Aims of the study

The Communities of Practice framework, originally developed by Jean Lave and Etienne Wenger (Lave &Wenger Reference Lave and Wenger1991; Wenger Reference Wenger1998), was adopted by sociolinguists working within the discourse constructionism paradigm (Eckert &McConnell-Ginet Reference Eckert and McConnell-Ginet1992, Reference Eckert and McConnell-Ginet1999) because it firmly grounds identity in social practice and can cater to a range of identity claims and constructions that have not been assigned a priori to interactants (Bucholtz Reference Bucholtz1999). The concept refers to a group of individuals who come together to perform certain activities and who, through the development and establishment of linguistic and social practices, come to constitute a community (of practice).

Lave & Wenger's initial conceptualisation of the framework is based on notions of commonality and homogeneity and includes references to common goals and shared repertoires. Holmes & Meyerhoff (Reference Holmes and Meyerhoff1999) explored variation within communities of practice and discussed the value of correlating frequency of occurrence of linguistic features and individual degree of integration into a community. Based on this assessment, members can be core, peripheral, or marginal (Wenger Reference Wenger1998). Despite identifying heterogeneity within communities of practice, this type of categorisation, based on quantitative analysis of individual language use, offers a static representation of a community.

Other sociolinguistic studies have challenged, or revised, the centrality of unity. Eckert (Reference Eckert, Warner, Ahlers, Bilmes, Oliver, Wertheim and Chen1998) explored how positions within communities can be co-constructed by individuals and their interactions with, and reactions to, their respective groups. Rock (Reference Rock, Barton and Tusting2005) focused on a specific activity in an institutional setting—the delivery of the right to silence by officers in the police force—suggesting that a “complex of shared and different cautioning practices and representations of those practices makes for intriguing challenges for the community of practice position” (Rock Reference Rock, Barton and Tusting2005:100).

Increasingly, researchers are paying more explicit attention to individual participation in communities of practice, as the well-boundedness of linguistic and social concepts (including identity and language) is progressively questioned, especially in the light of increased mobility and technological advances. This article develops this perspective, and further challenges “assumptions about homogeneity which remain implicit in much communities of practice work” (Barton &Tusting Reference Barton, Tusting, Barton and Tusting2005:8), by examining individual practices amongst multilingual British-born Greek-Cypriot speakers, members of a youth community group in London. The study offers a unique point of entry into research on individual language use within communities of practice, by providing an in-depth comparison of language use of the same individuals in two distinct contexts: official youth-group meetings that take place every month at a community centre, and an extended social event—a dinner, which took place in an Italian restaurant. The study also identifies the extent and nature of gender variation in each context.

This article considers how individuals use the same resources to engage in identity practices, though in different ways and to different extents. It stresses the importance of acknowledging, identifying, and accepting different ways of participating in a community, without making decisions regarding core or peripheral membership. It problematizes and further develops existing concepts used to describe variation within communities of practice, including core and periphery and their associations with full and partial participation. It also offers a more enhanced, or different, perspective on the key concept practice, stressing the importance of focusing on both production of speech, but also silent participation in constructions of ethnic identity (for example an acknowledgement of cultural personas and behaviours, rather than active participation in teasing sequences or joke-narration).

This study contributes to the new sociolinguistics of multilingualism (Gardner & Martin-Jones Reference Gardner and Martin-Jones2012)—an area of study that questions identity ontologies, deconstructs boundaries between languages, embraces process-orientated views of identity, and describes identity as situated practice—by exploring the relations and tensions between individuals, code-switching (CS) practices, larger community values, and discursive identity (re)constructions in a multilingual, multicultural, and heterogenous community of practice. Frequency tests identify whether speakers use the Greek-Cypriot dialect and code-switch, to different extents and for different (socio)pragmatic purposes. As we see below, a consideration of language use in different contexts allows access to a wider range of situated practices, enabling a more comprehensive account of differences observed. Complex interactions transpire between a range of local and global variables including genders, community-specific setups and contexts, individual preferences, discourse roles, and discourse types, further demonstrating the importance of an ethnographic approach in the study of identity and language use. While this community of practice does not constitute a uniform entity, each individual contributes in different ways, both quantitatively and qualitatively, to the construction of their community of practice, and ultimately their common cultural identity. At times these contributions, however, seem to be shaped within parameters set by wider community values and (gendered) expectations.

British-born Greek-Cypriots: A community of practice

Eckert & McConnell-Ginet (Reference Eckert and McConnell-Ginet1992) introduced the Community of Practice framework into sociolinguistics and provide the following definition:

Examples of such communities include choirs, friendship groups, and academic departments. Lave & Wenger (Reference Lave and Wenger1991) regarded the framework as a natural and productive environment that could enhance the learning process. They considered learning a social process achieved through participation in community-centred activities. The framework was seen as an essential import into the study of language and gender as it addresses the fact that gender is not an independent phenomenon but something that needs to be studied within social practices (Eckert & McConnell-Ginet Reference Eckert and McConnell-Ginet1992).

Wenger identified three main dimensions considered to be the basis for the creation of a community of practice (Wenger Reference Wenger1998; Holmes & Meyerhoff Reference Holmes and Meyerhoff1999). First, mutual engagement is required: regular interaction between speakers (e.g. meetings, encounters in corridors). Second, speakers need to be part of a jointly negotiated enterprise characterised by features such as shared goals, contributions from participants during on-going processes of negotiation, an understanding of roles of individuals, and mutually defining identities. Third, speakers develop a shared repertoire: linguistic resources and routines that accumulate over time and are subject to negotiation. Wenger (Reference Wenger1998:125–26) cites as examples of constitutive features of a community of practice: shared stories, insider jokes, knowing laughter, styles recognised as displaying membership, and a shared discourse that reflects a common outlook. As speakers spend more time together, they share an increased amount of information and knowledge and through established ways of interacting develop a common sense of identity.

The group of speakers in this study clearly exhibits these three dimensions of Communities of Practice as identified by Wenger (Reference Wenger1998). First of all, the speakers regularly interact with one another in a variety of situations including youth group meetings and Greek parties. Second, the speakers are part of a joint-negotiated enterprise: they have shared goals relating to the general objectives of the organisation (e.g. to keep the younger generation in contact with their cultural and ethnic roots). During their meetings, the members of the youth group contribute to different extents, and in different ways, to this enterprise (to be discussed in detail below). Each member has a role: official (e.g. president, secretary, etc.) and/or unofficial (e.g. main provider of jovial atmosphere). Finally, as this study demonstrates, a range of teasing sequences and disagreements are consistently employed by speakers to discursively (re)construct their position vis-à-vis the older Greek-Cypriot generation and British society at large. Their linguistic practices, which form a shared repertoire, enable the participants to exhibit familiarity with, and knowledge of, Greek cultural associations and images, and establish themselves as members of the Greek-Cypriot community (see also Georgakopoulou & Finnis Reference Georgakopoulou and Finnis2009).

The Community of Practice framework is most useful for the specific group of speakers as it enables the identification, description, and exploration of new communities and identities that reside outside existing formal social structures (Bucholtz Reference Bucholtz1999). These communities occur within particular domains such as the classroom and the workplace, but “can be seen as distinct from the formal structures of these domains” (Barton & Tusting Reference Barton, Tusting, Barton and Tusting2005:2). These speakers, through encountering each other on a frequent basis, create their own ways of speaking and interacting, their own community of practice. The participants in this study are living a different sociocultural experience compared to the older members of the Greek-Cypriot community. They are also leading a life that incorporates a culture and language other than English. They do not construct meaning and identity simply through the reproduction of existing inherited (Greek-Cypriot and British) sociocultural structures. In fact, at times they overtly express disagreement and discontent with many aspects of both Greek-Cypriot and British cultural and behavioural patterns (Georgakopoulou & Finnis Reference Georgakopoulou and Finnis2009). Instead, they reappropriate given cultural knowledge and structures locally during their interactions, and by coming together every month, they are relearning and reproducing what it means to come from a Greek-Cypriot background and live in London.

The Greek-Cypriot community in London: Politics, language and identity

Identity is an especially flagrant and complex issue in the Greek-Cypriot community in the UK due to its highly politicised make-up. The composition of the community includes an interplay of elements associated with Britishness on the one hand, and the at times turbulent relationship between Standard Modern Greek and the Greek-Cypriot dialect (GCD) on the other.

Migration to the UK from Cyprus started early in the twentieth century, flourishing in the 1950s to 1970s due to social and political turbulence in Cyprus—the result of the struggle for independence from British rule in the 1950s and the Turkish invasion in 1974. Britain was chosen as a host country because of work opportunities and the “colonial connection” (Josephides Reference Josephides, Rex, Joly and Wilpert1987:43). Greek-Cypriots in Britain are highly involved with the political situation in Cyprus relating to the relationship between Cyprus, Greece, and Turkey. Many individuals within the community are members of political parties that hold different views towards the relationship between Cyprus and Greece.Footnote ¹ Organisation of, and participation in, political marches and demonstrations relating to the invasion of 1974 are very common.Footnote ²

Newton (Reference Newton1972) suggested that the intelligibility of GCD to speakers of Standard Greek is comparable to that of Dutch to speakers of German. GCD comprises various subdialects that may deviate from standard GCD to such an extent as to become unintelligible to someone living in a main town. When the more standard form of the dialect is spoken, it is easy for the average speaker of Standard Greek to identify familiar lexical items and grammatical constructions. The Greek-Cypriot community of Nicosia, the capital, can be described as diglossic: the standard, used in the media and in parliament, constitutes the official government language spoken in formal domains. In order to have access to education, it is imperative to have a good command of both written and spoken Standard Greek. GCD is restricted to interaction in informal domains such as the family, and has no overt prestige. It has no official written form, although poetical works have been written in the dialect (Kitromilides Reference Kitromilides1981).

The first immigrants to arrive in London spoke only GCD, their competence in the standard being limited due to a lack of education. Subsequent generations were brought up learning GCD as their mother tongue, some having the opportunity to learn standard Greek at Greek schools run by the Greek-Cypriot Education Mission and the church. As in other migrant contexts (see Sebba & Wootton Reference Sebba, Wootton and Auer1998 for a discussion of British-born Carribeans), the British born Greek-Cypriot speakers' fluency in the ethnic variety varies. Greek-Cypriot families living in London are exposed to the dialect in domains exclusive to the community: satellite television, radio, banks, community centres, restaurants, and nightclubs. Many members of the older generation are linguistically reliant on the community, having never acquired English as their lives revolve(d) around community activities and services.

Like all other immigrant communities, the Greek-Cypriot community is changing as time progresses. As with the Chinese community in Newcastle (Milroy & Li Wei Reference Milroy, Wei, Milroy and Muysken1995), the London community constitutes a close-knit network providing support for families to integrate to a greater extent into urban life. Anthias (Reference Anthias1992) suggests that individuals who complete higher levels of education may detach themselves from Cypriot identity and social life. As Greek-Cypriot families in London become more financially able, they make education for their children a priority (Christodoulou-Pipis Reference Christodoulou-Pipis1991). Members of the younger generation are no longer expected to take over the families' businesses, but can secure higher-status jobs and have developed contacts outside the community.

Many researchers have touched on the issue of attitudes towards the use and preservation of the dialect, especially amongst members of the younger generation who are dominant in English. Papapavlou, Pavlou, & Pavlides (Reference Papapavlou, Pavlou; and Pavlides1999) challenge any suggestions that the younger speakers are experiencing an identity crisis (the result of living within two cultures). Kelly (Reference Kelly and Liebkind1989:80–84) observes of second generation Greek-Cypriots and Muslim Pakistanis in London that they “use the fusing of such ideologies as a source of both strength and potential for an ongoing process of ethnic redefinition”. Gardner-Chloros, McEntee-Atalianis, & Finnis (Reference Gardner-Chloros, McEntee-Atalianis; and Finnis2005) suggest that members of the community in London regard both Standard Greek and the Greek-Cypriot dialect as part of their cultural heritage and want to preserve these varieties; however, they do not regard them necessary for involvement in the Greek-Cypriot community. Their findings also showed that members of the community see each variety they speak as fulfilling different functions and thus as having the capacity to co-exist within the community.

The Greek-Cypriot youth organisation

Data consists of recordings of natural speech of members of a Greek-Cypriot youth organisation during their monthly meetings in North London and during one post-meeting dinner that took place in an Italian restaurant. Recordings of eight meetings were carried out throughout 2001, each roughly two hours long. The dinner lasted for just under three hours and took place in honour of a visiting fellow Greek-Cypriot from the equivalent committee in the United States. Apart from constituting different contexts, the two types of events also contain different discourse types: for example, arguments (and the code-switching strategies associated with them) only take place during the meetings, whereas the activity of joke-telling only takes place during the dinner. I attended the meetings as a silent observer, and the dinner as a participant observer.

The group consists of a president, vice president, secretary, treasurer, and the thirteen members of the executive committee. These positions are the result of elections, and can only be occupied by Greek-Cypriots between the ages of eighteen and thirty. While Menelaos, the president, is the main speaker, the meetings are relatively informal, consisting of interplay between formal meeting discourse and more informal teasing and chatting. The ages of the participants are between twenty-one and twenty-nine. Fourteen individuals attended the meetings more or less regularly. Out of these, the president, Menelaos, and the secretary, Lila, were the only two that were present for every recording (eight meetings and one dinner). The twelve remaining members are six males and six females, whose attendance at the meetings varied.

The respondents are nearly all university graduates and most of them work in banks/accounting, a common profession in the community for the younger members. Some of the parents themselves work in banks while others own restaurants. Overall, however, parent occupations are diverse, while the youths are more homogenous in this respect. The members of the youth group sometimes socialise in contexts other than the meetings, although this is not regular, because they are busy with their own professions. Despite the fact that overall they do not socialise frequently, their closeness to each other is clear: a substantial amount of ritual teasing and humorous exchanges take place. Newcomers are treated as if they have been members for years and are made to feel part of the group from their very first meeting. Some have been involved with the Federation for many years and thus regard themselves as being very well acquainted. In addition, some of the male participants are friends through sporting activities sometimes organised by the youth society itself.

Ethnicity is by default a collective point of reference in this cohort as the purpose of the meetings is to discuss issues and organise activities that centre around the community. The core aims of the organisation are to bring together the younger generation of the community, to educate them with regard to the political situation in Cyprus, and to preserve and maintain cultural and ethnic roots and identity. One of the tasks of the youth committee is to notify its members of any political demonstrations taking place in London, encouraging them to participate in order to sustain awareness of the past and present political developments in Cyprus. The organisation, being part of a global Greek-Cypriot youth organisation (NEPOMAK—World Federation of Overseas Cypriot Youth), constitutes the UK representative at conferences on global Greek-Cypriot youth. The point of these conferences is to discuss issues related to Greek-Cypriot youth around the world, to seek solutions to any problems, and to nourish relations between the communities at the global level. Another important objective for the youth group is to put together social events in London, including an annual party at a night club and community football tournaments.

Coding the data

In this article code-switching refers to “the juxtaposition within the same speech exchange of passages of speech belonging to two different grammatical systems or sub-systems” (Gumperz Reference Gumperz1982:59). Similarly to De Fina (Reference De Fina2007), code-switching is used as “an umbrella expression for different phenomena including word insertion and intrasentential and intersentential switching” (2007:380).

Frequency tests were carried out in order to identify whether speakers (a) use more or less Greek-Cypriot than each other, (b) code-switch more or less than each other, and (c) switch to Greek-Cypriot for different (socio)pragmatic purposes. Therefore the data was coded in order to identify the extent of use of GCD and code-switching, and the (socio)pragmatic force of each switch, that is, to ascertain whether the switch occurred when the speaker was being humorous or amidst an argument. (Socio)pragmatic force here refers in a general sense to the interactional function of the switch (e.g. whether humorous, argumentative, reproachful), which may influence or redefine the relationship between the interactants. Sociopragmatic force, then, can have an interpersonal function that at times may be at odds with the propositional goal (Thomas Reference Thomas1995): for example, a speaker may be reproachful, but the reproach, embedded within humorous discourse, can function to enhance interpersonal relations. Such an analysis helps us understand how speakers engage with the same resources, but in different ways and to different extents.

The following were used as a means of identifying the (socio)pragmatic force of an utterance (a) identification of appropriate verbal and nonverbal cues both on behalf of the speaker and the hearer (see also Pichler Reference Pichler2006), (b) consideration of whether propositional content of contributions are supported or challenged, and (c) sociocultural background knowledge of the researcher (see also Habib Reference Habib2008). An utterance can have more than one pragmatic function (Boxer & Cortes-Conde Reference Boxer and Cortes-Conde1997; Holmes Reference Holmes2000; Norrick Reference Norrick2003; Pichler Reference Pichler2006; Rogerson-Revell Reference Rogerson-Revell2007). Therefore, sometimes the same utterance was coded, for example, as both humorous and also direct. The following functions of code-switching were identified and coded: humour, directness, reproach, arguments, and mild disagreements.

Humour relates to all nonserious utterances (including teasing and joke-telling). The cues used to identify such discourse include laughter, exaggerated intonation and accent, formulaic expressions, and nonverbal cues such as exaggerated arm movements. Cultural knowledge was also employed when deciding whether an utterance was playful or not (e.g. knowledge of certain lifestyles and behavioural patterns that are at odds with the general tone of the meeting and that contribute to the creation of a playful frame). Example (1) is coded as humorous. During one of the meetings, Claudia is mocking the Greek-Cypriot tradition of arranged marriages by suggesting that the youth group should include the theme global arranged marriages at a global conference they are organising.Footnote ³

1. (1)

   1 Claudia: I know what, why don't we do a global proxenia [‘arranged marriages’] xx people

   from England xx people from Australia?!

   ((All laugh))

Directness, which describes candour and also aggression (see also Boxer & Cortes-Conde Reference Boxer and Cortes-Conde1997; Norrick Reference Norrick2003), is identified as such by raised volume, disruptive interruption, use (or not) of redressive strategies, and content of the utterance. The reaction of the hearer was also taken into consideration when deciding whether an utterance, even if potentially humorous, had a bite (Boxer & Cortes-Conde Reference Boxer and Cortes-Conde1997). In example (2), Poly inserts a direct GCD utterance amidst English discourse during one of the meetings, ordering the participants to hurry up and terminate the meeting.

1. (2)

   1 Menelaos: Hello!

   ((To Poly who was not paying attention))

   2 Poly: Ne, signomi! Ade telione. In na vgoume exo!

   ‘Yes, sorry! Hurry up and finish. We are to go out!’

Reproach was also identified, sometimes coded alongside directness, to describe an utterance that contained an explicit criticism or reproof of someone whether present or absent. In the following utterance, Themis uses GCD to critique and undermine the attitudes of the older generation by making reference to their origins, which he assumes reflects their outmoded opinions.

1. (3)

   1 Themis: At the end of the day they come from horka [‘villages’]

Finally arguments and mild disagreements were identified, with the latter reflecting some form of dissent/discordance on behalf of the speaker when expressing difference of opinion or when correcting the previous speaker. The decision to code an utterance as mild disagreement rather than argumentative was based on a consideration of tone and style of speaker (i.e. no raised volume or disruptive interruption), propositional content, and the reaction of the hearer (in cases where the disagreement was considered “mild” the hearer accepts the speaker's utterance without arguing back). Example (4) is an on-off disagreement where Menelaos switches to GCD to defend himself against Lila's direct utterance. This is considered “mild” as it is not a prolonged exchange; it also contains specific “over-the-top” expressions, including a taboo item (“hell”) and the exaggerated Greek expression “There's no need for you to eat me”, which seem to mark the exchange as informal and mitigate any potential threat.

1. (4)

   1 Lila: I've got the old banner

   2 Demetris: Guys, I wouldn't do a new banner if we're having mega changes and things coming up

   3 Menelaos: Is it in one piece? ((To Lila))

   4 Lila: Of course it's in one piece! What the hell do you think I did with it!

   5 Menelaos: Endaxi, eroto mana mou! Then in anangi na me fais, kiolas!

   ‘Ok, I'm just asking (“my mother”)! There's no need for you to eat me!’

   ((Change of topic follows))

In example (5), which demonstrates an argument, Menelaos and Claudia are discussing the reason why an email was sent out too late to inform people of a meeting (EFEKA and the Protoporia Committee are organisations within the overall Greek Cypriot Federation).

1. (5)

   1 Claudia: That's because that speech wasn't by EFEKA that speech was organised by LSE Cypriot society, we had nothing to do with it, we were notified later, that's why the email went out late...

   2 Menelaos: So? Then eshi, then eshi kanenan mes to simvoulio tis Protoporias pou pai sto LSE?

   ‘Isn't there anyone, isn't there anyone from the Protoporia committee that goes to LSE?’

   3 Claudia: That's got nothing to do with it because everything still has to get passed through...

   4 Menelaos: [Anyway.

In this slightly more heated exchange, the style used (compared to example (4)) is more formal and the exchange is slightly more prolonged. Menelaos switches to GCD in line 2 to challenge Claudia's reason for the lateness of the email. He subsequently dismisses her clarification in line 4 with the utterance “anyway”.

Individuals, code-switching, and (socio)pragmatic meaning

The results indicate that while all speakers incorporate GCD in their speech at some point in their interactions, individuals do it to different extents in each context and for different reasons. Menelaos, the president, stands out as dominating the interactions to a large extent (he dominates just over a third of utterances transcribed: 35.14%), which is expected, especially in the meetings, due to his position as chair. He also uses more GCD than anyone else, as he utters over 40% from the total of GCD uttered by all speakers (in fact, Menelaos' use of GCD is such that his speech can actually provide information on his fluency in GCD, something not always possible to deduce from the speech of others). These findings initially seem to suggest that Menelaos is a core, if not the core member of this community of practice, due to his extensive use of the minority variety, a variety that one could argue, directly indexes a we identity (Gumperz Reference Gumperz1982).

However, as my findings indicate, while he dominates the meetings due to his role as president, he does not dominate other domains associated with this community of practice: other participants dominate particular practices, such as joke-narration and teasing sequences, during the dinner. These findings suggest that it is not fruitful to define boundaries between core and peripheral members before carrying out more in-depth analyses of individual language use in more than one single context. Communities of practice extend into more than one domain, and it is important to study language use in more than one setting in order to access a fuller range of practices used.

Findings indicate that, apart from producing different amounts of GCD, speakers also code-switch to different extents and for different purposes, with Menelaos code-switching to GCD substantially more than the other individuals (see Figure 1; Table 1 contains information on names and gender). It could be the case that the more one talks, the more one code-switches. No correlations were found, however, between quantity of code-switching and overall linguistic output. Instead, as we see below, other variables including discourse type are more directly responsible for differences observed.

Figure 1. Individuals and amount of switches to GCD.

Table 1. Participants' names and gender.

An examination of functions and sociopragmatic significance of code-switching indicated that individuals differ as to when and why they bring GCD into their interactions (see Table 2). Menelaos seemed to use code-switching for all sociopragmatic functions identified, and to use each one to a greater extent than the other speakers. Most individuals used GCD more for humorous purposes than for attacking aspects of the interlocutor's face. This was not the case for all speakers, however; for example, practically the only use of GCD for Luka involved directness, reproach, a rude word, argument, and disagreement. By contrast, both Aphrodite and Apostolis use GCD almost exclusively for humour: neither use GCD to engage in argument or disagreement. Vaggelis uses GCD both for humour and for directness, argument, and reproach. In sum, all participants use GCD/CS to engage in a set of social practices—practices that include humorous and argumentative discourses and that characterise this cohort of speakers as a community of practice. However, the use of GCD, or CS, has different sociopragmatic significance for each speaker as participants engage with these practices to different extents and in different ways.

Table 2. Individuals and sociopragmatic functions.

Individuals and contexts

A comparison of language use in two different contexts can provide a unique insight into the practice and the function of code-switching. A greater use of GCD occurred during the dinner: 37% of the total Greek-Cypriot words uttered in all nine data sets were used for the dinner only. So just over one-third of the community variety was produced in one single recording. This result at first glance seems to confirm findings from studies that interpret language choice in the context of fixed sets of social and situational factors, such as use of the community language to index ingroupness when the situation is more informal (such as a dinner). A more in-depth examination of data, however, suggests that a combination of other variables, including gender and discourse type, are more directly responsible for the increased use of the minority variety in the context of the dinner.

Despite the fact that proportionally more GCD words were used for the dinner, results indicate that fewer switches (to GCD) took place: during the dinner nine switches to GCD occurred every 100 GCD words, while in the meetings sixteen switches to GCD occurred every 100 words. In conclusion, when the participants were dining, they used a larger amount of GCD proportionally to the meetings, but code-switched almost half as much, that is, the switches were fewer, but longer: each utterance contained a longer stretch of GCD.

These findings can be directly related to the types of discourse present in each context: the meetings contain arguments and disagreements and the dinner contains more jovial discourse, including a series of joke narrations. However, more significantly, detailed analysis shows that this finding is not consistent across individuals. Menelaos, Lila, and Vally display a substantial decrease in their use of the Greek-Cypriot dialect in the context of the dinner: Menelaos used 79% of his total GCD usage in the meetings, and both Vally and Lila only used GCD during the meetings (they both attended the dinner). These findings suggest that not all speakers automatically increase their use of the ethnic variety in informal contexts (see Figure 2). Menelaos engages in argumentative discourse during the meetings, which increases his own use of GCD and CS for this data set as he skilfully employs GCD and CS in various ways to construct, and respond to, opposition.

Figure 2. Individual use (percentage) of GCD in each context.

In contrast to these findings, results from a consideration of language use of other participants (who, interestingly, are all males; see the discussion in the section below) show a substantial increase in GCD usage. For example, Apostolis and Harry's usage of GCD for the dinner comprises 99% of their total use of GCD in all recordings respectively. The reason for this is that they were both engaged in the activity of joke-telling, which was prominent during the dinner. This activity seems to constitute an important practice in this cohort of speakers. The jokes are based on cultural knowledge, such as knowledge of stereotyped behaviours, and establish rapport between the speakers who are familiar with these cultural behaviours and stereotypes; while serving to bond the speakers, the process of narrating jokes provides a means of mocking Greek/Greek-Cypriot personas, and repositioning the younger participants with respect to traditional aspects of the Greek-Cypriot culture (see also Georgakopoulou & Finnis Reference Georgakopoulou and Finnis2009).

Joke narration in this group is a practice that requires increased use of GCD overall. Switches are used to structure jokes (also found in Woolard Reference Woolard and Heller1988), with the punchlines sometimes being delivered in a different variety, or used to import cultural images that can create incongruous, and hence humorous, frames (Georgakopoulou Reference Georgakopoulou, Georgakopoulou and Spanaki2001), and confirm solidarity and ingroupness (Stolen Reference Stolen1992). An example of a typical joke produced during the dinner is given in (6) below.

1. (6) Shut the door

   1 Menelaos: I've got some funny xxx I've got this one. I think this is courtesy of Luka. A sensitive Cypriot says to his wife “re gineka evarethika na se thoro mes ti kouzina na vasanizese. Klis tin porta.” [‘(hey, you) wife I am tired of seeing you in the kitchen torturing yourself. Shut the door’]

   ((All laugh))

   2 Aphrodite: A panagia mou! [‘Oh my Virgin Mary/oh my’] Horrible!

Menelaos, after introducing and initiating the joke in English, switches to GCD to represent the voice of the Greek-Cypriot husband. One of the females present, Aphrodite, responds to the joke uttering a typically Greek expression, ‘My Virgin Mary,’ used commonly by Greek speakers to express shock or disapproval. She further enhances her disapproval by switching to English to assume a more objective voice, again of disapproval, describing the joke as “horrible”.

Gender

Findings from initial analyses suggest that in this community of practice, the male speakers use more GCD in their speech than the female speakers. Even with Menelaos' results omitted (having uttered 42.37% of the total Greek-Cypriot words uttered in all data sets, he could be considered an outlier skewing the overall results), the female speakers still only produce just over half the speech uttered by the males: the male speakers uttered 53% of the total of Greek-Cypriot words uttered by both sexes, and the females 28%.Footnote ⁴

A consideration of language use during the dinner revealed that only one female, Aphrodite, used GCD, though she used it only slightly more than one of the male speakers. The other two females present, Lila and Vally, spoke only in English. In this particular context of the dinner, the male participants engaged extensively in joke-narration, and CS and GCD were used in various ways and for various reasons in this type of activity, including structuring discourse and adopting cultural personas. The females in this study seem to refrain from this practice and, as we see below, tend to use GCD for different sociopragmatic reasons.

The findings support research that suggests that in some communities, women do not use monological humour—joke-telling—as much as men (Bryant, Comisky, & Zillmann Reference Bryant, Comisky; and Zillmann1979; Boxer & Cortes-Conde Reference Boxer and Cortes-Conde1997). It is important to note that the jokes produced in the current study are obscene, revolving mainly around sexual intercourse, and can be considered derogatory to women. Kaplan (Reference Kaplan and Cameron1998) suggests that it is taboo in many cultures for women to narrate jokes, especially obscene ones. One would expect that the narration of jokes that celebrate the male as a prolific sexual entity would be more common amongst males. It could be argued, moreover, that women, especially Greek-Cypriot women who come from relatively traditional backgrounds, may choose not to contribute to such sexist and audacious discourse. This suggestion receives some support from Boxer & Cortes-Conde who suggest that it is more common for males to “tease about bodies”, including those of others, because “For women … to tease about bodies is to touch something that we have been trained to take seriously” (1997:291). Indeed humour, as a monological act of individuals, seems to be an almost exclusively male domain in this community of practice, with the females not expected to actively participate in the construction and maintenance of ingroupness through this practice.

Despite the fact that the women in this particular group do not narrate jokes, they play a key role in the process by constituting a (mock) disapproving audience for the males, thus providing them with a stage, and participating in the acknowledgement of familiar cultural frames. During the process of joke-narration, female and male roles are in complementary distribution with the women and men being passive and active respectively (Coser Reference Coser1960). Therefore, humour is used to construct, and also maintain, gendered roles (Crawford Reference Crawford2006).

While indeed females do not narrate jokes (in this particular community of practice) they engage, equally or slightly more than the men do, in other types of discourse that are mediated to a large extent by the minority variety: directness and expressions of reproach, and hence actively participate, albeit differently, in identity practices in this community (see Table 3 which contains Menelaos' results in a separate column to those of the other males, as his results are significantly higher than any other speaker). Indeed, Tannen & Kakava (Reference Tannen, Kakava, Georgakopoulou and Spanaki2001) focused on disagreement in the Greek culture and suggested that in this society, disagreement is frequently accompanied by markers of solidarity, for example, use of affectionate forms of names at disagreement points. They conclude that expressing disagreement can actually have a solidarity function in itself in the Greek culture. Therefore, a switch to a language that is more tolerant of directness (Sifianou Reference Sifianou1992; Tannen & Kakava Reference Tannen, Kakava, Georgakopoulou and Spanaki2001) can potentially soften the impact of the speech act enabling the female speaker to “get away with it” without coming across as being inappropriately rude or direct (Gardner-Chloros & Finnis Reference Gardner-Chloros and Finnis2003). In this study, the female participants seem to be “allowed” to produce this discourse type, which, on par with humorous discourse include joke-narrating, constitutes an important practice in this community. Therefore, female participation in this community of practice is restricted by wider community setups and cultural norms.

Table 3. Gender and sociopragmatic function.

Excerpt (7) is an example of a switch to GCD when being direct. The speakers are deciding upon a location for a day trip to take place during the global Greek-Cypriot youth conference in Nicosia. Lemesso (Limassol in English) and Protara are both locations in Cyprus. In this example, Claudia questions and dismisses her interlocutor's suggestion of going to Limassol for the beach trip, and uses the Greek expression mana mou, which is commonly used in disagreements in the Greek culture and, while literally translated as ‘my mother,’ is roughly equivalent to the English ‘mate.’

1. (7) Beach trip

   1 Claudia: You know they do, you know they gonna do the trip, they do a whole day trip xxx last day, can they do can they do a beach trip.

   2 XXX: Yeah, Lemesso!

   3 Claudia: No, not Lemesso, mana mou xxx pame sti Lemesso xxx.

   Protara! [‘my mother xxx let's (not) go to Limassol xxx Protara!’]

In this example, English and Greek-Cypriot combine to achieve directness. English is used to initially index the utterance as oppositional through the use of “No”. Claudia then switches to GCD to utter mana mou, which is commonly used in arguments in Greece. Indeed, Tannen & Kakava (Reference Tannen, Kakava, Georgakopoulou and Spanaki2001) mention that in Greek interactions, the use of the particle re sometimes accompanies a lexical item such as ohi ‘no,’ which is most closely association with disagreement, resulting in the construction ohi, re ‘no, you.’ While Claudia does not utter re here, she does switch to GCD after uttering “no, not Lemesso” in English, and produces a similar item (to re): mana mou. She then proceeds to specify her own preferred destination in GCD. Claudia is thus being impolite (Culpeper, Bousfield, & Wichmann Reference Culpeper, Bousfield; and Wichmann2003; Culpeper Reference Culpeper2005) attacking different aspects of her interlocutor's face, including his quality face: his desire to be valued in terms of his competence and his desire to preserve his self-esteem (Spencer-Oatey Reference Spencer-Oatey2002; Cashman Reference Cashman2006). She is, however, using Greek linguistic markers when expressing disagreement and is thus skilfully employing both varieties to ensure that rapport is maintained.

Similarly to joke-narration, being direct and expressing reproach is used by the speakers to establish and maintain rapport, and gives the female speakers the opportunity to actively participate in the process of constructing and maintaining ethnic identity. Therefore, men and women in this study seem to contribute to the construction and maintenance of their ethnic identity in different ways, with men engaging in teasing sequences and joke-narrations, and women in expressing reproach and directness. Both are exhibiting their familiarity with common cultural elements whilst at the same time adhering to cultural norms and gendered expectations in the community.

While this study does identify associations between particular discourse types and genders, these associations are not definitive. For example male speakers also switched to GCD to engage in argumentative discourse. Therefore, this study is not dismissing the suggestion that discourse type, rather than gender, can be directly responsible for differences observed in language use (Freed & Greenwood Reference Freed and Greenwood1996). Instead, it is highlighting the need to take into consideration (a) community-specific setups and gender roles (see Sebba & Wootton (Reference Sebba, Wootton and Auer1998) for a similar discussion of code-switching in the Caribbean community in London), and (b) the fact that different genders may contribute differently to the establishment and maintenance of their respective communities of practice.

It is important to also acknowledge that communities, including gender roles, do change, and while this change may not be captured by more general analyses of group language use, a more fine-tuned analysis of individual speech may reveal certain marked linguistic choices. In this study, one particular female, Claudia, used an English taboo item, fuck, excessively during one of the meetings, something that arguably both reflects, and potentially constructs and propagates, changes in gender roles in this community. While some studies have found that men use more taboo words than women (e.g. Bailey Reference Bailey1985; McEnery & Xiao Reference McEnery and Xiao2004), other studies have confirmed that this pattern is not consistent (Risch Reference Risch1987; De Klerk Reference De Klerk1992). In the current study, while none of the males ever used any swear or taboo words in English (and incidentally used relatively few in GCD), Claudia used the word fuck twenty-three times in different grammatical contexts, including “for fuck's sake”, “fuck it”, and “fucking”. Claudia did not utter any taboo items in GCD, however. This could indicate that each individual has a preference for which social and pragmatic information to convey in GCD (and which in English). In addition, while it seems acceptable for the men to swear in GCD, and the women not to, Claudia's increased use of swearing, albeit in English, could constitute an initial step towards change, with potentially taboo items creeping more readily into the GCD repertoire of female speakers. It is of course important to acknowledge the possibility of individual factors that could affect Claudia's increased use of taboo items, including personality and upbringing. Claudia was the only member of the group who did not reside in London. She was an undergraduate Law student at Oxford University. Therefore, both her geographical location and her language choices position her as a peripheral member of the community of practice. Peripheral members can either gradually become core members, adopting the language uses of the other core members, or continue on an outwards trajectory. By contrast, peripheral members, identified by their nonconformist language uses, can also be responsible for perpetuating change within the community of practice. Claudia does indeed seem to be challenging constructions of femininity in this community. I am not suggesting that one person's use of taboo items will ultimately be responsible for changes in gender expectations in the larger community. It is important, however, to have access to information on individual usages and identify nonconformist behaviours to obtain a greater understanding of the nature of the community of practice, its relation to wider communities (including British and Greek-Cypriot communities) and possible future developments within it.