the species-typical range for their reference class, and within that range to (the bottom of) the particular capability level which was the patient's genetic birthright. Interventions which take people to the top of their personal potential (like athletic trairung) or beyond their own birth range (like growth hormone), or to the top of the range of their reference class, or to the top of the species-typical range, or beyond(!), are all to be counted as enhancements and fall

130

Eric T. Juengst

successively further beyond the domain or responsibility of medicine or health care. The advantage of the Normal Function account is that it provides one relatively unified goal for health care, towards which the burdens and benefits of various interventions can be relatively objectively titrated, balanced, and integrated. By contrast, theories of "positive" health that would set the goals of medicine higher than the bottom of the individual's species typical capability range face the problem of having to compare different visions of human flourishing. As Boorse puts it, under these approaches: Not only is there no fixed goal of perfect health to advance towards, but there is also no unique direction of advance. This point reflects the familiar fact that realizing one potential is often inconsistent with realizing others. To enhance one function to the fullest, e.g., strength, may inhibit others e.g., speed Along any one functional dimension, comparative judgments of positive health can easily be made. What one cannot say is whether an advance along one dimension is healthier than an advance along another. ... As it is, in negative health all therapeutic programs converge towards one goal; in positive health they diverge further the greater the net improvement.... [Moreover] If the pursuit of positive health forces a choice between incompatible excellences, it requires an evaluative decision - by client, physician, or society - about what life goals are worthy of pursuit. ... [thus] ideals of positive health are not discoverable, but only advocable. Their advocacy raises familiar ethical dilemmas about the good life for man, yet no medical procedure can possibly resolve them The trouble with calling physical or mental or moral excellence health is that it tends to unite under one term a value neutral notion - freedom from disease - with the most controversial of all prescriptions: the recipe for an ideal human being (Boorse, 1977, pp. 570-572).

There are several implications here that are important for our purposes. First, the problem with "positive health" accounts of health care needs is not that they cannot draw a line between what they would count as restorative treatment and illegitimate enhancement: they can just as easily specify the top of the personal capability range, or the top of the species typical range for some fimction as their goal. The real problem with the positive health views is an epistemic one: positive health advocates cannot use the lines they would draw between treatment and enhancement, because they can never tell when they've crossed them. Daruels recognizes this when he complains that: Although the capability and welfare models capture basic moral insights, we believe that they cast too broad a net and pose severe problems for administra-


131

tion and cost. How are we to judge when the conditions for health insurance are met? If individuals invoke the capability model... do we simply take their assessment of need at face value? If not, how are we to investigate the claim? We have little idea of how to delve into questions like this (Sabin and Daniels, 1995, p. 11).

Normal functionalists can use physiology to determine when they've achieved the species typical range and clinical histories to determine when they've brought a patient up to the baseline of his or her personal capability range. Thus Daniels stresses that, unlike positive health models of health care entitlements, the boundaries set for medicine by the Normal Function model are "the result of a highly public process open to scientific scrutiny, field testing and repetitive criticism over time" (Sabin and Daniels, 1995, p. 10). At the very le:ast, death serves as clear indicator that one has left the range of normal function. But how does one know when one is at the top of a personal capability range, or the top of the species typical range? As far as we know, we've never seen the tops of the ranges. Performing better than one ever has before, or better than anyone ever has before, might just be a first step towards the true ceiling. In any case, we would not know the ceiling if we did see it, because there is nothing that sets a limit like death does at the other end of the range. To that extent, the positive health advocates carmot make a distinction with a difference between treatment and enhancement. This epistemic point about the need to be able to know when one has crossed a line is important to keep in mind in thinking about what separates genetic prevention from genetic enhancement, because in that context it applies equally forcefully to the Normal Function accoimt as well. B. Prevention and Normal Functioning. All the adherents to Normal

Function accounts of health and health care needs include and endorse disease prevention as a legitimate health need and a proper goal for medical practice. Indeed, given the place of prevention and preventive efforts in the history and ideology of medicine, it would be a severe problem, for their reconstruction of health care if it could not include prevenfive interventions within the sphere of legitimate medical practice. Unfortimately, without conceptual help, that is just what the Normal Function account cannot do in the case of preventive interventions that act by strengthening an organism's normal defenses against disease, like the gene transfer protocols mentioned above.

132

Eric T Juengst

Daniels' recent work using the Normal Function accoimt and the Treatment/Enhancement distinction to discuss questions about the bounds of legitimate medical practice is useful in explicating this limitation. Daniels summarizes his account by saying that: The central purpose of health care is to maintain, restore, or compensate for the restricted opportunity and loss of fimction caused by disease and disability. Successful health care restores people to the range of capabilities they would have had without the pathological condition, or prevents further deterioration (Sabin and Daniels, 1995, p. 10).

If we rewrite this passage to focus on the preventive aspects of health care, we get: The central purpose of preventive health care is to maintain the range of opportunity and functional efficiency threatened by disease and disability. Successful preventive health care preserves for people the range of capabilities they have in the absence of pathological conditions, or prevents further deterioration.

Thus, for example, in genetic medicine, providing a low phenylalanine diet to individuals who lack the (species typical) capacity to metabolize phenylalanine will prevent the loss of normal mental capabilities that phenylketonuria otherwise typically will cause. Similarly, by artificially supplying individuals with hypcholesterolemia with the low-density lipo-protein receptors they lack, gene therapists can prevent further fimctional deterioration in the circulatory system from being caused by that genetic disease. Thus, both of these genetic interventions should count as legitimate medicine for the Normal Functionalists. But what about preventive efforts that act by actually enhancing species-typical human health maintenance functions? Consider the case that LeRoy Walters and Julie Palmer make for including some genetic erJiancements within the domain of legitimate medical needs. They start with the paradigm of a non-genetic preventive intervention: immunization against infectious disease: In current medical practice, the best example of a widely-accepted health-related physical enhancement is immunization against infectious disease. With immunizatior« against diseases like polio or hepatitis B, what we are saying is in effect, "The immune system that we inherited from our parents may not be adequate to


133

ward off certain viruses if we are exposed to them." Therefore, we will enhance the capabilities of our immune system by priming it to fight against these viruses. From the current practice of immunizations against particular diseases, it would seem to be only a small step to try to enhance the general function of the immune system by genetic means. ... In our view, the genetic enhancement of immune system function would be morally justifiable if this kind of enhancement assisted in preventing disease and did not cause offsetting harms to the people treated by the technique (Walters and Palmer, in press, 1997, p. 110).

In fact, both normal immunization and the new "genetic immunizations" are hard cases for the Normal Function account. Susceptibility to infection by the polio virus is not a deviation from normal species typical fimctioning. Similarly, an individual's need for immunization does not require the loss of a fimctional efficiency or a disability. It simply reflects a human inability: the inability of the individual's immune system effectively to resist infection upon irutial exposure to this virus (cf. Clouse et al., 1981, p. 198). Yet in these cases, an intervention is performed that not only increases an individual's internal disease-resisting capability above their personal range of God-given abilities, but even enhances his or her capability beyond the functional range typical for the person's whole reference class within the species (e.g., the class of as yet xmexposed children of either gender).^ The problem here for the Normal Functionalists is the same problem that Daniels identifies in efforts to call normal short stature a "handicap" in order to secure medical treatment. That is, medicine would be faced with "the need for a clear notion of handicap" and with "the challenge of trying to choose from amongst the many human differences which should count as legitimate targets for medical amelioration" (Daniels, 1992, p. 48). If the bare notion of a handicapping vulnerability were enough to secure preventive interventions, medicine would be faced witti the need for some means of determining which human liabilities (aging?) it should attempt to prevent, and how one would know when one had succeeded in making an individual (or the species) impervious to them. In other words, the Normal Function account is here faced with the same kind of limitation that "positive health accounts" faced in trying to distinguish legitimate treatment from improper enhancements: it can posit a line between prevention and treatment, but it cannot indicate, on its own, when that line is being

134

Eric T. Juengst

crossed. This is why the "normal function" accoimt seems blind to the difference between strengtherung the body to resist disease and strengthening the body to gain other advantages.^

III. THE NEED FOR A MEDICAL ONTOLOGY

It is interesting that Daniels frames his criticism of the "handicapping" strategy as a set of "serious objections to dropping the reference to disease in drawing this version the treatment/enhancement distinction" (Daniels, 1992, p. 47). In fact , references to disease and diseases play an even more important role in allowing medicine to distinguish between prevention and erihancement than Daniels seems to acknowledge. In his discussions of bovindary problems in medicine, Daniels distinguishes loss of function and opportimity from the diseases, disabilities and pathological conditions that cause them. He stresses that the Normal Function model "makes the health sector responsible for correcting only those conditions which — in DSMIV terms - can be diagnosed as 'a s5Tnptorn of a dysfunction,' that is as mental disorders" (Sabin and Daniels, 1995, p. 10). This is odd, because that distinction is not otherwise part of his general account of health care needs, which simply equates health problems with "deviations from the natural ftmctional organization of a typical member of the species" (Daniels, 1986). Indeed, Daruels continues to cite and paraphrase this definition even as he casts diseases in these discussions as something distinct from the reductions in function they cause.^ This makes his recurrent causal coupling of disease and function loss in these passages vmusually (for Daniels) redundant. Forttmately, this unusual redundancy itself is a clue, which points us back along Daniel's citation trail to the source of his definitions in Boorse's fuller account of health and its definition of disease as "internal states which depress a functional ability below species-typical levels."^ The key point is that for Boorse, apparently, diseases do exist independently of the functional depression or deviance they can cause, as the "internal states" that cause them. That is still pretty mysterious, but such "internal states" do seem to be the suppressed causal concept in Daniels' short-hand summaries of Boorse's account that appear in his discussions of


135

medicine's boundary problems under the labels "pathological conditions," "diseases," and "diagnosable disorders." Moreover, Boorse says something important on this score in the process of setting u p his own analysis. He says that: Diseases in the sense to be here analyzed - e.g., cystic fibrosis, bronchial asthma, trichinosis, - are universals, or types of unhealthy condition that occur in more than one person. Roughly corresponding to philosophical debates over universals, there have been recurrent controversies in the history of medicine about whether diseases or only ill patients are real. Sometimes the issue was whether diseases were independently existing external entities; at other times it was whether disease taxonomy is artificial or natural. Our legacy from these controversies is the term "disease entity", used to mean a natural unit of disease classification. Today, the strictest definition of a disease entity would be a constellation of signs, symptoms and pathology with specific etiology and prognosis... Our generic notion of disease looks wider than that of a disease entity. Fortunately, if our goal is to understand health as the absence of disease, we can abstract entirely from this problem of individuating diseases. Complete freedom from disease is the same however the field of disease is split up into units. ... In this respect alone, we make no attempt to be faithful to the customary extension of "disease." The reader should bear this restriction in mind (Boorse, 1979, pp. 552-553). Daniels can also skirt the challenge of providing a fuller ontology of disease for his purposes, in defending the Normal Function model of medical needs for social policy purposes. ("Our endorsement of the normal function model does not rest on a view of disease or disability as ultimately real in a metaphysical sense" (Sabin and Daniels, p. 12)). But if Boorse and Daniels can avoid the problem of disease ontology for their initial purposes, they cannot avoid it in attempting to distinguish prevention from enhancement. As their own language suggests, distinguishing effectively between prevention and enhancement depends less on the range of species-typical functioning than on the kind of challenge an intervention is designed to address: what distinguishes prevention from enhancement, in other words, is that medical prevention seeks to improve the body's defenses against disease. Moreover, to do any work, "disease" in this context must mean more than deviation from species typical functiorung. To be helpful in drawing a publicly defensible boundary for medical

136

Eric T. Juengst

genetic engineering, it has to provide an independent epistemic ground for including an intervention within medicine's domain, above and beyond its ability to keep a person from succumbing to some species tjîcal vulnerability and losing functional efficiency. In the next section, I argue that it is medicine's traditional concept of robust, objectively knowable disease entities - as "metaphysically real" as any natural biological process - that is what the Natural Function account needs in order to make the distinction necessary to defend the domain of medicine.

IV. TAKING DISEASES SERIOUSLY

What I've been calling the "Normal Function" account of health care needs and the accoimt of health and disease it rests on are all descendants of an ancient tradition within Western medicine's 2000 year dialogue about its proper sphere. They reflect the perspective of what historians of medicine call the "physiological" school of thought about disease (Risse, 1978). "Physiologists" in pathology have always believed that only their patients were real, and diseases should be understood as relatively arfificial descriptions of their pafients' (malfunctioning) bodily processes. Different bodies, being constructed similarly and obe5dng the same natural laws, could be expected to go wrong in the same ways, and thereby appear to display the "same" disease, or even to 'communicate' it to other bodies. But the sophisticated physician would know those appearances to be illusions caused by the regularity with which biological systems operate: diseases need not be "metaphysically real" in themselves. The various incarnations of this perspective often led medical scholars to argue that, in principle, medicine could forego the concept of disease altogether, and reduce its explanation to the underl)dng laws of physiology, without compromising its ability to perform rationally and effectively (just as adherents of the Boorsian "normal function" account do today: Cf. Hesslow, 1993). The other historical camp has traditionally been the "ontologists" of disease (Risse, 1978). Ontologists have always pointed out that, in terms of building a solid basis of reliable empirical knowledge for medical science, it makes more sense to take the observable constellafions of complaints, physical signs, natural histories, pathogenetic processes and etiologies that constitute

The Enhancement/Prevention Distinction specific diseases as the primary data of medical science, since we are more confident about the reality of those systems than we are about the natvire of the physiological laws they are supposed to reflect. Beyond that, the nature of that reality has been controversial for ontologists. Ontological thinking about disease has been pulled one direction and then another by theories of disease causation, because of its inclination to look for the specific pathognomic thread that ties particular constellations together as systems. Ontological reductionism usually takes the form of assuming that medical explanations can be reduced, ultimately, to accounts of the behavior of these specific causes: germs, poisons, lesions, and genes. I have argued elsewhere that causal explanation in medicine actuallyrequires both ways thinking about health problems, and I think the effort to distinguish genetic prevention from enhancement is another piece of evidence for that thesis (Juengst, 1994). Diseases are real in the same way as the "functionally organized systems" (e.g., the circulatory system) they inhibit and inhabit are real: as decompositions of more complex biological systems that can be discerned by multiple corroborating techniques, from multiple levels of biological organization and across multiple cases (Whitbeck, 1977). In other words, they are susceptible to the same kind of robustness analysis that the natural sciences use to isolate natural entities from the flux of phenomena, even if they cannot exist independently of the patients who display them (Wimsatt, 1981). This epistemic robustness is what gives efforts to prevent diseases the intersubjective accessibility tiiat grounds our confidence that such efforts are meeting legitimate medical needs. If proponents of a particular genetic enhancement intervention cannot describe an adequately robust story of the pathological process their improvement will defend against, the risk increases that they will not be able to tell when they have successfully forestalled or interrupted the process. If they cannot do that, as we have seen, they will not be able to avoid becoming embroiled in questions which "we have little idea of how to delve into" to the ultimate detriment of medicine and society.^ This also makes a good bit more sense out of our paraphrase of Daniels' description of the purpose of health care. Now it should read: The central purpose of preventive health care is to maintain the range, of oppor-

137

138

Eric T. Juengst

tunity and functional efficiency threatened by a disease, e.g., a biological process that moves from discoverable causes (genes, germs, or environmental insults) through a robustly confirmable process of pathogenesis that yields characteristic signs and symptoms that, in turn, reduce function below species-typical norms.

V THE CONSEQUENCES OF PREVENTING DISEASES

In conclusion, let me suggest three of the major implications of my analysis for further work. Most of these points have come out along the way, but all deserve better explication before their significance will be fully clear. a) Reviving a robust concept of disease does seem to help resolve some "hard cases" for the currently dominant approach to medicine's boundary problems. It not only allows the "normal ftmction" approach to explain why patients in equivalent levels of distress and facing equivalent hardships due to their short stature, gynecomastia, or depression may have different claims to medical care (because of the relative certainty of one's endpoints in cases of diagnosable disease), but it also helps the approach include paradigmatic preventive medical services like vaccination that would otherwise be serious anomalies for the model. Vaccinations against infection are now acceptable, because they are efforts to strengthen the body against specific diseases. b) Incorporating a robust concept of disease thus does improve the boundary defenses of medical practice, allowing health professionals to carry their commitment to prevention as well as cure into genetic medicine without operung the door to illegitimate enhancements. This suggests progress in the professional ethics of gene therapy, at least, if not in either "private morality" or social policy. c) On the other hand, giving the concept of disease a crucial role in delimiting medicine's domain does increase the possibility of medical hegemony with respect to genetic engineering practices in our society, and preserves a traditional opening through which cultural values and prejudices can influence medical reasoning. Thus, medicine regains the burden of inter-subjective proof in setting the limits of its own practice through the discovery and delineation of new disease entities. This makes it objective and regulatable, but it also makes it possible for medicine to spin


139

stories of disease to suit various cultural or social values. For example, Kathy Davis, in her sociological study of cosmetic surgery, recounts the occasion in which she heard a well-known plastic surgeon rationalize performing a nose bob on a teen-age Morrocan girl living in Holland by diagnosing her with an "inferiority complex due to racial characteristics" (Davis, 1995, p. 2). d) Finally, this analysis, by illustrating one important theoretical role played by the robust concept of disease, suggests that, far from being ready to dispense with the concept of disease, as some have been suggesting, philosophers of medicine should pay even more attention to the theoretical role that pathological entities play in the ways in which medicine frames its domain. When new issues return us to old problems, does it suggest that we have been traveling in a circle, or a spiral? I have argued that the best way to draw a distinction for policy purposes between preventive and enhancement uses of medicine's emerging gene transfer techniques is to embrace the very old notion of a "disease entity" and to test proposed interventions against the robustness of the diseases they would forestall. Where the human problems anticipated by an intervention cannot be tied together into a diagnosable disease entity, with its recognizable coristellation of subjective symptoms, physical signs and causes, it should not be adopted as a proper part of medical practice. If this helps, it may be that we are ready for next turn in the helix: Is anything wrong with genetic self-improvement pursued without the help of a doctor?

NOTES: 1 The treatment/enhancement distinction is almost always used, and discussed, in the context of germ-line gene transfer, where it is used to demarcate the limits of permissible practice. Where the distinction is raised in the context of somatic cell interventions, it is much more likely to be raised as a question of resource allocation and legitimate medical needs. ^ But perhaps this is just a problem of specifying the right reference class? One might say that in this case the age of the individuals is irrelevant: unexposed humans of any age are equally susceptible to infection. Further, one might argue that the unexposed do fall below the resistance typical for the whole species, because immunity to polio is a trait of all those who survive their first exposure, whether by virtue of superior capabilities, luck or vaccine. By definition, at any one time the immune class will be the vast majority of living members of the

140

Eric T. Juengst

species, and will represent the statistical norm. Against this norm, the non-immunized suffer, if not a dysfunction, a distinct handicap, which they might claim help in overcoming (cf. Allen and Fost, 1990 for a similar argument regarding short stature). One way to analyze this argument would be to pursue a discussion of the relative roles of statistical and teleological conceptions of "normality" in biomedical thinking (Wachbroit, 1994). Without even pursuing that, however, we can note that the immimized gain their immunity much more efficiently than those who must survive infection by the disease, and indeed than would be normal for the species as a whole if left to its own natural devices. Even though vaccination preserves species-typical functioning, it results in a net gain for the immunized: an enhancement of their ability to respond effectively to infection over that provided by the human immune system's "natural functional organization." ^ For some other interesting challenges that the body's natural defense mechanisms raise for the "normal functiori" accovmt of health, see Nordenfelt, 1987, pp. 23-35. ^ Elsewhere Daniels makes this definitional link explicitly in the text: "We have obligations to provide services whenever someone desires that a medical need be met. Generally this is taken to mean that the service involves the treatment of a disease or disability when disease and disability are seen as departures from species typical normal functional orgaruzation or functioning." See Daniels, "Growth Hormone," ibid., p. 46. ^ Or more formally, "A disease is a type of internal state which is either an impairment of normal fvmctional ability, i.e., a reduction of one or more functional abilities below typical efficiency, or a limitation on functional ability caused by environmental agents" (Boorse, 1977, p. 567). * Interestingly, support for this view of the role of "disease entities" in defining the scope of "prevention" in genetic medicine can be found in the way in which the field of preventive medicine already describes its own domain: Prevention is said to have three components: primary, secondary and tertiary. Primary prevention means preventing the occurrence of disease or injury, for example, by immimization against infectious diseases and by the use of safety equipment to protect workers in hazardous occupations. Secondary prevention means early detection and intervention, preferably before the condition is clinically apparent, and has the aim of reversing, halting or at least retarding the progress of a condition. Secondary prevention is epitomized in screening programs in which people with early, often preclinical manifestations of disease are identified and offered a regimen to prevent its progression. Tertiary prevention means minimizing the effects of disease and disability by surveillance and maintenance aimed at preventing complications and premature deterioration (Last, 1986, p. 3). Each of the three traditional components or forms of prevention corresponds to one of the major explanatory levels in a robust disease entity. Primary prevention addresses the etiologic level, by preventing exposure to or infection by the causes


141

of disease. Secondary prevention attempts to interrupt a disease process by interfering with the cascade of broken parts and dysfunctional processes that constitute the pathogenesis of the disease. Tertiary prevention is aimed primarily at preventing the known signs and symptoms of a mature disease process. The field is deployed against the organizing principle of the ontological disease entity because it is what provides the clues to intervention and the criteria for success. This, after all, is what being "medically defined" or "diagnosable" means!

REFERENCES Allen, D. and Fost, N.: 1990, 'Growth hormone therapy for short stature: Panacea or Pandora's Box?' Journal of Pediatrics, 117, pp. 16-21. Anderson, W. F.: 1989, 'Why draw a line?' Journal of Medicine and Philosophy, 14, pp. 681-693. Baird, P.: 1994, 'Altering human genes: Social, ethical and legal implications,' Perspectives in Biology and Medicine, 37, pp. 566-575.

Berger, E., and Gert, B.: 1991, 'Genetic disorders and the ethical status of germline gene therapy,' Journal of Medicine and Philosophy, 16, pp. 667-685. Boorse, C: 1977, 'Health as a theoretical concept,' Philosophy of Science, 44, pp. 542-573. Clouser, K.D., Culver, C, and Gert, B.: 1981, 'Malady: A new treatment of disease,' Hastings Center Report, 11, pp. 29-37. Culver, K.: 1993, 'Current status of human gene therapy research,' The Genetic Resource 7, pp. 5-10. Daniels, N.: 1992, 'Growth hormone therapy for short stature: Can we support the treatment/enhancement distinction?' Growth, Genetic and Hormones, 8 (Suppl. 1), pp. 46-48. Daniels, N.: 1986, Just Health Care, Cambridge University Press, New York. Davis, K.: 1995, Reshaping the Female Body: The Dilemma of Cosmetic Surgery,

Routledge, New York. Dubos, R.: 1959, Mirage of Health: Utopias, Progress and Biological Change, Harper &

Row, New York. Engelhardt, H.T. Jr.: 1986, The Fountiations of Bioethics, Oxford University Press, Oxford. Gardner, W.: 1995, 'Can enhancement be prohibited?' Journal of Medicine and Philosophy, 20, pp. 65-84. Hesslow, G.: 1993, 'Do we need a concept of disease?' Theoretical Metlidne, 14, pp. 1-14. Juengst, E.: 1993, 'Causation and the conceptual scheme of medical knowledge,' in CD. Delkeskamp-Hayes and M. A. Gardell Cutter (eds.). Science, Technology and the Art of Medicine, Kluwer Academic Publishers, Dordrecht, pp. 127-152. Krimsky, S.: 1990, 'Human gene therapy: Must we know where to stop before we start?' Hitman Gene Therapy, 1, pp. 171-173.

142

Eric T. Juengst

Last, J.: 1986, 'Scope and methods of prevention,' in J. Last (ed.). Public Health and Preventive Medicine, 12th ed., Appleton-Century Crofts, New York. Moseley, R.: 1991, 'Maintaining the somatic/germ-line distinction: Some ethical drawbacks,' Journal of Medicine and Philosophy, 16, pp. 641-649. Nordenfeit, L.: 1987, On the Nature of Health: An Action-Theoretic Approach, D. Reidel, Dordrecht. Risse, G.: 1978, 'Health and disease: History of the concepts,' in W. Reich (ed.). The Encyclopedia ofBioethics, MacMillan, New York, pp. 579-585. Rosenberg, S., Kasid, A, Anderson, W.R, et al: 1990, 'TNF/TIL human gene therapy clinical protocol,' Human Gene Therapy, 1, pp. 443-462. Sabin, J. and Daniels, N.: 1995, 'Determining 'medical necessity in mental health practice," Hastings Center Report, 24, pp. 5-13. Silvers, A.: 1994, 'Defective agents: Equality, difference and the tyranny of the normal,' Journal of Social Philosophy, 25, pp. 154-175. Wachbroit, R.: 1994, 'Normality as a biological concept,' Philosophy of Science, 61, pp. 579-591. Walters, L.: 1986, 'Ethics of human gene therapy,' Nature, 320, pp. 225-227. Walters, L., and Palmer, J.: 1997, The Ethics of Human Gene Therapy, Oxford University Press, New York. Whitbeck, C: 1977, 'Causation in medicine: The disease entity model,' Philosophy of Science, 44, pp. 619-637. Wilson J., Grossman, M., Roper, S., et al: 1992, 'Ex Vivo gene therapy for familial hypercholesterolemia,' Human Gene Therapy, 3, pp. 179-222. Wimsatt, W.: 1981, 'Robustness, reliability and overdetermination,' in M. Brewer and B. Collins (eds.). Scientific Inquiry and the Social Sciences, JosseyBass, San Francisco, pp. 124-163.