Exploring the complexity of high school students

Linguistics and Education 45 (2018) 10–19

Contents lists available at ScienceDirect

Linguistics and Education journal homepage: www.elsevier.com/locate/linged

Exploring the complexity of high school students’ beliefs about language variation Mike Metz University of Missouri – Columbia, 303 Townsend Hall, Columbia, MO 65211, United States

a r t i c l e

i n f o

Article history: Received 3 October 2016 Received in revised form 3 February 2018 Accepted 22 February 2018 Keywords: Language ideologies Dialects Critical Language Awareness Student beliefs

a b s t r a c t This study explores the knowledge and beliefs about language variation from high school students in the San Francisco Bay Area. Using quantitative analysis of a survey of language beliefs, combined with a thematic analysis of student interviews, the study explores the language ideologies demonstrated by students from a wide range of sociocultural backgrounds. Key findings include that neither race nor linguistic background predict whether students hold dominant language ideologies that frame Standardized English as the correct form of English, or critical language ideologies that uphold the value of all English varieties. The key characteristics that predict language ideology are parents’ language ideology and students’ awareness that they speak more than one variety of English. The findings support previous theorizing that suggests language ideologies shift with context and purpose. Students describe the racialized nature of language although there is great diversity of language ideology within racialized groups. As suggested by Kroskrity (2010) increased awareness of linguistic diversity and language ideologies is related to contestation of dominant language ideologies. Altogether the findings paint a picture of students with wide ranging knowledge and beliefs about language variation that will complicate teaching about language variation in school. Teachers will do well to assess what their students know and believe about language before teaching them. Researchers are encouraged to continue to explore student understandings of language variation as this area remains underdeveloped. © 2018 Elsevier Inc. All rights reserved.

I think because there’s so many different forms of English, like, I couldn’t choose which one is the correct one. My idea is there’s many correct ones, but, according to my teachers and stuff, or other people, Standard English is correct. But I don’t think that’s, like, correct for myself. - Raul, 10th Grade 1. Introduction Work on culturally relevant pedagogy has long emphasized the linguistic aspects of culture. Given that one of the primary tenants of culturally relevant pedagogy is the incorporation of student knowledge and experiences in instruction, it is important to understand what students know and believe about language. As the opening quote suggests, students may be wrestling with understanding their own linguistic experiences in light of school and societal messages about the value of different language varieties. Existing literature on dialect awareness instruction emphasizes teachers’ knowledge and beliefs about language variation with

E-mail address: [email protected] https://doi.org/10.1016/j.linged.2018.02.003 0898-5898/© 2018 Elsevier Inc. All rights reserved.

little exploration of students’ knowledge and beliefs. To support teachers’ implementation of culturally relevant pedagogy and associated dialect awareness curricula there is a need for information about students’ understandings – and misconceptions – about language variation. This study helps to paint a picture of student understanding of language variation by asking: What do students know and believe about language variation? How do students understand the prevailing social narratives about speakers of different dialects? To what degree do students accept the prevailing narratives and to what degree do they espouse counter-narratives? With this knowledge in hand, teachers will be better equipped to anticipate the needs of their students and the range of responses to teaching about language variation in English classrooms.

2. Situating the study 2.1. Research on language variation in U.S. schools Research on English language variation in U.S. schools has a history almost as long as research on language variation itself, however students’ understanding of language remains largely absent from

M. Metz / Linguistics and Education 45 (2018) 10–19

this research. Studies as far back as the 1930s document the use of Pidgin English in Hawaii and the responses of teachers to the use of this English dialect (Reinecke & Tokimasa, 1934). Building on the rich body of sociolinguistic research produced in the 1960s, educational research on language in schools during the 1970s focused on the degree to which teacher’s beliefs and biases about language impacted their expectations of student achievement with consequences for student learning (Smitherman, 1977; Taylor, 1973; Williams, Whitehead, & Miller, 1972). In the 1980s and 1990s, alongside research on multicultural education and other resource pedagogies, research on language variation in U.S. schools began focusing on pedagogical approaches that leverage students’ language varieties as resources for learning (Ball & Lardner, 1997; Ladson-Billings, 1992; Perry & Delpit, 1998; Rickford, 1999). One branch of this research considered how to address teachers’ attitudes toward language through teacher education (Bowie & Bond, 1994; Byrnes, Kiger, & Lee Manning, 1997). The emphasis on pedagogical approaches and teacher attitudes is necessary, but not sufficient, for preparing teachers to enact critical language pedagogies. Much recent research on student language variation focuses on student use of different language varieties. This research highlights the value of students’ multi-lingual and multi-dialectical skills and emphasizes ways to promote those skills in the academic environment (Alim, 2005; Gutiérrez, Baquedano-López, & Tejeda, 1999; Young & Martinez, 2011). Even so, this research often falls short of assessing what students already know and understand about their language use. Important findings in recent scholarship endorse teaching meta-awareness of language variation to students (Brown, 2006; Christensen, 2011; Godley & Loretto, 2013; Janks, 2009; Mallinson & Charity Hudley, 2010) but how students respond to instruction around language variation may differ depending on their prior knowledge and beliefs. Notable exceptions to the trend described above include Godley and Escher (2012) and Martínez (2013), who center their studies on students’ understanding of their own language use. Godley and Escher (2012) explore students’ attitudes toward spoken dialects through interviews with a group of students that were part of a larger study of critical language pedagogy. They find that their African-American students possess a range of beliefs about what dialects (Standard English or African American English) should be spoken in school, as well as articulating a wide range of rationales for those beliefs. Martínez’ (2013) study explores middle-school students’ understanding of their own use of Spanglish. In his study, Martínez highlights the contradictions between how students explicitly name their beliefs about Spanglish (articulated beliefs) and the underlying beliefs demonstrated through their use of the language variety (embodied beliefs). Like Godley & Escher, Martinez shows the influence of pervasive societal messages about the value of different varieties of English on students’ understanding of their own language use. Outside of the U.S. context, McKinney (2017), as part of a more extensive exploration of language ideologies in contemporary South African schooling, attends to both teachers’ and students’ beliefs about language. She explores the tension around ways that students uphold and resist discourses that value “a narrow range of linguistic resources associated with prestige varieties of English” (McKinley, 2017, p. 80). Using the construct of anglonormativity – which McKinney describes as a privileged, yet invisible, white ethnolinguistic repertoire – she describes how students both reproduce these racialized linguistic patterns and contest them at the same time. Importantly, McKinney analyzes students’ explanations of their language beliefs as well as examples of students’ language practices. Research exploring student beliefs about language variation is important for two reasons. First as Martínez suggests,

11

quoting Kroskrity (2004), research on students beliefs helps reverse “a longstanding scholarly tradition of delegitimating common people’s views of language” (p. 507). Second, an understanding of students’ knowledge and beliefs about language variation is necessary to improve the teaching about language variation recommended in the research on pedagogical approaches cited above (Metz, 2018). Knowledge of students’ understanding is a primary component of pedagogical content knowledge (Grossman, 1990; Shulman, 1987) and is particularly important for work with students who have been historically underserved by school (Lee, 2007). As Grossman explains, “To generate appropriate explanations and representations, teachers must have some knowledge about what students already know about a topic and what they are likely to find puzzling” (1990, p. 8). Lee points out that valuing students’ understanding “is made more challenging when teachers and curriculum makers must overcome deficit assumptions about the nature of routine practices and attendant belief systems of students” (2017, p. 35). Previous research on language variation in schools provides rich resources for content knowledge and pedagogical knowledge, but there is little basis for developing teachers’ knowledge of students’ understandings of language variation. This study, in conversation with studies like those by Martínez, McKinney, and Godley & Escher, provides a resource for teachers to consider the types of understanding and beliefs they may contend with in their own classrooms as they teach students about language variation.

2.2. Language ideologies This study explores students’ beliefs through the frame of language ideologies. There is an extensive and growing body of literature on language ideologies with no consistent definition (See, for example, Irvine, 1989; Kroskrity, 2010; Milroy, 2001; Rosa & Burdick, 2017; Woolard & Schieffelin, 1994). However, within the complementary and sometimes conflicting approaches, there is a fundamental consensus that language ideologies describe sets of beliefs about what language is and how it works in society. Recent scholarship on language ideologies asserts that there is no “objective” understanding of language but rather that all understandings of language are ideological (Rosa & Burdick, 2017). For the purposes of this study, I adopt Irvine’s definition that equates language ideologies with “the cultural (or subcultural) system of ideas about social and linguistic relationships, together with their loading of moral and political interests” (Irvine, 1989). The explicit naming of moral and political interests allows for a grouping of language ideologies for analytic purposes. For the first part of this study I follow Martínez (2013) in bundling a set of language ideologies under the label dominant language ideologies. This bundle of ideologies includes the broad standard language ideology (Milroy, 2001) that centers a variety of English based on standardized conventions of written English. It also includes the political/moral belief that Standardized English1 (SE) is the correct and best language variety. This set of beliefs positions other varieties of English as deficient, sloppy, lazy, and lacking value in situations of consequence. Dominant language ideologies connect these beliefs about language to the speakers of the language (Wortham, 2008), thus asserting that speakers of other

1 I follow other critical language scholars in using Standardized English rather than Standard English, to signal that standardization is a social process and not an inherent aspect of any particular language variety. In the survey, I used the more common term Standard English to avoid confusing students. In this paper, I use the term Standard English when I refer to the survey construct.

12


varieties of English are lazy, uneducated, and intellectually and morally deficient. For the purpose of this study, beliefs that counter dominant language ideologies fall under the mantel of critical language ideologies, also described as a counter-hegemonic language ideologies (Martínez, 2013). This constellation of beliefs includes the understanding that all varieties of English are systematic and rule governed, capable of expressing complex reasoning, and possess inherent value. Critical language ideologies encompass beliefs that associate positive characteristics with speakers of historically stigmatized varieties; they are intelligent, educated, trustworthy, kind, etc. The survey analysis makes use of the bundles of language ideologies – dominant and critical – described above as heuristics, to help illustrate broad patterns. Students, and speakers in general, hold multiple and shifting language ideologies (Kroskrity, 2004), thus the use of the labels dominant and critical apply to constellations of beliefs which may mix and shift fluidly. Student talk about language makes visible the much more complex realities of language ideologies necessitating a more nuanced conceptualization. While standard language ideologies position languages and language varieties as stable, bounded objects (Milroy, 2001), this understanding is increasingly contested in post-modernist, poststructuralist views of language (Blommaert, 2010; Garcia, Flores, & Spotti, 2017; Makoni & Pennycook, 2007). A post-structuralist view of language describes unbounded linguistic repertoires focused on language in practice, rather than language as a decontextualized object. While the survey instrument relies on commonly available discourses about language based in a structuralist tradition, analysis of the fluid practices students describe in interviews calls for post-structuralist conceptions of language-in-use (Pennycook, 2017). Student explanations of language use often center on what identities are indexed by particular ways of using language (Bucholtz & Hall, 2004; Wortham, 2008). Thus, while the analysis of survey data uses the broad concepts of dominant and critical language ideologies, the analysis of interview data parses those concepts more carefully to examine ties between language and identity. The concept of raciolinguistics (Alim, 2016; Flores & Rosa, 2015; Rosa & Flores, 2017) helps illustrate the tight intersections between racialized cultural ways of being and the racialization of linguistic forms. The emerging field of raciolinguistics explores the way the linguistic practices of racialized bodies are interpreted through a racialized lens. Hence, the same linguistic forms, used by speakers racialized in different ways, will be interpreted differently according to how the speaker is positioned. For example, historically stigmatized forms of African American English, when spoken by white speakers, lose much of their stigma. The privileging of white language users and white language forms is captured in McKinney’s term anglonormativity (McKinney, 2017). Anglonormativity foregrounds the globalized power dynamics that normalize Anglo (white) ways of using language, particularly in school contexts. The concepts of raciolinguistics and Anglonormativity encompass additional language ideologies that suggest language is not simply reflective of identity but that language and identity are mutually constituted. Thus it is not only that “language ideologies systematically associate types of language use with socially located types of people” (Wortham, 2008, p. 45), but that types of people and language use construct one another in an ongoing and continually contested process (Rosa & Flores, 2017). The complex relationship between language, identity, and power contained in the language ideologies underlying raciolinguistics and the construct of anglonormativity show up in the ways students talk about language and identity, and thus these concepts help make sense of students’ talk.

3. Methods The analyzed data comes from a study of dialect awareness teaching in high school English classrooms in the San Francisco Bay area. One data set consists of student responses to a survey about beliefs regarding language varieties. A second data set comprises interviews with a subset of the survey respondents. 3.1. Survey methods The survey was given to 169 high-school students in 7 racially and linguistically diverse classrooms across 5 schools and 3 districts in the San Francisco Bay Area. The survey consisted of 15 Likert scale items based on constructs used in previous surveys on language attitudes (Taylor, 1973; Williams et al., 1972; Smitherman & Villanueva, 2000). The Likert scale was a six-point, forced-choice scale from 1-Agree Strongly to 6-Disagree Strongly. Eight additional questions asked about the respondents’ own language use and demographic information. Because the responses for each item are not continuous variables, the tests for significance between groups required non-parametric tests. When testing the statistical significance of differences between two groups (i.e. gender categorized as male or female) I used A Mann–Whitney U test. When testing the statistical significance of differences between three or more groups (i.e. four race categories) I used a Kruskal–Wallis H test. In all cases the distribution of scores across categories supported comparison of median scores, which are included in the analysis below. 3.2. Interview methods Interviews were conducted with 5–8 students from each of 6 classes that were observed during the study, a total of 37 students. Students were selected in collaboration with the classroom teachers, privileging a mix of gender, race, language background, and general engagement with the class. The interview followed a semi-structured protocol beginning with questions asking students to reflect on the survey statements and with follow up questions depending on student answers (Table 1). Student responses were coded for evidence of language ideologies. The codes, grouped into composite ideologies, are shown in Table 2.

Table 1 Characteristics of the student sample. Gender: Grade: Race/Ethnicity:

Male 50.9% 10th 11.8% Latino 48.5%

female 47.3% 11th 70.4% Black 26.6%

other 1.8% 12th 17.8% Asian 20.7%

White 4.1%

Table 2 Sub-codes for each language ideology. Ideology code

Dominant language ideology

Critical language ideology

Sub-codes

• SE is correct • SE is “normal” English • Only SE allows for complex reasoning – OVE is deficient • Speakers of SE have positive characteristics – speakers of OVE have negative characteristics

• There are multiple correct Englishes • OVE is “normal” English • OVE speakers have positive characteristics


13

Fig. 1. Student responses sorted by agreement with Standard English ideology.

In addition, codes were created based on two strong themes that arose from the data: the racialization of language, and the fluidity of multiple language ideologies. The coded interviews were examined in relation to the findings from the survey responses. 4. Findings 4.1. Overall survey results The survey analysis describes what students believe about language variation and examines those beliefs across student characteristics. Fig. 1 shows the frequency distributions for the 15 language belief items on the survey. The bars are centered on zero with percent disagreement to the left and percent agreement to the right. The four survey items indicated with an *R were reverse coded so that across all items, agreement indicates alignment with dominant language ideologies and disagreement signals alignment with critical language ideologies. The dashed line in Fig. 1 indicates the point at which 50% of the students indicate agreement with a traditional Standardized English ideology. The six items above the dashed line show greater agreement with statements that privilege Standard English and disagreement with statements that privilege Other Varieties of English. For the nine items below the dashed line a majority of students agree with statements that privilege Other Varieties of English and disagree with statements that privilege Standard English. This figure also shows the range of students’ beliefs. While the top and bottom items trend heavily toward one side or another, the majority of items demonstrate a spread across agreement and disagreement. Nearly 90% of the students agree that OVE is sometimes more useful than SE, and that SE is necessary for professional

settings. Likewise, more than 80% of students agreed that SE is the correct form of English and the same number disagreed with the statement that people who speak OVE are lazy. However, the survey items in the middle of Fig. 1, such as “To be successful, students should only use Standard English” show a broad distribution of student responses from strongly agree to strongly disagree. To help understand the distribution of responses, I examined differences between groups of students along six student characteristics. These included demographic items as well as responses to statements about students’ own language use. The percentage of students from the overall sample in each of these categories is shown in Table 3. 4.2. Survey responses by student characteristics Race, and ‘I grew up speaking.’ In contrast to the expected findings, there were no significant differences in any of the 15 items based on race or the languages students grew up speaking. This finding is distinct from suggesting that language ideologies are not racialized, a topic explored in detail during the interviews. Gender. The distribution of responses was the same across both gender categories for all the items except two. For “Standard English is the correct form of English,” males (Mdn = 2) agreed more strongly than females (Mdn = 3), U = 4217, z = 2.151, p = .031. For “People who speak other varieties of English are intelligent,” females (Mdn = 2) agreed more strongly than males (Mdn = 3), U = 2800.00, z = −2.489, p = .013. In both of these cases males and females share the same general belief, however males are more closely aligned with a Standard English ideology than females. I speak more than one variety of English. Students responses to the statement “I speak more than one variety of English” were collapsed to agree and disagree for this analysis. The analysis identified statistically significant differences on six of the fifteen

14


Table 3 Percent of students by student characteristics. Gender Race I grew up speaking I speak more than one variety of English My parents want me to speak SE

Male (51%) Latino (49%) English only (28%) Yes (78%) Yes (67%)

Female (49%) Black (27%) English Mostly (15%) No (22%) No (33%)

Asian (21%) English & Other equally (37%)

White (4%) Another mostly (15%)

Another only (5%)

Table 4 Results of Mann–Whitley U test comparing student responses by “I speak more than one variety of English.”. Item

Other varieties of English are sometimes more useful than Standard English Speakers of other varieties of English are intelligent Speakers of other varieties of English are lazy. Standard English is necessary for professional settings Teachers think poorly of students who speak OVE Most high school students believe that if you speak Standard English you are trying to sound smart.

“I speak more than one variety of English.”

U value

z-score

p value

Median (I agree)

Median (I disagree)

2

3

3326.5

3.536

≤0.001

2

3

3008

2.215

0.027

5 2

4 3

1676.5 3337

-3.078 3.58

0.002 ≤0.001

3

4

3241

3.098

0.002

2

4

3156

2.775

0.006

Responses are on a Likert scale from 1 = strongly agree to 6 = strongly disagree. Thus scores from 1 to 3 indicate lessening agreement and scores from 4 to 6 indicate increasing disagreement.

survey items. The results for the significant items are shown in Table 4. The medians of each group showed similarities in agreement or disagreement for the first four items in Table 5. Students who speak more than one variety of English felt more strongly about each statement. While the median for students who do not speak more than one variety of English is always agree slightly (3) or disagree slightly (4) the students who speak more than one variety of English Agree (2) and Disagree (5). For the item “teachers think poorly of students who speak OVE,” students who indicated they spoke varieties of English felt teachers did judge students for speaking dialects (median = 3), while students who did not speak varieties felt that teachers did not pass judgment (median = 4). An even larger difference appeared in the way students responded to “Most high school students think if you speak Standard English you are trying to sound smart.” Students who speak OVE agreed with this statement (Median = 2) while students who only speak SE disagreed (Median = 4). My parents want me to speak Standard English. One important difference between students and adults is the role of parents in influencing student beliefs. For this reason, all students were asked to respond to the statement “My parents/guardians want me to speak Standard English.” For the purpose of analysis, the scale was collapsed to agree and disagree. Analysis identified statistically significant differences on eight of the survey items. The results for the significant items are shown in Table 5. Students answered eight of fifteen, or 53%, of the survey items statistically significantly differently based on whether their parents want them to speak Standard English. For this sample, that makes the parents’ language ideology the most influential factor shaping students’ opinions of language. Further, as might be expected, students whose parents wanted them to speak Standard English aligned more closely with a Standard English ideology on each of the eight survey items. For six of the eight items, there was alignment in whether both groups agreed or disagreed, with the difference between groups showing up in the strength of agreement or disagreement. However, for two of the items, “People who speak standard English are well educated,” and “To be successful you must speak standard

English,” students whose parents wanted them to speak Standard English agreed, while others disagreed. Recap of survey findings. Key findings from the survey data suggest that race, ethnicity, and gender do not predict language ideology. Neither does the language spoken at home. In contrast, student awareness of speaking more than one variety of English was a significant predictor of language ideology with students who agreed that they spoke more than one variety of English aligning more strongly with a critical language ideology and also indicating feelings of judgment by teachers and peers. The strongest predictor of student language ideology was students’ belief that their parents wanted them to speak Standard English. Students who indicated that their parents wanted them to speak Standard English aligned more strongly with a dominant language ideology on a majority of items. 4.3. Interview findings This section describes two main themes from the interview data that help unpack the findings from the survey. (1) Although beliefs about language do not correlate with race, students understand language to be highly racialized. (2) Students hold multiple language ideologies and distinguish prevailing ideologies in different contexts and for different purposes. 4.3.1. Language and race That the survey revealed no correlation between race and language ideology is distinctly different than suggesting that students do not hold racialized views of language. In fact, 32 of the 37 students interviewed (86%) spoke about language in racialized terms. Students’ descriptions of racialized language fit into three categories: (a) explanations of racial stereotypes related to language, (b) descriptions of personal experiences of racial bias related to language, (c) contestation of racialized stereotypes related to language. Language, race, and speaker characteristics. Students described a range of racial stereotypes associated with language. Although these students draw relationships between racialized identity and language use, they describe the direction of the relationship differently. The differing descriptions of the


15

Table 5 Mann–Whitley U test for comparison of student responses by “My parents/guardians want me to speak Standard English.”. Item

Standard English is the correct form of English. Other varieties of English are inappropriate for school. People who speak standard English are well educated. To be successful, students should only speak standard English Standard English is necessary for professional settings Other varieties of English are sloppy forms of English. Standard English is the best language for complex reasoning. Most high school students think if you speak standard English you are trying to sound smart.

“My Parents want me to speak standard English”

U value

z-score

p value

3 5

3949.5 4130

2.885 3.429

0.004 0.001

3

4

3847.5

2.475

0.013

3

4

4076

3.215

0.001

2

2

3835

2.473

0.013

4

4

3719

2.013

0.044

3

3

3999.5

3.008

0.003

2.5

3

3789

2.244

0.025

Median (I agree)

Median (I disagree)

2 4

Responses are on a Likert scale from 1 = strongly agree to 6 = strongly disagree. Thus scores from 1 to 3 indicate lessening agreement and scores from 4 to 6 indicate increasing disagreement.

relationship between racialization and language use emphasizes the mutually constituting nature of racial and linguistic identity (Rosa & Flores, 2017). The analysis below shows how students grapple with this complex relationship. Thomas, an African American2 11th grader, described racialized linguistic stereotypes succinctly, “people tend to think black people talk ghetto, Mexican people have heavy accents and no one really can understand them; and then Asian people, they talk gibberish, and like, you know, stuff like that.” Thomas’s description of the relationship between race and language starts with racial groups and follows with particular ways of talking associated with those groups. Other students flipped that association and made the connection between ways of talking and particular racialized identities. Sonia, a Mexican-American 11th grader, explains racialized linguistic stereotypes this way, Like if you talk with a lot of slang, you’re looked at like, oh, you’re ghetto. If you talk really proper, oh, you’re white-wash. If you talk, like, Spanglish, you know a mixture of Spanish and English, then you’re looked at as like, really Mexican. Sonia uses the terms “slang,” “proper,” and “Spanglish” to describe three language varieties, and she uses the terms “ghetto,” “white-wash,” and “really Mexican,” to describe kinds of people. In contrast to Thomas’s description, Sonia is not referencing a stereotype that all people of a certain race talk a certain way, but rather is highlighting racialized labels that get put on individuals regardless of race, based on how they speak. Sonia’s labels link language, race, and characteristics of speakers to assert that her MexicanAmerican friends may be considered “ghetto,” “white-wash,” or “really Mexican” based on their use of language. Katrina, a white, 11th grade student, highlights the complex interactions of multiple categorizations alongside language. She starts with perceptions of intelligence, and then creates an assemblage of educational level, linguistic features, race, and class. Because when you’re smart, when you’re looked at smart, you’re expected to speak a certain way. And if you’re looked at as not smart, or uneducated, you have a ghetto accent. Or people will be like, why do you talk black? And that’s like, to me, saying blacks are stupid. That’s not true. Like, I think that’s when class

2 All racial labels for students come from an open-ended question on the survey that asked “How do you identify racially or ethnically?”

comes in, if you talk ghetto then – then you’re not smart or educated, and that’s not true. Katrina’s description exemplifies the way “categories are intersectionally assembled and communicatively co-constituted” (Rosa & Flores, 2017, p. 636). Katrina makes the association between being “looked at as not smart, or uneducated” and having a “ghetto accent.” She then equates having a “ghetto accent” with talking “Black.” Next, she critiques the perceived implication that a black identity equates with stupidity. Katrina both reifies and contests the idea that a “ghetto accent,” “Black” talk, and a Black identity are all one construct, and that this construct indexes being “not smart.” To try to disentangle these ideas, Katrina adds the idea of social class saying, “I think that’s where class comes in.” Thus, her implicit argument is that Black identity should be removed from this indexical chain, while she explicitly says that “ghetto” talk should not be equated with a lack of intelligence or education. The descriptions from each of these three students demonstrate what Rosa and Flores (2017) describe as a process of raciolinguistic enregisterment. The racial and linguistic categories are mutually constituted such that race signals language simultaneously with language signaling race. Thus, although the three students describe the connection between race and language from different directions – Thomas explains the perception that people of a certain race talk a certain way; Sonia explains that certain kinds of talk signal certain racialized identities; Katrina explains that certain characteristics involve an expectation of a certain kind of racialized speech – they are all describing the same entangling assemblage of mutually constituted raciolinguistic categorization. The students in this study did not have the meta-language to articulate these ideas precisely, still the key concept is that students are aware of, and contend with, larger societal narratives about the racialized nature of language, as well as what language and race signal about the characteristics of speakers. Differing responses to common racialized linguistic experiences. The survey responses suggest that students within any given racial group hold a range of language ideologies. This is consistent with previous research that describes the inherently multiple nature of language ideologies (Kroskrity, 2010). To explore this finding further, the analysis below shows how students of a single racial group describe similar racialized language experiences from distinct perspectives. Ten of the 16 African-American students interviewed described being told they “talk white,” or “white-washed.” When asked to describe what that meant, all the students explained the association

16


between “talking white” and the use of “big words” and “proper language.” The phrase carried the connotation of trying to “act smart” in an effort to be perceived as intelligent or educated, aligning with the privileging of a white ethnolinguistic repertoire that McKinney (2017) refers to as anglonormativity. Three female African-American students provided more detail on how they experienced and responded to being told they talk “white-wash.” The differences in their answers demonstrate the diversity of language ideologies within a single racial-gender group. Darcel described her experience in detail, [My step sister] would always say, “Why are you trying to sound smart?” like, “Why are you talking white?” And I’m like, “I’m not talking white. I’m talking like I’m educated, because I am. Like, I’m not trying to impress you. I’m not trying to, like, show off or anything. This is just how I speak sometimes.” . . . And I remember I had cousins – we weren’t really close – and sometimes they would call me an “Oreo” because of the way I acted, . . . And, when they called me that, like, it hurt my feelings. . . . but now it’s just, like, “I don’t care how you perceive me. That’s just how I am. I go to school. I pay attention in school. I like the way I speak in school, and so that’s just a part of who I am.” In her description, Darcel contests the idea that her speech indexes a white racial identity, (I’m not talking white”) but she accepts the idea that her speech indexes education (“I’m talking like I’m educated”). In this way, Darcel contests one aspect of the dominant language ideology (SE is white speech), while accepting another (SE is educated speech). Darcel responds to the name calling3 and pressure from her relatives by acknowledging the pain she felt at having her racial identity questioned, but persisting in her way of using language anyway. Jerlanne describes her experience and response differently, indicating a differing set of beliefs about language, Say you’re around your group of friends and you speak more like with big words and words that they don’t understand, they say like, “Oh you’re talking whitewash,” or you’re this and that. And like, you gotta kind of lower yourself to where they understand what you’re saying. Like Darcel, Jerlanne experiences a critique of her racial identity based on her language use. With her group of friends, the use of “big words” is associated with white speech and a white racial identity. Unlike Darcel, Jerlanne’s response is that “you gotta kind of lower yourself to where they understand what you’re saying.” The use of the phrase “lower yourself” shows Jerlanne’s adherence to a linguistic hierarchy (Alim, 2005) where some languages are “higher” and others are “lower.” Further, her adjustment of her language signals not only an effort to communicate but also an effort to perform a particular racialized identity through that language use, hence avoiding the “whitewash” label. This performance of a particular racialized Black identity may be more complicated for Jerlanne as her parents are immigrants from Sierra Leone and speak what Jerlanne described as Creole. Thus, although Jerlanne is phenotypically Black, linguistically, African American English is not her home language. Negotiating this raciolinguistic identity with her peers carries additional challenges that may impact her choice to accommodate rather than contest her peer’s expectations (Blake, 2016). Ashley presents a third approach to thinking about racialized language experiences. She explains,

3 The term Oreo is used as a derogatory metaphor to describe someone who is phenotypically black, but acts in ways typically associated with whiteness – the Oreo cookie is black on the outside and white on the inside.

When I’m with my Caucasian friends, I switch up really; like, I have a different vocabulary and tone of voice. It kind of switches up. And like, if I bring some of my other friends around them, they’re like, “You really change in front of them.” And you kind of have to, because then it’s like, there’s no way of communicating with them [Caucasian friends] if you speak the same way you speak with other people. So, I kind of switch up depending on who it is. Ashley’s description reveals an entirely different way of thinking about language use than that described by Darcel or Jerlanne. While Darcel talks about maintaining her way of speaking across contexts, Jerlanne and Ashley both take up the communicative burden (Lippi-Green, 2012), adjusting their speech to accommodate the expectations and needs of the listener. Jerlanne describes “lowering” her speech, while Ashley talks about using a “different vocabulary and tone of voice.” In contrast to Jerlanne, Ashley’s explanation does not assign a value to either way of speaking. The use of descriptive terms - “I have a different vocabulary and tone of voice” – instead of value-laden terms – “you gotta kind of lower yourself” – counters prevailing societal patterns for describing language use. Further, Ashley highlights the communicative limitations of her Caucasian friends (presumably because they are monolingual SE speakers) saying “there’s no way of communicating with them if you speak the same way you speak with other people.” By framing herself as linguistically capable and the SE-only monolingual speakers as limited in their communicative ability, Ashley is demonstrating a powerfully counter-hegemonic view. Instead of normalizing the monolingual white ethnolinguistic repertoire, Ashley normalizes her more expansive multi-dialectical repertoire. The three sets of divergent beliefs about language expressed by these three African-American young women confirm the finding in the survey that race and gender do not correlate with language ideology. These descriptions also prevent any misinterpretation of the findings that might suggest students do not hold racialized views of language. Contesting Racialized Linguistic Stereotypes. A final important theme that arose during students’ talk about race and language centers around contestation of racialized language stereotypes. Students identified the racialized language discourses that mutually constituted whiteness and language use valued in school. A number of students explicitly spoke about disrupting this link. Clinton rebutted the indexical meaning of “talking white” saying, “If I just use big words, it don’t mean I’m talking white; it just means I’m using big words, you know,” and similarly Nyah explained, “I’m not acting white, I’m just – that’s just what I know. Like, it’s not that I’m being white, because I’m Black.” Both Clinton and Nyah make ideological moves to assert their Black identity in conjunction with their language use. This bid for a micro-level reinterpretation of their language use as Black language based on their Black identity is the kind of discourse that can be empowered when amplified in the institutional spaces of schools. In a complimentary way, Thomas and Tracee contest the simple racial-linguistic indexicality suggested by terms like ‘whitewashed’ and ‘ghetto’. Thomas said, “I just think they’re [other students] stupid when they say that-like, ‘whitewashed’ and ‘ghetto.’ I mean, if they talk that way, a lot of people talk that way.” Thomas’s explanation that “a lot of people talk that way” was echoed by Tracee who asserted, in reference to “talking ghetto,” “Not just black people talk that way.” By contesting the iconization (Irvine & Gal, 2000) of particular language forms taken for granted in the larger society, these students demonstrate the ideological complexity students bring to discussions of language variation in schools. Several students identified the racial and linguistic diversity around them as a source of their counter-hegemonic language


ideologies. These students talked about personal experience with friends of different racial and/or linguistic backgrounds that served as counter narratives to easy stereotypes. For example, Joy, a Filipina-American in AP Lit, says of people who speak historically stigmatized language varieties, “I don’t think they’re ignorant; a lot of them are my friends, and I think they’re pretty smart.” And Brea explained, “They don’t think I’m ignorant because of the way I talk, and I don’t think the same thing of them.” Many, but not all, students acknowledged and demonstrated understanding of macro-social raciolinguistic ideologies, while advocating new indexical relations based on their micro-level interactions.

4.3.2. Awareness of multiple ideologies A second finding from the interviews supports the general finding of the surveys that student beliefs about language are often conflicting; In the interviews, 34 of 37 students (92%) expressed beliefs aligned with both dominant and critical language ideologies. Only one student did not express any aspects of a critical language ideology and only two students did not express aspects of a dominant language ideology. The interview data provides important insights into the inherently multiple nature of language ideologies (Kroskrity, 2004), students’ awareness of these divergent ideologies, and how students make sense of their co-existence. The Correctness of both SE and OVE. While nearly all students demonstrated a belief in multiple language ideologies, threequarters of students expressed an awareness of these competing ideologies. Kroskrity (2010) describes awareness – consciousness of language beliefs – as correlating with contestation of dominant ideologies. The very concept of Critical Language Awareness (Alim, 2005; Fairclough, 1992; Janks, 2009) rests on the idea that when students can identity and distinguish the details of language, as well as the belief systems tied to those language features and language processes, they gain the power to critique and resist dominant discourses. During the interviews, students provided insight into their awareness of competing ideologies and how they reconciled these potentially conflicting beliefs. The most prevalent pattern indicated students separated what they were told was right and what they believed was right, or they separated what they knew other people believed and what they believed themselves. The interviewed students tended to espouse a critical ideology for themselves while describing the need to conform to a dominant language ideology in school or “in a professional setting.” Students attendance to the dominant language ideology most often was attributed to school, but was also attributed to parents, and sometimes peers. This finding aligns with Blommaert’s (2010) concept of polycentricity, the idea that speakers orient themselves to different addressees when they speak, and that those different addressees, with different amounts of power, value different indexical meanings. Without stating it in linguistic terms, a number of students communicated the idea that “the packaging of topic, place, style, and people makes up the indexical direction of communication” (Bloomaert, 2010, p. 40). In other words, what language means, and thus what kind of English is considered correct, depends on the context. When asked about whether standard English was correct, 27 of 37 students (73%) answered yes, while also asserting that there are multiple correct Englishes. Typical responses from these students are included in Table 6. Several students elaborate on the co-existence of these multiple ideologies. Joy’s comment exemplifies these responses, I guess, like, depending where you come from, they speak different kinds of English, so I don’t think there’s really, like, an established Standard English, but we’re taught, like- we’re brought up to think that way.

17

Table 6 Responses demonstrating contextualized beliefs about correct English. Student

Multiple ideologies

Darcel

When I think, like, structured-wise, like, in an English class, I would say, “Yeah,” but then at the same time, I’m kind of, like, “Well, no.” Because, like, there’s different forms of English, and I don’t think there’s just, like, one correct way. Well I think it’s [Standard English] like the correct form. I think there’s a lot of other forms where it’s correct English, but I guess Standard English is in a way so, I guess I agree with that. Standard English would be the proper way probably for more professional use to whereas that’s correct. Other types of English would also be correct ‘cause that’s who they are and there’s nothing wrong with that.

Reyna

Kenny

Joy claims a critical stance by saying “I don’t think there’s really, like, an established Standard English.” She then acknowledges the larger social push toward a dominant language ideology by adding, “but we’re taught, like- we’re brought up to think that way.” Joy suggests that the socialization toward a language ideology occurs both in school (“we’re taught”) and at home (“we’re brought up”). By saying, “we’re brought up to think that way” Joy highlights her awareness that this is a way of thinking about language rather than an inherent truth. Taken as a whole, Joy’s comment shows an awareness of two potentially conflicting ways of thinking about language in her larger belief system. The survey data highlight that there is a range of beliefs, and the interview data shows the degree of nuance and understanding contained within that range. The following sections discuss the relevance of the findings and lay out implications for teaching students about language variation as well as implications for future research. 5. Implications and conclusion The findings of this study suggest important implications for teaching and research about language variation in schools. The discussion of findings is framed in terms of implications for teaching and implications for further research. 5.1. Implications for teaching This study highlights the complexity of students’ knowledge and beliefs about language variation. Supporting the findings of Kroskrity (2004), results of this study show that students hold beliefs about language that align with both dominant and critical ideologies. These beliefs are differentiated more strongly by parents’ beliefs about language than they are by racial/ethnic group membership or the language spoken at home. Students who indicate that their parents want them to speak Standard English are more likely to hold beliefs aligned with a Standard English ideology such as the belief that ‘Standard English is the correct form of English’ and that ‘to be successful students should only use standard English.’ Teachers need to be cautious of drawing conclusions about students’ language beliefs based on race or home language. Learning what students know and believe about language will require more careful assessment of prior understandings. Perhaps equally importantly for teaching about language variation, students who indicated awareness of their own use of multiple language varieties were more likely to express a belief that they are being judged by others, including teachers and peers, for their language use. A relationship between knowledge of language variation and a feeling of being judged creates potential problems for teachers. If teachers raise students’ awareness of language variation, and the associated stigma attached to certain language varieties, but do not also provide solutions for countering those language

18


prejudices, they run the risk of leaving students feeling more marginalized and more vulnerable. The examples of students, like Raul, Manny, and Ashley, who expressed a deep understanding of linguistic prejudice, but also a strong belief in the value of their own language varieties serve as evidence that high school students are capable of grappling with the nuances and complexities of social narratives surrounding language. Thus, while awareness of language variation may lead to a feeling of being judged, a more complete knowledge of language variation that draws on sociolinguistic facts may allow students to both acknowledge and contest that judgment. Knowledge of linguistic meta-language to describe the concepts students wrestled with would have allowed for more precise description. Students defaulted to “big words” and “slang” to describe language differences in peers. Without the meta-language to talk about language features students are hindered in their ability to describe the experiences they contend with every day. Other scholars exploring language variation in schools suggest the value of this meta-language (Godley & Loretto, 2013; Mallinson & Charity Hudley, 2010; Reaser & Adger, 2007). The complicated mix of language, vocabulary, and identity that students associate with terms such as “talking white-wash” or “talking ghetto” may be cleared up by a few salient terms, such as dialect and register, tied to linguistic concepts. Together the survey and interview data show that students are thinking about these issues and are experiencing the consequences of a linguistic hierarchy. Systematic teaching about language variation could allow students to organize these beliefs and experiences to help them navigate language use in multiple contexts. Critical language knowledge could also help students resist or work to change the existing language hierarchy. 5.2. Implications for research This study provides an initial glimpse into the way students understand language variation and the sources of those understandings. The sample of students in this study all came from extremely heterogeneous community, school, and classroom contexts. These students are immersed in a wide range of language practices in all aspects of their lives. This exposure to speakers of many language varieties likely influenced students’ beliefs, as Jhayme’s comment suggests, “a lot of them are my friends, and I think they’re pretty smart.” The field would benefit from additional studies that explore students’ beliefs about language variation in a range of contexts, including more homogeneous contexts. In particular, it would be good to know what students who are white SE speakers believe as they were the most underrepresented group in this study. Likewise, since parents’ beliefs about language were the most influential in distinguishing student’s language ideologies, further examination of the relationship between parent and student language ideologies is called for. Teachers, who need to address the concerns of parents as well as students, will be better equipped to teach about language variation if they can preemptively address the ideologies parents may communicate. To conclude, this study demonstrates the complex and nuanced beliefs students hold about language as well as the range of metaawareness students possess about those beliefs. More critically aware students are capable of reconciling the tensions in conflicting language ideologies as they distinguish their own beliefs from the societal messages around them. The growing body of research on teaching about language variation in schools will benefit from increased attention to students’ understandings of language. The complexity of students’ language beliefs reflects our linguistically complex society. Researchers interested in helping teachers address linguistic complexity in productive ways need to continue

to unpack the nuance in students’ understandings of language. The more accurately we understand students’ knowledge and beliefs, the better equipped we will be to tailor our teaching in ways that respond to students’ varied needs. Funding This work was supported by a Stanford University Diversity Dissertation Opportunity Grant and a Stanford Graduate School of Education Dissertation Support Grant. References Alim, H. S. (2005). Critical Language Awareness in the United States: Revisiting issues and revising pedagogies in a resegregated society. Educational Researcher, 34(7), 24–31. http://dx.doi.org/10.3102/0013189X034007024 Alim, H. S. (2016). Introducing raciolinguistics: Racing language and languaging race in hyperracial times. In H. S. Alim, J. R. Rickford, & A. F. Ball (Eds.), Raciolinguistics: How language shapes our ideas about race (1st ed., pp. 1–31). Oxford; New York: Oxford University Press. Ball, A. F., & Lardner, T. (1997). Dispositions toward language: Teacher constructs of knowledge and the Ann Arbor Black English Case. College Composition and Communication, 48(4), 469–485. http://dx.doi.org/10.2307/358453 Blake, R. (2016). Toward heterogeniety: A sociolinguistic perspective on the classification of Black people in the twenty-first century. In H. S. Alim, J. R. Rickford, & A. F. Ball (Eds.), Raciolinguistics: How language shapes our ideas about race (1st ed., pp. 153–169). Oxford; New York: Oxford University Press. Blommaert, J. (2010). The sociolinguistics of globalization (1st ed.). New York, NY: Cambridge University Press. Bowie, R. L., & Bond, C. L. (1994). Influencing future teachers’ attitudes toward Black English: Are we making a difference? Journal of Teacher Education, 45(2), 112–118. http://dx.doi.org/10.1177/0022487194045002005 Brown, D. W. (2006). Micro-level teaching strategies for linguistically diverse learners. Linguistics and Education, 17(2), 175–195. http://dx.doi.org/10.1016/j.linged. 2006.08.001 Bucholtz, M., & Hall, K. (2004). Language and identity. A Companion to Linguistic Anthropology, 1, 369–394. Byrnes, D. A., Kiger, G., & Lee Manning, M. (1997). Teachers’ attitudes about language diversity. Teaching and Teacher Education, 13(6), 637–644. http://dx.doi.org/10. 1016/S0742-051X(97)80006-6 Christensen, L. (2011). Finding voice: Learning about language and power. Voices From the Middle, 18(3), 9–17. Fairclough, N. (1992). Critical language awareness (1st ed.). London/New York: Routledge. Flores, N., & Rosa, J. (2015). Undoing appropriateness: Raciolinguistic ideologies and language diversity in education. Harvard Educational Review, 85(2), 149–171. Garcia, O., Flores, N., & Spotti, M. (2017). A critical poststructual perspective. In Oxford handbook of language and society. pp. 1–16. New York, NY: Oxford University Press. Godley, A. J., & Escher, A. (2012). Bidialectal African American Adolescents’ beliefs about spoken language expectations in English classrooms. Journal of Adolescent & Adult Literacy, 55(8), 704–713. http://dx.doi.org/10.1002/JAAL.00085 Godley, A. J., & Loretto, A. (2013). Fostering counter-narratives of race, language, and identity in an urban English classroom. Linguistics and Education, 24(3), 316–327. http://dx.doi.org/10.1016/j.linged.2013.03.006 Grossman, P. L. (1990). The making of a teacher: Teacher knowledge & teacher education. New York: Teachers College Press. Gutiérrez, K. D., Baquedano-López, P., & Tejeda, C. (1999). Rethinking diversity: Hybridity and hybrid language practices in the third space. Mind, Culture, and Activity, 6(4), 286–303. http://dx.doi.org/10.1080/10749039909524733 Irvine, J. (1989). When talk isn’t cheap: Language and political economy. American Ethnologist, 16(2), 248–267. http://dx.doi.org/10.1525/ae.1989.16.2.02a00040 Irvine, J., & Gal, S. (2000). Language ideology and linguistic differentiation. In P. V. Kroskrity (Ed.), Regimes of language: Ideologies, polities, and identities (Vol. School of American Research advanced seminar series (pp. 35–83). Santa Fe, NM: School of American Research Press. Janks, H. (2009). Literacy and power (1st ed.). New York: Routledge. Kroskrity, P. V. (2004). Language ideologies. In A. Duranti (Ed.), A companion to linguistic anthropology (pp. 496–517). Malden, MA: Blackwell Publishing. Kroskrity, P. V. (2010). Language ideologies – Evolving perspectives. Society and Language Use, 7, 192. Ladson-Billings, G. (1992). Culturally relevant teaching: The key to making multicultural education work. Research and Multicultural Education: From the Margins to the Mainstream, 106–121. Lee, C. D. (2007). Culture, literacy, & learning: Taking bloom in the midst of the whirlwind. New York, NY: Teachers College Press. Lippi-Green, R. (2012). English with an accent: Language, ideology and discrimination in the United States (2nd ed.). New York, NY: Routledge. Makoni, S., & Pennycook, A. (2007). Disinventing and reconstituting languages. Buffalo, NY: Multilingual Matters. Mallinson, C., & Charity Hudley, A. H. (2010). Communicating about communication: Multidisciplinary approaches to educating educators about language variation.

M. Metz / Linguistics and Education 45 (2018) 10–19 Language and Linguistics Compass, 4(4), 245–257. http://dx.doi.org/10.1111/j. 1749-818X.2010.00190.x Martínez, R. A. (2013). Reading the world in Spanglish: Hybrid language practices and ideological contestation in a sixth-grade English language arts classroom. Linguistics and Education, 24(3), 276–288. http://dx.doi.org/10.1016/j.linged. 2013.03.007 Metz, M. (2018). Pedagogical content knowledge for teaching critical language awareness: The importance of valuing student knowledge. Advance online publication, Urban Education, http://dx.doi.org/10.1177/0042085918756714 McKinney, C. (2017). Language and power in post-colonial schooling: Ideologies in practice. New York, NY: Routledge. Milroy, J. (2001). Language ideologies and the consequences of standardization. Journal of Sociolinguistics, 5(4), 530–555. Pennycook, A. (2017). Language policy and local practices. In N. Flores, M. Spotti, & O. Garcia (Eds.), Oxford handbook of langauge and society (pp. 125–140). New York, NY: Oxford University Press. Perry, T., & Delpit, L. D. (1998). The real ebonics debate: Power, language, and the education of African-American children. Boston: Beacon Press. Reaser, J., & Adger, C. (2007). Developing language awareness materials for nonlinguists: Lessons learned from the do you speak American? Curriculum development project. Language and Linguistics Compass, 1(3), 155–167. http:// dx.doi.org/10.1111/j.1749-818X.2007.00011.x Reinecke, J. E., & Tokimasa, A. (1934). The English dialect of Hawaii. American Speech, 48–58. Rickford, J. R. (1999). African American vernacular English: Features, evolution, educational implications. Hoboken, NJ: Wiley-Blackwell.

19

Rosa, J., & Burdick, C. (2017). Language ideologies. In The oxford handbook of language and society. pp. 103–123. New York, NY: Oxford University Press. Rosa, J., & Flores, N. (2017). Unsettling race and language: Toward a raciolinguistic perspective. Language in Society, 46(5), 621–647. http://dx.doi.org/10.1017/ S0047404517000562 Shulman, L. S. (1987). Knowledge and teaching. Harvard Educational Review, 57(1), 1–22. Smitherman, G. (1977). Talkin’ and testifyin’: The language of Black America. Detroit, MI: Wayne State University Press. Smitherman, G., & Villanueva, V. (2000). Language knowledge and awareness survey: Final research report. Urbana, IL: Conference on College Composition and Communication. Taylor, O. (1973). Teachers’ attitudes toward black and nonstandard English as measured by the language attitude scale. Language Attitudes: Current Trends and Prospects, 174–201. Williams, F., Whitehead, J. L., & Miller, L. (1972). Relations between language attitudes and teacher expectancy. American Educational Research Journal, 9(2), 263–277. http://dx.doi.org/10.2307/1161687 Woolard, K. A., & Schieffelin, B. B. (1994). Language ideology. Annual Review of Anthropology, 23, 55–82. Wortham, S. (2008). Linguistic anthropology of education. Annual Review of Anthropology, 37(1), 37–51. http://dx.doi.org/10.1146/annurev.anthro.36. 081406.094401 Young, V. A., & Martinez, A. Y. (2011). Code-meshing as World English: Pedagogy, policy, performance. Urbana, Ill: National Council of Teachers of English.