Home / Articles / Volume 8 (2011) / “On for drinkies?”: Email cues of participant alignments
Document Actions



One of the main preoccupations of linguistic studies of computer-mediated communication (CMC) has been with the medium-conditioned relations of its linguistic choices with spoken and written language. The underlying assumption of this inquiry is that “electronic discourse is a unique form of discourse which exists on a continuum between the context-dependent interaction of oral conversation and the contextually abstracted composition of written text” (Foertsch, 1995, p. 301). More specifically, CMC is recognized as combining qualities which are typically associated with face-to-face interactions, e.g., immediacy and informality of style, transience of message, reduced planning and editing, rapid (or immediate) feedback with properties of written language, e.g., lack of visual and paralinguistic cues, physical absence of the addressee, and written mode of delivery. CMC is thus uniquely positioned in the intersection of written and oral discourse, displaying some of the linguistic features that have been associated with certain forms of written language and others that are more prototypical of spoken language (hence the terms “interactive written discourse,” Ferrara, Brunner, & Whittemore, 1991, and “written speech,” see Yates, 1992). This intermingling of spoken and written (literate) features on CMC has provided further evidence for the widely endorsed view that spoken and written discourse themselves should not be treated as a dichotomy but as a continuum of numerous overlapping and intersecting cases that cut across various uses of language, which are in turn shaped by different situational and sociocultural contexts (Biber, 1988; Tannen, 1984; for a discussion of the place of CMC research in relation to inquiry into the spoken vs. the written, see Georgakopoulou 2006).

While at the time of this study research on the spoken and written features of CMC had shed light on the discourse composition of various text types of CMC (Reid, 1991, for e-chat; Du Bartell, 1995; Gruber, 1996; and Herring, 1996, for email; Yates, 1996, for e-conferencing; Amiran & Unsworth, 1991, for e-journals, etc.), there was still much scope for inquiry into the complex ways in which such text types invoke and are shaped by their contexts of occurrence, involving not only the medium, but also the roles and relationships of the participants, the purpose and functions of communication, etc.1 Such inquiry, informed by advances in sociolinguistic and discourse studies, is aimed at providing a deeper understanding of the ways in which linguistic devices in CMC act as meaningful semiotic resources for the pursuit of various local interactional projects. The present study2 adds to this rapidly expanding line of work by drawing on the frameworks of conversation analysis and interactional sociolinguistics3 in its exploration of the dynamics of participant relations in email discourse. Both frameworks emphasize the importance of closely observing the sequential organization of utterances in order to unravel the ways in which actions locally situated and accomplished through discourse implicate certain identities, roles, relations, rights, and obligations for interactants. Within this approach, discourse is seen as invoking, enacting, but also (re)constituting its context of occurrence (Duranti & Goodwin, 1992). In accordance with this, participant identities and relations are not viewed as externalized and static notions that are simply reflected in discourse activities, but rather as fluid, negotiable, and dynamic concepts, locally occasioned and methodically produced by means of, and in, language (Davies & Harré, 1990). Their discursive construction provides continual indices of who the speakers are and what they want to communicate.

With the above assumptions as its starting point, this article focuses on a corpus of private email messages in Greek and in English. The aim is to bring to the fore the linguistic strategies by which participants locally construct and make sense of the positions, stances, or alignments (Goffman, 1981) that they take towards their addressees. The identification of these strategies is based on the criterion of unmarked context-bound use (Ochs, 1992), in other words, on the search for recurrent sequential patterns in the observed and analysed corpus. This study will focus on how specific resources, captured here by the term “code-alternations,” are strategically drawn upon by participants to meet local context requirements and how their situated use provides for certain understandings of the performed activity of email.


The data for this article departed at the time of the study from most studies of email discourse in two ways: First, the participants know one another personally, their email interactions thus mediating past and future face-to-face interactions; second, the main dataset is not English (for recent studies of CMC in languages other than English, see Danet & Herring, 2007). To take up each issue separately, when this reseach was conducted, the primary data sources for studies of email were various public bulletin boards and “listserv” interactions, in which email messages were exchanged among participants who were unknown to one another and (at times) non-specific. Private email communication between intimates largely remained an uncharted territory on the grounds that it was a less accessible and convenient data source, in particular in view of the privacy issues involved (Yates, 1996, p. 30). The present study accessed such data by securing permission for the messages analysed from both their senders and receivers. This was felt to be a worthwhile expedition since prior observation had suggested that, despite being least explored, communication between intimates is a very frequent and salient type of email. Furthermore, CMC between “real” as opposed to “virtual” friends provides interactional contexts that are more comparable with everyday informal communication, especially as regards participant roles and relations (e.g., lack of anonymity, intimacy, shared assumptions, history of previous interactions). It thus allows for deeper insights into the differences in the construction of participation frameworks between CMC and face-to-face communication. A point to note is that comparative studies of email and spoken interactions have largely overlooked the need for controlling confounding variables; as a result, it is not always clear whether certain findings are medium-specific or the outcome of other contextual parameters (in interaction) in different discourse types. Furthermore, little is known about aspects of interpersonal communication that are equally salient in CMC and everyday interactions, e.g., linguistic strategies for enhancing symmetrical or asymmetrical relationships.

The data for this study come from the first phase (1995-1997) of a multi-phase data collection. The first phase involved messages which I had received from friends and colleagues, as well as messages received by six email users (three men and three women), all native Greek speakers, whom I approached using Milroy’s (1987) “friend of a friend” technique of data collection.4 This article is based on 600 such messages, namely 300 messages from my personal corpus and 50 messages from each of the six participants (300). None of the six recipients had known in advance that their messages were to be analysed, so there is no “observer’s paradox” involved. My own emails have not been included in the analyses, except for when they were actively invoked by and included in somebody else’s message. Though this study is not intended as comparative, in order to bring in further insights and validate some of its claims, it will occasionally include in the discussion (British) English email messages. Those were obtained (500 messages in total) using the same procedure as for the Greek data.

The senders and recipients of both the Greek and English messages, aged between 23 and 35, were at the time of data collection residents (temporary, in the case of certain Greek participants) in the U.K., most of them in London; they were students or recent graduates who could be further subdivided into academics and professionals. The sample has equal representation of females and males. In all cases, the participants were involved in intimate and egalitarian relations. Most messages exemplify the dyadic scheme of partipant roles, namely addresser-addressee; the data contained very few group-directed messages. In terms of content, the messages are mostly socially oriented (interactive) rather than information oriented (see Herring, 1996), their main purpose being to maintain and promote interpersonal relations. Numerous messages function like quick, local phone-calls (given that very few messages were sent outside the U.K.):5 These involve rapid exchange of news or social arrangements and contain speech acts such as thanks, congratulations, and apologies. Messages that are closer to the transactional end of the continuum involve some form of academic or professional co-operation, such as requests for papers, invitations to lectures and seminars, or co-participation in projects. In all these cases, the messages present the format of personal notes or, more frequently, conversational turns (cf. Herring’s interactive schema of email, 1996), in particular since parts of a previous message that is responded to can be imported in the reply. Such imported portions (quotations) are arguably a simulation of the orientation of participants in conversations to specific prior turns (see Severinson-Eklundh, Part I of this issue).

The time lag between sending a message and receiving a reply to it was normally short, ranging from half an hour to one day. As discussions with email users and my own experiences indicated, email messages of this kind were preferred over e-chat, on the grounds that they allowed the addressees to respond at their convenience. Rapid and frequent exchange of email messages is closely related to brevity of content. Longer messages (i.e., over 200 words) in the corpus as a rule serve a “catching-up” function and mostly encode the addresser’s recent (personal experience) narratives. In face-to-face communication, narratives suspend the turn-taking rules of conversation, granting floor-holding rights to the teller. Comparably, email messages that encode narratives are not organized in the non-narrative format of conversational turns.6

As already noted, the data for this study were collected in the mid-1990s. A comparative study between this earlier phase of data collection and the second phase (which was completed in 2004 and includes data from the same participants) is beyond the scope of the present article. It is worth noting, however, that some processes in the later data suggest a consolidation of textual and social norms on email that was not found in the 1990s. As a result, by 2004 there were certain conventionalized expectations regarding what was tellable and what was not tellable on email (for a discussion, see Georgakopoulou 2004). Another difference is the fact that in 2004 the participants engaged in more systematic ways in a wider range of technologically-mediated communications (e.g., mobile phone calls, text messaging, instant messaging, social network sites), making for an increased density and frequency of interactions with their friends, even if they were not geographically close (see Baym, 2010). Elsewhere (Georgakopoulou, 2005), I have discussed how this nexus of mediated and non-mediated interactions is abundantly invoked on email by means of cross-referencing and intertextual links between past, present, and future interactions.

Contextualization Cues in Email: The Role of Code Alternations

Close sequential analysis of the messages with regard to how participants display and construct their interactional alliances brought to the fore a set of recurrent linguistic devices that are broadly characterized here as “code-alternations.” The term has been more narrowly employed in the literature to refer to the alternating use of two different languages. Auer (1995) specifically postulated code-alternation as a hyperonym for code-switching and transfer and defined it as “the relations of contiguous juxtaposition of semiotic systems such that the appropriate recipients of the resulting complex sign are in a position to interpret this juxtaposition as such” (p. 116). The term reserved in the literature for changes between social or stylistic varieties of the same language is that of style shifts. This study, however, endorses the widely held view in sociolinguistics that code-switching and style shifting are similar phenomena with regard to their socio-psychological dynamics and functions in discourse. Both are points on a continuum with fuzzy boundaries, which presents numerous gradations and manifestations of contact of varieties with different degrees of linguistic independence and discreteness (Fasold, 1984; Milroy, 1987; Siegel, 1995). Code-alternation is thus reinterpreted here as a broadly defined umbrella term that encompasses a continuum of switches of variety, more or less rapid, occurring in the same or different turns. Following Auer (1995), we note that gradual transitions from one variety into the other do not qualify as code-alternations, the essence of which is to be found in rapid, unexpected, and even incongruous departures from the surrounding co-text. The categories of code-alternations identified in the data are as follows:

a.    Word-internal alternations between a Standard Modern Greek or a low variety base and a katharevousa (high variety in the former diglossic situation in Greece) bound morpheme; similarly, a (socially or regionally) marked variety ending added to a word.7 All code-alternations discussed here are very difficult to render in the translated text, and the signposting in the English text is by no means ideal. Bold denotes switches to katharevousa or, generally, formal varieties, while underlining denotes switches to non-standard, informal varieties.

(1)    >Elpizo na diavazis akoma to email sou8
        Alimonon, ksekollame potes apo thn mhxanh mas gia na ksekollhsoume kai tvra?
        Sou euxomai lampron xeimvna kai fusika eimai on line.
        >I hope you still read your email ...
        Alas, as if I ever log off.
        I wish you a glorious winter and of course I’m on line.
        [end of the message]

In the above example, there are two katharevousa endings [alimono-n (alas), lampro-n (glorious)] and one marked regional variety ending [pote-s (never)].

b.   Turn-internal alternations between low and high varieties (including katharevousa). These involve different linguistic levels, e.g., lexical, morphological, grammatical.

         (2)    To de peristeri ... eihe katahesei to sympan ... kai mas ekane to grafeio (o theos na to kanei grafeio) therino. Den mou eipe h Maria telika poios to edioxe, giati perase se allo thema syzitisis, moy eipe mono oti den apepeirathi kan na to dioxei i idia, gia na min tis humixei (os gnoston, ta peristeria einai epikindyna kai sarkovora zoa); ant’autou, etrexe allofron sti grammatea pros voitheian. De ftanei poy giname rentikolo, exoume kai ti Maria na mas leei ti ftaio ego? ...

And the said pigeon ... had crapped all over the place ... and it turned our office (sad excuse for an office) into a sty [lit: summery, open-air]. Maria didn’t tell me who kicked it out, cos she moved into a new topic of conversation, she only told me that she had not endeavoured to kick it out by herself, out of fear that it might attack her (as widely held, pigeons are dangerous and carnivorous animals); instead [in the original prepositional phrase in katharevousa], she ran in a state of dementia to the secretary for assistance [in the original, prepositional phrase in katharevousa]. As if it wasn’t enough that we became a laughing stock, we have Maria on our back saying “it’s not my fault” ...

The extract above owes much of its humorous effect to rapid shifts to formal varieties, the contextual unexpectedness of which is accentuated by their incongruity with the story’s theme (a pigeon’s “desecration” of the narrator’s office).

The above categories of alternations are inextricably bound with the aftermath of the Greek diglossic situation which was resolved in 1976; at that time, katharevousa was officially abolished for education and government affairs in favour of demotic (the then low variety). The resolution of diglossic situations is, as Ferguson (1959) aptly pointed out, a particularly slow process involving a gradual leakage of functions of one variety into those formerly reserved for the other. In the case of Greek, it has bequeathed a fair amount of paired items (doublets) and co-existing elements at different linguistic levels denoting register variation. While Standard Modern Greek as currently shaped inevitably involves a considerable mixture of demotic and katharevousa, the mixing occurring in the examined messages is easily recognizable as exaggerated, incongruous, and comic.

A similar type of code-alternation is also to be found in the English messages, in the form of switches to high/formal or low/informal varieties. Consider, for instance, the use of lexis derived from (socially and/or regionally) low varieties in example 3 below: drinkies, old swinger, bevy, fags, wee, trouble and strife (Cockney for ‘wife’). These varieties seem to possess a covert prestige (Milroy, 1987) for the participants involved. In the example, three friends are trying to organize a social meeting (the interaction between Mark and Shriman has been forwarded to Alex). Excerpts from Shriman’s message which have been imported in Mark’s reply are indicated with “>“):

(3)   Subject: Re: On for drinkies?
       most likely, cc’ing to the trouble and strife
       On Friday, 16 May 1997, Shriman X wrote:
        ________________ Forwarded message ________________
       From: Mark X
       To: Shriman X
       >hello you old swinger, how’s things?
       Great! Just come back from a weekend in Wales.
       >we should meet up for a bevy sometime, maybe with Alex as well?
        A drink would be good. How about the 1st week of June (or the Fri. 23rd May)?
       >We’ll talk about life, women, men, football but not about computers nor
       >libraries. ok by you?
       Can’t talk about football cos Boro have been relegated and will probably lose the Cup
       this Saturday. Now women that’s a different matter ...
            How’s the company of love? [The company that Shriman works for is called Venus]
       Working you hard? Burgers and fags til9 the wee hours of the morning?

Reminiscent of the story in example 2, the code-alternations of the following example are mostly in the direction of formality (e.g., underwater explorations, escapologists, daringly, surreptitiously, venerable, referenced). In this case, too, the body of the messages which are aptly titled “Correspondence catch up” presents a narrative format. The unexpected shifts towards more formal lexis enhance the self-deprecatory tone of the stories (in particular the first one) and add to a humorous, animated narration.

(4)   >Hi Chris + eileen,
       >sorry for not responding earlier (the bt boss 12 hour response time maxim don’t
       >apply here). Anyway, with respect to the mask and snorkel I’m afraid that was a
       >mixed success. the snorkel worked fine but the mask required ridiculous
       >amounts of tension to not let water in so i finally gave up on my underwater
       >explorations. We left successfully the following day [the two couples had spent
       >a week of holidays together in Rhodes, Greece], turned up in paradisi (not) and i
       >had a little accommodation adventure. basically, we went straight to anastasia’s
       >but were not at all impressed and promptly escaped to some rooms a block
       >away when the landlady turned her back. of course not being the greatest
       >escapologists we forgot cathy’s fave espadrilles in the room so we daringly
       >went back in, collected them, then surreptitiously left again (many litres of sweat later). ...
       Catching up again ... Spent the week or so in Cumbria, Glasgow and the M6
       north and south. Eileen’s sister was married last Saturday to an Irishman in a kilt.
       Eileen was a bridesmaid, I was chief dogsbody and bride’s mother’s transport -
       less glamorous and more stressful than budget close week at work. Ceilidh
       (Scottish country dancing disco!) in the evening with dances called the “Gay
       Gordons” and “Strip the Willow”. Hilarious - lots of two left feet, counting beats
       after venerable quantities of alcohol. Big EDS [the company that the story’s
       bridegroom works for], even referenced in the bride’s father’s speech for
       bringing them together - “E’xceptional D’ating S’ervice”. Ho Ho!

c.   Switches from Greek (matrix language frame, Scotton, 1993) to English (italicized in the examples),10 which have to be viewed in connection with the language contact situation experienced by the speech community in question (see the following switches to English in example 10: knowing how busy you are, we would be unreasonable if we asked for more, bloody foreigners, native speakers, male, middle class, etc., just kidding).

d.   Shifts to a specific style identified with a given discourse type (and the associated linguistic choices) and recognized as incongruous within the given local context. The direction of these shifts is not fixed or pre-determined; their role, however, is in most cases to parody or undermine the evoked style.

         (5)    i.    Skeptomai pragmati na syggrapso. Sti Naxo mou irthe mia diathesi, mia empneusi, alla tora mou efuge. Skeftika na grapso ena diigima me titlo “To arrosto gourouni”, opou tha afigoumai to taxidi mou sti Sandorini me to Giorgo, kai pos telika auto to arrosto kai aidiastiko gourouni ito to sumvolo mias kapoias ptosis, mias sapilas, to prominima mias dustuxias. Alla prepei na vro kai mia alfa ploki, den mporo na grapso 10-15 selides gia to gourouni

I am truly contemplating writing. In Naxos I had an insight, an inspiration, but it’s not there any more. I thought I should write a short story titled “The sick pig”, where I’d relate my trip to Sandorini with George, and how that sick and repulsive pig was the symbol of a certain decay, a degeneration, the omen of misfortune. But I have to find an X story, I can’t really write 10-15 pages on a pig ...

                 ii.    >time to get down to some real work and add some value!
looks like my main employers want me to become a database manager and get a bunch of lackies to maintain our dbs to boot. ascending the slippery slopes of management - will probably have to join you on one of those foot-stomping, value-adding, catalysing management courses soon!

The writer in example i. above appropriates the discourse of literary critics in Greece to convey a tone of sarcasm and self-deprecation (see also the discussion of humour below) for her ambitious goal of becoming a writer. Similarly, the management-speak in English example (ii) is evoked by both addresser and addressee to be parodied. It can also be seen as a device for disarming the addressee’s possible criticism that the addresser is boasting of his job promotion.

e.   Shifts to language play that frequently involves incongruous mixing of high and low varieties. The arrows → ← highlight the words involved in the play. (For more examples, see 8.ii.)

         (6)    perimeno dema me →xeirografo (i xeiroterografo)←
Awaiting a parcel with a →manuscript (or worsescript)←

[in the original, a play on the first word of the compound noun: xeiro means hand, and xeirotero means worse]..

f.   Intertextual references or allusions to (assumed to be shared) cultural texts (see Hodsdon-Champeon, Part I of this issue).

         (7)     i.   Subject: Paraskeyh boreis? Oxi, oxi!!
Subject: Can you make it on Friday? No, no!!!

The above is a reference to a well-known Greek song in response to the addressee’s original suggestion to meet up on Friday (see also the allusion to a Greek movie in example 11). Below, the writer emphasizes the news about the end of his military service with a switch to English (CITIZEN), which, combined with the shortening of his name (he is normally called Thanasis), can be read as an intertextual reference to the film “Citizen Cane” (Thanos is closer to Cane than Thanasis):11

        (8)    ii.    Ela... epitelous apoluthika!
Elpizw na ta poume arxes Iouliou

Hi ... I have at last finished my military service!
Hope to see you beginning of July

Mediated Contextualization Cues

All the above forms of code-alternations should be seen as points on a continuum of linguistic choices rather than as discrete and clear-cut categories; their separation here is in the interests of analytical convenience. In fact, their common denominator is that they capitalize on “bisociation” (Norrick, 1994), i.e., on bringing together incongruous linguistic forms that are felt to be inappropriate for use in the specific context or in their combination. They thus exploit symbolic meanings and indexical values underlying different linguistic codes. For instance, katharevousa, which commonly figures in code-alternations, carries for members of the Greek speech community the historical associations of a pretentious, pompous, vacuous, and authoritarian code; as a result, it has time and again been the object of parody within artistic and literary circles. Thus, the use of katharevousa in informal communication between friends renders the bisociation effect more salient. This frame-breaking juxtaposition and blending of different semiotic contexts typically introduces into the data a humorous element. As incongruity theories of humour have asserted, an unexpected switch from one frame to another, initially incompatible, provides an unmistakable cue to laughter (Raskin, 1985). Spontaneous conversational joking has been found to thrive on such a bisocation of conflicting frames (Norrick, 1994). The role of humour in participant relations on email is discussed below (see also Baym 1995b). Here, it is worth noting that, in terms of stylistic effect, the code-alternations in the data contribute to the creation of a multivoiced and pastiche or hybrid style; in Bakhtinian terms, they foreground the texts’ polyphony or heteroglossia. This can be aligned with the findings of previous studies, according to which mediated discourse is typified by a mixing (bricolage)—or even recasting and parody—of genres. In a similar vein, the abundance of playfulness, humour, and wit, particularly in synchronous modes of CMC, has been amply documented (see, e.g., the articles in Danet, 1995; Danet et al., 1997). Such phenomena have been mainly attributed to medium-specific factors, for instance, the disinhibition engendered by the novelty of the medium, which in turn gives rise to increased creativity and risk-taking. It has also been argued that the medium allows for an essentially postmodernist construction of social interactional worlds: worlds with attention to nuances of language and symbolism and concern for a realisation of the power of language (Reid, 1991, p. 32).

Such explanations can be critiqued for their disregard of the sequential and local context embeddedness of e-discourse. What they fail to shed light on is how instances of a hybrid style function as strategic activities in their contexts of occurrence, what their situated roles and meanings are, and how participants make sense of them (for a study of the in situ functions of humour on a discussion list, see Baym, 1996). Rather than treat code-alternations as predictable and generic patterns with a fixed discourse behaviour, this study argues that they are resources for the constitution of socially meaningful activities; more specifically, they are the main textual signals that allow participants to mark their interactional stances and positionings towards their addressees. In this respect, the roles and functions of code-alternations in the data can be encapsulated by the notion of contextualization cues (Gumperz, 1982, 1992): (mainly) non-referential, verbal or non-verbal signs that help speakers hint at or clarify a certain set of expectations, attitudes, and social actions associated with the discourse activity, thus helping listeners to make inferences about what is meant. As such, they comprise “all activities by participants, which make relevant, maintain, revise, cancel ... any aspect of context, which, in turn, is responsible for the interpretation of an utterance in its particular locus of occurrence” (Auer, 1992, p. 4). The notion of contextualization cues has been aimed at capturing processes of construction of meaning and inferencing in online, interactive discourse (e.g., Auer & Di Luzio, 1992). In particular, code-switching and style shifting have been amply documented as contextualization cues in face-to-face interactions which enable speakers to communicate social meanings and accomplish various interactional goals (Milroy & Muysken, 1995).

The present analysis attests to the validity and relevance of the notion of contextualization cues for email discourse as well. The contention is that the contextualization roles of code-alternations in email messages are modeled on conversations. Cues in conversation serve to highlight certain phonological or lexical strings vis-à-vis other similar units, thus functioning relationally and in context. To act as foregrounding mechanisms, they have to mark a departure from the text’s local norms. This also applies to the code-alternations of email messages. In addition, the inferencing of cues leads to interpretation by contrast (cf. the incongruity involved in the code-alternations at hand) or by inherent meaning potential, which restricts the number of plausible inferences as to what this might be about (Gumperz, 1992). The importance of conventionalized meanings is evidenced in the case of switches to katharevousa. As already discussed, in addition to the contrast that switches to katharevousa establish, they evoke a set of symbolic functions and meanings conventionally associated with the code.

Although modelled on conversations, email contextualization cues are also tailored to the functional requirements of the communicative context of email. The main contextualization cues of spoken discourse, which include prosody (e.g., stress, intonation), paralinguistic features (e.g., tempo, laughter), and other non-verbal signs, cannot easily be adapted in an email environment (but cf. Cho, Part I of this issue): This is despite the best efforts of the participants, who frequently attempt to overcome the constraints of email by resorting to the use of orthographic innovation, emoticons, and other graphics. Overall, the cues of the messages examined mainly operate at the level of code choices from among the options of a linguistic repertoire. As a result, code-alternations prove to be the most important and frequently used contextualization cues. Previous work on conversation has not indicated this pivotal role of code-alternation as a contextualization cue. Furthermore, it has not attested to the orchestration and synergy of different types of code-alternations, such as the ones reported here, in the contextualization process. Since they cannot depend on their addressees to catch non-verbal nuances, email writers seem to maximize the capabilities provided by manipulations of the verbal code. They thus convey through code-alternations much of the subtle information that would be conveyed in conversation by variations of the voice, gestures, and other non-verbal cues. In view of the above, a comparative study could use contextualization cues as a point of departure for exploring differences between face-to-face and email discourse. In the messages examined here, code-alternations as contextualization cues are at the heart of signalling symmetrical participant alignments.

Cueing Intimacy and Symmetrical Alignments

The data at hand are socially oriented in content, and their main purpose is to maintain and promote interpersonal relationships. Participants are intimate with each other, and they have already established an ingroup membership before entering email communication. Their shared assumptions not only involve their linguistic, social, and cultural identities (young people living and working in London) but also the history of their past interactions. It is thus not surprising that code-alternations as contextualization cues of participant roles and relations cue symmetrical, egalitarian, and solidarity-based relations or alignments. The term alignment is used here in its broad meaning to include not only agreement with a position expressed in the discourse, but also a participant’s affiliative orientation to the talk in progress and to the participants (Goodwin & Goodwin, 1990). In this way, alignments cover the speakers’ set of solidarity-based and friendly interactional stances or positions towards the addressees. In Goffman’s model (1974, 1981), from which “alignment” is adapted, two more notions provide links to the dynamics of interactional positions in communication: footing and frame. Briefly defined, footing encompasses the speakers’ and hearers’ positions vis-à-vis one another as well as oneself, as expressed in the way they manage the production and reception of an utterance (1981, p. 28). Frame is a relational concept that provides the addressees with instructions on how to understand a discourse message and the relation between the participants. It can thus be seen as a set of assumptions and expectations about the type of activity engaged in and the organizational or interactional principles by which the activity is defined and sustained as experience (Tannen & Wallat, 1987). Frames are linguistically signalled and realized by a variety of contextualization cues.

In the data, cueing alignments by means of code-alternations embraces the following acts: i) reframing participant alignments as symmetrical, ii) maintaining or protecting alignments which includes aversion of risk to them, and iii) indexing (i.e., displaying, (re)affirming) and enhancing alignments. These are discussed below.

Reframing Alignments

Reframing alignments mostly occurs in email messages with predominantly information-oriented and/or professional content. In such cases, code-alternations signal a rapid departure from a formal linguistic variety deemed appropriate for the issues discussed. They thus introduce rapid shifts from official to personal footings (Tannen & Wallat, 1987) which reframe the communication as an informal interaction between friends who happen to be involved in a professional exchange of some kind. In this way, they enable addressers to project themselves during formal talk in the capacity of a friend.

         (9)     i.   Egw proswpika vrisko oti den einai eparkos tekmiriomeni i anangaiotita tis synyparxis pollon dekton, se antithesi me dimeri programmata. Telos sta vivliografika zitimata kalo tha itan na ksanaskeftoume kapoia pragmata giati toulaxiston i germaniki vivliografia einai aneparkis.
       Ola ayta prepei na ta ksanaskeftoume sovara. Dhladhs, o dromos pou blepw egw ayth th stigmh einai na steiloume faxion stas Bryxellas osonoypw kai na rwtame pou allou mporoume na to xwsoume to kolokuthi. Auta prepei na ta kanw egw???

Personally, I do not think there is a strong enough case for the co-existence of numerous users as opposed to bilateral projects. Finally, with respect to the literature review, it would be wise to revise some of our current thinking because at least the german section is insufficient.
       We need to give more thought to all these issues. So, the way to go as I see it is to send a fax to Brussels as soon as possible asking where else we can stuff the thingy. Should I do these things??

                  ii.        Tha synesthna loipon na epikoinwnhsete me tous armodious sth xwra sas. Pantws, ta calls tha prepei na ginoun gyrw sta telh Maiou kai oi prothesmies na kleisoun gurw ston Augousto (gia osous programmatizoun diakopoules).
       Exoume dhladh ligon kairo gia na ***skeftoume*** oxi na anapautoume ... (kai oute na apautwthoume? ... polu moiazan metaksu tous aftes oi duo lekseis grammenes fragkolebantinisti).

       So, I’d suggest that you should contact the relevant people at your end. In any case, call will be out by the end of May and the deadline is some time in August (for those who were planning on a little holiday).
       So we have a little bit of time to ** think*** but not to rest ... (or to fool around... funny how similar those two words are written in Greek with Latin characters).

                 iii.  Katarxin se epishma keimena parakalo to onoma mou na anagrafetai ws XXXXX kai se epishma documents efoson etsi me kseroun oli tous edo. 2. I tipissa lei oti mas voleyei na katevoume san Ellhniko prama ...

First of all, I request that my name be written as XXXXX in formal documents since this is how I am known here. 2. The woman [in the original, the female equivalent of the English “dude”] says it’s better if we go on the pitch as a Greek team.

In the above examples, a sudden shift in variety relaxes the formal discourse and reaffirms the unmarked rights and obligations set of a symmetrical relationship (Scotton, 1993), which has been momentarily overshadowed by official footings. For instance, a jocular aside in 8.iii shifts participant relations towards the intimate and playful end of the continuum. The shift to an informal variety is clearly set off from the rest of the text by preposing the number 2 and is rendered more salient by immediately following a formally-phrased request. The style shift in example 8.ii is combined with a pun, based on the near-identical pronunciation of two words that are semantically unrelated. Its humorous effect helps to establish a familiar frame of reference that stresses intimacy.

Maintaining Alignments

The maintenance of alignments in any interaction involves the avoidance or mitigation of face-threatening acts (acts which by their nature run contrary to the face wants of the addressee and/or the addresser; Brown & Levinson, 1987). Code-alternations proved to be frequent qualifying devices or hedges of such acts, in that their occurrence enhanced or mitigated the acts’ force and, subsequently, the extent to which the speaker could be held accountable for them. As hedges, code-alternations are frequently part of the expression of requests. Requests normally pose a threat to the addressee’s negative face (i.e., the basic want to maintain claims of territory and self-determination). While face-threatening acts can be performed without any redressive action, baldly and on record, they are commonly accompanied by various devices that avert or counterbalance the potential face-damage (code-alternations, in this case):

        (10)  i.    Tha hthela na kserw mhpws exoyne tipotsi corporakia available. Prokeitai na kataskeyasoyn ti to paromoion?
I would like to know whether they have any little corpora available. Are they going to construct anything similar?

                ii.    To thlefono mou, just in case: 111111 (peri oikias omiloume).
Here’s my phone number, just in case: 111111 (we are discussing the number of our abode).

                iii.    den sas akouw sto e-mail.
einai mhpws giati oi mhxanes mas einai sunexeia ftwmata h giati →sigeite←?

                       I don’t hear you on email.
is it that your machines are dead or are you →silent← (literary register)?

The examples above encode instances of more-or-less indirectly expressed requests. The code-alternations accompanying them are aimed at mitigating the imposition of the act performed by introducing a jocular dimension and by cueing the assumptions the addresser shares with the addressee. Thus, the direction of the switch towards greater formality or informality introduced by code-alternations is not in itself significant. It is rather the act of switching and what that indexes for participants that are significant for a message’s facework (e.g., redress of impositions, attendance to addresser’s and/or addressee’s face). Switches to both formal and informal varieties are ultimately appeals to the participants’ shared ingroup knowledge and shared understandings of what counts as unexpected, incongruous, and lighthearted in the activity of email in which they are involved.

Code-alternations as part of facework are particularly prominent in the following extracts, which present an orchestration of humorous switches to English (mostly in the form of asides) with intertextual allusions and language play.

(11)   >Epishs legame poso tha htan efikto (knowing how busy you are) na mas
         >ekanes proof-reading kai genikotero sxoliasmo 1 kefalaiou (we would be
         >unreasonable if we asked for more
). Kathoti bloody foreigners kai
         >prospathoume ksereis vrume anthropus (native speakers, academics, male,
         >middle class etc, just kidding
) na tu riksun mia matia →ews dyo←. Gi ayto
         >→leeeew mporeiteeeeeeee leew (kyrie Samiwtakh lew)← an soy steiloyme
         >kati tis epomenes dyo vdomades na to diavasis mexri telh Aprili? An ne,
         >→let me know (email me know)←.
         nai. mono pou ego fevgo to savvato →leo gia ena taksidaaaaki← sto
         Swaziland kai tha gyriso th (dytiki) megali paraskevi, opote (my favourite
) to panepisthmio mas tha einai kleisto logo paskhatos kai
         ksananoigei kata →tin oxi kai toso megali← tetarti. opote tha argiso men alla
         nai poli efaharistos. ok? ola gia hari ton bloody foreigners (just kidding) afta
         perimeno dema me →xeirografo (i xeiroterografo)←
         >We were also saying how feasible it would be (knowing how busy you are)
         >for you to do some proof-reading and provide general comments on 1
         >chapter (we would be unreasonable if we asked for more). Being bloody
we are looking for people (native speakers, academics, male,
         >middle class etc, just kidding
) to give it a once →or twice over←. So I’m
         >→aaaaaasking could you I’m aaasking (Mr Samiotakis I’m asking),
         >[allusion to a popular Greek movie from the 1960s]← if we sent you
         >something in the next two weeks could you read it before the end of April?
         >If yes, →let me know (email >me know)←.
         Yes. Only I’m off on Saturday →I’m saying on a teensy weensy trip
         [allusion to the same movie]← to Swaziland and will return on the
         (occidental) good Friday, at which time (my favourite conjunction) the
         university will be shut for Easter and will reopen on the →not so good←
         wednesday. so I’ll delay a bit but yes of course with pleasure. Everything for
         the →bloody foreigners← sake (just kidding). That’s all. Awaiting a parcel
         with a →manuscript (or worsescript)←.

In the first message, code-alternations qualify the act of requesting. The adoption of a comparable style in the reply to the message is a signal that the cues of shared assumptions have been successfully recognized; their reiteration is ultimately a reaffirmation of the intimacy frame set by the first message.

Apologies, a potentially face-threatening act for the speaker’s positive and/or negative face, are also realized by means of code-alternations (for a discussion and examples, see Georgakopoulou, 1997a, pp. 154-155). Furthermore, code-alternations are found in dispreferred responses (i.e., the least expected and least preferred responses in an adjacency pair of turns; e.g., refusal (instead of acceptance) of an invitation, disagreement (instead of agreement) with an assessment; see Pomerantz, 1984). Dispreferred turns, being largely destructive of social solidarity and the relationship between addresser and addressee, have been shown to be accompanied by a wide range of markedness features in monolingual conversations and by code switches in bilingual conversations, aimed at weakening the threatening act and disarming the addressee’s potential criticism (Milroy & Muysken, 1995).

         (12)  fkaristo gia tim brosklisi, isos tou hronou, otan douleoun →ta apala eidh (software)←.

                  thanks for the invitation, but maybe next year, when the →soft-ware← [literal, nonsensical translation] is working.

The style shift above is arguably a hedge of the dispreferred act of rejection of an offer.

The redress realized by means of code-alternations in the above examples generally appeals to positive face concerns (i.e., the desire to be liked, appreciated, and approved by others); it satisfies the need for connection and involvement and stresses common ingroup membership. A preference for positive politeness strategies has been documented for the Greek society and culture as a whole (Sifianou, 1992); this contention, however, remains to be tested and empirically validated in a variety of interactional contexts. In the data at hand, the use of positive politeness strategies, instead of being treated as the outcome of sociocultural norms alone, must be seen in connection with the messages’ local contextual parameters, in particular the participants’ intimate relations and the communicative goals of reaffirming these relations. Code-alternations as positively polite strategies frequently appear in the English messages, too. In the example below, Michael and Andrey are in the process of a professional conflict resolution. As part of an attempt to patch things up, Michael invites Andrey to his place for dinner. A switch to German fittingly accompanies his request.

(13)   >So you are coming over on Freitag yeah?
         Yesh, I’ll see you on Friday!
         7pm is fine, innit!

The response, which also presents a code-alternation (innit), is framed as jocular (see discussion below), as attested to by the impersonation of Sean Connery (yesh) and the greeting (xo).

Humour: Indexing and Enhancing Alignments

Code-alternations on email can be brought together by the much sought-after function of participant bonding, which involves not only displaying but also strengthening intimacy and solidarity. The main vehicle for doing so is the association of code-alternations with an element of joking and levity, which render the social contacts in question more enjoyable. It is widely accepted that situational humour helps to establish familiar frames of reference (e.g., Boxer & Cortés-Conde, 1997). In particular when it relies on bisociation, as is the case in these data, it acts as a test of intelligence for the addressees: It sets a play frame against a backdrop of shared assumptions, thereby inviting the addressees to search their repository of sociocultural knowledge and prove their group membership by accurately inferring what is being signalled. In Norrick’s (1994) terms, this is a test that routinely aims to find common ground, to retrieve insider knowledge and past history: Passing it means sharing in the payoff of amusement and increased solidarity.

At another level, humour builds affiliative alliances by contributing to the addresser’s positive self-presentation: It carries connotations of urban sophistication, prestige, and ability to access and participate in multiple discourses and social roles. Furthermore, conveying a sense of humour is a virtue in numerous societies, not least the western ones, especially when it is combined with an element of self-teasing and self-denigration. In the latter case, the speaker communicates the ability to be self-critical, to laugh at problems and overcome them (Jefferson, 1984). Although metalinguistic in essence, the humour springing from code-alternations frequently puts the writer in the centre of verbal playing and teasing.12 Even at first glance, most of the examples already presented in this study incorporate instances of the writer’s self-deprecation (see, e.g., the stories in 2 and 4, in 5.i where the writer parodies her ambitious wish to become an author, in 5.ii where the writer, by means of self-deprecation, attempts to show modesty about his job promotion, and the self-deprecatory remark, bloody foreigners, in example 10 as part of hedging a request).

According to Boxer and Cortés-Conde (1997), self-denigrating joking bonds without biting, that is, it strengthens participant alignments without endangering them. By contrast, while teasing the addressee constitutes an activity with stronger bonding results, it is a higher risk-game that can “bite” (annoy, offend). Teasing the addressee is combined in the data with code-alternations: This can be seen as a hedged way of bonding with biting, i.e., as a device which qualifies the bite of the teasing act. In example 10, for instance, the language play of the closing section qualifies the teasing of the addressee concerning the bad quality of her manuscript. Consider also the following examples in English and Greek, respectively:

        (14)  i.    So, what’s new? Ich bin Deutschlanderbonken? hee hee [reference to the addressee’s recent trip to Germany]
[later on] Can I get your thesis (personally autogrpahed?) in paperback when it comes out?

                ii.    Epitelous epikoivwvnsame! Anoigoume kai ligo to rimadi to computer. Ase bre paidi mou xtes trigurizovtas stous diadromous tou economics eida poster me to onoma sou Dr X, egrafe. ... [later on] Mias kai to efere o logos to Sabbato kati kavovizoume gia clubbing. Fusika dev vomizw va sas endiaferei esas kati toso potapon, alla pote den gvwrizei kaveis eis aytov tov vtounia. ... [later on] Kane mou thn timh na mou apadhseis giati allivs ua se mastigvsv anhlevs

                iii.    At last, we’ve gotten in touch! You should switch your bloody computer on every now and then. Geez kid [lit. “my child”: affective with teasing overtones in this case] while walking down the corridors of economics yesterday, I saw your name on all these posters saying Dr X. ... [later on] Which reminds me, we’re planning to do some clubbing this Saturday. Of course I don’t think you’re [ironic use of polite plural] interested in something as menial, but one never knows in this world. ... [later on] Do me the honour and reply to this or else I will whip you relentlessly ...

The teasing tone is mitigated in both examples by means of of code-alternations (the humorous code-switch, Deutschlanderbonken, in i; switches to katharevousa and formal varieties in ii).

Concluding Discussion

This article has focused on the ways in which participants locally construct and make sense of the positions, stances, or alignments towards their addressees in a corpus of private email messages. In particular, it was argued that the main linguistic strategies for contextually cueing participant relations involve various types of code-alternations (Greek-English code-switches, style shifts, intertextual references, etc). As summarized in Table 1 below, in terms of discoursal functions, these code-alternations have in common an ability to introduce into the text incongruous associations between elements from different varieties and contexts. In terms of interactional functions, they establish footings of symmetrical alignments encompassing the acts of indexing, enhancing, and maintaining intimate relations. At the heart of these different constitutions of alignments is the establishment of humorous and playful frames that appeal to the participants’ shared assumptions and group membership and attend to the addresser’s positive self-presentation.

Table 1

Table 1. Summary of findings

One of the main aims of this study was to draw on and apply tools and methods from areas such as conversation analysis and interactional sociolinguistics whose impact has not been adequately felt or explored in studies of CMC. More specifically, the goal was to demonstrate the need for exploring the complex ways in which CMC invokes and is shaped by its contexts of occurrence. The point of departure was thus that email discourse, like any other form of discourse, is not detachable from its context, which involves not only the medium but also sociocultural stances, attitudes, and values; participant identities, roles, and relationships; communication functions; locally emergent interactional projects, etc. From there on, the discussion attempted to demonstrate how, in the data at hand, the use of linguistic strategies (code-alternations, in this case) would not be fully comprehensible if contextual parameters were not taken into account. The most important parameter in these data was the participants’ intimate relation, which also involved a high degree of shared knowledge and assumptions and a history of previous interactions. This relation was interconnected with the socially-oriented nature of the messages’ topics and with their interactive organizational format. In a similar vein, unlike public email messages, the data at hand presented a dyadic participation framework which rendered them comparable with conversational interactions, in that interactants were accountable to their email partners rather than to a whole group (Baym, 1996). Private email messages of this sort, rather than forming a self-contained universe of discourse with no grounding in communication realities outside the virtual ones, follow upon and supplement face-to-face interactions in various ways. A case in point is that the “real-life” personal and sociocultural identities of the participants are well known to one another, so the shared context is from the outset significantly enriched compared to other CMC fora, such as Internet Relay Chat and Usenet (see Paolillo, this issue).

The types and uses of code-alternations in the data are also intimately linked with the linguistic and sociocultural features of the specific discourse communities of email users. The Greek participants are in a language contact situation with English, which accounts for the switches to English. Both Greek and British participants are educated and highly computer literate, with an arguably good metalinguistic awareness of language subtleties and register variation in their language. In addition, their young age could be claimed to underlie an age-linked predilection for the use of marked (i.e., non-standard) varieties (Widdicombe & Wooffitt, 1995).

The impact of medium-specific constraints on the code-alternations in the data has to be viewed in interaction with the above context features. It is important to note that these context features render the data akin to face-to-face conversations between friends. This was reflected in the interactive organizational format of the messages, which was modeled on the turn-taking structure of conversations or on storytelling. In view of the above, it is hardly surprising that various types of the code-alternations discussed here, e.g., language play, code-switches and style shifts, have also been found to act as rapport-building strategies in conversations and, thus, cannot be claimed to be specific to email. Nonetheless, previous work does not indicate that the specific orchestration of code-alternating devices, with its richness and variety, is particularly common or pivotal in face-to-face interactions. In contrast, research attests to the importance of non-verbal (e.g., prosodic, gestural) contextualization cues, frequently co-occurring with one another to achieve the desired level of redundancy, which facilitates the addressees’ inferencing. It is therefore plausible that the increased reliance on a code-centered contextualization cueing in the data is intimately linked with medium specificities. These come to bear not on the choice of the devices per se, but on the ways in which the devices are put to use to structure activities and create an emergent context for the participants.

In particular, two communicative context features of email can be argued to shape decisively the devices’ patterns of use: the addressees’ physical absence, and their inability to provide online feedback due to the asynchronous nature of the communication. Although more specified than in most types of written discourse, the physical and temporal context of interaction is still not shared by participants on email. Meaning cannot be co-constructed and negotiated on-the-spot, as most of the non-verbal cues of oral interaction are not available. Email users cannot fully replicate the presence of the addressee, which, among other things, allows for immediate feedback as to the successful uptake of intended meanings and subtleties. As a result, they carry through various types of code-alternations and strategic manipulations of the verbal part of the message what would otherwise be delegated to different (mainly non-verbal) signals in conversations (cf. Baym, 1995b).

The above is also evident in narrative, where reliable comparisons with everyday storytelling can be made on the basis of my previous study of Greek oral narratives (Georgakopoulou, 1997b). Narratives on email base their evaluation component (Labov, 1972), i.e., the encoding of tellability, on code-alternations. Manipulations of the code can undermine, parody, or enhance the teller, the (absent) parties talked about, and the discourse activity itself. In addition, they can build affiliative alliances with the addressees by enacting shared assumptions in witty, humorous, and playful ways. This heavy reliance on code-alternations is inextricably bound with the medium and the expectations associated with the discourse activity of email. Specifically, the activity was found to be cued by the participants as one which capitalized on multivoicedness and strategically appropriated discourse; likewise, the mood (key) in which it was performed was cued as jocular, lighthearted, and informal. By contrast, evaluation in everyday storytelling between intimates in Greece is deeply embedded in narrative action; similarly, participant alignments are constructed through highly dramatized deliveries. Participants thus bond through the creation of visual and aural performances which activate prosodic and gestural cues.

In addition to providing insights into the interaction between linguistic strategies and the context of email, this study’s findings shed light on local occasionings of CMC shared discourse norms by specific linguistic and sociocultural groups. Within a framework of background expectations about and knowledge of the type of activity engaged in, specific user communities and individual participants draw upon their linguistic resources in order to create the emergent contexts of their interaction. The Greek email writers in this study were found to use English and shifts to katharevousa among their resources for cueing symmetrical alignments. Greek discourse communities that lack those resources would arguably draw upon other, comparable discourse strategies. What can be hypothesized on the basis of this study is that in all those cases, code-centered phenomena would still play a vital role in the construction of participant alliances. This was in fact the case in the British messages, which also exemplified playful and incongruous shifts to informal and formal varieties as indexes of participant alignments. Further research could explore the extent to which comparability in the pragmatic uses of code-alternations can be established cross-linguistically and cross-culturally in email discourse.13

As a final note, it is worth reiterating the importance and necessity of contextual approaches to the study of computer-mediated modes.14 Sensitivity to the context is a prerequisite for a full understanding of a medium that is ultimately defined by an impressive multiplicity of arenas for linguistic communication.


  1. Exceptional examples of work with such a focus include Baym (1995a, 1995b) and Cherny (1999). For an overview of relevant studies and trends in CMC research, see Georgakopoulou (2003); for more recent studies with this focus, see the chapters in Thurlow and Mroczek (2011).

  2. An earlier version of this chapter, with a focus on conventionalized generic features of email, can be found in the International Journal of Applied Linguistics, 7(1997), reprinted in A. Georgakopoulou and M. Spanaki, Eds. (2001).

  3. Conversation analysis arose primarily through the insights of Sacks (1972) in collaboration with his colleagues (e.g., Schegloff 1979), while interactional sociolinguistics originated in the work of Goffman (1974, 1981) and Gumperz (1982). For a discussion of both approaches to discourse, see Schiffrin (1994).

  4. To study language use in social networks, Milroy was introduced to participants as a “friend of a friend.” In the present study, permission to use the examined messages was granted by all the participants involved in the exchange of the messages. All examples cited here make use of pseudonyms except for cases in which permission to use the real names was granted by the participants.

  5. Since the first phase of data collection, mobile phone use and text messaging have exploded. In the second phase (end of 2000-2004), however, it transpired that the given sample of participants (by then well into their thirties) still preferred email over text messaging but mostly rang their friends on their mobile phones.

  6. For the division of discourse types into narrative and non-narrative and the relationship of the two modes, see Georgakopoulou and Goutsos (1997, 2000).

  7. Note that Greek is a highly inflected language.

  8. The participants could not have typed their messages using the Greek alphabet (computers at the time required significant customization for users to be able to use Greek fonts). As a result, they resorted to transliteration. The transliteration conventions employed in the data present a striking variability even within a single message. The data are presented here exactly as they were originally typed.

  9. Typing and spelling mistakes are left uncorrected in the messages.

  10. One-word language transfers, e.g., “deadline,” “clubbing,” and “email,” which are frequently used in the examples, are not included in this type of code-alternation, since they are more or less integrated loans in Greek.

  11. In a post-hoc interview, the addresser verified that he had intended for the closing formula to be read as an allusion to “Citizen Cane.”

  12. A study with a different emphasis could shed light on gendered aspects of this choice, given that women have been found to capitalize more on self-deprecation in various face-to-face interactional contexts (see, e.g., Coates, 1996). Some evidence that women have a greater tendency to self-deprecate in CMC as well can be found in Herring (1994).

  13. For a current state-of-the-art discussion of code-switching in CMC, see Androutsopoulos (forthcoming).

  14. Since the research for this study was conducted, contextual approaches to CMC have increasingly taken hold (as attested to, e.g., in the articles in Androutsopoulos, 2006 and the contributions to Thurlow and Morczek, 2011).


Androutsopoulos, J., Ed. (2006). Sociolinguistics and computer-mediated communication. Special Issue, Journal of Sociolinguistics, 10.

Androutsopoulos, J. (forthcoming) Code-switching and CMC. In S. C. Herring, D. Stein, & T. Virtanen (Eds.), Handbook of the pragmatics of computer-mediated communication. Berlin: Mouton de Gruyter.

Auer, P. (1992). Introduction. John Gumperz' approach to contextualization. In: P. Auer & A. Di Luzio (Eds.), The contextualization of language (pp. 1-38). Amsterdam/Philadelphia: John Benjamins.

Auer, P. (1995). The pragmatics of code-switching: A sequential approach. In: L. Milroy & P. Muysken (Eds.), One speaker, two languages: Cross-disciplinary perspectives on code-switching (pp. 115-135). Cambridge, UK: Cambridge University Press.

Baym, N. K. (1995a). The emergence of community in computer-mediated communication. In: S. G. Jones (Ed.), Cybersociety: Computer-mediated communication and community (pp. 138-163). Thousand Oaks, CA: Sage.

Baym, N. K. (1995b). The performance of humor in computer-mediated communication. Journal of Computer-Mediated Communication, 1(2). http://jcmc.indiana.edu/vol1/issue2/baym.html

Baym, N. K. (1996). Agreements and disagreements in a computer-mediated discussion. Research on Language and Social Interaction, 29, 315-345.

Baym, N. K. (2010). Personal connections in the digital age. Cambridge, UK: Polity Press.

Biber, D. (1988). Variation across speaking and writing. Cambridge, UK: Cambridge University Press.

Boxer, D., & Cortés-Conde, F. (1997). From bonding to biting: Conversational joking and identity display. Journal of Pragmatics, 27, 275-294.

Brown, P., & Levinson, S. (1987). Politeness: Some universals in language usage. Cambridge, UK: Cambridge University Press.

Cherny, L. (1995). The MUD Register: Conversational modes of action in a text-based virtual reality. Ph.D. Dissertation, Stanford University.

Cherny, L. (1999). Conversation and community: Chat in a virtual world. Stanford, CA: CSLI.

Coates, J. (1996). Women talk: Conversation between women friends. Oxford, UK: Blackwell.

Danet, B., Ed. (1995). Play and performance in computer-mediated communication. Special issue, Journal of Computer-Mediated Communication, 1(2). http://jcmc.indiana.edu/vol1/issue2/

Danet, B., Ruedenberg-Wright, L., & Rosenbaum-Tamari, Y. (1997). "HMMM...WHERE'S THAT SMOKE COMING FROM?" Writing, Play and Performance on Internet Relay Chat. Journal of Computer-Mediated Communication, 2(4). http://jcmc.indiana.edu/vol2/issue4/danet.html

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

Davies, B., & Harré, R. (1990). Position: The discursive position of selves. Journal of the Theory of Social Behaviour, 20, 43-63.

Du Bartell, D. (1995). Discourse features of computer-mediated communication: "Spoken-like" and "written-like". In: B. Warvik, S. K. Taskanen, & R. Hiltunen (Eds.), Turkuensia Anglicana: Organization in Discourse (pp. 231-239). Turku, Finland: Turun Yliopisto.

Duranti, A., & Goodwin, C. (Eds.). Rethinking context: Language as an interactive phenomenon. Cambridge, UK: Cambridge University Press.

Fasold, R. (1984). The sociolinguistics of society. Introduction to Sociolinguistics v. 1. Oxford, UK: Blackwell.

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

Ferrara, K., Brunner, H., & Whittemore, G. (1991). Interactive written discourse as an emergent register. Written Communication, 8, 8-34.

Foertsch, J. (1995). The impact of electronic networks on scholarly communication: avenues to research. Discourse Processes, 19, 301-28.

Georgakopoulou, A. (1997a). 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.

Georgakopoulou, A. (1997b). Narrative Performances: A study of Modern Greek storytelling. Amsterdam/Philadelphia: John Benjamins.

Georgakopoulou, A., & Spanaki, M., Eds. (2001). A reader in Greek sociolinguistics. Berlin: Peter Lang.

Georgakopoulou, A. (2003). Computer-mediated communication. In: J. Verschueren, J-O. Östman, J. Blommaert, & C. Bulcaen (Eds.), Handbook of Pragmatics (2003). Amsterdam/Philadelphia: John Benjamins.

Georgakopoulou, A. (2004) To tell or not to tell?: Email stories between on- and off-line interactions. Language@Internet, 1, article 1. http://www.languageatinternet.org/articles/2004/36

Georgakopoulou, A. (2006). Postscript. Computer-mediated communication in sociolinguistics. Journal of Sociolinguistics, 10, 548-557.

Georgakopoulou, A., & Goutsos, D. (1997, 2nd ed. 2004). Discourse analysis. Edinburgh: Edinburgh University Press.

Georgakopoulou, A., & Goutsos, D. (2000). Revisiting discourse boundaries: The narrative and non-narrative modes. Text, 20, 63-82.

Goffman, E. (1974). Frame analysis. New York: Harper and Row.

Goffman, E. (1981). Forms of talk. Philadelphia: University of Pennsylvania Press.

Goodwin, C., & Goodwin, M. H. (1990). Interstitial argument. In: A. D. Grimshaw (Ed.), Conflict talk: Sociolinguistic investigations of arguments in conversations (pp. 85-117). Cambridge, UK: Cambridge University Press.

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

Gumperz, J. J. (1992). Contextualization and understanding. In: A. Duranti & C. Goodwin (Eds.), Rethinking context: Language as an interactive phenomenon (pp. 229-252). Cambridge, UK: Cambridge University Press.

Herring, S. C. (1994). Politeness in computer culture: Why women thank and men flame. In: M. Bucholtz, A.C. Laing, L.Sutton, & C. Hines (Eds.), Cultural Performances: Proceedings of the Third Berkeley Women and Language Conference (pp. 278-294). Berkeley, CA: Berkeley Women and Language Group.

Herring, S. C. (1996). Two variants of an electronic message schema. In: S.C. Herring (Ed.), Computer-mediated communication: Linguistic, social, and cross-cultural perspectives (pp. 81-105). Amsterdam/Philadelphia: John Benjamins.

Jefferson, G. (1984). On the organization of laughter in talk about troubles. In: J. M. Atkinson & J. Heritage (Eds.), Structures of social interaction. Cambridge, UK: Cambridge University Press.

Labov, W. (1972). Language in the inner city. Philadelphia: University of Pennsylvania Press.

Milroy, L. (1987). Observing and analysing natural language. Oxford, UK: Blackwell.

Milroy, L., & Muysken, P., Eds. (1995). One speaker, two languages: Cross-disciplinary perspectives on code-switching. Cambridge, UK: Cambridge University Press.

Norrick, J. (1994). Repetition as a conversational joking strategy. In: B. Johnstone (Ed.) Repetition in discourse: Interdisciplinary perspectives, Vol. 1 (pp. 162-75). Norwood, NJ: Ablex.

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

Pomerantz, A. (1984). Agreeing and disagreeing with assessments: Some features of preferred/dispreferred turn shapes . In: M. Atkinson & H. Heritage (Eds.) Structures of social action (pp. 57-101). Cambridge, UK: Cambridge University Press.

Raskin, V. (1985). Semantic mechanisms of humor. Dordrecht: Reidel.

Reid, E. (1991). Electropolis: Communication and community on Internet Relay Chat. Honours Thesis, University of Melbourne. http://www.irchelp.org/irchelp/communication-research/academic/academic-reid-e-electropolis-1991.html

Romaine, S. (1989). Bilingualism. Oxford, UK: Blackwell.

Schiffrin, D. (1994). Approaches to discourse. Oxford, UK: Blackwell.

Scotton, C. (1993). Social motivations for codeswitching: Evidence from Africa. Oxford, UK: Clarendon Press.

Siegel, J. (1995). How to get a good laugh in Fijian: Code-switching and humor . Language in Society, 24, 95-110.

Sifianou, M. (1992). Politeness phenomena in England and Greece. Oxford, UK: Clarendon Press.

Tannen, D., Ed. (1984). Written and spoken language: Exploring orality and literacy. Norwood, NJ: Ablex.

Tannen, D., & Wallat, C. (1987). Interactive frames and knowledge schemata in interaction: Examples from a medical examination interview. Social Psychological Quarterly, 50, 205-216.

Thurlow, C., & Mroczek, K. (2011). Digital discourse: Language in the new media. Oxford/New York: Oxford University Press.

Widdicombe, S., & Wooffitt, R. (1995). The language of youth subcultures. New York: Harvester Wheatscheaf.

Yates, S. J. (1992). Computer conferencing: A written or spoken environment? CITE Report No. 151. Milton Keynes: Open University.

Yates, S. J. (1996). Oral and literate linguistic aspects of CMC discourse: A corpus based study. In: S. C. Herring (Ed.), Computer-mediated communication: Linguistic, social, and cross-cultural perspectives (pp. 29-48). Amsterdam/Philadelphia: John Benjamins.

Biographical Note

Alexandra Georgakopoulou [ alexandra "dot" georgakopoulou "at" kcl "dot" ac “dot” uk] is Professor of Discourse Analysis & Sociolinguistics at King's College London. She researches new media with narrative analysis, in particular small stories (2007, John Benjamins), and language and identities-in-interaction approaches. Her latest book (co-authored with Anna De Fina) is Analyzing narrative: Discourse and sociolinguistic perspectives (2011, Cambridge University Press).


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.