“ Dolls or teddies ? ” Constructing lesbian identity through community-specific practice

The concept of ‘community’ often presents a problem for queer linguists. ‘The gay community’ is often viewed as an impossible site for research due to its imagined status, 
whilst local communities of gay people have been considered too heterogeneous and idiosyncratic to draw conclusions from. In this article, however, it is argued that both of these aspects of community can, and should, be a central focus of an investigation into language and sexual identity. Through the analysis of a conversation emerging from a lesbian group, using a sociocultural linguistics framework, it is argued here that the community of practice 
approach can play a crucial role in understanding how ideologies from ‘the gay community’ are used to construct a coherent sexual identity on a local level. The analysis reveals how the group engages in practices that enable them to construct micro-level personas in direct response to broader, ideological structures of heteronormativity.


Introduction
Questions of how language and community relate to one another have, to some extent, fallen from the agenda of language and sexuality research.The concept of 'the gay community' is often taken simply as a metaphor for gay culture, whilst the speech of those identifying as lesbian, gay or bisexual (LGB) may be analysed in isolation from the local communities that they inhabit.This article argues that researchers should instead make speakers' community memberships and identifications central to any investigation of language and sexual identity.It posits that an analytical approach that can combine local instantiations of community identity with broader ideological notions of collectivity is required for such an endeavour.It will be argued, here, that the combined use of two theoretical approaches, one from variationist sociolinguistics (the community of practice) and one from interactionist sociolinguistics (sociocultural linguistics), enables a coherent analysis of the relationship between language and sexual identity.The article begins with a discussion of how community has been dealt with in general language and sexuality research, before moving on to outline the community of practice and sociocultural linguistic frameworks.An ethnographic analysis of a lesbian community of practice is then provided, in which its members' styles and practices are considered in relation to their identity construction as a group.Finally, a brief extract of data is analysed, in which members of the lesbian community of practice engage in identity work together.Combined, the ethnographic discussion and linguistic analysis provide a useful illustration of how and why 'community' is an important concept to consider when exploring language and the construction of sexual identity.

Language, sexuality and community
In some of the earliest approaches to language and sexuality, 'the gay community' (Rudes & Healy 1979) or 'the homosexual subculture' (Farrell 1972) were central concepts, though they were used largely as convenient ways of describing all gay and lesbian people as a homogenous group.Indeed, early attempts to provide glossaries of gay slang or 'homosexual argots' (such as Rodgers 1972) relied on this notion of homogeneity in order to refer to something tangible which could be reported upon.As Penelope and Wolfe (1979: 1) argued at the time, however, no homogenous community could possibly transcend all geographical and social boundaries.This argument mirrors that of researchers in other disciplines, where the phrase 'the gay community' is often used to refer to something which is imagined rather than rooted in a physical reality (see Boyd 1997;Harvey 2000;Hughes 2008 for example).This references Anderson (1983), who argues that 'imagined communities', such as those related to nationalities, are created largely through the media's construction of idealised, supposedly homogenous identities.
There is ample evidence for this process occurring within language and sexuality research, too; Turner (2008) demonstrates how language is used in the British lesbian magazine Diva to construct a shared sense of community, by focusing on presumed homogeneity between its readers and by othering non-lesbians.Similarly, Wong and Zhang (2000) describe the Tongzhi community, an imagined community of gay and lesbian people in East Asia, as it is constructed by a magazine catering for this readership.Such research demonstrates how existing linguistic resources are used by writers to create a new discourse for gay and lesbian people, one which communicates an 'ideal', global image of what it means to be gay.In this way, magazines and other media aimed at 'the gay and lesbian community' may function to actively construct their readers' subjective sense of self through the communication of culturally relevant ideologies.This, in turn, feeds the notion of an imagined community by emphasising the supposed homogeneity between its apparent members.Though early researchers hoped to capture a linguistic variety that was specific to one tangible gay community, the reality of geographical and social variation between gay speakers made this an impossible goal.

Language in gay communities
As scholars began to move on from investigations, which aimed to identify 'the language of the gay community', research into language and sexuality began to consider actual communities within given locations.Some of the first research into this, such as that of Chesebro and Klenk (1981), considered gay spaces -discos, bars and so on -in order to observe the language used by gay men when they were amongst one another.In this way, linguistic enquiries into sexuality began to shift towards contextually-specific language use, with 'community' defined as a space where people interact rather than as an imagined construct.As social geographers Valentine and Skelton (2003: 849) outline, urban spaces have played a significant role in the construction of gay culture since the late nineteenth century, when bars and bathhouses emerged in urban locations in the USA, and since the middle of the last century when whole neighbourhoods began to become visibly identified as 'gay'.The role of actual, tangible communities cannot be ignored, therefore, when we consider the relationship between language and sexuality.Indeed, in the UK, a specific language known as Polari emerged from contexts in which gay men socialised together up until the 1960s, demonstrating the significance of these spaces for their users.Polari, a vocabulary that allowed gay men (and some lesbians) to communicate in a coded way, allowed its speakers to construct a shared sense of collectivity and belonging.As Baker (2002: 85) puts it, the language allowed gay people to not only "recreate themselves, it also allowed them to recreate their whole world" in a time when homosexuality was illegal in Britain.Given that such a language emerged as a result of face-to-face interaction between individuals, it is clear that gay and lesbian communities do exist in a literal, physical setting.
More recently, such gay-specific communities may be more typically referred to as 'scenes' or as gay spaces.Knopp (1998: 173), who reviews several urban locations with gay spaces, argues that the modern 'gay scene' often serves to allow gay people a means of survival, but yet is also typically constructed with primarily commercial interests.
Importantly, the people served by such scenes are typically relatively privileged; they are often white, middle-class, able-bodied and male (Noble 2012: 23).Gay scenes in urban locations may include support or activist groups, sports teams or book clubs, but most prominently tend to revolve around a dominant bar culture.They can involve a firm network of organisations and businesses, such as in the Castro area of San Francisco, North America, or the Canal Street area of Manchester, England.They may be somewhat disparate and loose in their structure, but gay scenes of this nature tend to play an important role in shaping the identities of those that frequent them.In the search for a 'Gay English', Leap (1996) draws on this idea by arguing that gay men might use a specific language (one which includes metaphors about women, for example) when in gay-specific contexts, but a more coded one when outside of these.The relevance of looking at these contexts is clear when we consider the presence of coded languages such as Polari, yet a certain amount of criticism was levelled at such an approach as the following conclusion was drawn: 'gay language' may be spoken by some gay people in some gay contexts, but that does not, in itself, make it a 'gay language' (Darsey 1981: 63, Graf & Lippa 1995: 233).As Kulick (2000: 265) argues, any language that can be identified as being used by gay men or lesbians can also usually be found in the language of other subcultural groups, or even in more mainstream contexts.
Another issue with the investigation of language use within given lesbian and gay communities or scenes is that, although they may exist on a local or regional level, not all gay people in that area will necessarily belong to them.That the term 'non-scene' can be found in gay media and dating advertisements (Thorne & Coupland 1998) illustrates this fact.
Similarly, as Barrett (1997: 118) argues, not all those who exist under the umbrella term of 'the gay community' are included within it -bisexual people, those who do not identify with the label 'gay', those who have not yet come out or are straight may all belong to a particular scene-based group without being gay.Arguably, of course, lesbian women may also not be included when we talk of 'gay' establishments, which are often exclusively male in their customer base.Furthermore, although modern technology means that queer individuals around the globe can use social networking sites to develop online communities where they interact together, not everybody has access to these resources.Similarly, those who are unable or unwilling to take part in the drinking culture of a particular urban gay scene will not necessarily view themselves as part of that city's 'gay and lesbian community', and those who do engage with a scene will be unlikely to include such people in their conceptualisation of that community.Factors of age, socioeconomic class, ethnicity, religion, gender, physical ability and age will all play a role in how involved people can be in a gay scene or network.Koller (2008) makes this point, showing that not all lesbian women necessarily perceive themselves to be part of a lesbian community simply by virtue of experiencing same-sex desire.Those (typically white, middle-class) lesbians who identified as feminists in the 1970s and 1980s, for example, who tried to set a clear political agenda for 'the lesbian community,' were often opposed by other lesbian women from ethnic minority and working-class backgrounds, who identified with their sexual orientation in very different ways (Koller 2008: 86).To suggest that there was one coherent way of identifying as a lesbian during these decades would be no more accurate then than in the present day.This suggests that, if we attempt to investigate the language used within a given lesbian and gay community, we must also remember that the community some people feel they inhabit will not be recognised by others just because they share a broad sexual orientation, or even the same space.

Language in gay contexts
It is evident, then, that the concept of 'community' in relation to lesbian and gay people is fraught with difficulties.One cannot argue that there is some shared community between all homosexual people, irrespective of personal background, location or preference, and equally one cannot claim to thoroughly investigate language use in regionally-specific communities (or scenes) because of the variation that exists within them.However, the research outlined above shows that the use of language specific to a person's sexuality may be restricted to relevant contexts.Whilst not all people will use a particular language by virtue of being gay or being part of a gay scene, individuals might adopt culturally-prevalent, symbolically-loaded ways of doing things when in a 'gay context'.That is to say, language use might be better viewed not as reflecting some pre-defined, subjective, personal sense of self which somehow exists between all gay speakers, but instead as being indicative of characteristics that hold cultural salience in a particular imagined community.In this sense, 'identity' relates here to the cultural categories that exist in a given society and have ideological meanings attached to them.Such categories include 'lesbian' and 'gay man'; these labels have stereotypical characteristics and ideas associated with them, which result from the prevalent cultural norms of Western society, and particularly from assumptions of heteronormativity (the idea that it is both typical and expected for people to be straight).It is possible, then, for a gay woman or man to use language which allows them to present themselves in line with a culturally-accepted, ideological image of what a gay woman or man is (such as by eschewing heteronormative ideals); in doing so, they may align themselves with a particular cultural identity.It is quite likely that speakers will do this when they are in a context where their sexuality is relevant to the way in which they wish to position themselves to others, as will be outlined below.
The role that language can play in this positioning process can be explained through the theory of indexicality (Ochs 1992: Irvine & Gal 2000: Silverstein 2003;Bucholtz & Hall 2004).In this process, particular ways of using language can become associated with specific ideological identities, such as that of 'gay man' or 'lesbian'.When language is used in a particular way by somebody who associates themselves with a given identity, it can point towards -or index -their membership to that cultural category.Importantly, this is not to say that a linguistic form always holds a particular ideological meaning; its indexical quality will depend on the person who is using it and the context in which they use it, as all linguistic forms are multifunctional (Ochs 1992: 340).Indexical meaning is therefore produced as speakers project the aspects of themselves they wish to in given situations.The theory of indexicality draws on the concept of performativity, whereby an identity is constituted through certain symbolically-meaningful acts (such as gestures and parts of speech) and an illusion of an essential, 'natural' self is created (Butler 1990: 270).Podesva's (2007) study of one man's language use in different contexts provides an excellent illustration of this.
Podesva shows a falsetto voice quality being used by a gay man when in a situation with gay friends, allowing him to project a 'diva' persona, yet the same speaker barely uses the feature when in a workplace setting.The male diva is one of many stereotypes about gay men, and the use of falsetto enabled this speaker to embody that stereotype.In doing so, he conveyed social meanings specific to the ideological concept of 'gay community,' and signalled his sexuality.Similarly, Queen (1998) suggests that the communication of particular concepts from gay and lesbian culture, such as the subversion of gender norms or the problem of homophobia, can index membership to an imagined gay and lesbian community and allow interlocutors to reproduce that community notion for themselves (1998: 210).In this sense, whilst a homogenous community of gay and lesbian speakers may be no more 'real' than a shared language that all lesbian or gay people use, it does exist as an ideological concept which may be reproduced within social interaction.The relevance of given social identities to speakers should not be assumed, therefore, but considered in direct relation to a local community; it is from this standpoint that those speakers can construct a sexually-specific self.
Such consideration of individuals or small groups of speakers may ascertain which broader ideologies are indexed through which linguistic moves, instead of attempting to provide an account of all people within an arbitrary set of physical boundaries.This enables us to take a bottom-up approach that avoids essentialised, pre-defined categories and embraces variation.This approach also enables us to recognise that language can produce a contextually-salient identity that is underpinned by broader, ideological structures and systems, but which may remain specific to that context.A group of young, white middleclass lesbians may draw on the same broad ideological notions of lesbian identity as a group of older, black, working-class lesbians, for example, but the specific personas that are produced as a result will differ in line with other intersecting aspects of their identities and experiences.I use 'persona' here to mean a type of 'identity image,' which is specific to a given context (Coupland 2007: 237).This differs to the ideological construct of identity, which may be attached to broad categories such as 'lesbian'.A persona relates instead to a group-specific sense of self that is produced collaboratively, and on a local level.Whilst a group of women may draw on the same ideological construct of 'lesbian', then, their specific reworking and negotiation of that construct will result in a unique persona, which reflects their understanding of what it means to be a lesbian.
It is at this group level that it can be useful to reconsider what is meant by 'community'; whilst there may be no such thing as one homogenous gay and lesbian community, it is evident that smaller, localised collectives of gay and lesbian people do exist.
As suggested above, ideological constructs such as 'the gay and lesbian community' may be reified within interaction by those who index a culturally-recognisable social identity.If we link the interaction of individuals who are homosexual with the broader cultural context of an imagined gay and lesbian community and the contextual significance of any gay scene or group with which they are engaged, we are better able to understand the meanings that exist within their language use.It is vital that we understand these meanings because their reproduction and reappropriation will reveal the attitudes and stereotypes which continue to constrain, as well as facilitate, the identity construction of LGB people.This is not to suggest that sexuality will bear relevance to every gay person's personal identity, of course.Instead, it is to argue that those who identify with a minority sexual orientation are often likely to seek out support and solidarity from those they perceive as being similar to them.In this sense, we need to consider 'community' as something which is meaningful on a very local scene-based level, and to view language use as representative both of that local community and of broader ideological concepts of 'the' lesbian and gay community.

Community in sociolinguistics
One way of achieving this balance is to use the community of practice (CoP) approach.The community of practice was introduced to sociolinguistics by Eckert and McConnell-Ginet (1992).It came from a theory of learning established by Lave and Wenger (1991), and was posited as an alternative to the traditional speech community model.The speech community is often used in variationist sociolinguistic work, where the usage of particular linguistic variants, such as locally-salient vowels, are correlated with speakers' membership to given social categories or groupings.Speech communities may be defined as a geographically-placed group of people sharing "a set of evaluations about the speech of [their] community" (Britain & Matsumoto 2005: 5).As Barrett (1997: 186) argues, however, such research often overlooks non-normative identities in favour of those speakers who fit a given demographic; the speech community positions individual speakers with potentially distinct identities as one homogenous group, and rarely is sexuality taken into account.
Considering the variation that exists within lesbian and gay communities, even if sexuality were taken into account in sociolinguistic work of this kind, it would be difficult to find a 'typical' language used by gay or lesbian speakers.Rather than forcing speakers into homogenous groups and ignoring their differences, then, Barrett argues that we should focus on language use across lines of social differentiation, drawing on Pratt's (1987: 60) argument for a 'linguistics of contact'.
The 'linguistics of contact' approach is a conscious shift away from the traditional sociolinguistic model of the speech community and towards a model which can deal with the reality of speakers moving in and out of different speech contexts, relating to different imagined communities, and adapting their language accordingly.Employed by Barrett (1997) in his analysis of African American drag queens' language use, the linguistics of contact approach enables a focus on how language can index membership to a range of different imagined communities (including, in his study, 'the' gay community as well as 'the' African American community).In this important research, Barrett demonstrates that speakers do not fit one community alone, but instead creatively produce their identities based on the context in which they are situated at that time.As Coupland (2003: 425) argues, people engage in multiple communities and, as a result, have multiple identities; the speech community model cannot adequately deal with this.As researchers, Coupland suggests, we ought to move from attempting to explain language use as a result of speakers' positions in "a predetermined social structure" and instead focus on that which is "locally negotiated in reflexive and strategic communicative practice " (2003: 426).This reflects the aims of the community of practice approach.A CoP concerns speakers who engage together in something in a mutual way, which, over time, leads to shared ways of doing things, or practices (Eckert & McConnell-Ginet 1999: 185).It allows for a focus on language as one of many social practices (others may include style of dress or haircut, for example), which combine to create something meaningful.Rather than looking primarily at structures across a speech community and showing language to reflect those structures, then, the CoP approach looks at the language that is used as a tool to create smaller-scale structures within a speech community.
Importantly, the CoP approach views speakers as agentive in their use of language and as performatively constructing shared, group-specific personas that are informed by broader cultural ideologies and identities.Variationist researchers have used the CoP to demonstrate how shared practices emerge from interaction and become symbolic of a mutual sense of self.Eckert's (2000) work in an American high school is perhaps the most wellknown example of this, as she reveals how the language practices of students identifying as 'Jocks' (in contrast to 'Burnouts') allowed them to index a high-achieving, academicallyoriented persona, which reflected both their experiences within the local community and their middle-class aspirations.Other variationist accounts (for example Lawson 2011;Mendoza-Denton 2008;Moore 2006) have continued to make use of the CoP, and have shown that it is not merely demographic category membership, but identification with particular cultural norms or ideals, which can lead to language use within a given speech community.For queer linguists, the CoP also offers a way of dealing with the variation that exists amongst those identifying as lesbian or gay; it facilitates explorations of how language gains meaning on a local level and is indexical of an imagined gay and lesbian community on a more global, ideological level.A focus on individual CoPs allows for a more nuanced understanding of the relationship between its members and these wider contexts.

Applying the community of practice approach
For qualitative researchers (such as Bucholtz 1999), the CoP enables a focus on speakers' engagement together in order to see how the language used in given groups gains its symbolic meaning.Clark (2012), for example, explores the interaction of female university hockey players, finding that their definition of success was formed not only from their sporting achievement, but from the extent to which they were perceived as heterosexual.
In order to construct an 'achieving' persona, for example, part of the practice of the young women was to emphasise their attractiveness and sexual desirability.In this sense, being a successful woman for members of this particular CoP was conflated with being heterosexual, and discourses of homophobia were often employed as part of this process.In unravelling these themes from her data, Clark demonstrates that small, local-level groupings play an important role in the reproduction and maintenance of broader, heteronormative ideologies.
Conceiving of this group as a CoP, importantly, allows her to explain how those ideologies became naturalised through the everyday practices of the hockey team, and the extent to which they became ingrained within the performative construction of a shared sense of self.
Despite the evident usefulness of the CoP model for the qualitative analysis of smallgroup identity construction, however, the CoP framework has not yet been widely employed by scholars of language and sexuality.A considerable amount of work into language and sexuality has concerned small groupings of queer speakers, certainly, in order to see how their interaction reveals some construction of identities (such as Bland 1996;Coates & Jordan 1997;Johnsen 2008;Kitzinger 2005;Thorne 2013) but, typically, the CoP framework has not been employed within these analyses.Notable exceptions to this include Morrish and Sauntson (2007), Sauntson and Morrish (2012), and Queen (1998).In Queen's data, a CoP of lesbian and gay friends are shown to actively produce a joint persona based upon their shared understanding of gay culture, whilst in Morrish and Sauntson's (2007: 43) data, interactions between a CoP of lesbian friends are shown to switch between the production of a collaborative lesbian persona (achieved via the rejection of heterosexual norms and the use of reclaimed epithets such as 'dyke') and one which is more generically 'female' (drawing on stories about mutual female friends and their children) when a straight friend joins them.As Queen (1998: 204) argues, the CoP offers a useful alternative to 'community' in such research because of its focus on "shared knowledge, cultural stances and the ideological expressions of those stances" rather than regional coincidence or demographic category.
Implementation of the CoP in this kind of qualitative, interactionist work, as Mills and Mullany (2011: 71) point out, tends to make use of Bourdieu's (1991) notion of 'habitus', the social and cultural environment within which one establishes certain beliefs and dispositions.
Habitus is central to Bourdieu's interpretation of social life: that our understanding of the world constrains the identities and ways of being that are available to us.When analysing the significance of language to a CoP, then, a consideration of habitus reminds us that the sociocultural context from which the language emerges must be accounted for if we are to understand the meaning of that language use.It is also important, of course, to focus on the moment in which language is produced.
In a fleeting instant, speakers may make use of a particular turn-taking style or word choice, which enables them to jointly negotiate a value that is core to their sense of shared being, though that style or phrase may never be used again.Through the use of an ethnographic methodology, in which the analyst is a participant observer of a CoP for a period of time, it is possible to interpret the language used as part of a broader system of meaning.In other words, one must become familiar with the typical practice of a group in order to understand what ideologies or norms the speakers are responding to in a given moment, and what significance that has to the cultural identities that they are drawing on and the personas that they are working together to construct.Though an interaction may be momentary, and though the syntax, intonation or lexical choice may never feature again in the conversations of those under study, it may still be considered significant by an analyst because of their understanding of the sociocultural context in which it occurred, and the way in which the language ties in to the broader practice of the CoP.As Coupland (2007: 218) puts it, the relationship between language and meaning is not direct or static, but is constantly being renewed, reshaped and revalorised; the only way to understand that relationship is to understand the ideological conditions that define the context of its production.By using the CoP framework in qualitative sociolinguistics, one focuses not on variants that are used in a consistent and statistically meaningful way, but on these individual, transitory moments, which have been captured, and their relationship to the broader ethnographic context that is under constant redefinition.
As with the linguistics of contact approach, the CoP first and foremost focuses on those with whom speakers engage, rather than relying on categories that the analyst has imposed upon them.Similarly, the CoP must not stand alone as the end point of any analysis.
As Rampton (2000: 10) argues, a CoP persona cannot only be viewed as "co-participation in locally embedded practice", since analysis "also has to extend to the way in which 'community' serves as a symbol and sign itself."In relation to sexual personas, then, it is useful to understand what happens in very local contexts, within CoPs that respond in some way to a geographically-based gay and lesbian scene (which, in turn, are responding to a heteronormative system whereby gay and lesbian people are 'othered').But it is also useful to recognise that any identity work happening within that context ties into a broader cultural context and habitus.It is argued here, of course, that this broader cultural context is an imagined gay and lesbian community, and that those who interact on a very local level do so in some kind of response to that imagined community.In other words, any analysis of sexual identity construction within a CoP should relate to how global ideologies are instantiated in local contexts.One way to ensure that data is viewed in a rigorous, detailed way, which allows for a focus on identity as it emerges in interaction, is to employ a framework established by Bucholtz and Hall (2005): the sociocultural linguistics approach.

Sociocultural linguistics
'Sociocultural linguistics' is an umbrella term, established by Bucholtz and Hall and designed to draw a contrast with variationist sociolinguistics.The framework they propose takes identity as "the social positioning of the self and other" (2005: 586), viewing language as a resource that holds social meaning and operates on a cultural level in order to produce that identity.Bucholtz and Hall establish their framework based on existing research which they identify as sharing a sociocultural linguistics agenda; that which focuses on the intersubjective construction of identity within interaction, as well as that which employs an ethnographic methodology in order to make links between the local context of an interaction and the broader sociocultural context surrounding it.In this sense, the framework very neatly aligns with the CoP, since both allow analysts to draw conclusions about the meaning behind language as it is used in interaction.The theoretical framework offers a set of principles concerning emergence (that identities are not pre-defined, but emerge through interaction), indexicality (that semiotic links between language and social meaning are created in interaction but also depend on locally-prevalent cultural ideologies), relationality (which concerns the fact that identities are intersubjective), partialness (that identities are everchanging, fluid concepts, and are in this sense never 'full'), and positionality (that the ways people position themselves in conversations constructs a contextually-specific identity).The principles outlined by Bucholtz and Hall (2005) are, importantly, not rigid rules which state that all identity construction works in the same way.Instead, they are the result of their review of interactionist research into language and identity, and they therefore provide an extremely useful and strategic way of approaching work in this area.By employing the principles as a framework when observing CoP interactions, it is possible to enhance a discourse analytic approach to data collected.
The positionality principle, in particular, demonstrates the usefulness of a sociocultural linguistic framework for CoP research as, according to Bucholtz and Hall (2005: 592), identity construction works on three levels.Firstly, they suggest that macro-level demographic categories are important to consider; these may be linked to the broad, cultural identity categories such as 'lesbian' and 'gay man' mentioned above.Secondly, Bucholtz and Hall argue that local cultural positions, which are specific to the ethnographic context exist; these may be aligned with the personas that are produced as a result of CoP interaction.
Finally, they argue that both of these aspects are realised via temporary roles and moves, which are taken by participants in an interaction.That is to say that, whilst local personas emerge from fleeting and momentary interactions, that interaction is situated within a broader series of ethnographic norms that make up local practice.In turn, these practices carry meaning that is specific to that local context, yet are also informed by broader cultural identity categories and their corresponding ideologies.A sociocultural framework, then, is centrally concerned with the relationship between what happens in a given moment, how that ties into the practices which are significant to speakers on an ethnographic level, and how broader ideological discourses and structures constrain or facilitate the subsequent personas that are produced.It is on this ethnographic level that the CoP is a useful construct, as it provides a clear way of understanding how the practices of speakers "articulate with the wider world, and with wider discourses of gender and sexuality" (Eckert & McConnell-Ginet 2007: 28).Before moving on to demonstrate the usefulness of the CoP in combination with a sociocultural linguistics framework, however, it is important to recognise and address some of the problems that linguists have identified with the CoP as a model.

Challenges to the community of practice
Despite the apparent benefits that it offers, there has been considerable critique of the CoP within the sociolinguistic literature, which may, in part, explain why it has not been put to more significant use within language and sexuality research.Largely, the critique concerns the CoP's definition, which may be viewed as overly rigid; it does not always seem to fit all speakers or contexts.For example, Eckert and McConnell-Ginet (2007) argue that, since CoPs typically emerge in response to a context such as the school, they "cannot be understood without viewing their relation to the institution" (2007: 33) and in relation to other CoPs within it (2007: 29).Indeed, the CoP was initially developed to enable a view of active learning through participation within institutional contexts (Lave &Wenger 1991), and one reading of the CoP as it has been implemented in sociolinguistics would be that the approach is restricted to schools, workplaces, or similar organisations.In this sense, a CoP, as the concept has been employed within quantitative sociolinguistics, must be defined partly by its emergence from an institutional setting.In essence, this excludes groups of speakers who regularly engage in shared practices, but who do so in a self-composed, non-institutional setting, as is the case for many LGB people.
If we deconstruct the meaning of an 'institution', however, we may find a less rigid way of understanding how CoPs that are unconnected to schools, businesses or organisations emerge.For example, I would argue that discourses emerging from ideologies such as patriarchy and heteronormativity play an important, structural role in terms of how they both constrain and define people.Citizens respond to these structures, and groupings of people with shared experiences (and conceptions of those experiences) are formed in response to them -just as members of a Jock CoP (Eckert 2000) respond to the structures of their school.
In a CoP of lesbian women, for example, we might see heteronormative discourses surrounding the sexual allure of make-up and cosmetics challenged and reworked, such as by stating that such 'beauty' products are unattractive (as in Jones in press).We might also find a group of students at a bisexuality support group constructing their own norms about what is 'authentically' bisexual by responding to stereotypes of bisexuals as promiscuous, or as going through a phase (shown in Thorne 2013).In both cases, we would see groups of people who have come together due to a shared need or desire to belong, which has emerged due to the prevalent ideological structures surrounding a fundamental aspect of their sense of self: their sexuality.As stated in Eckert and McConnell-Ginet's (2007: 28) own definition of the CoP, indeed, "day-to-day practice at the local level both feeds, and is structured by, larger social constructs and discourses".In this sense, broader ideologies surrounding gender and sexuality may be dealt with on a local level, where practice (everything from the clothes worn to the linguistic choices made) can both reinforce and challenge prevalent discourses.
Concerns have also been raised, however, about the process by which individuals may become a part of a CoP.Gee (2005: 590), for example, argues that participants in a CoP must be aligned with ideologies that pertain to that CoP before they can gain access to it, whilst Davies (2005: 567) suggests that individuals cannot become members of a CoP without first matching the social make-up of it; they must gain legitimacy as a potential member before becoming a member, meaning that they must already engage in some of a CoP's practices in order to 'fit in' and become accepted.Furthermore, Davies suggests, there must be some hierarchy within a CoP, which maintains its boundaries and controls its internal structure (2005: 571).Such critique suggests that CoPs are not entirely self-constitutive, and that they are in fact restrictive; one cannot simply opt to join a CoP, but must be accepted into it after having proved oneself legitimate to do so.It may be argued that this model would not fit well with studies hoping to investigate the role of language in constructing queer personas.CoPs that are based on sexual or gender identity may be secretive and formed at a grass-roots level, and the members of those CoPs may be more defensive or protective of their safety than other, more mainstream, groups.If, to join a CoP that will serve as a supportive or social network, one must share some of their practices in order to be accepted, but one cannot discover those practices without becoming accepted, how is it ever possible to join them and become engaged in the construction of a mutual sense of self?
Eckert and Wenger's response to this critique offers a solution: The question 'I do all the practices, why aren't I part of the community?' belies a confusion about the notion of practice with activity.One doesn't 'do' a practice excised from the community.A practice is a way of doing things, as grounded in and shared by a community….Legitimacy in any community of practice involves not just having access to knowledge necessary for 'getting it right', but being at the table at which 'what is right' is continually negotiated.(Eckert & Wenger 2005: 583, emphasis in original) This positions the CoP not as an entity that prevents others from joining it, but as a group of people who want to engage with others in practices, which relate to their own sense of self.
As Corder and Meyerhoff put it, to "analyse interaction and language within a community of practice framework is to study the emergence of norms and the gradual fixing of their social meaning through the dual dynamics of participation and reification" (2007: 444, emphasis in original).A CoP is ever-changing, not static, and the practices which are used within it emerge as a result of its members' interaction.They are not, in this sense, pre-defined.That is not to say that some activities or styles that are salient to a given CoP may not be engaged in by those outside of that CoP, however, but to argue instead that they will not carry the same symbolic meaning within a CoP until they have been negotiated by its members.
Below, these points will be illustrated via a discussion of a lesbian CoP.

The Sapphic Stompers
The remainder of this article concerns a discussion of ethnographic and linguistic data collected from a CoP called the Sapphic Stompers, 2 a lesbian hiking group based in the North of England.I spent fifteen months with the Sapphic Stompers, who met two to four times a month to spend a full day hiking in the nearby countryside.I approached the group as a researcher, and was accepted on this basis.I carried out interviews with the most regular members of the group, spent time with the group whenever they would allow it, and made audio recordings of their interactions after a year of ethnography with them.The hiking group was managed by six women who broadly represented the group in terms of their demographic profile: lesbian women who were typically white, in their fifties or sixties, were university educated, who held (or were recently retired from) professional, middle-class jobs, and who identified as feminist.Whilst I fit in with the group in most senses, I was in my mid-twenties at the time of my research (between 2006 and 2007) and was therefore much younger than the other women.Though they treated me as an 'apprentice' (Lave & Wenger 1991), this fact in combination with my status as a researcher meant that I would always be somewhat on the outside of the group.Despite this, the ethnographic nature of my research meant that I was often involved in the recorded discussions that took place.Therefore, I subjectively analyse my own input in the interaction that follows, as a result.
As outlined above, certain ideologies surrounding cultural identity categories may be drawn upon by individuals in order to index their membership to those categories.Indeed, the Stompers presented themselves in line with a number of cultural stereotypes about lesbians; they typically wore their hair short, avoided pastels and pink colours, did not wear make-up and wore unisex clothing when on their hikes.This is not to suggest that the women only cut their hair short in order to be legitimately recognised as a Stomper.In fact, most of the women, quite separate to the group, had been in some way involved in gay rights and women's liberation politics when they were younger.They had all, to differing extents, embraced some of the practices of those movements (such as the rejection of certain symbols of femininity) and, in this sense, their experiences as feminists overlapped with their identification as lesbians.Upon reaching their fifties and sixties, a number of these women found that they wanted a social network which depended less on activism and more on a way to stay fit and active with like-minded (typically left-wing, feminist, middle-class, middleaged and lesbian) people.An inevitable consequence of this shared outside experience was a similar style of dress, outlook and perspective, one which made the women quite recognisable as lesbians.In this sense, there were certain boundaries around who could become a Stomper member; as outlined above, membership to a CoP depends on some shared purpose and mutual participation in practice.
These shared styles had a specific CoP meaning for the Stompers, in that they functioned as markers, which defined what it meant to be a member of that group.Engaging in these styles was a wider part of the women's CoP practice.One of the Stompers' strategies was to produce an ideological binary between what was authentically lesbian (styles or practices which they referred to positively as being 'dykey') and those that were synonymous with heteronormative femaleness (which they negatively evaluated when employed by lesbians, and which they viewed as 'girly' and inauthentic for the Stomper group).In this sense, what is authentic can be defined by a given group in accordance with what they consider to be more legitimate or genuine, which in turn will depend on the overall norms of the CoP.For the Stompers, this involved the construction of two personas that lay on either side of an ideological binary, and they took stances (see Jaffe 2009) in line with the 'authentic' side of this binary in order to construct a dykey persona.During my ethnography with the women, I often witnessed conversations in which the women actively defined what it meant to be an authentic or inauthentic lesbian, with the authentic 'dyke' persona typically being achieved by eschewing feminine practices.Such practices included talk in which the women rejected traditionally feminine styles, including doing housework, wearing skirts and make-up, and shaving one's legs; these symbolised a 'girly' identity due to their association with traditional female gender roles and styles.
Though certainly reminiscent of, and enabled by, broader stereotypical identity categories of butch and femme, which are prevalent within the imagined lesbian community and realised within certain lesbian scenes, the construction of the 'dyke' and 'girl' personas by the Stompers carried meanings which were specific to their CoP.Being dykey, for example, was about more than looking like a butch lesbian; it included a feminist political perspective and a serious approach to the hobby of hiking.Being girly did not simply mean styling oneself in an ideologically feminine way; it included 'choosing' to be a lesbian rather than being born as such.By viewing the Stompers as not just a social group but as a CoP, then, it is possible to see not only how the women indexed broader ideological categories such as 'lesbian' through their interaction, but also to witness how they were negotiated by the group into something meaningful for them on a local level.
By viewing speakers as a collective who are mutually engaged in meaning-making, then, it is possible to see how practices and styles that are specific to a given group relate to the broader identity categories that are being indexed.This is particularly important for language and sexuality researchers due to the pervasiveness of heteronormative ideologies, and subsequent notions of queer authenticities, within gay and lesbian culture in the West.By analysing the ethnographic context of a CoP, such as that of the Stompers, we are able to see how their group-specific practices index identity categories, and in turn it is possible to understand the locally specific versions of that identity -the personas -that are being constructed.When we consider how broader ideological categories relating to sexual identity are reworked on a local level, it is possible to develop our understanding of how those ideologies constrain or shape the ways in which, as individuals, speakers realise their own sense of self.
In addition to ethnographic description, it is useful to engage in close, detailed analysis of interactions in order to identify the moments in which personas are (re)produced.
Below, a brief extract of data from one of the Stompers' conversations together is presented.
In this, the women can be seen to reproduce the binary between what is 'authentic' as a lesbian and what is inauthentic, using a discussion of childhood toys to do so.This is then analysed using the sociocultural linguistics framework and via the conceptualisation of the Stompers as a CoP.This data is intended to provide some insight into the process by which community-based identity work can occur.The overall conclusions drawn emerge from the ethnographic data -an extended account of which is provided in Jones ( 2012) -as much as they do from the discourse presented below.Whilst the extract itself is brief, then, it provides a useful illustration of how we may benefit from a focus on CoPs in language and sexuality research.

Dolls or teddies?
The data considered here emerge from an evening during which several members of the Stomper CoP were relaxing after a day-long hike.Not all of the women spent time together in non-walk contexts such as this, but those featured in this interaction were core to the CoP in that they organised and managed the regular hikes that the wider group ran and attended.During dinner, one of the women (Claire) mentioned a television programme she had seen which showed twin toddlers, both boys, choosing to play with different toys.The programme showed that one boy preferred stereotypical girls' toys, such as dolls and a makebelieve kitchen, while the other preferred stereotypical boys' toys like guns and trucks.

Constructing the binary
In the above interaction, Eve, Sam and Claire take key roles in setting up the brief discussion of dolls and teddies.Claire's reporting on the television programme allows her to set up the first crucial aspect of the ideological binary by suggesting that a boy who likes dolls may be more likely to grow up to be gay (line 1), but it is Eve who responds to this personally by claiming that she 'hated dolls' (line 2).In doing so, Eve begins to construct an ideological binary between herself (as a gay woman) and the boy in the television programme (an apparently gay boy).She positions him as liking dolls because he is gay and herself as not liking dolls for the same reason.By asking the rhetorical question 'does that explain everything?', she presents the binary as a logical construct.It is telling that this is Eve's only turn in this interactive moment.She has made her stance immediately clear and has set up an ideology for the other women to follow, and seemingly does not need to play a further role.
Though Eve sets up a very clear stance against dolls, however, it is Claire in line 3 who introduces teddies and sets up the opposition between the two types of toy, using language tactically to create a norm based on the assumption of homogeneity.By presenting her 'theory', for example, she makes the concept of lesbians disliking dolls (and preferring teddies) something that needs to be proven, or at least tested in some way.
When the women are each required to state their stance towards dolls, then, they are essentially being asked to state whether they feel Claire's theory is correct or not.The subtle pressure on them to agree with her theory is evident, of course, by Claire's use of the determiner 'all' (line 3) to premodify 'lesbians.'This lexical choice serves to present lesbianism as a homogenous experience and, given the context, seems to also suggest that hating dolls is somehow an inherent quality of gay women.In this moment, she reifies the imagined lesbian community by drawing on cultural stereotypes from it, indexing an authentic persona for herself in relation to this.Her theory that lesbians liked teddies 'and not dolls' is also significant as it allows for no manoeuvre or flexibility -either teddies or dolls could be preferred.Given this, it is again unsurprising that all of the women in the interaction then claim to have preferred teddies to dolls.The subjective stance taken by Eve in line 2 ('I hated dolls') is mirrored exactly by Hannah in line 5, with an emphasis on 'I', sharing the stance and collaborating with Eve whilst positioning herself as a lesbian due to also 'hating' dolls.Sam shares this stance in line 3, choosing a less strong construction in 'never liked' (rather than 'hated'), positioning herself as a dyke but doing so with a direct reference to the theory itself: 'I'd go along with that one'.

Authenticating dykeyness
It is telling that neither Jill nor I explicitly claim to have hated dolls.We simply say that we preferred teddies.It may be argued that this allows us to position ourselves on the 'correct' side of the binary, without explicitly claiming something that was untrue (as it was in my case; as I reveal in line 10, I did have dolls as a child).It is noteworthy that this is not accepted as a direct-enough stance move by Sam, who later puts the question to me again (in line 7).Indeed, in line 6, Sam interrupts Claire in order to directly challenge the two participants (me and Jill) who have not already strongly articulated their stance against dolls.
It is evident, however, that Sam is working from the knowledge that Jill and I are already lesbian, in that she counts up how many of us (as lesbians) preferred teddies.She is not, therefore, seriously offering us a chance to authenticate our sexuality in line with the newly constructed ideology, but rather is testing Claire's theory.
In this sense, there is no challenge to our dykeyness, but Claire's theory is being playfully employed as a way to reinforce core CoP ideologies -that lesbian womanhood is distinct from heteronormative womanhood (in which, presumably, dolls would be preferred).
Sam's direct challenge ('dolls or teddies?' in lines 7 and 8) is addressed first to me, and both mine and Jill's answers are minimal given the limited choice of options presented to us; both of us answer with the appropriate response, in a single word ('teddies'), to back up the test.
Sam then employs scientific jargon, 'QED' to seemingly suggest that our responses are 'proof' of Claire's theory.This allows Sam to play a key role in the construction of this temporary ideology, and thus to further authenticate herself.Though, in lines 8-9, I raise the point that we 'have to say that' to demonstrate my awareness that this is all achieved on a fairly superficial level, this is largely ignored by the other women in favour of the convenience of the identity work in which we have all engaged.It is also likely that my relative outsider status in the group (both in terms of my age and my role as a researcher) prevented me from being taken seriously in this moment, hence the women not acknowledging my contribution.
Claire's turns in this extract show her keenness to give anecdotal evidence.She offers a narrative event whereby she positions herself as 'other' in order to both back up her theory and authenticate herself by constructing a non-heteronormative female persona.In lines 5-7, for example, she claims that she was thought to be strange for pushing a teddy bear around in a pram, rather than a doll.How unusual this would really be is questionable, however, and suggests that Claire is consciously putting herself forward as unusual in this moment, reinforcing the notion of lesbians as 'other' in a heteronormative culture and therefore strengthening the binary that she has presented.A little probing of her claim reveals the problematic nature of it; pushing a pram around could in itself index femininity due to its relationship with motherhood and the fact that toy prams are marketed primarily at girls.In this moment, Claire's interlocutors could challenge her on this point, as it seems to directly conflict with the implied construct she has presented about lesbians not liking girls' toys, but they do not.Sam does interrupt Claire in line 6, however, as she provides the crucial element of her narrative (that people would look at her strangely, thus reinforcing her 'othered' identity), suggesting perhaps that Claire's assertion was thought to be rather weak.
When Hannah returns to the idea of a pram in line 10, in contrast, her turn is more successful; she makes lexical choices such as referring to 'babies' rather than 'dolls', which explicitly defines dolls as symbolic of babies for the first time and strengthens the symbolic property of dolls as 'girly' in contrast to prams.She also suggests that she used her toy pram not as a vehicle for teddies but as a repository for coal (line 11), removing any quality of femininity from it; she claims that she 'chucked' the dolls from the pram (a colloquial turn of phrase which suggests a lack of care for the dolls) and filled it instead with 'coal from the coal bunker'.Arguably, this indexes a somewhat 'tomboy' identity -coal is dirty and semantically associated with coal-mining, which is a symbolically male, working-class occupation.Certainly, Hannah's turn here allows her to position herself as fundamentally against dolls and 'girly' toys as a youngster, and it serves to close the interaction -the topic then moves on to a more general discussion of childhood.

Interpreting the binary
If we consider the context in which this binary is produced, the idea that lesbians would dislike dolls is quite logical.Dolls are essentially toy babies, as Hannah implies, and have long been marketed at little girls to enable them to play at being mothers.In this sense, dolls can be seen to symbolise a very heteronormative ideal -that women have a maternal instinct and that girls will grow up to be mothers -and the women are constructing an apparently historical identification as lesbians by stating that they rejected them as children.
As discussed above, the Stomper women queered notions of womanhood through much of their practice together, such as through the rejection of practices that they positioned as 'girly.'The rejection of a symbol of motherhood -the epitome of ideological womanhoodis therefore unsurprising here.It can be very clearly connected to the ethnographic context of the group (in which such binary opposition was the norm) and the broader sociocultural context of an imagined lesbian community (in which a rejection of heteronormative practice is typical).In another lesbian group, it is likely that the construction of this binary would not necessarily occur, however.The Stompers' shared experience as middle-class, second-wave feminist women combines with their identity as lesbians to produce a very specific 'dyke' persona, one which involves resistance to the mainstream interpretation of the category 'woman.'The reference to dolls and teddies is quite unique to this group, then.A different CoP of women -those of a different generation, ethnicity or class, for instance -would not necessarily produce the same version of lesbianism as these women.For the Stompers, however, the construction of a childhood resistance to dolls affords them an essentialised authenticity as lesbians.
It may also be argued, of course, that the oppositional positioning of teddies to dolls is not entirely logical because it only makes sense within this specific context.If one were to draw a stereotypical binary between girls' toys and boys' toys, one might expect to find cars, trucks and toy soldiers opposing such items as dolls.Instead, in this moment, teddies are presented as being opposite to dolls.Teddies, however, are toys that are designed to be cuddled and played with in a similar way as dolls.Teddies are perhaps more gender-neutral, in that they may be given to boys and girls when they are very young, whereas dolls have a very clear link to girls' play, but they nonetheless would not typically be found as the binary opposite to dolls because they are not ideologically masculine.This highlights the fact that stereotypical lesbian identity, such as butchness, is not about masculinity.It is about a reworking of the category 'woman' and often involves the rejection of overtly feminine styles (see Jones in press for a more extensive discussion of this).Importantly, one can reject femininity without embracing 'masculinity' as it is so typically aligned with men, and this is revealed in this moment: instead of a toy like a gun or a sword, teddies are used to indirectly index a dykey persona.They are presented as the alternative to dolls for lesbian children, then, not because they somehow embody masculinity, but because they are not dolls.By taking a stance in favour of teddies, the women are able to take a stance against dolls, and it is this that enables them to project a contextually-specific, authentic lesbian persona.In this moment, then, working within the parameters that Claire has set for them, the women embrace an alternative to that which is stereotypically feminine and thus heteronormatively female, demonstrating that binaries exist on many different levels than simply 'male/female' or 'masculine/feminine'. Though it would perhaps be unusual to hear a woman claim that she was authentically lesbian because she liked teddies outside of this context (and therefore inaccurate to suggest that teddies are somehow symbolic of a wider lesbian or gay community), within this interaction such a claim enables the positioning of oneself as a dykey, legitimate Stomper.

Discussion
In the analysis of the interaction outlined above, I have claimed that the Stomper women's purported preference for teddies is significant only inasmuch as they are presented as the opposite of dolls.Nonetheless, the indexicality principle of Bucholtz and Hall (2005) is evident in that the women construct complex, indirect indexical links between lesbians and teddies, and between dolls and heterosexual females.Though the interaction begins with an opposition being constructed between lesbians and gay males, the more typical practice of the Stompers was to reject heteronormatively feminine cues and styles and, indeed, this is the main thrust of their identity work in this moment.The women here are engaged in the dialogic construction of lesbian authenticity via this rejection of heteronormativity.Through the joint production of an ideological binary, and the mutual taking of stances, which align them with an imagined, homogenous lesbian community, they index a dykey persona.
The fact that they do this dialogically is crucial to an understanding of how this identity work takes place, because the performance of lesbian authenticity in this moment occurs not as an individual pursuit but as a community endeavour.The fact that it is a supposed preference for childhood toys that is used by the women in constructing their identity is not as relevant here as the binary between what counts as genuine versus inauthentic.It does not, indeed, particularly matter if the women actually hold these apparently entrenched, historical preferences for teddies over dolls.What matters most is that they jointly construct, in this moment, a shared history which allows them to align themselves with a CoP-specific persona; the dyke, who eschews that which is feminine and girly, such as dolls.The concept of 'community' is central, then, as we observe the Stomper women as a collective with shared identity practices.This interaction must be understood, therefore, as part of their CoP practice and as a collective construction of a mutual sense of self.Through this, they are able to reify aspects of lesbian stereotypes and position themselves as a part of a wider, ideological, imagined lesbian community.
The positionality principle is also particularly clear in this extract, as the women's stance-taking positions teddies as temporarily indexical of authentic lesbian practice.By positioning lesbians firstly in opposition to gay males due to their supposedly inherent femininity, and then later in opposition with heteronormative femaleness itself, the women in this interaction are able to position themselves as authentic lesbians.This makes sense only within this interaction, however, reflecting Bucholtz and Hall's emergence principle; the mutual personas that they are producing are borne out of the specific, fleeting conversation occurring at that time.The relationality principle is highlighted, too; the indexical link between teddies and lesbians only exists because of broader ideologies surrounding girls and babies, and because of the ideological binary that the women are constructing for themselves as lesbians versus heterosexual women.Without the ideological link between dolls and heteronormative femininity, and the additional ideology of lesbians as being somehow gender inverted, teddies would not hold the symbolic meaning that they do here.
In this sense, Bucholtz and Hall's argument that identities are only ever partial is also apparent.Evidently, these principles stand as a useful way of explaining how a group of women's brief claiming of a preference for teddies enables them to produce an authentically lesbian persona.Importantly, these conclusions may only be drawn due to our awareness of the women's membership to the Stomper CoP.The construction of this indexical relationship may only be explained in light of the typical practice of the group -in this case, the construction of binary categories.By linking the identity work being undertaken by the CoP members in this one, fleeting moment to the sociocultural principles of Bucholtz and Hall (2005), it is possible to analyse the interaction both in terms of the typical ethnographic context of the Stompers, but also in light of the broader cultural ideologies making up the imagined lesbian community.This demonstrates that the particular stances taken or roles assumed in a given interaction need not be long-lived or repeated, but must instead be relevant to the overall mutual identity work that results from shared engagement in joint practices.

Conclusion
In this article, it has been argued that 'the gay community' can, after all, be a useful concept in studies of language and sexual identity research.Whilst a homogenous gay community itself might not exist in any tangible sense, it has been argued here that the community of practice model allows us to analyse how locally-based communities of LGB speakers may draw on and rework discourses from this imagined construct in order to produce culturally-salient personas.It is possible to interpret the shared aspects of identity construction between a group sharing a sexual orientation in light of their typical practices, then, without making assumptions about how they identify as individuals.Rather than disconnect our research with LGB groups from the ideological construct that our participants may identify with, then, scholars of language and sexuality may benefit from actively questioning the relevance of an imagined community to those producing identities associated with it, and indeed by distinguishing between the locally-salient personas and the broader, cultural identities that are invoked within CoP interaction.Through the use of the sociocultural linguistics framework, we can do this in a way that explicitly recognises the imagined nature of the gay community, focusing on how members of local CoPs draw on discourses from it by looking at the indexical meaning behind their interactional moves and stances.By focusing on an imagined gay and lesbian community, in the form of a global, sociocultural context, as well as the ethnographic context of a local CoP, and by investigating through close discourse analysis the ways in which identities are indexed and constructed through a variety of linguistic and non-linguistic practices, it is therefore possible to draw links between language and sexual identity.
Though this article has addressed some of the concerns and critiques of the CoP construct, it has also demonstrated that the qualitative, interactionist employment of the model allows researchers interested in language and sexuality to do a number of things productively.It enables us to look at local groups as they construct themselves in relation to broader ideologies associated with 'the gay community.'Future language and sexuality research which makes use of the CoP model might focus on how ideologies of sexual attractiveness and desire, authentic queer citizenship, or homonormativity, for example, are negotiated between queer speakers in local contexts.Discourse analysts hoping to understand how discourses of homophobia or heterosexism impact on the identity construction of LGBT speakers might also benefit from focusing on queer CoPs.The construct allows us to see how such discourses are resisted, embodied, reworked and negotiated in order to produce a mutual identity, which is meaningful within a given context.In other words, the use of the CoP allows language and sexuality scholars to consider small, local groups of queer people in relation to a broader cultural context, rather than as isolated groups with no connection to something bigger than themselves.This enables us, in turn, to explain similarities in practice that might be found across otherwise unconnected groups of queer speakers, and to understand the stereotypes and ideals which both facilitate and constrain their identity construction.The combined use of the sociocultural linguistics framework with the CoP also assists us in drawing these connections, as concerned as sociocultural linguistics is with not just the local community context, but the ethnographic setting of an interaction and the interaction itself.By explicitly linking the practices of a CoP to an interaction such as that outlined above, and by recognising the indexical relationships between those practices and broader sociocultural ideals and values, we can therefore return to a focus on communityboth in local and imagined terms -in language and sexuality research.
could be gay (.) the one who likes (.) the dolls.
And they'd they'd look at you like that <pulls face>] S [you? <To Lucy> Dolls or teddies?]L Teddies.