Home / Articles / Volume 2 (2005) / The interaction of generic structure and interpersonal relations in two-party e-chat discourse
Document Actions

 


Generic Structure and Computer-Mediated Communication

The question of genre and generic structure is of central importance in discourse analysis, since it encapsulates the "systematic co-patternings between the form, content, function and context of our discourse activities" (Georgakopoulou & Goutsos, 2004, p. 33). Thus, as early as 1987 discourse boundaries and their linguistic signals have been identified in 'service encounter' texts (Ventola, 1987), while work within the Systemic Functional framework has developed notions such as schematic structure and generic structure potential (see, e.g., Halliday & Hasan, 1985). Similarly, the work of John Swales (e.g., 1990) has emphasized the importance of moves and steps for language learning and teaching.

A similar concern with the identification of typical segments and boundaries is absent in the field of computer-mediated communication (CMC), mainly because of the concentration of most studies on micro-level or situational features (e.g., Cherny, 1999; Danet et al., 1997; Werry, 1996; Yates, 2001). In a recent overview of CMC, Georgakopoulou (2003) calls for further inquiry "away from quantitative measures of numerous micro-level features to a close scrutiny of the macro-level constitution and contextualization of discourse styles" (p. 5). Whereas many studies of CMC have taken up the latter suggestion by shifting their focus onto the roles and relationships of participants and the purpose and functions of communication (e.g., Baym, 1995; Cherny, 1999; Ferrara et al., 1991; Jones, 1997; Reid, 1991; Rheingold, 1993), the call for a closer scrutiny of the "macro-level constitution" has still remained unheeded.

The need for focusing on the structure of CMC genres is indispensable in order to describe them in their own terms rather than in terms of how they differ from other, spoken or written genres, as was the case with earlier research. After some considerable time of familiarization, users of CMC nowadays seem to be, as Baron (1998) puts it, "increasingly relaxed about the technological limitations of the medium" (p. 165). This increasing tolerance of technological limitations allows for the development of new conventions for genres of technologically mediated communication that do not draw from traditional spoken or written resources. Emoticons are perhaps the best-known example of these new conventions (see the section on Structure Signals and Strategies in IRC Exchanges below), exploiting visual resources to complement the traditional written and spoken channel. On a larger scale, it has been found that in other technologically-mediated texts such as messages left on answering machines, a clearly defined schema has emerged, with some variation in the realization and sequencing of signals and moves (Goutsos, 2001). As a result, we can safely argue that CMC is a unique discourse type, existing on a continuum between oral conversation and written text (Collot & Belmore, 1996; Foertsch, 1995, p. 301). This view concurs with findings of discourse analysis (e.g., Biber, 1988; Georgakopoulou & Goutsos, 2004) suggesting that the distinction between spoken and written discourse cannot be captured in absolute terms as a rigid dichotomy, but must be seen as a continuum with intermediate points, in which texts can be situated according to their (more or less) prototypical features.

At the same time, this emphasis on generic structure should be placed in the context of previous studies such as Cherny (1999), where it is claimed that CMC interactions are more amenable to description in terms of register than genre, i.e., linguistic variation rather than overall text structure. The reason given for this is that "the interaction is usually highly participatory, focused on high sociability rather than pre-determined goals, generally free-form rather than highly structured" (Cherny, 1999, p. 27). Although these characteristics seem to be shared by many CMC genres, including the two-party IRC exchanges studied here, they should rather be seen as complementary to the emergence of an overall generic schema to which they variously contribute. Thus, our inquiry still has to include basic questions such as how sociability is signaled and jointly achieved in CMC, how elements of structure are combined with elements of free form and how medium and genre constraints interact with less 'pre-determined' individual planning. These questions cannot be answered without a detailed analysis of the specific organization of CMC genres.

In addition, by studying new genres in languages other than English, we will be able to establish which features of the new genres are typical of the medium they exploit and which are accidental or due to language-specific or cultural preferences. Our findings will thus contribute towards "separating out the contributions of the medium from those of human users," which Herring (1996, p. 4) includes among the key issues in studying CMC.

This article addresses the need identified above in a genre of synchronous CMC, namely two-party exchanges from the Internet Relay Chat (IRC), an electronic discussion (e-chat) program. The data includes messages which use both English and Greek, involving participants from both the same and different, cross-linguistic backgrounds. [1] I will try to establish the typical generic structure for two-party IRC exchanges, by focusing on how participants are oriented towards an ideal schema of phases and acts, as well as on how their interpersonal concerns contribute to the shaping of this schema. In particular, the proposed generic schema is related to linguistic signals of particular segments with corresponding strategies or speech acts on the part of the initiator and the responder. It is suggested that the identification of central concerns and strategies in two-party IRC exchanges is a necessary step for an understanding of their purpose and function.

The following section presents the data in more detail, while more information about the characteristics of IRC is given after that. The main discussion of the findings starts with a general schema of generic structure for IRC exchanges, before moving on to the main signals and strategies found in relation to this structure and the specific speech acts associated with them.

Data

The texts studied in this article come from a corpus of 34 e-chat IRC exchanges (approximately 10,000 words in total). These exchanges lasted from 1 minute to 1 hour and 45 minutes. [2] The participants were three female Greek speakers, who kindly provided me with the data, and twelve male and female Greek speakers, an Israeli female speaker (nickname: weirdo) and a Maltese male speaker (nickname: Skywalker18), who were their interlocutors. Only four exchanges involve participants using English as their main channel of communication and these can be used to compare with exchanges in Greek. In most exchanges, participants engaged in IRC communication with each other for the first time, although in ten cases interlocutors had already communicated with each other in the past. However, all Greek participants were quite experienced users of IRC and the specific exchanges studied here constitute only a small portion of their daily communication through this medium. Data come from two different periods of interaction (October 1998 and October 1999). Although e-chat programs allow for multi-party conversation, my data comes exclusively from two-party interactions, which constitute the focus of this study. [3]

Although basic features of the speakers' identity can be gathered from their contributions, it must be noted that, by e-chat convention, all participants use a nickname and so there is no way one can check upon the truthfulness of the information they give about themselves. The information gathered from the data itself and the profile of my informants suggests that most participants belong to a homogeneous 18-30 age group. In addition, most exchanges involve female to male interaction. Further details about the identity of the participants could only be gathered in an ethnographically-oriented study like that of Cherny (1999), although it must be pointed out once more that CMC is well-known for the ability it offers to participants to construct virtual identities, which may not bear any relation to their actual social characteristics (e.g., Jones, 1997; Rheingold 1993). As a result, sampling for age, gender and social class/education is not easy, unless in experimental situations at the expense of naturalness. Furthermore, it is not always clear whether these characteristics become relevant for the participants themselves [4] or whether they can be disentangled from generic characteristics, granted that most CMC genres are exclusively employed by young people in their everyday interactions. Thus, a fuller discussion of social parameters would require a more extensive research outlook.

IRC Characteristics

Internet Relay Chat is one of the most popular interactive services on the Internet, offering a worldwide multi-user network, where people using a nickname can communicate with each other by participating in real-time conversations. By using an IRC program, one can exchange interactive text messages with other people of any age, occupation, nationality etc., irrespective of location. There are several IRC programs which connect to the same chat networks, including mIRC, Pirch and Virc, as well as various other e-chat programs that might also involve the transmission of voice (VAX Phone, Macintosh Broadcast).

When logged into a chat session, one can converse by typing messages that are instantly sent to other chat participants. While chatting, users spend their time in one or more windows, each representing a different channel or user. The window is split into two panes: the viewing area, where incoming messages appear, and the composing area, for outgoing messages. Thus, participants read in one area and type their messages in another. Because of this, IRC is not quite real-time conversation, since what you type does not appear until you press the "Enter" key or click on the "Send" button. There is, in other words, a chance to edit what one sends. In practice, however, users dash off a comment, question or reply, then send it quickly for the sake of staying with the conversation. In these terms, IRC could be seen as closer to the synchronous end of the CMC continuum (Yates, 2001, p. 97), although in strict terms it should be characterized as quasi-synchronous.

Because of these contextual features, IRC exchanges contrast with spoken and written exchanges in interesting ways (Werry, 1996; Yates, 1996). Both addressee and addresser are physically absent in each other's world, since they do not share the same space context (co-presence). In addition, e-chat discourse makes use of the written channel of communication and thus lacks the visual and paralinguistic cues of face-to-face communication such as body language, gestures, facial expressions, hesitations and variation in intonation. E-chat is also characterized by relative permanence and editing, as seen in cases where conversation time does not coincide with communication time. As a result, there are indications of pause or 'overlapping' in our records of IRC exchanges (pace Werry, 1996, p. 51), as in the following: [5]

11.6

<athinagor>

pou eise koukli?

'where are you doll?'

11.7

<Gipsy>

epestrepsa

'I'm back'

11.23

<athinagor>

me tiasxolise?

'what do you do?'

11.24

<Gipsy>

a/s/l please

As can be seen from the examples above, e-chat is spontaneous and requires immediate feedback, something which may not always happen. As will be shown below, interruptions of continuity contribute to the shaping of the typical structure of two-party e-chat interactions. In addition, the exchange is relatively transient, although not to the extent that face-to-face interaction requires it to be. As Cherny (1999, p. 155) clarifies for another CMC genre (MUDs), it is synchronous but not two-way, so that we do not have real overlap but only interruption.

Generic Structure of Two-Party IRC Exchanges

A prominent feature of IRC exchanges noted in the literature is the frequent and abrupt introduction of new topics and endings. Werry (1996) suggests that "successive, independent speech acts are simply juxtaposed, and different topics interwoven" (p. 51), resulting in rapid shifts. Collot and Belmore (1996) also consider that this "easy interaction of participants and alternation of topics" (p. 14) characterize electronic interaction in general. However, this should not lead us to conclude that IRC exchanges are totally haphazard or lack organization. In contrast, participants seem to follow an implicit orientation to structure, which can be summarized in Table 1:

Table 1 shows the skeleton of generic structure, which will take flesh by the end of the discussion (see Table 2 in the conclusions). As is suggested, the complexity of IRC exchanges is due to a combination of structural parts with free elements. E-chat messages show an orientation towards an opening, a main body and a closing phase, of which only the first is found in all exchanges. These phases consist of parts with a fixed position and free elements that may occur anywhere in the structure. The latter appear in all exchanges and have two different functions: to engage in conversational play and check the channel of communication.

It must not be assumed from Table 1 that any exchange would exhibit this generic structure in full. A full analysis of the structural schemas of all 34 exchanges shows that most e-chat interactions contain only a few of the parts indicated above. (An analysis of a complete IRC exchange according to the generic schema above is given in Appendix 1). This suggests an interesting parallel with other genres of CMC, such as, e.g., Herring's (1996) finding about email messages that "participants are aiming at an ideal message schema" (p. 90). Table 1 reflects precisely such an "ideal message schema" for two-party IRC exchanges in its fullest realization. Although there are obvious and significant differences between asynchronous and synchronous CMC, what this schema points to is the common pre-occupation of participants in CMC genres with the organization of their contributions. The generic structure of a CMC genre should thus be seen as the constellation of medium and other constraints as well as of individual creative acts into a schema reflecting the participants' orientation towards organization.

To further discuss generic structure, the only obligatory phase in the schema is the opening of the interaction, which revolves around the only obligatory part of the structure, self-identification. This part usually follows a routine specific to IRC communication, according to which participants ask each other the stereotypical "a/s/l?" (age/sex/location). This part necessarily appears in conversations taking place for the first time, that is, in cases where the participants did not have a previous chat, and stems from the requirements of the medium. It may also occur later, as in exchange 12 in our data, where real names are exchanged. In the opening phase, it is also common to have a greeting part (cf. Cherny, 1999, p. 204 ff.) and, less commonly, a self-description part, which elaborates on self-identification and might lead to the first segment of the interaction. Reference to personal information including physical characteristics etc. is found in all messages in which participants were previously unknown to each other. [6]

The main body and closing phases are optional and can be significantly shortened or even left out altogether. This is, no doubt, surprising, at least for the main body, but can be accounted for by the informal character of e-chat, which allows interlocutors to withdraw from the interaction whenever they want. As a result, only the initial, "investigative" phase may occur before an aborted conversation. The main body phase opens with an offer of introduction by one participant, which is taken up or (less commonly) rejected by the other participant. There are roughly five to seven introductions per exchange.

I deliberately avoid calling the segments opened by introduction 'topics', because of the complexity surrounding the notion (see Goutsos 1997). This, of course, does not mean that we cannot identify recurrent themes or foci of interest such as plans for the evening, common interests or news, films the participants have seen, places they have been to in their holidays etc. However, the relation between these themes is spurious, as has also been found for email messages (Georgakopoulou, 1997, p. 146) and segment changes are not motivated by thematic closure but, as will be argued below, by largely interpersonal concerns.

To move on, in the closing phase there is usually a pre-closing announcement by one of the participants (cf. Scollon's 1998 'pre-emptive closure') and a closing greeting. Quite often but not in all exchanges, there is also a part concerning arrangements for a future interaction either through e-chat or by telephone. Although there is an attempt to establish a closing exchange even in exchanges with no main body (e.g., exchange 3), IRC exchanges can break off at any point (see 9). In these cases, the closing phase involves an abrupt ending, where no participant signals the ending or one participant does but the other does not respond. Alternatively, it may involve a perfunctory ending of two turns or lengthier endings that may be up to 11 turns.

This skeletal presentation of the generic schema points to the fact that participants orient their contribution towards an ideal structure schema, in which IRC interlocutors are primarily concerned with establishing contact with each other. The fact that the main body phase may be absent would seem unthinkable in other genres such as telephone conversation, but can be easily understood in the context of e-chat discourse. The generic schema indicates that self-identification is a prerequisite for the occurrence of the other parts; the latter may be omitted, if the interlocutors do not want to continue with the interaction. Similarly, whereas the closing phase can be left out altogether, any exchange that does not have a well-developed opening phase is problematic. In conclusion, the arrangement of phases and parts reflects a primarily interpersonal concern for engaging in or disengaging from CMC, whereas conversation 'itself' i.e., the (ideational) exchange of thematic content seems to be a less pressing need.

The interpersonal orientation of two-party IRC exchanges is also apparent in the combination of the basic structure with the two free elements, conversational play and channel check, which can appear in all places in the basic schema. Conversational play has already been noted as a dominant element of CMC (Cherny, 1999, p. 96; Danet, 1995, p. 2). In this genre, it mainly consists of an elaboration on other parts of the schema, alongside the main line of interaction, involving, e.g., details about self-identification, phatic turns or ironic, flirtatious and humorous comments on the co-participant's previous turn.

Channel check constitutes a correction mechanism, at points where the line of communication seems to be cut off, e.g., when one of the participants seems not to be responding (see section 5). This element is also found in other CMC genres, where it has been explained as a means of maintaining a sense of co-presence or awareness in a conversation (Cherny, 1999, p. 198). As we will see below, channel check in IRC exchanges also relates to the predominant interactional concerns of the participants.

The appearance of the two free elements partly accounts for the extraordinary flexibility of e-chat exchanges. It also concurs with Scollon's (1998) findings about the structure of business telephone calls, which have been summed up in the following "maxims of stance:" a) attend to the channel, b) attend to the relationship, and c) attend to the topic. As Scollon (1998) clarifies, the frames of channel, relationship, and topic form an implicational hierarchy, with channel on the outside and topic in the inside. This is not an order preference, since "all three frames must be understood as being sustained throughout the interaction" (p. 71). Scollon points out that these maxims "provide a framework in social practice for the discussion of the negotiation of identity" (p. 75). In other words, we should expect success or failure in e-chat communication to be intimately related to the management of these frames, in which interpersonal relations figure prominently. In order to examine this, it is necessary to first identify the means by which the essential tasks of e-chat communication are achieved.

Structure Signals and Strategies in IRC Exchanges

The tasks involved in e-chat communication are achieved in IRC exchanges with the help of a variety of signals. Because of the particularities of the medium, it is the first interlocutor who initiates each part and organizes what will happen, while the second to take the turn usually responds, either positively or negatively, to the former's contribution. This is true for the free element of channel check, which is always initiated by the first participant, as in exchange 1:

11.6

<athinagor>

pou eise koukli?

'where are you doll?'

11.13

<athinagor>

pali efiges koukli?

'are you gone again doll?'

11.32

<athinagor>

pou eise pali?

'where are you again?'

11.38

<athinagor>

pou eise?

'where are you?'

A typical structure is the one used in exchange 13 below, where the phatic éla is used to check that the interlocutor is still online (cf. Georgakopoulou, 1997, p. 146):

13.38 [and 3.59]

<Progomfios>

Ela. ti egine?

' éla. what happened?'

The responder has the option of either continuing the disrupted introduction, as in the first example that follows, or explicitly acknowledging the check and, optionally, apologizing for the interruption, as in the second and third examples below:

11.32

<athinagor>

pou eise pali?

'where are you again?'

11.33

<Gipsy>

den tha vgeis apopse?

'you're not going out tonight?'

4.60

<mple>

me grafis?

'are you ignoring me?'

4.61

<Gipsy>

oxi. Eimoun sto telefono. Sorry

'no. I was on the phone. Sorry'

4.62

<mple>

ok re [a]

[a] The items re and vre are phatic markers of familiarity with no neat equivalents in English.

16.8

<PIGASOS>

kala tora exipnises?

'did you really just wake up?'

16.9

<Gipsy>

sorry. eimoun sto tilefono.

'sorry. I was on the phone.'

16.10

<PIGASOS>

kala

'alright'

Other examples of communication breakdown occur after explicit requests for a pause or break in discourse. In this case, signals such as brb (be right back), gtg (got to go), ena lepto please ('a moment please') or se 3 lepta tha imai edo ('I'll be back in 3 minutes') interrupt the flow of the exchange and result in the need to start over when contact is resumed. This could be indicated by a signal like: 7.12 lipon ksana apo tin arxi ('well, starting all over again'). A special case is found in exchange 10, where a large portion of the interaction (21-50) is devoted to an extended channel check, which also involves a clarification of the relationship between the interlocutors (see discussion in section 6). The fact that the first speaker in all messages initiates channel check reveals an asymmetry in the way the participants view their relationship: there is an initiator of the communication, who has the responsibility to keep the channel open and a responder, who does most of the replying.

As noted above, the expression of interpersonal relations is a central task in e-chat, since getting to know each other is central in IRC discourse. In the absence of paralinguistic clues, the attitude towards the topic of communication and the interlocutor is signaled by a multiplicity of orthographic strategies. Punctuation is used to make the written mode look and "sound" like oral language. Capital letters, periods, repeated question and exclamation marks in a row, commas, semicolons, colons etc. help the interlocutors 'hear' the intonation of the text. Also, in order to give emphasis, express irony and imitate prosodic features and paralinguistic cues, words may be underlined, italicized or capitalized. Examples include: 7.6: YES!!, 7.71: NO, 7.100: Aha, 8.7: aaaa…., 8.47: emm…., 8.50: …xmmm…, 9.38: keeping soooo many conversations, 9.60: started being very rood etc. This manipulation of graphic features, regarded by Werry (1996) as a basic feature of e-chat communication, should be regarded as one of the main resources available to IRC interlocutors for the achievement of interpersonal and other goals in their interaction.

The main conventional signals employed by IRC users in e-chat communication are the so-called emoticons. This "electronic paralanguage" (cf. Cherny, 1999, p. 110) includes, for instance, p: for showing the tongue, :) for ☺ (smile, happy face) [e.g., 7.9, 15, 21, 23, 41, 46, 50, 58 etc.], :( for ☹ (sadness, feel sorry) etc. In e-chat, all these signals reflect the interlocutors' attitude towards their message, i.e., are employed as stance or attitude markers. Their use is related to the tendency of e-chat towards brevity (cf. Werry 1996: 53), which is also found in the use of abbreviations such as: m for 'masculine', f for 'feminine', a/s/l for 'age/sex/location', cu for 'see you', thx for 'thanks' and lol for 'laugh out loudly'. IRC-specific devices further include signals of involvement also occurring in other genres. These include negative encoding in instances like: 7.43 never tried, 7.64 nobody listens to rock, 8.66 it wasn't VERY popular, 12.88 i still don't like it, 19.61 not all guys are like that, and exaggeration or hyperbole in examples like: 7.12 Fathers of rock, 7.26 Cant live without it, 7.99 It's hell, 18.12 Esi theteis ta meizona h elassona themata 'you are the one who poses the major or minor issues', 18.29 H prosopopoihsh tis ipomonis 'the personification of patience' etc.

The analysis of IRC generic structure shows that most interactional signals make an appearance in the conversational play part of IRC exchanges, revealing the informal relationship that exists between the participants. The informality of communication, along with the use of a nickname and the interlocutors' willingness to get to know each other develop what can be labeled a conversational playing field. The interaction of involvement signals (e.g., terms of address 'megale', 'koukli') in the construction of the conversational playing field can be seen in the following extended example between the female <Gipsy> and the male <athinagor>:

11.2

<Gipsy>

kai pou 3ereis megale oti eimai koukli???

'and how do you know, mister, I'm a doll???'

11.3

<athinagor>

then eise?

'aren't you?'

11.4

<Gipsy>

an sou po tha me pistepsis!!!

'will you believe me if I tell you!!!'

11.5

<athinagor>

then exw logous na mhn se pistepsw

'I've no reasons not to believe you'

11.9

<Gipsy>

esi eisai koukli??

'are you a doll??'

11.10

<athinagor>

san arxaios theos!!!!!!!!!

'like an ancient god!!!!!!!!!'

11.11

<Gipsy>

nai,se fantazome

'yes, I can imagine you'

Very briefly, the ironic remarks occurring in the conversational play part show the intimacy between the interlocutors: thus, in 11.9-11.11, <athinagor> exaggerates in his ironic reply. A similar example of the development of a conversational play field can be seen in the longer exchange 12.53-12.69 presented in Appendix 2, where a pseudo-misunderstanding takes place, involving an ironic negotiation.

Another prominent feature in the conversational playing field is the occurrence of intimacy markers contributing to the participants' attempt to find a common ground. Examples of these include terms of address such as koukli 'dolly' (11.2, 11.9), glikia mou 'my sweet one', megale 'big guy/man' (11.2), re (see note 5) and idiomatic phrases like: aman! Ti sxesi exei!! 'for God's sake!', Ntropi sou 'shame on you' (12.53), ante pali… 'here we go again' (3.64), etsi pes!!! 'now you're talking' etc. The medium of interaction seems to reinforce informality between the interlocutors, who would certainly not use these intimacy markers when talking to strangers.

The signaling of interpersonal relations is pervasive throughout e-chat discourse. This is not independent from the achievement of the tasks that relate to the structure of IRC exchanges. As mentioned above, the most important part of the structure, self-identification, is achieved by a variety of interpersonal devices. Discourse markers also seem to predominate in the signaling of sequential relations i.e., relations of continuity and discontinuity (Goutsos, 1997) throughout the exchanges. In Greek exchanges, the most common marker is lipón, which is used with multiple functions, including the following: to indicate return to self-identification after a pause, to signal new introduction and to function as a pre-closing device. [7] In the exchanges using English, a similar but less extensive role seems to be played by so.  [8] Another common marker for introduction is ce 'and' (e.g., in 11.2 above), emphasizing the loose, paratactic connection between segment. Metalinguistic expressions are also found for segment introduction (den mou les 'tell me something', seira sou 'your turn') and pre-closing (re ela na sou po 're let me tell you something'). These may include anaphoric expressions (ti allo 'what else') and direct or indirect requests (rota me kai esy kati 'you ask me something now', ti tha eleges gia ligo koutsompolio 'what would you say about a little bit of gossip'). In cases where the flow is interrupted, explicit requests for continuation are noted, including phrases like: tipota allo 'what else' and so…ti allo 'so…what else'. The occurrence of meta-linguistic expressions of topic change like the above is another feature which differentiates e-chat from face-to-face communication.

In the main body of the IRC exchange, one of the techniques used by the responder in order to contribute to the interaction is to acknowledge the initiator's move and then shift to a new segment. This is achieved in exchanges in both languages by the use of OK, whose function is also multiple, including introduction, pre-closing and indication of agreement. [9] A characteristic example is the following, in which there is an explicit request for segment change:

2.38

<elena>

as allaxoume ligo thema

'let's change the subject, shall we?'

2.39

<nick>

OK

One of the most common techniques for pre-closing is to show emergency, e.g.,: prepei na fygw 'have to go' or den boro na mino poli giati mou exoun etoimasei trapezi 'can't stay longer because I'm invited to dinner'. In this case, closing may be individually achieved by the responder or may be collaboratively achieved by both initiator and responder. Finally, closing can be achieved through a combination of greeting and reason as in the following:

15.36

<elen3>

loipon nikola

'lipón nikola [Name]'

15.37

<nickF>

ne?

'yes?'

15.38

<elen3>

xarika pou ta ipame alla prepi na pao gia ipno

'I'm glad we talked but I have to go to bed'

In sum, IRC exchanges seem to have developed a range of devices in all parts of the generic structure for handling the tasks of achieving interpersonal and sequential relations. Some of these exploit devices common with other genres, while others are particular to e-chat (formulae, graphic devices etc.). The overall generic structure is achieved both individually and collaboratively, on the basis of turn-taking and negotiation. Whereas one participant acts as segment initiator (caller), the responder (callee) has the option to accept or reject the initiator's move or shift the conversation towards a different part. Interpersonal relations also predominate in clearly identifiable parts such as conversation play and are found extensively throughout the structure of IRC exchanges. The following section focuses more closely on the achievement of interpersonal relations by identifying the main speech acts associated with the generic schema presented above.

Speech Acts in IRC Communication

As can be gathered from the examples above, the dominant type of speech act in IRC communication involves the question-answer adjacency pair, well known from face-to-face conversation. It is through this that the participants construct their contribution at individual points in collaborative or non-collaborative ways. However, IRC chat crucially deviates from the co-constructive norms of face-to-face conversation (Cherny, 1999, p. 196). According to Herring (1999), this seems to be due to the lack of overlap between messages, which precludes simultaneous feedback, as well as the absence of certain audio and visual cues, and the frequent disruption of the conversation-analytic structure of adjacency pairs, caused by the channel disruptions observed above. Our data confirms these findings but also points to the occurrence of a number of speech acts present in IRC communication that are reminiscent of conversational contexts. These are feedback, agreement-disagreement pairs, self-clarification and self-identification.

Firstly, the appearance of an optional feedback move is parallel to feedback in the teaching exchange (see, e.g., the Birmingham School analysis of discourse: Georgakopoulou & Goutsos, 2004, p. 75). Feedback appears in the channel check element, which, as noted above, is the responsibility of the caller. The callee has the option of acknowledging the interruption in communication and apologizing for it or ignoring it and proceeding with her contribution. Similar options are at the disposal of the responding interlocutor at all points in the IRC exchange. First, the responder can provide feedback, which may or may not be acknowledged by the initiator. This feedback can take the form of acknowledgment of the previous turn, as in:

7.35

<Skywalker18>

i play in a band too

7.36

<Skywalker18>

lead guitar

7.37

<led-zep>

wow thats great

10.13

<weirdo>

dont know that island very well – is it a nice place?

10.14

<anastasia>

its ok its nice for 18 years people specially at night

10.15

<weirdo>

ok nice and safe

11.19

<Gipsy>

isia i sgoura mallia?

'straight or curly hair?'

11.20

<athinagor>

sgoura

'curly'

11.21

<Gipsy>

opa kai edo sifonoume!

'opa [marker of surprise] we agree on that too!'

11.22

<athinagor>

xairomai

'I'm glad'

Feedback can also appear as an aside, a parenthetical comment on the interlocutor's contribution, as in the following:

13.63

<Progomfios>

Ti na sou kano… Akolouthise tis odigies akrivos

'what can I do … Follow the exact orders'

13.64

<Progomfios>

Ante pali…

'Here we go again …'

Finally, feedback can occur as a mechanism of backchaneling:

7.66/7.67

<led-zep>

:) :)

7.95

<Skywalker18>

it's hell

7.96

<led-zep>

aha

8.26

<thetiko>

kai den mou les vre file?

'and tell me something vre friend?'

8.27

<wet>

:)

Another 'conversational' speech act that is found in IRC exchanges is agreement, indicated by interlocutors by an opinion or statement, either directly or indirectly, as in the following:

12.7

<Progomfios>

Protimo ta games!!

'I prefer games!!'

12.8

<Loucy>

Sigoura!!!!

'Sure!!!!

12.41

<Loucy>

Eytixos pou se exw esena kai mou ta les ola auta

'thank God I have you to tell me all this'

12.42

<Progomfios>

Akrivos.

'Exactly.'

Less commonly, the initiator explicitly elicits agreement from the responder:

4.28

<mple>

ine romantiko

'it's romantic'

4.29

<mple>

den sifonis?

'don't you agree?'

As expected from pragmatics studies (e.g., Levinson, 1983, p. 332 ff.; Pomerantz, 1984), in the IRC exchanges under analysis, unqualified disagreement is rare. In the following example, the responder resorts to a personal opinion:

10.41

<anastasia>

its not that difficult if you want it

10.42

<weirdo>

i find it very difficult

More often, disagreement is qualified in several ways through the use of mitigation devices, as in the following:

8.12

<wet>

ti tha eleges gia ligo koutsompolio?

'what would you say about a little bit of gossip?'

8.13

<thetiko>

..apw9hmena brbrb

'…what a complex brbrb'

13.20

<Progomfios>

Nai, tha to do…ankai den mou aresoun ta "Amerikanika" erga.

'yes, I'll watch it …though I don't like "American" films.'

13.23

<Gipsy>

exei to americaniko to stixio, alla san ergo einai poly pragmatiko

'it's got the American element, but it's very realistic'

13.24

<Progomfios>

Pantos ego yia ton Hanks tha to do

'anyway I will see it for Hanks'

Another option of the responder is to challenge the initiator's move. The reply from the initiator can be in the form of redress:

11.27

<athinagor>

me ti asxoleise koukli?

'what do you do doll?'

11.28

<Gipsy>

Doulevo

'I work'

11.29

<athinagor>

pou?

'where?'

11.30

<Gipsy>

aman! ti sxesi exei!!

'aman! [marker of annoyance] what do you care!!'

11.31

<athinagor>

apla rwtisa glikia mou an then thes na mou peis then peirazei ok

'I just asked my sweety if you don't want to tell me it doesn't matter ok'

However, in most cases, challenge triggers a conversational play element, where it is answered by further challenge, as in:

8.5

<wet>

a/s/l?

8.6

<thetiko>

aaaa ..gewgrafikes erwthseis prw prwi deyteras file..:)

'aaaa ..geography questions so early monday morning, my friend..:)'

8.7

<wet>

h geographia de se travaei e?

'geography is not very attractive to you e?'

8.18

<wet>

siga re! Tora tha mou peis oti den koutsompoleueis pote!!!!!!!!!!!

'c'mon re! don't tell us you never gossip!!!!!!!!!!!'

8.19

<thetiko>

..dokei moi oti touto or9on esti !

'...methinks 'tis true' [a]

8.20

<wet>

re mpas kai eisai h mentepsixosh tou Socrati?

're you're not Socrates' reincarnation, are you?'

8.21

<thetiko>

isws…

'maybe…'

8.22

<thetiko>

esy ti mporei na eisai?

'what can you be?'

8.34

<wet>

mpravo re file, vrikes ena thema gia deytera proi! (irony)

'well done re friend, that's a subject for Monday morning! (irony)'

8.38

<wet>

periergo treno eisai

'you're a funny train'

8.39

<thetiko>

..xmmm…oxi kai traino ..syrmos (isws)

'...hmmm…not a train ...a wagon [= fashion] (perhaps)'

[a] In the original, there is code-switching into Ancient Greek (cf. Georgakopoulou, 1997).

The above example shows how the development of a conversational play field can rely on successive challenges, which, however, may be qualified as here by a signal of hedge in 8.34 and modality in 8.21, 8.22 and 8.39.

The third possibility for both initiator and responder is a move of self-clarification, as in the following examples:

11.39

<athinagor>

loipon me ti allo asxolise endiaferonta sou ennoo

'well what else do you do, your hobbies I mean'

12.11

<Loucy>

Siga siga, eipame eimai arxaria egw.

'take it easy, we've said it, I'm a beginner.'

A request for clarification is also found—usually, but not exclusively—in conversation play elements:

3.8

<Gipsy>

me stisane oloi!!!!

'they all stood me up!!!!'

3.9

<Progomfios>

Otan les oloi?

'When you say all?'

3.10

<Gipsy>

oi filoi mou!

'my friends!'

4.28

<mple>

ine romantiko

'it's romantic'

4.30

<Gipsy>

romantiko! Diladi?

'romantic! Meaning?'

A long section of self-clarification is found in exchange 10, starting with:

10.21

<anastasia>

i dont think you wann talk with me

10.22

<weirdo>

I do but having trouble keeping soooo many conversations going at the same time

The same line of development continues up to 10.40, indicating that concern with this task is central in the participants' perception of structure to the point that it may hinder any other topic development.

Finally, the central task of IRC exchanges, as pointed out in the ideal generic schema, is the achievement of self-identification by both participants. This move is not fulfilled in all cases. In failed exchanges, the self-identification section is missing or is very brief. In these cases, challenges or channel problems crucially affect self-identification and thus hinder successful further introduction and development.

Summing up, a number of speech acts or moves are taken up by e-chat participants. Initiators may elicit information or agreement, perform a channel check and issue or redress a challenge, engaging thus in conversational play. Responders may offer information, provide feedback by acknowledging a previous turn, making an aside, agreeing or, less commonly, disagreeing. Responders may also challenge a previous turn, instigating the development of a conversational play field. This range of options available to responders suggests ways in which the asymmetry of responsibility between them and the initiators can be reduced.

The accomplishment of the ideal IRC generic schema crucially depends on the participants' individual, creative acts for its realization. Thus, in successful exchanges the tasks of self-identification and self-presentation are achieved, while challenges are redressed, clarified or lead to friendly conversational play. Both disagreement and self-presentation are usually hedged and there is positive evaluation and backchanneling. In failed exchanges, on the other hand, the self-identification section is missing or is very brief and challenges remain unanswered or lead to non-friendly (ironic) conversational play. As a result, further development of the main body is hindered. The ideal schema towards which the participants seem to be oriented relies in its realization on the acts of negotiation performed by the interlocutors.

Conclusions

We have identified above the ideal generic schema for IRC two-party messages, as well as the main signals indicating discourse boundaries and the speech acts related to them. Our discussion can be summarized in the following table:

Table 2: Summary of moves, signals and speech acts in IRC messages

What is not immediately apparent in Table 2 is the influence of the interpersonal concerns of the interlocutors on the accomplishment of the generic schema. First, as noted above, self-identification is the most essential task of interlocutors, occupying a central place in the interaction. Thus, IRC two-party messages foreground the concern with 'introductions' or 'becoming acquainted', which seems to constitute their main purpose or function. In addition, interpersonal concerns are also apparent in the occurrence of the two free elements of conversational play and channel check, which reflect the priorities of speakers. Channel checks are resorted to for restoring the line of communication and thus helping to establish common ground between the participants. Once this has been achieved, a conversational play field can be developed. This element predominates in e-chat interaction past the 'becoming acquainted' phase and is responsible for its unique repartee of quip. At the same time, conversational play elements are also employed to resolve misunderstandings and achieve a more complete self-presentation.

The upshot of our discussion is the possibility to incorporate interpersonal concerns into a model of structural analysis. There seems to be an inextricable link between generic structure and interpersonal relations in two-party e-chat conversation. The negotiation of interpersonal relations crucially depends on the successful signaling and manipulation of a range of speech acts, while the interplay between these acts accounts for the success or failure in the realization of the ideal generic schema. The efficiency of communication via IRC is thus related to negotiation strategies that involve a positive response from interlocutors and their involvement in what is discussed. To this effect, e-chat combines signals found in other genres such as evaluation devices or intimacy markers with conventions developed specifically in its own context such as manipulation of graphic forms and spelling for the achievement of interpersonal and sequential tasks. At the same time, the flexibility of the generic schema reflects the conflicting needs and concerns of the participants. In this sense, the typical structure of IRC communication wavers between elements of stability and elements of play, diverging from traditional schemas of other genres.

Because of its laconic character and its focusing only on the most important elements for achieving interpersonal contact, e-chat discourse can be a useful test-bed for pragmatic theories, revealing the most central aspects of interpersonal communication. With its stripped-down content and lack of paralinguistic and visual clues, synchronous CMC comes as close as possible to an experimental communication setup, without losing ecological validity. As this article suggests, however, before we are able to formulate any pragmatic hypotheses for this kind of discourse, it is necessary to investigate the generic schema and its linguistic accomplishment in terms of dyadic interaction. This suggestion implies that questions of identity construction, cognitive constraints and social or community patterns can be most fruitfully discussed on the basis of close textual analysis.

References

Androutsopoulos, Jannis 1998. Orthographic variation in Greek e-mails: a first approach. Glossa 46, 49-67.

Androutsopoulos, Jannis 2000. Latin-Greek spelling in Greek e-mail messages: Usage and attitudes. Studies in Greek Linguistics 19, 75-86.

Baron, Naomi S. 1998. Letters by phone or speech by other means: The linguistics of email. Language and Communication, 18, 133-170.

Baym, N. 1995. The emergence of community in computer-mediated communication. In S. Jones (ed.) Cybersociety: Computer-mediated Communication and Community. London: Sage. 138-163.

Biber, Douglas 1988. Variation across Speaking and Writing. Cambridge: Cambridge University Press.

Cherny, Lynn 1999. Conversation and Community. Chat in a Virtual World. Stanford: CSLI Publications.

Collot, Milena and Belmore, Nancy 1996. Electronic language: A new variety of English. In S. C. Herring (ed.). Computer-mediated Communication: Linguistic, Social and Cross-cultural Perspectives. Amsterdam/Philadelphia: John Benjamins. 13-28.

Condon, Sherri L. and C'ech, Claude G. 1996. Functional comparisons of face-to-face and computer-mediated decision making interactions. In S. C. Herring (ed.). Computer-mediated Communication: Linguistic, Social and Cross-cultural Perspectives. Amsterdam/Philadelphia: John Benjamins. 65-80.

Danet, B. (ed). 1995. Play and performance in computer-mediated communication, Special Issue, Journal of Computer Mediated Communication 1. Available: http://jcmc.indiana.edu/vol1/issue2/index.html

Danet, B., Ruedenberg-Wright, L. and 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. Available: http://jcmc.indiana.edu/vol2/issue4/danet.html

Foertsch, Julie 1995. The Impact of Electronic Networks on Scholarly Communication: Avenues for Research. Discourse Processes. 19 (2), 301-328.

Ferrara, K., Brunner, H. and Whittemore, G. 1991. Interactive written discourse as an emergent register. Written Communication 8 (1), 8-33.

Georgakopoulou, Alexandra 1997. Self-presentation and interactional alliances in e-mail discourse: The style- and code-switches of Greek messages. International Journal of Applied Linguistics, 7 (2), 141-164.

Georgakopoulou, Alexandra 2003. Computer-mediated communication. In Verschueren, Jef, Jan-Ola Östman & Crhis Bulcaen (eds.) Handbook of Pragmatics. (2001 Instalment). Amsterdam/Philadelphia: John Benjamins. 1-20.

Georgakopoulou, Alexandra and Goutsos, Dionysis 2004. Discourse Analysis: An Introduction. Second edition. Edinburgh: Edinburgh University Press.

Georgakopoulou, Alexandra and Goutsos, Dionysis 1998. Conjunctions versus discourse markers in Greek: the interaction of frequency, position and functions in context. Linguistics 36 (5), 887-917.

Goutsos Dionysis 1997. Modeling Discourse Topic: Sequential Relations and Strategies in Expository Text. Norwood, NJ: Ablex.

Goutsos, Dionysis 2001. Sequential and interpersonal aspects of English and Greek answering machine messages. Pragmatics 11 (4), 357-377.

Goutsos, Dionysis 2002. English in Cypriot Greek conversations: A preliminary investigation. In M. Makri-Tsilipakou (ed.) Selected Papers on Theoretical and Applied Linguistics from the 14th International Symposium, April 20-22, 2000. Thessaloniki: Aristotle University of Thessaloniki. 97-110.

Halliday, M. A. K. and Hasan; Ruqaiya 1985. Language, Context and Text: Aspects of Language in a Social-semiotic Perspective. Oxford: Oxford University Press.

Herring, Susan C. (ed.) 1996. Computer-mediated Communication: Linguistic, Social and Cross-cultural Perspectives. Amsterdam/Philadelphia: John Benjamins.

Herring, Susan, C. 1996. Two variants of an electronic message schema. In S. C. Herring (ed.). Computer-mediated Communication: Linguistic, Social and Cross-cultural Perspectives. Amsterdam/Philadelphia: John Benjamins. 81-105.

Herring, Susan C. 1999. Interactional coherence in CMC. Journal of Computer-Mediated Communication 4. Available: http://jcmc.indiana.edu/vol4/issue4/herring.html

Jones, S. 1997. Virtual Culture: Identity and Communication in Cybersociety. London: Sage.

Levinson, Stephen 1983. Pragmatics. Cambridge: Cambridge University Press.

Pomerantz, A. 1984. Agreeing and disagreeing with assessments: Some features of preferred/dispreferred turn shapes. In Atkinson, J. M. and Heritage, J. (eds.) Structures of Social Action: Studies in Conversation Analysis. Cambridge: Cambridge University Press. 57-101.

Reid, E. 1991. Electropolis: Communication and community on Internet Relay Chat. Adapted from a B.A. Honors thesis, University of Melbourne, Australia. Available: http://www.irchelp.org/irchelp/misc/electropolis.html

Rheingold, H. 1993. The Virtual Community. Reading, MA: Addison-Wesley.

Scollon, Ron 1998. Mediated Discourse as Social Interaction: A Study of News Discourse. London: Longman.

Swales, John 1990. Genre analysis. Cambridge: Cambridge University Press.

Ventola, Eija 1987. The Structure of Social Interaction: A Systemic Approach to the Semiotics of Service Encounters. London: Pinter.

Werry, Christopher C. 1996. Linguistic and interactional features of Internet relay Chat. In S. C. Herring (ed.). Computer-Mediated Communication: Linguistic, Social and Cross-Cultural Perspectives. Amsterdam/Philadelphia: John Benjamins. 47-64.

Yates, Simeon J. 1996. Oral and written linguistic aspects of computer conferencing. In S. C. Herring (ed.). Computer-mediated Communication: Linguistic, Social and Cross-cultural Perspectives. Amsterdam/Philadelphia: John Benjamins. 29-46.

Υates, Simeon J. 2001. Researching internet interaction: Sociolinguistics and corpus analysis. In Wetherell, M., Taylor, S. and Yates, S. Discourse as Data: A Guide for Analysis. London: Sage Publications. 93-146.

Appendix 1: Analysis of generic structure in a two-party IRC exchange

Session Start: Fri Oct 16 22:06:08 1998

Session Ident: athinagor

11.1

<athinagor>

hi koukli

greeting OPENING

'hi, doll'

11.2

<Gipsy>

kai pou 3ereis megale oti eimai koukli???

challenge

'and how do you know, mister, I'm a doll???'

11.3

<athinagor>

then eise?

'aren't you?'

11.4

<Gipsy>

an sou po tha me pistepsis!!!

'will you believe me if I tell you!!!'

11.5

<athinagor>

then exw logous na mhn se

pistepsw

'I've no reasons not to'

11.6

<athinagor>

pou eise koukli?

channel check

'where are you, dolly?'

11.7

<Gipsy>

epestrepsa

'I'm back'

11.8

<athinagor>

loipon?

'well?'

11.9

<Gipsy>

esi eisai koukli??

attempt at self-description

'are you a doll??'

11.10

<athinagor>

san arxaios theos!!!!!!!!!

'like an ancient male god!!!!!!!!!'

11.11

<Gipsy>

nai,se fantazome

'yes, I can imagine'

11.12

<athinagor>

loipon tha mou peis esi pos eise

glikia mou?

'well, are you going to tell me how

you are honey?'

11.13

<athinagor>

pali efiges koukli?

channel check

'did you go again doll?'

11.14

<Gipsy>

melaxrini,megala matia

kaiwraio kormi

self-description

'dark-haired, big eyes and nice body'

11.15

<athinagor>

xroma matiwn ipsos ?

'eye color height?'

11.16

<Gipsy>

kastana/1.66.Seira sou

'brown/1.66. Your turn'

11.17

<athinagor>

mine opos eise mhn allaxseis tipota

self-description

'stay as you are don't change anything'

11.18

<athinagor>

180cm 70 kg kastana malia kastana matia

'180cm 70kg brown hair brown eyes'

11.19

<Gipsy>

isia i sgoura mallia?

'straight or curly hair?'

11.20

<athinagor>

sgoura

'curly'

11.21

<Gipsy>

opa kai edw sifonoume!

'well we agree on this too!'

11.22

<athinagor>

xairomai

'I'm glad'

11.23

<athinagor>

me tiasxolise?

attempt at introduction1

'what do you do?'

11.24

<Gipsy>

a/s/l please

self-identification

11.25

<athinagor>

28 m u?

11.26

<Gipsy>

24/f

self-identification

11.27

<athinagor>

me ti asxolise koukli?

introduction1 MAIN BODY

'what do you do doll?'

11.28

<Gipsy>

Doulevo

'I work'

11.29

<athinagor>

pou?

'where?'

11.30

<Gipsy>

aman!ti sxesi exei!!

challenge

'for God's sake! what do you care!!'

11.31

<athinago>

apla rwtisa glikia mou an then thes na mou peis then peirazei ok

'I just asked my sweety if you don't want to tell me it doesn't matter ok'

11.32

<athinago>

pou eise pali?

channel check

'where are you again?'

11.33

<Gipsy>

den tha vgeis apopse?

introduction2

'are you not going out tonight?'

11.34

<athinago>

then xserw eimai ligo kriomenos u?

'I don't know I got a bit of a cold u?'

11.35

<Gipsy>

varieme ligo

'I'm a bit bored'

11.36

<athinago>

giati glikia mou?

'why honey?'

11.37

<Gipsy>

kai ligo kourasmenei

'and a bit tired'

11.38

<athinago>

pou eise?

channel check

'where are you?'

11.39

<athinago>

loipon me ti allo asxolise endiaferonta sou ennoo

introduction3

'well what else do you do, your hobbies I mean'

11.40

<Gipsy>

mou aresei to cnema

'I like cinema'

11.41

<Gipsy>

prepei na figw,bye

pre-closing greeting CLOSING

'got to go, bye'

Session Close: Fri Oct 16 23:20:34 1998

Appendix 2: Development of a conversational play field

12.53

<Loucy>

Oste thelis na me parisireis!Ntropi sou

12.54

<Progomfios>

:))....Na sou po..........parasiresai....._4efkola...?????????

12.55

<Loucy>

Parasirome otan thelw egw!!

12.56

<Progomfios>

....Kalo afto.......!!

12.57

<Progomfios>

Kai..........thelis na se _13parasiro???

12.58

<Loucy>

Na me parasireis se TI??

12.59

<Progomfios>

Hehehehe........

12.60

<Loucy>

Hehe??poniro??

12.61

<Loucy>

MIn pareksigithoume

12.62

<Progomfios>

_4Hehehehehehehe........................

12.63

<Progomfios>

(Mi mou pis oti pareksigithikes??)

12.64

<Loucy>

Oxi vre

12.65

<Progomfios>

A.

12.66

<Loucy>

Aplos ithela na to ksekatharisw,astievomaste

12.67

<Progomfios>

A.

12.68

<Loucy>

Oi kaloi logariasmoi kanoun tous kalous filous!

12.69

<Loucy>

Eime poli piraxtiri egw

12.53

<Loucy>

so you want to lead me astray! Shame on you

12.54

<Progomfios>

:))....let me tell you..........are you.....easily led astray...?????????

12.55

<Loucy>

I am when I want to!!

12.56

<Progomfios>

....that's good.......!!

12.57

<Progomfios>

and..........do you want me to lead you astray???

12.58

<Loucy>

in WHAT??

12.59

<Progomfios>

Hehehehe........

12.60

<Loucy>

Hehe??sly??

12.61

<Loucy>

let's not get offended

12.62

<Progomfios>

_4Hehehehehehehe........................

12.63

<Progomfios>

(Don't tell me you were offended??)

12.64

<Loucy>

no, vre

12.65

<Progomfios>

A.

12.66

<Loucy>

I just wanted to make it clear, we're joking

12.67

<Progomfios>

A.

12.68

<Loucy>

better be safe than sorry!

12.69

<Loucy>

I'm a teaser

Submitted: 09.03.2005

Review results sent out: 06.05.2005

Resubmitted: 28.05.2005

Accepted: 29.06.2005

.



Notes

[1] In the examples that follow all utterances which are not glossed appear in English in the original interaction. For lack of space, discussion of the code-switching between the languages used in my data as well as between Standard Modern Greek and the Cypriot dialect is left for another occasion (see, however, Goutsos 2002 for a starting point). An equally important question that cannot be discussed here concerns the transcription conventions for Greek, since IRC does not yet allow for non-Latin characters (see, e.g., Androutsopoulos 1998, 2000).

[2] As will be mentioned later, it must not be assumed that the participants interacted throughout the time of connection, since there were (shorter or longer) delays in response. The total number of words is thus not distributed in each exchange according to their duration, so that long messages may involve only a few words.

[3] It must be noted that when my informants engaged in IRC communication, they were not aware of the purposes of my research and, naturally, nor were their interlocutors. For a discussion of ethics in CMC research, see Cherny (1999: 297 ff.).

[4] A case in which gender, for example, becomes relevant for interactants in my corpus concerns a major thematic preoccupation with flirting. Most free elements of conversational play (see, e.g., the text in Appendix 1) revolve around the question of sex, suggesting that when participants find a gender difference they are oriented towards using IRC as a means of enticing.

[5] In the examples, the first number refers to the message and the second to the line. The data are given exactly as they appear in the log file of the relevant communications. An interlingual gloss in English follows Greek data. As noted above, because of technical restrictions, e-chat Greek is written in the Latin alphabet mostly in arbitrary, individualized transliteration.

[6] The structure of the opening phase concurs with the high occurrence of acts like 'being in physical/geographical space' and 'appearance' in personal identity moves in Yates's (2001) study.

[7] This does not mean that lipón cannot have an ideational function as in 2.37, where it indicates a conclusion (cf. ara: 2.40). For a full description of its functions, see Georgakopoulou and Goutsos (1998).

[8] It is interesting to note, however, that not many discourse markers are used, in comparison with other technologically-mediated forms of communication such as answering machine messages (see Goutsos 2001; cf. Condon and C'ech 1996).

[9] Cf. also the use of wraia 'fine' in 3.14.

License

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.

Fulltext