Home / Articles / Volume 11 (2014) / Projecting social and discursive identities through code-switching on Facebook: The case of Greek Cypriots
Document Actions



Two varieties of Greek are spoken on the island of Cyprus: the local dialect, namely the Greek-Cypriot Dialect (GCD), and Standard Modern Greek (SMG). English is also influential, as Cyprus was an English colony until 1960. The dialect is rarely employed for everyday written purposes; however, it is now evident in computer-mediated communication (CMC). As a contribution to the field of code-switching in writing, this study examines how Greek-Cypriot internet users employ GCD, SMG, and English in their Facebook interactions. In particular, we investigate how identities (discursive and social) are performed and indexed through the linguistic choices of Greek-Cypriot internet users. The findings indicate that switches to GCD add a humorous tone and express solidarity and informality. SMG is mostly used for ‘official’ statements, and it is preferred by mature internet users, while English is used with expressions of affect and evaluative comments.


The internet was developed in the 1960s in the United States, and at this initial stage English was the language exclusively used online. Since then, the phenomenal growth of the internet has promoted its availability to the public and, consequently, its globalisation. Although still the dominant language, English has ceased to be the only language used online, as nowadays the internet has shifted from being a mostly monolingual to a highly multilingual network (Danet & Herring, 2007). The actual degree of multilingualism online is still unknown, due to the lack of comprehensive research and censuses of internet users (Paolillo, 2007). Findings from research suggest, however, that communication online takes place not just in official, standard languages, but also in varieties which are generally not used for writing purposes (Androutsopoulos & Ziegler, 2004; Berjaoui, 2001; Danet & Herring, 2007; Deuber & Hinrichs, 2007; Rajah-Carrim, 2008).

The ‘multilingual internet’ has attracted scholarly interest in code-switching (CS) in CMC. Some studies have emerged since the mid-1990s exploring CS in CMC, using data from ethnic minority groups, diasporic populations, youth groups, local communities, and more (see Androutsopoulos, 2013). This literature reveals that CS in CMC is not random; it may have discourse functions (e.g., emphasis, challenge, mitigation, topic shift, greeting, reported speech, etc.) and pragmatic/identity functions (e.g., to index social or ethnic identity). Androutsopoulos (2013) argues, however, that CS online still remains a less researched topic compared to other linguistic processes in CMC, and it is not yet well understood. It is important therefore to study online CS in order to better grasp its forms and functions.

The focus of the present study is the Greek-Cypriot Dialect (GCD), which Themistocleous (2008, 2010a, forthcoming) found is widely used in synchronous CMC. Our aim is to investigate CS among GCD, Standard Modern Greek (SMG), and English on Facebook by focusing on the kind of social and discursive identities that are projected through the linguistic choices of Greek-Cypriots.

GCD, SMG, and English in the Republic of Cyprus

GCD is a non-standard variety of Greek, spoken on the island of Cyprus. GCD and SMG are used interchangeably by the same speakers; the former is generally used in informal oral communication, whereas the latter enjoys more prestige and serves formal functions. Numerous English loanwords began to enter GCD when Cyprus became an English colony in 1925, and many are now well assimilated into the linguistic repertoire of Greek Cypriots (Papapavlou, 1994).

Research shows that Greek Cypriots have differing attitudes towards the two varieties of Greek. SMG, for example, is associated with schooling, professionalism, prosperity, and modernity (Papapavlou, 1998), whereas GCD, the native language of Greek Cypriots, is deprecated by its speakers in formal settings (Papapavlou, 2001) and even in contexts where social class issues and formality are not salient (Sophocleous, 2009). When it comes to social attractiveness and solidarity-related traits (e.g., warmth, humour, friendliness, kindness), dialect speakers often receive more positive evaluations than their SMG counterparts (Sophocleous, 2009). Interestingly, young Greek Cypriots seem to have positive attitudes towards GCD nowadays; this might be because GCD is now more evident in local TV series, and also because young Greek Cypriots use hyperdialectal forms of GCD as part of youth slang, which is emblematic of youth identity (Tsiplakou, 2003).

GCD does not have a standard official orthography, and dialect writing has traditionally been restricted to folk literature, local play scripts, advertisements (Pavlou, 1992), and cartoons (Themistocleous, 2010b). All other kinds of writing are generally produced in SMG. Yet Themistocleous (2008) found that, with the emergence of CMC, the dialect is now widely used among Greek-Cypriot teenagers and young adults in synchronous CMC. Moreover, many internet users use a Romanised version of GCD online, a practice similar to ‘Greeklish’ (Androutsopoulos, 2009; Tseliga, 2007), although Greek Cypriots employ additional transliteration solutions for sounds that do not exist in SMG (i.e., post-alveolar fricatives, post-alveolar affricates, and geminates). It is noteworthy that young Greek Cypriots still make use of Romanised GCD online, despite the fact that Greek fonts are now available, whereas GCD with Greek fonts is preferred by members of the older generation (Themistocleous & Sophocleous, 2012).


We understand CS in its broadest sense, as a form of language practice where individuals employ combinations of two or more linguistic varieties (languages or dialects) to accomplish conversational purposes, but at the same time to convey aspects of their personal and social identities (Dabène & Moore, 1995; Heller, 1995; Gardner-Chloros, 2009). Terms also exist to refer to standard-dialect switching, such as ‘code-shifting,’ ‘code fluctuations,’ and ‘style shifting’ (Auer, 1998; Giacalone Ramat, 1995). A crucial point that needs to be clarified when examining standard-dialect switching is whether the linguistic situation under scrutiny involves switches among separate systems (e.g., SMG vs. GCD vs. English) or switches within a single system with internal variation (e.g., within GCD). The discussion below addresses how to employ terms such as ‘code-switching’ and ‘style shifting’ in the linguistic context of the Republic of Cyprus.

GCD and SMG are genetically related varieties, yet they differ in terms of pronunciation, grammar, and lexicon (Yiakoumetti, 2007). In addition to a High (SMG) and a Low (GCD) variety, there exist numerous other levels of GCD positioned between the two ends on a dialectal continuum. Different registers/styles range from acrolectal forms closer to the standard, to basilectal forms closer to the dialect. At present no systematic work has been conducted to determine the number of these linguistic styles. Consequently, one encounters various proposals regarding the number of register levels in Cyprus (see Papapavlou, 2004; Sophocleous, 2006; Tsiplakou, Papapavlou, Pavlou, & Katsoyannou, 2006). However, preliminary testing with 200 native speakers found that there is a hierarchy of linguistic categories ranging from “heavily peasanty” to “SMG” (Tsiplakou et al., 2006, p. 273). This means that there is no clear distinction between SMG and GCD in Cyprus; in other words, it is not a clear case of diglossia (Ferguson, 1959), but it can possibly be described as a post-diglossic bi-dialectal continuum (Papapavlou & Pavlou, 2005; Tsiplakou et al., 2006; Yiakoumetti, 2007).

In this study we use the term ‘code-switching’ to refer to the alternation of SMG, GCD, and English and the term ‘style shifting’ for shifts where internal variation in GCD is observed. This encompasses shifts from acrolectal to mesolectal or basilectal linguistic styles.

Code-Switching in CMC

CS online has attracted the attention of linguists since the mid-1990s. Paolillo (1996, 2011) investigated CS between English and Punjabi on IRC and Usenet. He found that, although English was statistically predominant, Punjabi was used for discourse functions and also for identification with the Punjabi culture. Similar findings are reported by Hinrichs (2006), who investigated Jamaican Creole/English CS in personal emails. Hinrichs identified rhetorical uses of CS (e.g., double-voicing, stylization), with Jamaican Creole used to create stereotypical local speech and project Jamaican cultural identity. In a study exploring CS in German-based diasporic web forums (Greek, Persian, Indian), Androutsopoulos (2007) found that switches to the minority language served both discourse functions and indexed ethnic identity. Warschauer et al. (2007) used a questionnaire to investigate how young professionals in Cairo use language in formal and informal emails and in online chat. They found that English predominates in formal emails in the professional milieu, whereas English/Egyptian Arabic CS is widely evident in online chat. They argue that Egyptian Arabic is used to express local identity and culture in a global network.

Other studies have focused specifically on the language practices of young people looking at public forums and blogs. Data from Finland and Germany show that young bilinguals code-switch between English and the respective national language (e.g., German/Finnish) and also produce stylized representations of vernacular Englishes to index their orientation to music, media culture, and fan fiction (Androutsopoulos, 2004; Leppänen et al., 2009). Stylization is also evident in the work of Sebba (2003), who identified English/stylized Creole CS in bulletin boards devoted to the satirical fictional character Ali G. With regards to dialect/standard language switches, Siebenhaar (2006) found that in Swiss-German IRC chat rooms, teenagers prefer to use the Swiss German dialect whereas middle-aged users show a preference for standard German; thus through language choice each group indexes affiliation with their relevant age group.

With regards to the Greek language, Tseliga (2007) reports that Greek internet users frequently incorporate foreign language elements in their emails. Georgakopoulou (1997) focused on SMG/English CS in private emails between friends, taking also into consideration style-shifting from SMG into Greek dialects and sociolects. She found that switches functioned as contextualisation cues and were used for humour and solidarity, and that switches into English were used to mitigate potentially face-threatening acts. Tsiplakou (2009) investigated CS between Greek varieties and English in private emails between academics. A qualitative analysis revealed that expressions of affect and evaluative comments were mostly in English, whereas Greek was used for referential information. Style-shifting among varieties of Greek was used for humour and to mitigate potentially face-threatening acts. Another study examined GCD/SMG CS using data from the #Cyprus channel of IRC, where public synchronous chat takes place among teenagers and young adults who do not know each other in real life (Themistocleous, forthcoming). The study found that Greek-Cypriot IRC users switch between GCD and SMG in order to perform imaginary roles and play with their identity online.

The current study aims to provide additional insights in this under-researched area by analysing data from asynchronous interactions on Facebook. Facebook is an important context to investigate not only because it is popular, but also because people interact there for a variety of purposes. Moreover, Facebook is a new domain that has not been previously studied in relation to the use of GCD and the projection of identity. Furthermore, the findings arising from this study on language use in an asynchronous mode of CMC can complement those already observed in synchronous modes (Themistocleous, forthcoming). Paolillo (2011) predicts that synchronous modes of CMC will contain more conversational CS than asynchronous modes, because the interactions in the former are closer to spoken language. This prediction is supported by Lee (2007), who found that code-mixing was more frequent in data from ICQ than in the emails of the same users. However, Androutsopoulos (2013, p. 675) argues that “the synchronicity hypothesis should not lead us to assume that asynchronous modes lag behind synchronous ones in all aspects of CS.” It is important to study various modes of CMC in order to better understand the role that synchronicity plays in language use online.

Language and Identities

Language is one of the central means through which the presentation of the self is achieved. Through language we exhibit our acculturation in terms of beliefs, attitudes, lifestyles, values, and meanings (Baker & Galasiński, 2001). Identity is socially constructed (Hall, 1996; Kroskrity, 2000), and one’s sense of self can emerge as the result of communicative interaction with other people (Riley, 2007). Individuals also have social identities and use language to index their belonging in social groups, be they local (e.g., friends, family, colleagues) or global (e.g., gender, class, ethnicity, nationality). Membership in these social groups is manifested through shared linguistic behaviour; as Llamas and Watt (2010, p. 1) maintain: “All sociolinguistically competent language users can draw upon an array of linguistic resources for foregrounding different aspects of their identities in particular contexts at particular times.”

This means that identity is not fixed (Riley, 2007). Rather, it can be seen as a discursive-performative construct, a timeless essence that takes form in specific social and cultural contexts (Butler, 1993). People are composed of multiple identities (Tabouret-Keller, 1997) and at times contradictory ones; this enables them to shift identity – and consequently language use – depending on the social situations they encounter. Omoniyi (2006) makes reference to the “Hierarchy of Identities,” where all different identity options are co-present but are projected by means of hierarchization. In each social action the speaker assigns each identity a position on a hierarchy and projects the one that s/he considers most appropriate at a given moment.

Bucholtz and Hall (2010) propose various frameworks for the analysis of identity as constructed through linguistic interaction, which take into consideration both linguistic practices and aspects of culture and society. Two of these, the positionality principle and the indexicality principle, will be used for the analysis of the data in this study. The indexicality principle focuses on how linguistic forms are used to construct identity positions, looking at semiotic links between linguistic forms and social meanings (Ochs, 1992; Silverstein, 1985). An index is defined by Bucholtz and Hall (2010, p. 21) as “a linguistic form that depends on the interactional context for its meaning.” Indexical processes include (a) overt mentions of identity categories and labels, (b) implicatures and presuppositions regarding interlocutors’ identity positions, (c) evaluative or epistemic orientations to on-going talk and interactional footings, and (d) use of linguistic structures and systems that are ideologically associated with specific personas and groups. Linguistic structures may vary, from grammar, phonology, and lexis, to style, to entire linguistic systems such as languages and dialects. Each might be associated with particular social identities; for instance the use of a dialect in interaction might be indexically linked to a regional identity. Similarly, frequent use of slang might index youth identity (Androutsopoulos, 2003).

Using the positionality principle one can investigate the micro details of identity as expressed moment-to-moment in interaction. According to this principle, interactants display their identities in discourse through temporary roles and orientations that they engage in different circumstances. According to Goffman (1959), the idea of performance is useful for understanding how social categories are connected to discourse, as in all interactions participants are required to strategically perform their identities based on the interactional demand at hand (Bucholtz & Hall, 2005; Gumperz, 1982). Identity in this case is related to the “outcome of processes by which people index their similarity to and difference from others, sometimes self-consciously and strategically and sometimes as a matter of habit” (Johnstone, 2008, p. 151). Interactional positions and discursive performances of identity can have various purposes and effects such as ‘evaluator,’ ‘joke-teller,’ ‘overhearer,’ ‘teacher,’ and so on. In other words, in a given interactional event, interlocutors may adopt features of roles with which they would like to be associated.

Aims of the Study

We agree with Bucholtz and Hall’s (2010, p. 21) assertion that: “From the perspective of the analyst, it is not a matter of choosing one dimension of identity over others, but of considering multiple facets in order to achieve a more complete understanding of how identity works.” Thus, in this qualitative study we examine micro and macro dimensions of identity by focusing both on discursive and social identities evident in the interactions of Greek-Cypriot Facebook users. Specifically, the dual aim of this investigation is:

  1. to examine the ways in which Greek Cypriots use the various codes available in their linguistic repertoire (namely GCD, SMG, and English) in their Facebook interactions, and

  2. to investigate what kind of identities (discursive and social) are performed and indexed through the linguistic choice among GCD, SMG, and English.


Data for this study were collected by contacting Greek-Cypriot Facebook users from our list of friends and asking them if we could have access to their interactions with their friends which involved a minimum of four exchange messages, one responding to another. An explanation was given to all participants prior to their agreement regarding the purpose of the study, i.e., that we would be examining the use of GCD, SMG, and English in their interactions. One limitation was that many interactions were personal, and this meant that some participants were unwilling to provide them for the purposes of research, although we ensured confidentiality and anonymity. Using the friend-of-a-friend approach, we contacted 39 participants, who provided us with data for a total of 53 interactions involving 424 messages. These included message postings on participants’ own timelines (e.g., comments on status updates, comments on game results) and comments on photographs. The analysis was based on comments that status updates received, where the updates were the original trigger for the comments. It should be noted that the researchers did not take part in any of the interactions.

The participants’ ages ranged between 15 and 60 years old at the time of data collection. A previous study suggested that GCD is employed mostly by teenagers and young adults in synchronous CMC (Themistocleous, 2008). We envisaged that by obtaining data from older Facebook users also, we would be able to compare the linguistic choices of younger and older users in relation to the identities they project. All participants were born and raised in the Republic of Cyprus (this was evident in their Facebook profiles). All participants completed a consent form in September 2011, before data collection began. To ensure that participants’ linguistic practices were not influenced by the purpose of the study, data were collected from their past interactions that took place between May 2009 and August 2011. Finally, to maintain anonymity, the real names of participants have been replaced by pseudonyms, and participants’ personal information is not used in the study.


In the analysis that follows we present examples of interactions from our data. CS in the examples is indicated as follows: regular font for GCD, italics for SMG, and underlined text for English. The majority of the participants show a preference for Romanised GCD/Greeklish (examples 1, 2, 3, 4, and 5), although some use the Greek alphabet (examples 4 and 5). Emphatic repetitions of punctuation marks, repeated keystrokes, and emoticons are also evident in our data. As other studies suggest, such practices are employed by internet users in order to compensate creatively for the lack of prosodic features in digital network writing and recreate speech-like qualities (Danet, 2001; Dresner & Herring, 2010; Nishimura, 2007).

The interaction below takes place among seven female friends aged 15-16. This interaction began after Xenia posted an album with pictures from her brother’s recent wedding, where Xenia and her cousin Elena posed for the camera in their evening gowns outside the church.

Example 1

The participants in this interaction commented on Xenia and Elena’s dazzling appearance. The compliments came from very close friends, who either attend the same secondary school or belong to the same group of friends. These two common denominators reinforce solidarity bonds among them, which are expressed in GCD.

The close friendship these interactants share is further expressed in the use of dialectal lexical items such as ‘dete mas ke an exome aipi pete mas!!’ (‘look at us and if we have a flaw let us know!!’) (turn 10), where the interactant’s confidence about her and her cousin’s good looks is expressed in a rather boastful but, at the same time, humorous manner. Her humour is exemplified at the beginning of her utterance by the phrase ‘e hello ennoeite afto!’ (‘well hello, this goes without saying!’) (turn 10) in her response to her cousin’s previous comment on the beauty of their family (turn 9). The use of repeated keystrokes in ‘xxxx’ (kisses) and double and triple question and exclamation marks throughout this interaction shows enthusiasm, spontaneity, and informality.

It is also worth noting that even though the dialectal lexical item ‘aipi’ (‘flaw’) is used, it is accompanied by the SMG conjunction ‘ke (‘and’), whose formal style is enhanced by the use of ‘exome’ (GCD ‘we have,’ ‘exoume’ in SMG) (2nd.pers.pl.), which is very often used by Greek Cypriots next to basilectal words, as in this example ‘aipi,’ to signify a mesolectal style, a more polished GCD. This kind of style shifting is used by Elena humorously and proudly in turn 10 to invite current interactants, and possibly other friends who have not contributed to the discussion so far, to praise them on their appearance. Moreover, CS to SMG is evident in turns 2, 3, 6, and 10, where participants are observed to make open statements, such as praising appearance (turns 2 and 6), to add parenthetical comments secondary to the topic of discussion (turn 3), or to make open affirmations (turn 10); the commonality among all of these instances is the use of CS as a means to publicly and formally highlight a topic in the interaction.

The interaction in example 2 below is between two female friends (Lina and Elena), both aged 16. It appeared on Elena’s timeline after the friends completed a test on Facebook that purported to assess how well they knew their friends. Both interactants obtained excellent results in the test and received a ‘trophy’ as the prize.

Example 2

The interaction is in GCD; it begins when Lina humorously asks Elena to share the “trophy” with her (turn 1), as a means to indirectly inform her that she has also obtained excellent results in the test. Elena teasingly replies that Lina can have the whole “trophy” to herself, since she will obtain a better “trophy” for herself (turn 2). Thus far, solidarity between the two interactants is expressed in their coded interaction; the topic is unclear to outgroup members and can only be comprehended and developed by the interactants themselves. Elena’s insinuation about “a better trophy” (namely ‘Leontis’ - a Greek-Cypriot comedy character from Egia Fuxia 1) is, however, very clear to Lina, who at this point adopts Elena’s teasing approach and clarifies to her that Leontis is not to be shared (turn 2). Elena expresses her feelings in an equally direct manner, i.e., ‘en dikos mu oti tz na genei...’ (‘he is mine whatever happens…’) (turn 4), and then interestingly, Lina takes on the role of a character from the this sitcom, Orthodoxia, a prostitute who is also in love with Leontis (turn 5). Orthodoxia, the gossiper of the village in the series, constantly attempts to gain Leontis’ approval and get closer to him.

What is particularly interesting in this example is the successful use of stylization in GCD, through indexicality and positionality, by both speakers, who take on a new role (both desiring Leontis) to express their interest in the actor of the series. In this case, “stylizing a perceived ‘other’” who is absent from the interaction (Tsiplakou & Ioannidou, 2012) illustrates the interactants’ performativity (see also Virtanen, 2013 on performativity in CMC) and their attempt to perform a specific type of constructed identity. The adoption of Orthodoxia’s voice is evident in the use of several phrases in GCD in Lina’s last response (turn 5). The phrase ‘Apanagia mu panagia mu j pale panagia mu!,’ loosely translated as ‘Oh my God’; the phrase ‘Egw exw ena onoma se tt tin koinwnia’ (‘I have a good reputation to maintain in this society’), one that Orthodoxia often uses in the series to conceal her true profession; and ‘aaaax leonti mu’ (‘aahh, my Leontis’), a phrase she uses when she is thinking of him, are all typical phrases that television viewers associate with Orthodoxia. Towards the end of her last utterance, Lina employs Orthodoxia’s humorous but also abusive language to swear at Elena by insulting her that she is not capable of getting Leontis, as she cannot do the simplest of tasks such as stuffing a vine leaf. The act of performing here on the part of Lina is particularly successful (turn 5), since the type of stinging approach she used to respond to Elena’s statement (turn 4) is indeed characteristic of Orthodoxia’s temperament in Egia Fuxia. All this is achieved not only through style-shifting in GCD but also with the addition of the emoticons :-P and :P (sticking out their tongues) to express humour and to indicate that the criticism should be taken light-heartedly.

The exchange in example 3 involves six young adults, aged 29-30, who used to be classmates in primary school. Three males (Andreas, Panos, and Grigoris) and three females (Hara, Vera, and Ioulia) take part in this interaction. The discussion began when Andreas posted on his timeline a class photo, dated from the early 1980s, when the participants were around the age of eight.

Example 3

Vera posts the first comment underneath the photo. She uses English to express her surprise and then switches to GCD. Andreas posts the next comment using the same language mixture as Vera; this could signal consent and ratification of the first speaker’s activity. Another instance of a similar switch to English can be found in turn 13, where Ioulia thanks her friends for wishing her Happy Birthday (the wishes are not evident here, which means that Andreas and Vera posted their wishes on Ioulia’s timeline). Again, in this case, the switch to English is associated with an expression of affect, in which Ioulia is thanking her friends. There are various one-word switches to English in the above example: ‘indeed’ (turn 3), ‘supermarket’ (turn 5), ‘since then’ (turn 9), ‘timing’ (turn 13), ‘reunion’ (turn 13, 14, 17), ‘email’ (turn 17), and ‘date’ (turn 17). All these seem to involve loanwords that have been assimilated into GCD (Papapavlou, 1994), but at the same time their usage might be enhanced by the fact that most of the participants in this interaction studied in English-speaking universities, and some are still living abroad.

In turn 4, Andreas states that photographs of this kind are ‘dokumenta axias’ (‘precious documents’) and that they should not be kept in drawers. This post is in SMG, the code usually associated with professionalism and expertise (Papapavlou, 1998). It seems that, by switching to SMG, Andreas performs the role of the ‘responsible discoverer’ (an individual discourse identity) who managed to recover this important document.

The interaction that follows, including the rest of Andreas’ comments, is realised in GCD. As participants start to remember funny incidents from their early school years, a pleasant, humorous tone is adopted, which triggers the use of the dialect. The first topic of conversation revolves around Ms. Doxoulla, the primary school teacher. Panos (turns 5 and 7) and Andreas (turns 6 and 8) post their comments, which consist of recollections of different comical incidents. Hara joins in (turn 11) by saying that she does not remember any of these incidents. Yet, it is not just the teacher that they seem to ridicule, as self-critique is also evident. The first to make fun of himself is Grigoris (turn 10), who says that instead of looking at the photographer, he was looking at the floor. Hara makes jokes about her clothes (turn 11), and Andreas comments on his appearance, claiming that he stands out because of his looks (turn 12). As mentioned above, jokes, anecdotes, and humorous stories are usually expressed in a language that is close to the interlocutors’ heart (Papapavlou, 1998). Since GCD is in this case the participants’ native language, it is not surprising that they choose this code when they recall funny stories or self-critique their appearance. GCD is used in this case to express solidarity and project in-group identity. At a more local level, GCD seems to allow these interlocutors to perform a ‘joke-teller’ identity.

It is clear from the comments that the classmates have not seen each other for a long time; therefore, a reunion is suggested by Ioulia (turn 13). Ioulia switches to SMG to make her suggestion, as the topic now shifts from jocular to a more ‘serious’ one. This style-shift also projects an individual discourse identity for Ioulia, namely that of ‘reunion organiser/coordinator’ who takes on the responsibility of making arrangements for the rest of the classmates to meet. Ioulia makes suggestions and sets the plan (see also turn 17). SMG is employed in this case because the information imparted is factual and referential. The topic is more ‘serious’; hence, SMG can be regarded as more appropriate. Hara accommodates to SMG to agree with Ioulia, whereas Andreas and Vera switch between SMG and GCD. Vera’s switches to GCD are more marked (Andreas only switches for the verb ‘pete’ (‘say’) and the adverb ‘polla’ (‘many’)). Vera congratulates Ioulia for her idea and also makes a joke about everyone getting older, again projecting a shared group identity with the use of GCD.

This linguistic bricolage enables the old classmates to achieve not only interpersonal functions (indexing, for instance, solidarity and in-group membership through the use of GCD), but it also allows the participants to perform local identities, e.g., that of ‘joke-teller’ through the use of GCD, or ‘coordinator’ and ‘responsible discoverer’ through the use of SMG.

GCD Indexing Youth Identity

The use of GCD online also seems to be related to age. Research suggests that this new way of writing is prominent among teenagers and young adults, whereas mature Greek Cypriots use dialect writing much less in CMC (Themistocleous, 2008). A recent study conducted by Themistocleous and Sophocleous (2012) involved two focus groups, 18-21 and 51-63 year olds. Participants in the younger focus group were more positive towards the use of Romanised GCD in writing (both in online contexts and in handwriting) and reported using it widely. The majority of the mature participants, however, stated that they more or less accept Romanised SMG in CMC contexts, but that they prefer to use non-Romanised SMG in their digital writing. Mature participants also reported that they do not write in GCD, either in online or offline contexts, and they expressed strong negative attitudes towards Romanised GCD outside the domain of CMC. The following examples from our Facebook data support these findings.

In example 4 below, Athos, a 17-year old male, posts a comment on Mrs. Anna’s timeline in order to ‘catch up.’ Anna is Athos’ 60-year-old neighbor and a very close family friend.

Example 4

Athos’ form of address (mrs. anna) reaffirms the social relationship between the two participants and indexes the conventional expectation that since Mrs. Anna is older than Athos, she should be addressed in a formal way (notice also the formal form of address with the 2nd person plural form ‘sas (‘your’) in turn 5). His post is in GCD. In turn 2, Anna responds in English ‘alowwww’ (‘heeeelloooo’) (a typical greeting in Cyprus – the repeated letters indicate excitement) and invites Athos to her house, using SMG. Anna then switches to English saying to Athos that she has a surprise for him. This switch to English could be said to have an emotive function, as this surprise would make Athos very happy (the surprise is revealed in turn 6, when Anna explains that some common friends would be visiting). Anna then uses SMG and continues to do so throughout the interaction (turns 6 and 8; notice also the switch to Greek characters in line 8). Neither Anna nor Athos converge to the other’s code: Anna uses SMG, whereas Athos uses GCD. The only exception is when Athos accepts Anna’s invitation (turn 3). Athos’ brief switch to SMG could be an indication of politeness (notice also the use of the emoticon, expressing happiness). Yet Athos still uses elements of the code that is closer to his age group: The verb ‘erto’ (‘to come’) is in GCD. These linguistic choices by the two participants and the fact that neither of them accommodates to each other’s code indicate that they are projecting distinct social identities: A youth identity for Athos and a more mature identity for Anna.

Data from other interactions involving older participants show a similar pattern. The following is another example illustrating the use of SMG by older Facebook users. Nikolas is 51 years old, and Stella is 36.

Example 5

Even though use of SMG is generally the rule that we observed among our mature group of users, Nikolas’ use of SMG (as well as Greek characters) might be attributed to the fact that he used to be Stella’s high school teacher; hence, relations of role and normative social pressures might have especially encouraged Nikolas to write in SMG with Greek characters. Nikolas also employs SMG with Greek characters in almost all his interactions with others, however. Only one instance of writing SMG with Roman letters was observed on his timeline, in which a friend initiated an interaction in SMG using Roman letters, and Nikolas responded in a similar manner. It is possible that the emotional trigger generated by the message led Nikolas to respond spontaneously using SMG with Roman characters, or more generally led him to accommodate to his friend, suggesting solidarity.


Overall, we found that Facebook users employ numerous indexical processes and positionality means to index both discursive and social identities. This finding goes hand in hand with previous investigations, including Hinrichs (2006), Paolillo (1996/2011), Androutsopoulos (2007) and Warshchauer et al. (2007), who also found that non-standard, local varieties are used online to index identities.

Our analysis reveals that GCD, and especially style shifting within the dialect, is used to introduce a humourous note and incite laughter. Georgakopoulou (1997) and Tsiplakou (2009) also found that style-shifting in emails is used for these functions, and especially for humour. In terms of discursive performances of identity, we saw in example 3 that the use of GCD enables the participants to perform a ‘joke-teller’ identity, and in example 2, we saw Elena and Lina perform imaginary identities to imitate characters from a TV series. Playing with identity and performing imaginary roles was also evident in Themistocleous’ (forthcoming) study, who looked at data from synchronous CMC. This suggests that synchronicity and anonymity does not necessarily affect CS and language playfulness online.

We found that SMG is used for more formal and serious purposes, for example to make public and official statements, to add parenthetical comments, and to give referential information. This finding is consistent with that reported by Tsiplakou (2009). We also saw in example 3 that Ioulia projects an individual discourse identity (‘reunion organiser/coordinator’) through the use of SMG, while Andreas performs the role of the ‘responsible discoverer.’ A possible explanation for this language choice is because SMG is usually associated with professionalism and expertise (Papapavlou, 1998). Warschauer et al. (2007) also found that a dominant language (namely English) is preferred over Egyptian Arabic in professional emails.

Mature participants prefer to use SMG in their Facebook interactions. This is evident in examples 4 and 5, where Anna and Nicolas use SMG with younger participants Athos and Stella, respectively, and show a preference for Greek characters rather than Romanisation (Anna switches between Greek and Roman characters, however). Based on this, and consistent with Themistocleous and Sophocleous (2012), we argue that SMG is used to project a mature identity, whereas GCD is mostly associated with youth identity. Previous studies have also revealed that regional varieties are preferred by younger generations online (Androutsopoulos, 2004; Leppänen et al. 2009; Siebenhaar, 2006). At the same time, however, these mature users also adopt some typical features of CMC (e.g., repeated keystrokes, strings of exclamation marks) that denote linguistic innovation and playfulness. Participants in Themistocleous and Sophocleous’ (2012) study also mentioned that they would use such features in digital writing, although never in handwriting. This means that the use of such CMC features could project the performance of an online identity relevant to the medium (Herring, 2007; Themistocleous, forthcoming; Tsiplakou, 2009).

Switches into English were also found in our data. Some switches involve loanwords that are frequently used in free variation with their Greek equivalents. Their usage might be enhanced by the fact that some participants, especially in example 3, studied and still live abroad. In other cases, switches into English have emotive functions (examples 3, 4) and are used to introduce an element of surprise, to express affection, and in evaluative comments. This finding is consistent with Tsiplakou (2009), who also found that expressions of affect and evaluative comments in emails were realised in English.

It becomes clear, then, that CS allows Greek Cypriots to project various social and discursive identities on Facebook. Moreover, the asynchronous nature of Facebook does not seem to minimise the use of CS, as predicted by Paolillo (2011). It is important however to study various modes of CMC as well as public vs. private material, as social- and medium-related facets2 in addition to synchronicity (Herring, 2007) can affect the forms and functions of language use.


Although CS online has started to attract the attention of linguists, it has been an under-researched area (Androutsopoulos, 2013). This study contributes to this area by documenting language choice on Facebook, a recently developed online environment. GCD, SMG, and English are the three main codes used systematically in Greek Cypriots’ linguistic repertoire. Our data illustrate that Greek-Cypriot Facebook users employ all three codes, including style shifting within the dialect, quite skilfully in their social network site writing.

This study has shed light on the use of CS within the Greek-Cypriot online context. Further research needs to be carried out in order to better understand this phenomenon. This was a qualitative study; quantitative research is needed in order to identify how frequently each code is used within the interactions of Greek Cypriots. A combination of qualitative and quantitative methods should enable scholars to obtain an overall picture of the use of CS in Cyprus. It is also important to compare studies investigating CS in CMC and in face-to-face interactions in order to explore whether the processes are similar or different; this will enable better understanding of the influence of the internet on language choice. Finally, computer-mediated language samples could be compared systematically across different multilingual societies, in order to facilitate the development of an overall framework to investigate CS in CMC environments.


  1. Egia Fuxia is a popular Greek-Cypriot comedy television series. Its success resides mostly in the fact that even though the plot revolves around Greek-Cypriots’ lifestyle 100 years ago, the issues discussed are relevant to contemporary topics in Greek-Cypriot society. Dialect stylization in this sitcom is deployed for comic purposes, and it also projects Cypriot identities (Tsiplakou & Ioannidou, 2012).

  2. Herring (2007) proposes a classification of computer-mediated discourse based on social- and medium-related factors or ‘facets’ that potentially influence discourse usage in CMC environments. For instance, synchronicity (a medium-related factor) might influence language use, in that discourse may be more structured in asynchronous modes of CMC as internet users take time to respond to the sender and can edit their text. In synchronous modes of CMC internet users chat in “real time,” hence interactional behaviors are more evident. Social factors include aspects such as participation structure (public vs. private conversation), purpose (professional vs. social interaction), and tone (serious, fun, friendly, formal).


Androutsopoulos, J. K. (2003). Discourse constructions of youth identities: Introduction. In J. Androutsopoulos & A. Georgakopoulou (Eds.), Discourse constructions of youth identities (pp. 1-25). Amsterdam/Philadelphia: John Benjamins.

Androutsopoulos, J. K. (2004). Non-native English and sub-cultural identities in media discourse. In H. Sandøy (Ed.), Den fleirspråklege utfordringa [The multilingual challenge] (pp. 83-98). Oslo: Novus.

Androutsopoulos, J. K. (2007). Language choice in German-based diasporic web forums. In B. Danet & S. C. Herring (Eds.), The multilingual internet: Language, culture, and communication online (pp. 340-361). New York: Oxford University Press.

Androutsopoulos, J. K. (2009). ‘Greeklish’: Transliteration practice and discourse in a setting of computer-mediated digraphia. In A. Georgakopoulou & M. Silk (Eds.), Standard languages and language standards: Greek, past and present (pp. 221-249). Surrey, UK: Ashgate.

Androutsopoulos, J. K. (2013). Code-switching in computer-mediated communication. In S. C. Herring, D. Stein, & T. Virtanen (Eds), Handbook of the pragmatics of computer-mediated communication. (pp. 667-694). Berlin: Mouton de Gruyter

Androutsopoulos, J. K., & Ziegler, E. (2004). Exploring language variation on the Internet: Regional speech in a chat community. In B. L. Gunnarsson (Ed.), Language variation in Europe Papers from ICLaVE 2 (pp. 99-111). Uppsala: Uppsala University Press.

Auer, P. (1998). Code-switching in conversation: Language, interaction and identity. London: Routledge.

Baker, C., & Galasiński, D. (2001). Cultural studies and discourse analysis. London: Sage.

Berjaoui, N. (2001). Aspects of the Arabic orthography with preliminary insights from the Moroccan computer-mediated communication. In M. Beißswenger (Ed.), Chat-Kommunication: Sprache, interaction, sozialität & identität in synchroner computerver-mittelter kommmunikation (pp. 431-465). Stuttgart: Ibidem.

Bucholtz, M., & Hall, K. (2005). Identity and interaction: A sociocultural linguistic approach. Discourse Studies, 7, 585-614.

Bucholtz, M., & Hall, K. (2010). Locating identity in language. In C. Llamas & D. Watt (Eds.), Language and identities (pp. 18-28). Edinburgh: Edinburgh University Press.

Butler, J. (1993). Bodies that matter. London: Routledge.

Dabène, L., & Moore, D. (1995). Bilingual speech of migrant people. In L. Milroy & P. Muysken (Eds.), One speaker, two languages: Cross-disciplinary perspectives on code-switching (pp. 17-44). Cambridge: Cambridge University Press.

Danet, B. (2001). Cyberpl@y: Communicating online. Oxford: Berg.

Danet, B., & Herring, S. C. (Eds.) (2007). The multilingual interne: Language, culture, and communication online. Oxford: Oxford University Press.

Dresner, E., & Herring, S. C. (2010). Functions of the non-verbal in CMC: Emoticons and illocutionary force. Communication Theory, 20, 249-268.

Deuber, D., & Hinrichs, L. (2007). Dynamics of orthographic standardization in Jamaican Creole and Nigerian Pidgin. World Englishes, 26(1), 22-47.

Ferguson, C. (1959). Diglossia. Word, 15, 325-340.

Gardner-Chloros, P. (2009). Code-switching. Cambridge: Cambridge University Press.

Giacalone Ramat, A. (1995). Code-switching in the context of dialect/standard language relations. In L. Milroy & P. Muysken (Eds.), One speaker, two Languages: Cross-disciplinary perspectives on code-switching (pp. 45-67). Cambridge: Cambridge University Press.

Georgakopoulou, A. (1997). Self-presentation and interactional alliances in e-mail discourse: The style and code-switches of Greek messages. International Journal of Applied Linguistics, 7, 141-164.

Goffman, E. (1959). The presentation of self in everyday life. Garden City, NY: Doubleday Anchor Books.

Gumperz, J. (1982). Discourse strategies. Cambridge: Cambridge University Press.

Hall, S. (1996). Who needs ‘identity’? In S. Hall & P. du Gay (Eds.), Questions of cultural identity (pp. 1-17). London: Sage.

Heller, M. (1995). Code-switching and the politics of language. In L. Milroy & P. Muysken (Eds.), One speaker, two languages: Cross-disciplinary perspectives on code-switching (pp. 158-174). Cambridge: Cambridge University Press.

Herring, S. C. (2007). A faceted classification scheme for computer-mediated discourse. Language@Internet, 4(1). Retrieved February 9, 2013 from http://www.languageatinternet.org/articles/2007/761

Hinrichs, L. (2006) Codeswitching on the web. Amsterdam/Philadelphia: John Benjamins.

Johnstone, B. (2008). Discourse analysis, 2nd ed. Oxford: Blackwell.

Kroskrity, P. V. (2000). Identity. Journal of Linguistic Anthropology, 9(1-2), 111-114.

Lee, C. (2007). Linguistic features of email and ICQ instant messaging in Hong Kong. In B. Danet & S. C. Herring (Eds.), The multilingual internet: Language, culture, and communication online (pp. 184-208). New York: Oxford University Press.

Leppänen, S., Pitkänen-Huhta, A., Piirainen-Marsh, A., Nikula, T. & Peuronen, S. (2009). Young people’s translocal new media uses: A multiperspective analysis of language choice and heteroglossia. Journal of Computer-Mediated Communication, 14(4), 1080-1107.

Llamas, C., & Watt, D. (Eds.) (2010). Language and identities. Edinburgh: Edinburgh University Press.

Nishimura, Y. (2007). Linguistic innovations and interactional features in Japanese BBS communication. In B. Danet & S. C. Herring (Eds.), The multilingual internet: Language, culture, and communication online (pp. 163-183). New York: Oxford University Press.

Ochs, E. (1992). Indexing gender. In A. Duranti & C. Goodwin (Eds.), Rethinking context: Language as an interactive phenomenon. (pp. 335-358). New York: Cambridge University Press.

Omoniyi, T. (2006). Hierarchy of identities. In T. Omoniyi & G. White (Eds.), Sociolinguistics of identity (pp. 11-33). London: Continuum.

Paolillo, J. (1996). Language choice on soc.culture.punjab. Electronic Journal of Communication. 6(3). Retrieved June 29, 2014 from http://www.cios.org/EJCPUBLIC/006/3/006312.HTML

Paolillo, J. (2007). How much multilingualism? Language diversity on the internet. In B. Danet & S. C. Herring (Eds.), The multilingual internet: Language, culture, and communication online (pp. 408-430). New York: Oxford University Press.

Paolillo, J. (2011). “Conversational” codeswitching on Usenet and Internet Relay Chat. Language@Internet, 8(3). Retrieved June 29, 2014 from http://www.languageatinternet.org/articles/2011/Paolillo

Papapavlou, A. (1994). Language contact & lexical borrowing in the Greek Cypriot dialect. Athens: Grivas Publications.

Papapavlou, A. (1998). Attitudes toward the Greek Cypriot dialect: Sociocultural implications. International Journal of the Sociology of Language, 134, 15-28.

Papapavlou, A. (2001). Mind your speech: Language attitudes in Cyprus. Journal of Multilingual and Multicultural Development, 22, 491-501.

Papapavlou, A. (2004). Verbal fluency of bidialectal speakers of SMG and the role of language-in-education practices in Cyprus. International Journal of the Sociology of Language, 168, 91-100.

Papapavlou, A., & Pavlou, P. (2005) Literacy and language-in-education policy in bidialectal settings. Current Issues in Language Planning, 6(2), 164-181.

Pavlou, P. (1992). The use of the Cypriot-Greek dialect in the commercials of the Cyprus Broadcasting Corporation. Paper presented at the 37th Conference of the International Linguistics Association, Washington DC, 20-26 June.

Rajah-Carrim, A. (2008). Choosing a spelling system for Mauritian Creole. Journal of Pidgin and Creole Languages, 23(2), 193-226.

Riley, P. (2007). Language, culture and identity. London: Continuum.

Sebba, M. (2003). “Will the real impersonator please stand up?” Language and identity in the Ali G websites. Arbeiten aus Anglistik and Amerikanistik, 28(2), 279-304.

Siebenhaar, B. (2006). Code choice and code-switching in Swiss-German Internet Relay Chat rooms. Journal of Sociolinguistics, 10(4), 481-509.

Silverstein, M. (1985). Language and the culture of gender: At the intersection of structure, usage and ideology. In E. Mertz & R. Parmentier (Eds.), Semiotic mediation: Socio-cultural and psychological perspectives (pp. 219-259). Orlando, FL: Academic Press.

Sophocleous, A. (2006). Identity formation and dialect use among young speakers of the Greek-Cypriot community in Cyprus. In R. Kiely, P. Rea-Dickins, H. Woodfield, & G. Clibbon (Eds.), Language, culture and identity in applied linguistics (pp. 61-77). London: Equinox Publishing.

Sophocleous, A. (2009). Language attitudes towards the Greek Cypriot Dialect: Social factors contributing to their development and maintenance. Unpublished Ph.D. Thesis, Kingston upon Thames: Kingston University.

Tabouret-Keller, A. (1997). Language and identity. In F. Coulmas (Ed.), The handbook of sociolinguistics (pp. 315-326). Oxford: Blackwell.

Themistocleous, C. (2008). The use of Cypriot Greek in synchronous computer-mediated communication. Unpublished Ph.D. Thesis, University of Manchester.

Themistocleous, C. (2010a). Writing in a non-standard Greek variety: Romanized Cypriot Greek in IRC. Writing Systems Research, 2(2), 155-168.

Themistocleous, C. (2010b). ‘Bananas and other fruits’: Language and identity in PIN’s cartoons. Paper presented at the 18th Sociolinguistics Symposium, University of Southampton, UK, 1-4 September.

Themistocleous, C. (forthcoming). Digital code-switching between Cypriot and Standard Greek: Performance and identity play online. International Journal of Bilingualism. Prepublished December 12, 2013   as doi: 10.1177/1367006913512727

Themistocleous, C., & Sophocleous, A. (2012). Intergenerational attitudes towards texts written in the Cypriot Greek dialect: Greek vs. Roman characters. Paper presented at the 19th Sociolinguistics Symposium, Freie Universität Berlin, Germany, 21-24 August.

Tseliga, T. (2007). “It’s all Greeklish to me!”: Linguistic and sociocultural perspectives on Roman-alphabeted Greek in asynchronous computer-mediated communication. In B. Danet & S. C. Herring (Eds.), The multilingual internet: Language, culture, and communication online (pp. 116-141). New York: Oxford University Press.

Tsiplakou, S. (2003). Στάσεις απένταντι στη γλώσσα και γλωσσική αλλαγή: μια αμφίδρομη σχέση; [Attitudes towards language and language change: A two-way relationship?] In Proceedings of the 6th International Conference on Greek Linguistics (e-book), University of Crete. Retrieved December 15, 2013 from http://www.philology.uoc.gr/conferences/6thICGL/

Tsiplakou, S. (2009). Doing (bi)lingualism: Language alternation as performative construction of online identities. Pragmatics, 19(3), 363-393.

Tsiplakou, S., & Ioannidou, E. (2012). Stylizing stylization: The case of Aigia Fuxia. Multilingua, 31, 277-299.

Tsiplakou, S., Papapavlou, A., Pavlou, P., & Katsoyannou, M, (2006). Levelling, koineization and their implications for bidialectism. In F. Hinskens (Ed.), Language variation – European perspectives . Selected papers from the third International Conference on Language Variation in Europe (pp. 265-276). Amsterdam/Philadelphia: John Benjamins.

Virtanen, T. (2013). Performativity in computer-mediated communication. In S. C. Herring, D. Stein, & T. Virtanen (Eds.), Handbook of pragmatics of computer-mediated communication (pp. 269-290). Berlin: Mouton de Gruyter.

Warschauer, M., El Said, G., & Zohry, A. (2007). Language choice online: Globalization and identity in Egypt. In B. Danet & S. C. Herring (Eds.), The multilingual internet: Language, culture, and communication online (pp. 303-318). New York: Oxford University Press.

Yiakoumetti, A. (2007). Choice of classroom language in bidialectal communities: To include or to exclude the dialect? Cambridge Journal of Education, 37(1), 51-66.

Biographical Notes

Andry Sophocleous ( sophocleous.a@unic.ac.cy ) is Assistant Professor of Sociolinguistics at the University of Nicosia, Cyprus. Her research interests include bidialectalism, language and identity construction, style and sociolinguistic variation, language and education, language and culture, and computer-mediated communication.

Christiana Themistocleous ( c.themistocleous@reading.ac.uk ) is a Lecturer in Sociolinguistics at the Department of English Language and Applied Linguistics, University of Reading, UK. Her research interests include societal multilingualism, code-switching, dialect writing, discourse analysis, language and identity, and language use in CMC.


Any party may pass on this Work by electronic means and make it available for download under the terms and conditions of the Digital Peer Publishing License. The text of the license may be accessed and retrieved at http://www.dipp.nrw.de/lizenzen/dppl/dppl/DPPL_v2_en_06-2004.html.