Language in Society 28, 203–223.

Printed in the United States of America

“Why be normal?”: Language and identity practices

in a community of nerd girls
M A RY B U C H O LT Z
Department of English
Texas A&M University
College Station, TX 77843–4227
[email protected]

ABSTRACT

The introduction of practice theory into sociolinguistics is an important re-

cent development in the field. The community of practice provides a useful
alternative to the speech-community model, which has limitations for lan-
guage and gender researchers in particular. As an ethnographic, activity-
based approach, the community of practice is of special value to researchers
in language and gender because of its compatibility with current theories of
identity. An extension of the community of practice allows identities to be
explained as the result of positive and negative identity practices rather than
as fixed social categories, as in the speech-community model. The frame-
work is used here to analyze the linguistic practices associated with an un-
examined social identity, the nerd, and to illustrate how members of a local
community of female nerds at a US high school negotiate gender and other
aspects of their identities through practice. (Community of practice, gender,
discourse analysis, identity, social construction, social practice, speech com-
munity, adolescents, nerds)*

In sociolinguistics, social theory is rooted in the concept of the speech commu-

nity. As a language-based unit of social analysis, the speech community has al-
lowed sociolinguists to demonstrate that many linguistic phenomena previously
relegated to the realm of free variation are in fact socially structured. Thus Labov
1966 showed that the linguistic heterogeneity of New York City can be quanti-
tatively analyzed as the patterning of a single speech community, despite differ-
ences in New Yorkers’ language use based on sociological variables such as age,
social class, and gender.
Nonetheless, because the concept of speech community is indigenous to so-
ciolinguistics, it is not connected to any larger social theory. This theoretical
isolation, along with the fact that the speech community defines the social world
in strictly (socio)linguistic terms, has meant that sociolinguistic theory has largely
stood apart from theoretical advances in related disciplines. Meanwhile, within
sociolinguistics, the concept of the speech community has been hotly contested
© 1999 Cambridge University Press 0047-4045/99 $9.50 203
 M A R Y B U C H O LT Z

and continually revised as researchers have uncovered the limitations of previous

definitions.
The speech community presents special difficulties for researchers in the so-
ciolinguistic subfield of language and gender. The disciplinary autonomy of theory
based on the speech community is unproblematic for traditional sociolinguistic
research, which uses social information to account for linguistic phenomena such
as sound change. But when sociolinguists reverse the direction of analysis – ask-
ing instead how linguistic data can illuminate the social world, as language and
gender researchers seek to do – then connections to social theory beyond linguis-
tics become imperative. Moreover, the speech community model, which was de-
signed to analyze sociolinguistic phenomena at a macro level, is often inappropriate
and inadequate for the kinds of questions currently being asked in language and
gender scholarship. Central among these is the question of identity: How do speak-
ers use language to project their identities as gendered beings? And how are
gender identities interwoven with other social parameters?
This article draws on a theory of community and identity that avoids the prob-
lems associated with the speech community model. The new framework, the
community of practice, emerges from practice theory, an approach that has
currency in such disciplines as sociology, anthropology, and education. The con-
nections of the community of practice to these recent developments in other fields
allow sociolinguists to offer more fully theorized social explanations than were
possible with the earlier model. In addition, the community of practice over-
comes many of the faults that sociolinguists have found with the speech commu-
nity, and it therefore has wide applicability to the field’s central questions. The
theory’s broad range of use is especially evident in language and gender studies –
because, unlike the speech community, the community of practice was intro-
duced into sociolinguistics specifically to address issues of gender.
In this article, I build on the theory of the community of practice to develop its
potential as an analytic tool for the sociolinguistic investigation of gendered iden-
tities. The framework is applied to a social identity, that of the nerd, which has
remained out of bounds in traditional sociolinguistic research based on the speech
community. This identity is analyzed within the community of practice frame-
work because only this concept permits us to draw on the linguistic and social
information necessary to understand the production of nerd identity. I argue that
nerd identity, contrary to popular perceptions, is not a stigma imposed by others,
but a purposefully chosen alternative to mainstream gender identities which is
achieved and maintained through language and other social practices.

LANGUAGE AND PRACTICE THEORY

The idea that the social world is best viewed as a set of practices is not new.
Praxis is a foundational concept of Marxism, and more recently Giddens 1979

204 Language in Society 28:2 (1999)

 “WHY BE NORMAL?”: LANGUAGE AND IDENTITY PRACTICES

has offered a practice-based account as a way out of the impasse created by

social structure, on the one hand, and personal agency, on the other. Given the
focus of practice theory on enduring social activity, it was perhaps inevitable
that it should soon come to view language as a central object of social analysis.
Outside linguistics, this perspective has been most fully articulated by the French
sociologists Pierre Bourdieu (1978, 1991) and Michel de Certeau (1984). Both
Bourdieu and Certeau understand language in relation to other social practices,
and both scholars view language as a social phenomenon, rather than merely as
an abstract formal system. As a consequence, they explicitly align their work
with the sociolinguistic enterprise, broadly conceived; the litany of familiar
names they invoke includes Joshua Fishman, Erving Goffman, William Labov,
and Emanuel Schegloff.
For Bourdieu, the starting point of practice is habitus, the set of dispositions
to act (e.g. speak, walk, read, or eat) in particular ways which are inculcated in each
individual through implicit and explicit socialization. These dispositions are linked
to particular social dimensions such as class and gender. Habitus is also tied to the
body via hexis, the individual’s habitual and socially meaningful embodied stances
and gestures, and through other aspects of physical self-presentation. Language is
merely one practice in which habitus is embedded, and through which the indi-
vidual becomes socially locatable to observers. Thus non-linguistic social prac-
tices and language should be approached in analogous ways.As Bourdieu observes
(1991:89),
Not only are linguistic features never clearly separated from the speaker’s whole
set of social properties (bodily hexis, physiognomy, cosmetics, clothing), but
phonological (or lexical, or any other) features are never clearly separated
from other levels of language; and the judgement which classifies a speech
form as “popular” or a person as “vulgar” is based, like all practical predica-
tion, on sets of indices which never impinge on consciousness in that form.
Bourdieu here offers two important methodological insights to sociolinguists:
first, that non-linguistic practices may carry important linguistic information (and
vice versa); and second, that a complete sociolinguistic analysis must examine
multiple levels of language simultaneously. Yet, as a theorist, Bourdieu is less
useful to sociolinguists, and especially to language and gender scholars. His in-
sistence on the unconsciousness of practice reflects a general attenuation of agency
within his theory. Although speakers are not bound by their habitus, which is
inflected by the particular context in which it occurs, the tendency is to act in
accordance with what has been naturalized as appropriate. Bourdieu sees the
individual, then, more as a product of social structure than as a free agent. Prac-
tice at the local level – especially linguistic practice, which is embedded in the
class habitus of the standard and the non-standard – is primarily in the business of
reproducing existing social arrangements.

Language in Society 28:2 (1999) 205

 M A R Y B U C H O LT Z

For Certeau, by contrast, the individual is much more agentive, because the
focus of investigation is subversion as well as reproduction of the social order.
But like Bourdieu, Certeau finds inspiration for his project in linguistics:
Our investigation . . . can use as its theoretical model the construction of
individual sentences with an established vocabulary and syntax. In linguis-
tics, “performance” and “competence” are different: the act of speaking (with
all the enunciative strategies that implies) is not reducible to a knowledge of
the language. By adopting the point of view of enunciation – which is the
subject of our study – we privilege the act of speaking; according to that point
of view, speaking operates within the field of a linguistic system; it effects an
appropriation, or reappropriation, of language by its speakers; it establishes a
present relative to a time and place; and it posits a contract with the
other (the interlocutor) in a network of places and relations. These four char-
acteristics of the speech act can be found in many other practices (walking,
cooking, etc.). (1984:xiii; original emphasis)
Certeau here makes the link between language and other social practices even
more explicit than did Bourdieu before him. Certeau sees all social practices,
both linguistic and non-linguistic, as similar in their social effects. But where
Bourdieu considers practice to be a reproduction of social structure, Certeau views
it as an appropriation, an act of agency. The point, then, is to understand how
culturally shared resources (such as language) are made to serve the specific
social needs of individuals. These needs may enforce the social status quo, but
they may just as easily challenge or revise it.
A third theory of practice has been developed within anthropology by Ortner
1996, who criticizes earlier scholarship on the grounds that it fails to take seri-
ously the practices of women. Making the female agent central in the project of
practice theory, Ortner constructs a framework that has room for both structure
and agency. Although language is not a guiding concept in Ortner’s work as it is
for Bourdieu and Certeau, she views structure itself as textual in nature – the
“field of a linguistic system”, in Certeau’s words – within which an individual act
of speaking operates. Thus a complete analysis of gender, and especially of lan-
guage and gender, cannot focus on texts alone. As Ortner argues (1996:2),
Studies of the ways in which some set of “texts” – media productions, literary
creations, medical writings, religious discourses, and so on – “constructs” cat-
egories, identities, or subject positions, are incomplete and misleading unless
they ask to what degree those texts successfully impose themselves on real
people (and which people) in real time. Similarly, studies of the ways in which
people resist, negotiate, or appropriate some feature of their world are also
inadequate and misleading without careful analysis of the cultural meanings
and structural arrangements that construct and constrain their “agency”, and
that limit the transformative potential of all such intentionalized activity.
206 Language in Society 28:2 (1999)
 “WHY BE NORMAL?”: LANGUAGE AND IDENTITY PRACTICES

The possibility – and the reality – of such unified analyses within language and
gender studies is offered by the community of practice framework. More than any
previous approach in sociolinguistics, the community of practice allows research-
ers to examine, in a theoretically adequate way, both the actions of individuals
and the structures that are thereby produced and reproduced, resisted and subverted.

G E N D E R , T H E S P E E C H C O M M U N I T Y,
AND THE COMMUNITY OF PRACTICE

Ortner’s introduction of a feminist perspective was a relatively late development

in practice theory in anthropology and sociology. Likewise, the theory of the
community of practice, which emerged from education (Lave 1988, Lave &
Wenger 1991, Wenger 1998), was not applied to gender until it was imported into
linguistics by Eckert & McConnell-Ginet in a highly influential survey article
(1992). As an alternative to the speech community – a central analytic tool of
sociolinguistics – the community of practice requires language and gender schol-
ars to rethink traditional notions of community, identity, and gender. However,
Eckert & McConnell-Ginet do not offer an explicit critique of the speech com-
munity; although that concept has been widely debated (see Hudson 1980, Wil-
liams 1992), its particular limitations for language and gender research have not
been systematically addressed. I suggest six ways in which the speech commu-
nity has been an inadequate model for work on language and gender:
(a) Its tendency to take language as central.
(b) Its emphasis on consensus as the organizing principle of community.
(c) Its preference for studying central members of the community over those
at the margins.
(d) Its focus on the group at the expense of individuals.
(e) Its view of identity as a set of static categories.
(f ) Its valorization of researchers’ interpretations over participants’ own un-
derstandings of their practices.

Language vs. social practice

The speech community has been defined in many ways, but every definition
posits language as a primary criterion of community. What is taken as shared may
be the linguistic system (Bloomfield 1933:42–56); or shared linguistic norms
(Labov 1972, Guy 1988); the pattern of variation (Milroy 1992); or only a set of
sociolinguistic norms (Romaine 1982). The emphasis may be less on the linguis-
tic system, and more on shared interactional settings and norms (Hymes 1974,
Dorian 1982, Silverstein 1996). But in every case, the focus remains on language.
Even many scholars who advocate a more interactional approach understand in-
teraction to be a preeminently linguistic concept. Other forms of mutual en-
gagement – that is, all non-linguistic aspects of social activity – are marginalized
or ignored.
Language in Society 28:2 (1999) 207
 M A R Y B U C H O LT Z

By recognizing practice – the social projects of participants – as the motivat-

ing context for linguistic interaction, the theory of the community of practice
makes activity much more central to sociolinguistic analysis. Just as importantly,
whereas the speech community model understands language as fundamentally
disembodied – as detachable from the physicality of speakers – the community of
practice quite literally reincorporates language into the physical self. In this re-
gard, it echoes Bourdieu’s concept of hexis – a crucial connection for feminist
researchers, for whom the specificity of the gendered body is a theoretical start-
ing point.

Consensus vs. conflict

Another aspect of the traditional model that has received a great deal of criticism
is the idea that the speech community is constituted around shared sociolinguistic
norms. This definition was first proposed by Labov 1972, and was taken up by
many subsequent researchers. The postulate that speakers agree on and uphold
certain linguistic forms as normative, regardless of differences in social back-
ground, assumes a consensus model of society that is at odds with a long-standing
tradition of social theory. Several sociolinguists have critiqued the Labovian def-
inition of the speech community on these grounds (e.g. Rickford 1986, Milroy
1992). Moreover, the invocation of “norms” obscures the fact that these are suc-
cessfully imposed ideologies favoring the interests of the powerful (Bourdieu
1991). This arrangement has long been recognized by scholars of language and
gender who have worked to combat views of women’s language as deficient in
comparison to men’s (see Cameron 1992:42 ff.)

Central vs. marginal members

The language of norms also presumes that some members of the speech commu-
nity are central and others are marginal, and that it is the central members who are
of interest. To be sure, the structured heterogeneity of the speech community
improves on earlier models by recognizing the existence and systematicity of
heterogeneity; however, speakers who do not share the same norms (for example,
because they are recent immigrants or transplants from other regions) are ex-
cluded from the community.1 Thus, despite the model’s emphasis on heteroge-
neity, the focus is in fact on what speakers share. Marginal members rarely enter
the analysis, and when they do, they remain at the margins; their linguistic prac-
tices are used primarily to demonstrate how they fall short of central member-
ship. Language and gender researchers are acutely aware of the problems with
this approach. Because women may be defined, implicitly or explicitly, as mar-
ginal to the vernacular speech community, they may be underrepresented or sim-
ply misrepresented (cf. Morgan 1999). Speakers whose identities differ from those
of the wider community – especially those whose gender identities do not con-
form to community norms – are likewise omitted or obscured in research within
this paradigm.
208 Language in Society 28:2 (1999)
 “WHY BE NORMAL?”: LANGUAGE AND IDENTITY PRACTICES

The expectation of consensus in speech community norms also requires that

the system be closed to outside influence. The possibility of interaction between
speech communities is not important in the model.2 Hence researchers seek same-
ness, not difference; difference (e.g. in language use) is contained by interpreting
it as sameness at an underlying level (e.g. in shared sociolinguistic norms). With
this emphasis on analysis of the group as an autonomous system, phenomena
resulting from linguistic and cultural contact (Pratt 1987) may be overlooked.
For example, the focus on the internal workings of the speech community does
not accommodate investigations of gendered interaction across cultural groups.

Groups vs. individuals

Related to the problem of homogeneity in the speech community model is its
privileging of the group over the individual as the unit of analysis. In such an
approach, the role of the individual is merely to instantiate the practices of the
group. Individual actions result less from choice and agency than from a social
order that impinges on individuals from above. The traditional model’s strong
preference for structure over agency means that individual variation, or style, is
interpreted as the mechanical outcome of structural forces such as situational
norms. A more agentive view locates style in personal choices concerning self-
presentation (Johnstone 1995, 1996, Johnstone & Bean 1997). This perspective,
which also admits structural constraints on the individual, is well suited for gen-
der studies, given the field’s longtime recognition that individuals make purpose-
ful choices in the face of the limitations imposed on them by social structures. As
Ortner notes above, one of the benefits of practice theory is its ability to cope with
both aspects of women’s (and men’s) lives.

Identity categories vs. identity practices

The structural perspective is a static perspective, one in which the social order
remains largely unaltered. Changes in the practices of its inhabitants have the
effect of keeping the system in equilibrium. Nowhere is this more evident than in
the speech community model’s implicit theory of identity: Individuals are viewed
as occupying particular social identities throughout their lives by virtue of their
position in the social structure. Such an analysis is particularly problematic for
researchers of language and gender. The concept of identity is central to gender-
oriented research, but the version offered by the speech community framework
contradicts basic insights of recent feminist theory. Contemporary feminists view
identities as fluid, not frozen; they note that, although identities link individuals
to particular social groups, such links are not predetermined. Instead, identities
emerge in practice, through the combined effects of structure and agency. Indi-
viduals engage in multiple identity practices simultaneously, and they are able to
move from one identity to another. This process is not entirely unconstrained;
speakers may end up reproducing hegemonic identities more often than resisting
Language in Society 28:2 (1999) 209
 M A R Y B U C H O LT Z

them, as suggested by Holmes 1997. It is also important, however, to call atten-

tion to the previously unacknowledged flexibility of identity formation.
Top-down vs. bottom-up
For the specificity of identity to become visible, it must be examined from the
point of view of the individuals who enact it. Such a vantage point is not avail-
able within the speech community model, which privileges the analyst’s inter-
pretations over those of participants. Indeed, the speech community itself is an
analytic construct which may fail to correspond to its putative members’ own
perceptions. Nonetheless, many analyses are carried out under the belief that
the linguist has access to elements of speakers’ reality that are not available to
the speakers themselves.
An alternative to this top-down paradigm is ethnography, an approach that
is participant- rather than analyst-driven. Where the speech community frame-
work is skeptical of speakers’ perspectives on their own practices, ethnography
makes local interpretations central to the analysis. Gender does not have the same
meanings across space and time, but is instead a local production, realized dif-
ferently by different members of a community; thus an ethnographic orientation
yields particularly fruitful results for language and gender research.

NERDS, GENDER, AND THE COMMUNITY OF PRACTICE

The inadequacies of the speech community model for scholars of language and
gender are overcome in the theory of the community of practice as articulated by
Eckert & McConnell-Ginet 1992, 1995.3 Rather than investing language with a
special analytic status, the community of practice framework considers language
as one of many social practices in which participants engage. By defining the
community as a group of people oriented to the same practice, though not nec-
essarily in the same way, the community of practice model treats difference and
conflict, not uniformity and consensus, as the ordinary state of affairs. The in-
herent heterogeneity of the community of practice also brings marginal members
to the forefront of analysis. One reason for this shift to the margins is that some
peripheral members are recognized as novices, as in Lave & Wenger’s original
formulation (1991). More importantly, however, the community of practice, un-
like the speech community, may be constituted around any social or linguistic
practice, no matter how marginal from the perspective of the traditional speech
community. Likewise, by focusing on individuals as well as groups, the theory of
the community of practice integrates structure with agency. And because identi-
ties are rooted in actions rather than categories, the community of practice model
can capture the multiplicity of identities at work in specific speech situations
more fully than is possible within the speech community framework. Such nu-
anced description is also facilitated by Eckert & McConnell-Ginet’s intrinsically
ethnographic approach to language and gender research. The remainder of this
article draws on the above characteristics of the community of practice to dem-
210 Language in Society 28:2 (1999)
 “WHY BE NORMAL?”: LANGUAGE AND IDENTITY PRACTICES

onstrate the theory’s utility in the investigation of an understudied social identity

as it emerges locally in a high-school setting.
Eckert 1989a offers an account of the social organization of a typical suburban
US high school. She found that students’ social worlds and identities were de-
fined by two polar opposites: the Jocks (overachieving students who oriented to
middle-class values) and the Burnouts (underachieving students who were bound
for work, rather than college, at the end of their high-school careers). Yet the
dichotomy that separated these students also united them in what can be under-
stood as a single community of practice, since the ultimate goal of members of
both groups was to be cool. The difference lay in how each group defined coolness.
Not all high-school students, however, share the Jocks’ and Burnouts’ pre-
occupation with coolness. A third group, the nerds, defines itself largely in op-
position to “cool” students – whether Jocks, Burnouts, or any other social identity.
Nerds stand as the antithesis of all these groups, a situation that Eckert succinctly
captures in her observation, “If a Jock is the opposite of a Burnout, a nerd is the
opposite of both” (1989a:48). But despite the structural significance of the nerd
in the organization of youth identities, few researchers have examined its impli-
cations, and those who have tried have fallen far short of the mark in their analy-
ses. Thus the sociologist David Kinney, in a rare study of nerds (1993), argues
that, in order to succeed socially, nerds must undergo a process of “recovery of
identity” that involves broadening one’s friendship network, participating in ex-
tracurricular activities, and heterosexual dating: In short, they must become Jocks.
Another scholarly treatment (Tolone & Tieman 1990) investigates the drug use of
nerds in an article subtitled “Are loners deviant?” – in other words, are nerds
really Burnouts?
What both studies overlook is that being a nerd is not about being a failed
Burnout or an inadequate Jock. It is about rejecting both Jockness and Burnout-
ness, and all the other forms of coolness that youth identities take. Although
previous researchers maintain that nerd identity is invalid or deficient, in fact
nerds, like Jocks and Burnouts, to a great extent consciously choose and display
their identities through language and other social practices. And where other
scholars tend to equate nerdiness with social death, I propose that nerds in US
high schools are not socially isolated misfits, but competent members of a dis-
tinctive and oppositionally defined community of practice. Nerdiness is an espe-
cially valuable resource for girls in the gendered world of the US high school.
Elsewhere (Bucholtz 1998) I describe the social identity of the nerd and detail
the phonological, syntactic, lexical, and discourse practices through which nerd
identity is linguistically indexed. Here I propose a framework for the classifica-
tion of such practices. These linguistic indices are of two kinds: Negative iden-
tity practices are those that individuals employ to distance themselves from a
rejected identity, while positive identity practices are those in which indi-
viduals engage in order actively to construct a chosen identity. In other words,
negative identity practices define what their users are not, and hence emphasize
Language in Society 28:2 (1999) 211
 M A R Y B U C H O LT Z

TABLE 1. Linguistic identity practices of nerds at Bay City High School.

Linguistic Level Negative Identity Practices Positive Identity Practices

Phonology Lesser fronting of (uw) and

(ow) a
Phonology Resistance to colloquial Employment of superstan-
phonological processes dard and hypercorrect
such as vowel reduction, phonological forms (e.g.
consonant-cluster simplifi- spelling pronunciations)
cation, and contraction
Syntax Avoidance of nonstandard Adherence to standard and
syntactic forms superstandard syntactic
forms
Lexicon Avoidance of current slang Employment of lexical items
associated with the formal
register (e.g. Greco-
Latinate forms)
Discourse Orientation to language form
(e.g. punning, parody,
word coinage)

a
In Bucholtz 1998 I offer a fuller discussion of the phonological and syntactic patterns of nerds. The
present article focuses primarily on lexicon and on discursive identity practices. The variables (uw)
and (ow) are part of a vowel shift that is characteristic of California teenagers (Hinton et al. 1987,
Luthin 1987). It is stereotypically associated with trendy and cool youth identities.

identity as an intergroup phenomenon; positive identity practices define what

their users are, and thus emphasize the intragroup aspects of social identity. The
linguistic identity practices of nerds in the present study are shown in Table 1.
The negative identity practices listed here work to disassociate nerds from
non-nerds, and especially from cool teenagers. Each of these practices, which
mark nerdy teenagers as avowedly uncool, constitutes a refusal to engage in the
pursuit of coolness that consumes other students. Meanwhile, all the positive
identity practices listed contribute to the speaker’s construction of an intelligent
self – a primary value of nerd identity. These linguistic practices also have non-
linguistic counterparts in positive and negative identity practices of other kinds
(see below).
But linguistic practices can often reveal important social information that is
not available from the examination of other community practices alone. For ex-
ample, Eckert & McConnell-Ginet 1995 apply the theory of the community of
practice to Eckert’s study of Jocks and Burnouts. Linguistic analysis revealed
that the two groups were participating at different rates in the Northern Cities
Vowel Shift, with the most innovative vowels being those used by the “Burned-
Out Burnout girls”, the most extreme adherents to this social identity. Eckert &
212 Language in Society 28:2 (1999)
 “WHY BE NORMAL?”: LANGUAGE AND IDENTITY PRACTICES

McConnell-Ginet’s finding runs counter to the sociolinguistic tenet that “in sta-
ble variables, women use fewer non-standard variants than men of the same so-
cial class and age under the same circumstances” (Chambers 1995:112).4 The
researchers argue that the vowels employed by the Burned-Out Burnout girls are
resources through which they construct their identities as tough and streetwise;
unlike the boys, who can display their toughness through physical confronta-
tions, female Burnouts must index their identities semiotically, because fighting
is viewed as inappropriate for girls. Thus Burnout girls and boys share an orien-
tation toward toughness in their community of practice, but the practice of tough-
ness is achieved in different ways by each gender. By viewing language as
equivalent to other social practices like fighting, Eckert & McConnell-Ginet are
able to explain the ethnographic meaning of the Burnout girls’ vowel systems,
and to show how, as symbolic capital (Bourdieu 1978), language can acquire the
empowering authority of physical force itself.
Nerds, of course, attain empowerment in very different ways than either Burn-
outs or Jocks. One of the primary ways they differ from these other, more trend-
conscious groups is through the high value they place on individuality. Compared
to both Jocks and Burnouts – who must toe the subcultural line in dress, language,
friendship choices, and other social practices – nerds are somewhat less con-
strained by peer-group sanctions.
For girls, nerd identity also offers an alternative to the pressures of hegemonic
femininity – an ideological construct that is at best incompatible with, and at
worst hostile to, female intellectual ability. Nerd girls’ conscious opposition to
this ideology is evident in every aspect of their lives, from language to hexis to
other aspects of self-presentation. Where cool girls aim for either cuteness or
sophistication in their personal style, nerd girls aim for silliness. Cool girls play
soccer or basketball; nerd girls play badminton. Cool girls read fashion maga-
zines; nerd girls read novels. Cool girls wear tight T-shirts, and either very tight
or very baggy jeans; nerd girls wear shirts and jeans that are neither tight nor
extremely baggy. Cool girls wear pastels or dark tones; nerd girls wear bright
primary colors. But these practices are specific to individuals; they are engaged
in by particular nerd girls, not all of them.
The community of practice model accommodates the individuality that is par-
amount in the nerd social identity, without overlooking the strong community ties
that unify the nerd girls in this study. The community of practice also allows us to
look at nerd girls in the same way that Eckert & McConnell-Ginet 1999 view the
Burnout girls: as speakers and social actors, as individuals and members of
communities, and as both resisting and responding to cultural ideologies of gender.

IDENTITY PRACTICES IN A LOCAL NERD COMMUNITY

To illustrate the value of the community of practice framework, I will focus on a

single social group that displays the nerd social identity. Nerds at the high school
Language in Society 28:2 (1999) 213
 M A R Y B U C H O LT Z

in my study constitute a single community insofar as they engage in shared prac-

tices, but this identity is divided into particular social groups whose members
associate primarily with one another, and these groups form their own commu-
nities of practice. In communities of practice, unlike speech communities, the
boundaries are determined not externally by linguists, but internally through eth-
nographically specific social meanings of language use. As suggested above,
ethnographic methods therefore become crucial to the investigation of commu-
nities of practice.
The ethnographic fieldwork from which the data are taken was carried out
during the 1994–95 academic year at a California high school that I call Bay City
High. The social group of nerd girls that is the focus of this discussion is a small,
cohesive friendship group that comprises four central members – Fred, Bob, Kate,
and Loden – and two peripheral members, Carrie and Ada. (Ada does not appear
in the data that follow.) All the girls are European American except Ada, who is
Asian American. The same group also formed a club, which I will call the Ran-
dom Reigns Supreme Club.5
Random Reigns Supreme is more properly described as an anti-club, which is
in keeping with the counter-hegemonic orientation of nerd identity. It was created
by members in order to celebrate their own preferences, from Sesame Street to
cows to Mr. Salty the pretzel man. Members emphasize the “randomness” of the
club’s structure. It is not organized around shared preferences; instead, any indi-
vidual’s preferences can be part of the club’s de facto charter, and all six members
are co-presidents. This structure contrasts with the corporate focus and hierar-
chical structure of most school clubs, which bring together people who are other-
wise unconnected to perform a shared activity (Eckert 1989a). The Random Reigns
Supreme Club centers around members’ daily practices, not specialized activi-
ties. It has no goals, no ongoing projects, and no official meetings. Nevertheless,
members proudly take their place among the corporate clubs in the pages of the
school’s yearbook. The girls’ insistence on being photographed for the yearbook
has a subversive quality: The photo publicly documents the existence of this
otherwise little-recognized friendship group, and demands its institutional legit-
imacy on par with the French Club, the Backpacking Club, and other activity-
based organizations. Like their yearbook photograph, the language used by the
girls not only marks their nerd identity but also expresses their separation from
outsiders. As shown by the following examples (taken from a single interaction),
the details of interaction are important and contested resources in defining a
shared oppositional nerd identity within the club’s community of practice.

Positive identity practices

As indicated above, many positive identity practices in which nerds engage con-
tribute to the display of intelligence. The community value placed on intelligence
is reflected in non-linguistic identity practices oriented to the world of school,
books, and knowledge. This orientation is amply illustrated in the following.6
214 Language in Society 28:2 (1999)
 “WHY BE NORMAL?”: LANGUAGE AND IDENTITY PRACTICES

(1) 1 Carrie: Where where do those seeds come from?

2 ^points to her bagel&
3 ^laughter&
4 Bob: [Poppies. ]
5 Fred: [Sesame plants.]
6 Carrie: {But what do they look like?} ^high pitch&
7 Fred: I have no idea. hh
8 Bob: Sesame:.
9 Carrie: [Is anybody– h]
10 Fred: Ask me (.) [tomorrow. ]
11 I’ll look it up for you. h
12 Carrie: h Is anybody here knowledgeable about (.)
13 the seeds on top of bagels?/
14 Fred: /Sesame.
15 Bob: They’re sesame?
16 They’re not sunfl– ?
17 No,
18 of course they’re not sunflower.
19 Loden: Yeah,
20 [What kind of seeds are– ]
21 Carrie: [Because sunflower are those whopping ones?]
22 Bob: [Yeah.
23 Yeah.
24 I know. ]
25 ^laughter&

Carrie’s question in line 1 creates the conditions for intellectual display. Although
the humor of the question is acknowledged through laughter (line 2), it receives
immediate, serious uptake from two participants, Bob and Fred (lines 4–5). Car-
rie’s subsequent question (line 6), however, forces an admission of ignorance
from Fred (line 7).
Because knowledge is symbolic capital within the nerd community of prac-
tice, Fred’s admission results in some loss of face. She recovers from this (minor)
social setback by invoking the authority of a reference book (I’ll look it up for
you, line 11). In this way Fred can safely assure her interlocutor that, although she
does not yet know the answer, she soon will. She is also able to one-up Bob, who
has misidentified the bagel seeds (line 4) and continues to show some skepticism
about Fred’s classification of them (Sesame:, line 8). Fred tracks this indirect
challenge for five lines, through her own turn and Carrie’s next question; rather
than continuing to participate in the series of adjacency pairs that Carrie has
initiated (lines 12–13), she responds to Bob (line 14). Fred thus succeeds in dis-
playing both actual knowledge, about the type of seeds under discussion, and
potential knowledge, about the appearance of sesame plants.
Claims to knowledge are, however, often disputed in this community of prac-
tice. After Bob provides an incorrect answer to Carrie and receives a correction
from Fred, she continues to exhibit doubt about Fred’s knowledge (line 15). She
offers a second incorrect identification of the seeds in line 16, but this time she
interrupts herself and self-corrects (lines 17–18), in an effort to prevent further
other-correction. She does not succeed, however; and when Carrie explains why
Language in Society 28:2 (1999) 215
 M A R Y B U C H O LT Z

Bob is mistaken, the latter overlaps with her, offering three quick acknowledg-
ments that are designed to cut off Carrie’s turn (lines 22–24).
This passage shows several deviations from the preference organization of
repair in conversation (Schegloff et al. 1977), according to which self-initiation
and self-repair are preferred over initiation and repair by another. Bob twice
initiates dispreferred repairs of Fred’s turns (lines 8, 15), and she even begins to
carry out the repair itself in line 16. When Bob initiates a repair of her own
utterance through self-interruption in the same line, Carrie performs the repair
despite Bob’s efforts to prevent her from doing so (lines 21–24). The frequent
apparent violations of repair organization suggest that, in this community of prac-
tice, self-repair is preferred only by the speaker; the listener’s positive face (the
desire to be viewed as intelligent) wars against and often overrides consideration
of the speaker’s negative face (the desire not to be viewed as unintelligent).
Bob’s loss of face in ex. 1 leads her, in ex. 2, to initiate a new conversational
direction:

(2) 26 Bob: They come from trees.

27 They have big trees and they just
28 [ra:in down seeds]
29 [^laughter& ]
30 Carrie: [No they don’t. ]
31 Uh uh.
32 Why would little tiny seeds [come from– ]
33 Fred: [{into baskets.}] ^smiling quality&
34 Ye:p,
35 [({I’ve been there.})] ^smiling quality&
36 Carrie: [No:. ]
37 Loden: [No:. ]
38 Bob: [[Little tiny leaves come from trees, ]]
39 Fred: [[And the whole culture’s built around it,]]
40 like in: some countries,
41 All they do is like the women come out and they have ba(h)skets on
42 th(h)eir h(h)eads and they st(h)and under a [tree,]

Bob jokingly provides an authoritative answer to Carrie’s question (lines 26–28)

and thereby skillfully shifts attention from her own lack of knowledge to Carrie’s.
Fred eagerly joins in with the parody of scientific discourse, amplifying on the
theme while supplying invented anthropological details that invoke the didactic
style of a typical high-school classroom or public television documentary (33–
35, 39– 42). Such teasing episodes are frequent in this friendship group. But more
importantly, this exchange is a collaborative performance of nerd identity: The
participants collude in sustaining the frame of an intellectual debate, even as
laughter keys the talk as play. Nerd identities are here jointly constructed and
displayed.
In ex. 3, Carrie – who up to this point has mostly provided opportunities for
others to display their nerd identities, rather than participating herself (but see
below) – shifts the topic, which she sustains for the rest of the interaction:
216 Language in Society 28:2 (1999)
 “WHY BE NORMAL?”: LANGUAGE AND IDENTITY PRACTICES

(3) 43 Carrie: [My– ]

44 You sound like my crusty king,
45 I’m writing this (.) poem because I have to like incorporate these
46 words into a poem, and it’s all about–
47 ^interruption, lines omitted&
48 Fred: So what about this king?

Carrie’s discussion of a class assignment returns to a central value of nerdiness:

school. The topic is sustained for 56 lines and 26 turns; and although it is inter-
rupted immediately after Carrie introduces it (line 47), Fred prompts her to return
to the subject several minutes later (line 48). Carrie’s enthusiastic description of
her poem – and the eager participation of others in this topic – is rare among
students with cool social identities, but it is quite common among nerds, for
whom academic pursuits are a central resource for identity practices.
At the same time, however, Carrie’s selection of subject matter for her poem,
with its mildly scatological – or at least “gross” theme (line 80) – is playfully
subversive of school values and emphatically counter to traditional feminine top-
ics, as ex. 4 illustrates:

(4) 49 Carrie: He’s like (.) has this (.) castle,

50 (xxx: Is he xxx king?)
51 Carrie: No–
52 Yeah,
53 he is.
54 Loden: hh
55 Carrie: He has this–
56 {He has this castle right?
57 except it’s all crusty,}
58 ^rustling of lunch bag, clanging of aluminum can&
59 (Fred: Uh huh.)
60 Carrie: And so he lives on a boat [in the moat.]
61 Bob: [A crusty– ]
62 ^Fred crushes her aluminum can&
63 Kate: Who:a!
64 ^quiet laughter&
65 Bob: Is it really [crusty?]
66 Carrie: [He’s– ]
67 And so like the– like because– the people are trying to convince
68 him that like he should stay in the castle and he’s all,
69 {“No, it’s crusty!”} ^high pitch, tensed vocal cords&
70 [^laughter& ]
71 Carrie: [{“I’m in the moat!”}] ^high pitch, quiet&
72 right,
73 Bob: What’s wrong with [crusty castles?]
74 Carrie: [And so– ]
75 Well,
76 Would [you want to live ]5
77 Kate: [Crusty (castles). ]
78 Carrie: 5in a castle full of crust?
79 {[é@é]} ^noise of disgust and disapproval&
80 Kate: [How gross.]
81 Bob: [I mi:ght. ]
82 Carrie: Huh?

Language in Society 28:2 (1999) 217

 M A R Y B U C H O LT Z

Bob here enters into the unfeminine spirit of Carrie’s narrative, even outdoing
Carrie with her repeated insistence on her own immunity from “gross” subjects
like crustiness (lines 73, 81). A competitive tone is also evident in the multiple
challenges she issues to Carrie throughout the latter’s narrative (lines 65, 73). As
questions, these challenges echo Carrie’s earlier questions (lines 1, 6, 12–13); but
whereas Carrie’s appeared to be genuine information-seeking questions, Bob’s
are not. Carrie’s recognition of this fact is shown by her failure to respond at all
to the first question, and by her answering the second question with an equally
challenging question of her own (Would you want to live in a castle full of crust?,
lines 76, 78). Bob’s face-threatening response (I mi:ght, line 81) perpetuates the
jocular-combative tone. In ex. 5, however, this combativeness becomes not a
shared resource for joint identity construction, but a marker of social division.
The positive identity practices that dominate in the earlier part of the interaction
are replaced by negative identity practices, as community members experience a
threat not only to their face but also to their identities.

Negative identity practices

Example 5 is a continuation of Bob’s face-threatening questions to Carrie. This
final series of questions is unified through a shared template (like 1 adj 1 crust);
their syntactic similarity emphasizes that they are designed as a series, and it thus
produces an effect of unremitting interrogation.

(5) 83 Bob: What kind of crust?

84 Like,
85 bread crust?
86 Carrie: Like
87 Bob: Like [eye crust? ]
88 Carrie: [crusty crust.]
89 Like {boo:ts y} ^high pitch, tensed vocal cords&
90 crust.
91 ^laughter&
92 Bob: Oh.
93 Well,
94 Maybe if it’s bootsy,
95 I don’t know.
96 Fred: {Boot[sy! ]} ^falsetto, sing-song&
97 Kate: [^coughs&]
98 ^laughter&

These questions display Bob’s nerd identity through her use of puns on the word
crust (lines 85, 87). Punning, as a discourse practice that orients to linguistic
form, is characteristic of nerds’ discourse style (see Table 1). Carrie’s refusal (line
88) to participate in Bob’s punning thus constitutes a negative identity practice –
one which, moreover, indexes a rejection of nerd identity as it has been con-
structed through preceding interactional practices. The refusal is made more
evident by her exploitation (lines 86, 88–90) of Bob’s syntactic template. By
conforming to the syntactic form of Bob’s turn, while failing to conform to the
218 Language in Society 28:2 (1999)
 “WHY BE NORMAL?”: LANGUAGE AND IDENTITY PRACTICES

discourse practice of punning, Carrie separates herself from Bob at a point when
the latter is fully engaged in nerdy identity practices.
This analysis is confirmed by Carrie’s choice of upgraded adjective in line
89. Bootsy is a slang term with a negative evaluative sense; it is not used by
other members of the Random Reigns Supreme Club. The introduction of youth
slang into a group that explicitly rejects such linguistic forms is part of a strongly
negative identity practice, and the reactions of Carrie’s interlocutors are corre-
spondingly negative: Bob’s response (lines 92–95) jokingly concedes the point,
while underscoring that Carrie has violated the rules of nerdy argument by
appealing to the authority of cool youth culture. Fred’s mocking repetition of
the term (line 96) demonstrates that the use of slang is itself worthy of com-
ment. With Carrie’s narrative entirely derailed – it never becomes clear how it
is connected to the earlier discussion – she soon afterward moves away from
the group.
The complex interaction presented above reveals Carrie’s peripheral status in
this community of practice. As a non-core member, she moves between friend-
ship groups – in fact, the interaction occurred when Carrie approached the core
group in the middle of lunch period. Carrie’s social flexibility has made her a
cultural and linguistic broker for the Random Reigns Supreme Club, whose mem-
bers become aware of current youth slang in large part through contact with her.
Hence many slang terms that circulate widely in the “cool” groups are labeled by
club members as “Carrie words”.
Yet Carrie also demonstrates her ability and willingness to participate in the
group’s positive identity practices. She does so most obviously by engaging in
sound play in recounting her poem (crusty king, line 44; a boat in the moat, line
60). More significant, though, is the subtle shift in her speech practices at the
beginning of the interaction. Thus Carrie’s question Is anybody here knowledge-
able about (.) the seeds on top of bagels? (lines 12–13) draws on the formal
register through her choice of the word knowledgeable. Among nerds, this reg-
ister projects a speaker’s persona as smart and highly educated. But the use of the
formal register is strategic, not a mechanical result of membership in a particular
social category. This point is supported by the fact that Carrie employs the nerd
identity practice only after she asks two related questions in colloquial register
(lines 1, 6). Her unwillingness to overlap her turn with Fred’s (lines 9, 10) further
suggests that the question is a performance of nerdiness, not just a manifestation
of it; she does not produce her utterance until she is assured of an attentive audi-
ence. That is, Carrie is simultaneously displaying and commenting on nerd prac-
tice – showing her awareness of nerdy linguistic forms, and announcing her
willingness to enter a nerdy interactional space by carefully gauging her utter-
ance to match the group’s practices. Thus Carrie’s performance of nerdiness places
her within the community of practice; but her use of slang, as the other members
are quick to let her know, moves her outside it. Such adjustments at interactional
boundaries may reflect adjustments at community boundaries.
Language in Society 28:2 (1999) 219
 M A R Y B U C H O LT Z

CONCLUSION
Because all the participants in the above exchange are middle-class European
American girls from the same California city, the traditional sociolinguistic per-
spective would classify them unproblematically as members of the same speech
community. Such an analysis would overlook the details of greatest interest to
language and gender researchers: the performances of identity, and the struggles
over it, which are achieved through language. However, by viewing the inter-
action as the product of a community of practice, we can avoid this problem, as
well as others associated with the speech community model.
The ethnographic method brings into view the social meanings with which
participants invest their practices. These meanings emerge on the ground in local
contexts; thus what it means to display academic knowledge, or to use slang,
depends not on fixed identity categories but on where one is standing. Nor do
participants necessarily agree on the meanings of their actions; nerdiness, like all
identities, is a contested domain in which speakers struggle both over control of
shared values, via positive identity practices (Who’s better at being a nerd?), and
over control of identity itself, via negative identity practices (Who counts as a
nerd?). Such conflicts reveal the heterogeneity of membership in the community
of practice – its constitution through the work of central and peripheral members
alike. In this project, the interactional choices of specific individuals matter. Thus
Carrie’s identity is on display – and at risk – in a way that Loden’s, for example,
is not. These actions must be seen as choices, not as the outputs of interactional
algorithms. While some practices reproduce the existing local social structure (as
does Carrie’s use of the formal register), others undermine it (e.g. her use of
slang). Likewise, some nerdy practices (such as being good students) comply
with the larger social order, while others (such as rejecting femininity) resist it.
Linguistic practices, moreover, have no special status in this process. Instead,
they work in conjunction with other social practices to produce meanings and
identities. Bob’s interactional work to distance herself from hegemonic feminin-
ity, for instance, is part of her overall participation in anti-feminine practices and
her non-participation in feminine practices, as evidenced also by her physical
self-presentation.
For sociolinguists, the community of practice represents an improvement over
the speech community in that it addresses itself to both the social and the linguis-
tic aspects of the discipline. As a well-grounded framework with currency in a
number of fields, practice theory in general, in particular the community of prac-
tice, revitalizes social theory within sociolinguistics. What is more, it does so at
a sufficiently general level to accommodate multiple dimensions of social analy-
sis – including both structure and agency, both ideology and identity, both norms
and interactions. The community of practice also provides an avenue for a more
complete sociolinguistic investigation of identity. Although introduced for gender-
based research, the community of practice has never been restricted to the analy-
sis of a single element of identity. Indeed, it lends itself to the simultaneous
220 Language in Society 28:2 (1999)
 “WHY BE NORMAL?”: LANGUAGE AND IDENTITY PRACTICES

investigation of multiple aspects of the self, from those at the macro level – like
gender, ethnicity, and class – to micro-identities like Jocks, Burnouts, or nerds.
The framework also allows for the study of interaction between levels of identity.
The concepts of positive and negative identity practices, as proposed in this ar-
ticle, are intended as one way to develop the potential of the community of prac-
tice in this arena.
In addition to its benefits for social analysis, the community of practice offers
an integrated approach to linguistic analysis. By understanding all socially mean-
ingful language use as practices tied to various communities, the model enables
researchers to provide more complete linguistic descriptions – along with social
explanations – of particular social groups. Moreover, the community of practice
provides a way to bring qualitative and quantitative research closer together.
Because both kinds of linguistic data emerge from practice, both can be included
in a single analysis. This richly contextualized approach to both language and
society is one of the great strengths of the community of practice as a sociolin-
guistic framework.
The community of practice, having revolutionized the field of language and
gender almost as soon as it was first proposed, enables researchers of socially
situated language use to view language within the context of social practice.
Perhaps the most valuable feature is that the community of practice admits a
range of social and linguistic phenomena that are not analyzed in other theoretical
models. Local identities, and the linguistic practices that produce them, become
visible to sociolinguistic analysis as the purposeful choices of agentive individ-
uals, operating within (and alongside and outside) the constraints of the social
structure. To describe and explain such complexity must be the next step not only
for language and gender scholars, but for all sociolinguists concerned with the
linguistic construction of the social world.

NOTES

* My thanks to Janet Holmes, Chris Holcomb, Stephanie Stanbro, and members of the Ethnography/
Theory Group at Texas A&M University for comments on and discussion of the ideas in this article.
1
The work of Barbara Horvath on immigrants in Sydney’s speech community (Horvath 1985,
Horvath & Sankoff 1987) has done a great deal to correct this omission.
2
Santa Ana & Parodi’s model of nested speech communities (1998) is a recent attempt to address
this problem.
3
A fuller discussion of the advantages of practice theory for language and gender research is
provided by Bucholtz 1999.
4
Eckert 1989b calls this simple formulation into question; see also Labov 1990 for a response.
5
Though this is not its actual name, it preserves the flavor of the original. All other names are
pseudonyms chosen by the speakers.
6
Transcription conventions are as follows:
. end of intonation unit; falling intonation
, end of intonation unit; fall-rise intonation
? end of intonation unit; rising intonation
– self-interruption
: length

Language in Society 28:2 (1999) 221

 M A R Y B U C H O LT Z

underline emphatic stress or increased amplitude

(.) pause of 0.5 seconds or less
(n.n) pause of greater than 0.5 seconds, measured by a stopwatch
h exhalation (e.g. laughter, sigh); each token marks one pulse
( ) uncertain transcription
^ & transcriber comment; nonvocal noise
{ } stretch of talk over which a transcriber comment applies
[ ] overlap beginning and end
/ latching (no pause between speaker turns)
5 no pause between intonation units
The transcript emphasizes sequential organization in order to highlight speakers’ orientation to one
another. It excludes phonological detail that is necessary for a complete analysis of nerd identity
performance.

REFERENCES

Bloomfield, Leonard (1933). Language. New York: Holt.

Bourdieu, Pierre (1978). Outline of a theory of practice. Cambridge & New York: Cambridge Uni-
versity Press.
_ (1991). Language and symbolic power. Cambridge, MA: Harvard University Press.
Bucholtz, Mary (1998). Geek the girl: Language, femininity, and female nerds. In Natasha Warner
et al. (eds.), Gender and belief systems: Proceedings of the Fourth Berkeley Women and Language
Conference, 119–31. Berkeley: Berkeley Women and Language Group.
_ (1999). Bad examples: Transgression and progress in language and gender research. In Mary
Bucholtz et al. (eds.), Reinventing identities. To appear, New York: Oxford University Press.
Cameron, Deborah (1992). Feminism and linguistic theory. 2d edn. New York: St. Martin’s Press.
Chambers, J. K. (1995). Sociolinguistic theory. Oxford: Blackwell.
Certeau, Michel de (1984). The practice of everyday life. Berkeley: University of California Press.
Dorian, Nancy C. (1982). Defining the speech community to include its working margins. In Suzanne
Romaine (ed.), Sociolinguistic variation in speech communities, 25–33. London: Arnold.
Eckert, Penelope (1989a). Jocks and Burnouts: Social categories and identity in the high school. New
York: Teachers College Press.
_ (1989b). The whole woman: Sex and gender differences in variation. Language Variation and
Change 1:245– 67.
_ , & McConnell-Ginet, Sally (1992). Think practically and look locally: Language and gender
as community-based practice. Annual Review of Anthropology 21:461–90.
_ , _ (1995). Constructing meaning, constructing selves: Snapshots of language, gender,
and class from Belten High. In Kira Hall & Mary Bucholtz (eds.), Gender articulated: Language
and the socially constructed self, 459–507. London: Routledge.
Giddens, Anthony (1979). Central problems in social theory: Action, structure, and contradiction in
social analysis. Berkeley: University of California Press.
Guy, Gregory R. (1988). Language and social class. In Frederick J. Newmeyer (ed.), Linguistics: The
Cambridge survey, vol. 4: Language: The socio-cultural context, 37– 63. Cambridge & New York:
Cambridge University Press. 37-63.
Hinton, Leanne, et al. (1987). It’s not just the Valley Girls: A study of California English. Berkeley
Linguistics Society 13:117–28.
Holmes, Janet (1997). Women, language and identity. Journal of Sociolinguistics 1:195–223.
Horvath, Barbara (1985). Variation in Australian English: The sociolects of Sydney. Cambridge &
New York: Cambridge University Press.
_ , & Sankoff, David (1987). Delimiting the Sydney speech community. Language in Society
16:179–204.
Hudson, Richard A. (1980). Sociolinguistics. Cambridge & New York: Cambridge University Press.
Hymes, Dell (1974). Foundations in sociolinguistics: An ethnographic approach. Philadelphia: Uni-
versity of Pennsylvania Press.
Johnstone, Barbara (1995). Sociolinguistic resources, individual identities, and public speech styles
of Texas women. Journal of Linguistic Anthropology 5:183–202.

222 Language in Society 28:2 (1999)

 “WHY BE NORMAL?”: LANGUAGE AND IDENTITY PRACTICES

_ (1996). The linguistic individual: Self-expression in language and linguistics. Oxford & New
York: Oxford University Press.
_ , & Bean, Judith Mattson (1997). Self-expression and linguistic variation. Language in Soci-
ety 26:221– 46.
Kinney, David A. (1993). From nerds to normals: The recovery of identity among adolescents from
middle school to high school. Sociology of Education 66:1.21– 40.
Labov, William (1966). The social stratification of English in New York City. Washington, DC: Center
for Applied Linguistics.
_ (1972). The reflection of social processes in linguistic structures. In his Sociolinguistic pat-
terns, 110–21. Philadelphia: University of Pennsylvania Press.
_ (1990). The intersection of sex and social class in the course of linguistic change. Language
Variation and Change 2:205–54.
Lave, Jean (1988). Cognition in practice: Mind, mathematics, and culture in everyday life. Cam-
bridge & New York: Cambridge University Press.
_ , & Wenger, Etienne (1991). Situated learning: Legitimate peripheral participation. Cam-
bridge & New York: Cambridge University Press.
Luthin, Herbert W. (1987). The story of California (ow): The coming-of-age of English in California.
In Keith M. Denning et al. (eds.), Variation in language: NWAV-XV at Stanford, 312–24. Stanford,
CA: Department of Linguistics, Stanford University.
Milroy, James (1992). Linguistic variation and change: On the historical sociolinguistics of English.
Oxford: Blackwell.
Morgan, Marcyliena (1999). No woman, no cry: Claiming African American women’s place. In Mary
Bucholtz et al. (eds.), Reinventing identities (to appear).
Ortner, Sherry (1996). Making gender: The politics and erotics of culture. Boston: Beacon.
Pratt, Mary Louise (1987). Linguistic utopias. In Nigel Fabb et al. (eds.), The linguistics of writing:
Arguments between language and literature, 48– 66. New York: Methuen.
Rickford, John R. (1986). The need for new approaches to social class analysis in sociolinguistics.
Language and Communication 6:3.215–21.
Romaine, Suzanne (1982). What is a speech community? In Suzanne Romaine (ed.), Sociolinguistic
variation in speech communities, 13–24. London: Arnold.
Santa Ana, Otto, & Parodi, Claudia (1998). Modeling the speech community: Configuration and
variable types in the Mexican Spanish setting. Language in Society 27:23-51.
Schegloff, Emanuel; Jefferson, Gail; & Sacks, Harvey (1977). The preference for self-correction in
the organization of repair in conversation. Language 53:361–82.
Silverstein, Michael (1996). Encountering languages and languages of encounter in North American
ethnohistory. Journal of Linguistic Anthropology 6:126– 44.
Tolone, W. L., & C. R. Tieman (1990). Drugs, delinquency and nerds: Are loners deviant? Journal of
Drug Education 20:2.153– 62.
Wenger, Etienne (1998). Communities of practice. Cambridge & New York: Cambridge University
Press.
Williams, Glyn (1992). Sociolinguistics: A sociological critique. London: Routledge.

Language in Society 28:2 (1999) 223