Framing and face in Internet exchanges: A socio-cognitive approach

Hillary Bays (Paris)

1 Introduction

In a world in which technology can provide a large scope of technical materials and convincingly real looking simulated images (such as with HDTV, the WWW and video arcade games), many people are concerned about the notion of presence in these virtual environments. I feel that it is unnecessary to concentrate on the stimulating visual effects provoked in order to proceed with such studies, as they tend to overlook the fundamentals of our cognitive need for the feeling of presence in our everyday lives and interactions. If we can understand some of the underlying mechanisms which establish presence on a cognitive level by using a very perceptually stripped mode of interpersonal communication, we can then apply these concepts to research and development for these more stimulating environments - in which studies have shown that the feeling of presence increases as the perceptual details become more precise.

This paper will examine some conversational mechanisms in an emerging form of computer mediated communication (CMC) called Internet Relay Chat (IRC). The goal is to gain access to some cognitive structures of interaction through an examination of this mediated social domain. For this, I will concentrate on socio-cognitive manifestations of frame and face, notions based on the writings of Erving Goffman. Throughout this discussion of IRC, I assert that there are base cognitive structures of presence which are constructed through discursive conventions and strategies and inside of which we find parallels to ordinary conversation.

The specificity of IRC resides in its mix of both written and spoken characteristics. IRC strips the physical qualities of conversation found in face to face interactions such as vocal intonation, stress, and gestures. However, there IS presence. I propose that the participants suspend the notion of physical co-presence because there are other discursive strategies which fulfill the need for the underlying sense of presence. In the physical setting of the computer screen and the scrolling dialogue the frame establishes a valid interactive environment. Discursive devices such as third person self-referential speech and conversational strategies such as management of multiple simultaneous "conversations" and addressivity delineate perceptual barriers and allow the users to engage themselves personally in these electronic communities.

I will indicate how participants construct and maintain a social environment, studying first the construction of frame and then the dynamics of facework within IRC. Inside of this frame, face maintenance functions on a structurally basic level, while the surface manifestations are adapted to the medium.

2 The specificity of IRC as a form of computer-mediated communication

In order to begin a study about CMCs and specifically IRC, it is necessary to define their qualities and to delineate the field of study. Many of today's academics are already familiar with the most common forms of CMCs such as e-mail, the Minitel, the World Wide Web and perhaps even BBSs (newsgroups), IRC and MUDs. Fortunately for my study there is an ever growing mass of litterature (cf Herring, Jones, Donath, Rheingold, Schiano etc.) which addresses the novelty of CMCs and compares them to other modes of communication. For my study I have chosen to treat only IRC, first, because it is synchronic communication and, secondly, due to its relatively bare format and simplistic operating technology (in comparison to other forms of synchronic CMCs, such as MUDs or MOOs). The channels I observed were found on the Undernet, one of the newer networks which has less participants than the original EFfnet. Each channel had a minimum of six participants on the userlist and generally had no predefined explicit theme concerning topics such as sex or science fiction. I chose these less defined channels so that the dialogue would be more centered around the social interaction of the IRC encounter itself and less around a fantasy game. Indeed, because the channel is understood as a strictly "social" channel, this may motivate its users to maintian the social encounter above all, sacrificing content for the social ritual.

Although there are now more indepth studies of the specificity of IRC as a means of communication (cf above), it is useful to reiterate the most important characteristics of IRC in order to illuminate its particularities in terms of conversation analysis and my particular study. To explain briefly, "IRC is a multi-user chat system where people convene on "channels" (a virtual place usually with a topic of conversation) to talk in groups or privately." (1) When one first "connects" to a channel, he is put inside a setting where other conversations are already in progress. (2) To "join the conversation", participants type and send in their contribution to the computer which manages the channel and this text is posted on the screens of each participant in the order that the computer receives and retransmits them. So what one sees on the screen is a running scroll of a kind of dialogue made up of short sentences which are posted in relative real time (relative because there is always a slight delay between the moment that a message is typed and the moment it is posted on the screens of the users. The terminology, however, remains "real time" or "synchronic".).

What one sees is something like this, posted line after line as it is submitted to the interaction. (my numbers in the left margin)(3)

1. *** Channel #A created at 8 Octobre1996 6:38:47
2. so-hot: rofl (4)
3. Mamoo: kentucky
4. Tungzzz: 0 Home of Fried Chicken..
5. Mamoo: hi celery
6. celery: hello
7. so-hot 0,15 0,14 0,13 0,12 0,11 0,10 0,9 0,8 0,7 0,6 0,5 0,4 0,3 0,20,6 Hi There celery 0,2 0,3, 0,4 0,5 0,6 0,7 0,8 0,9 0,10 0,11 0,12 0,13 0,14 0,15
8. Birdie: mmm celery for breakfast... naaaa doesn't sound good lol
9. _sunnyman: lol (5)
10. _sunnyman: how about brocolli ???
11. so-hot: lol
12. Birdie: rofl no thanks!!!!!!!!!
13. _sunnyman: broccoli actually

 IRC exchanges are textual (There exist, however, some limited functions which allow participants to send eachother sounds such as "Ping" or a burping sound).(6) Meaningful physical signs and prosodique cues such as gaze, gesture, intonation and proxemics must all be condensed into this one form of transmission. So from a first impression, one could assume that IRC is something like a correspondance, like e-mail or BBSs. But there are some fundamental differences which distinguish IRC from written interaction. First (unlike the two forms previously mentioned), the exchanges on IRC are technically not intended to be permanent as the server does not save and stock more than about an hour's worth of interaction. An individual (like a reseacher) can save and print the exchanges, but this does not capture the unfolding aspect of these exchanges, and one unfortunately sacrifices all of the meaningful cues that temporality (that is the time and rhythm of the postings)

(7) provides, such as pauses between responses and computer lag time.

Secondly, unlike mail or a phone call, IRC conversations usually involve more than two people. I have seen up to 60 connected simultaneously - far more than a typical face to face conversation as well. Exchanges in populated channels are often broken down into conversations between pairs or groups of three interactancts who send private messages to eachother, much like the Minitel.

On the screen one sees multiple conversations and overlapping conversation threads happening simultaneously. It often takes a few minutes of observation before one can distinguish the various conversations from eachother and decide which one to join, if any.

Additionally, unlike relationships maintained through a correspondance (except for pen-pals and in some business situations), the participants of IRC do not generally know eachother in the physical world, and do not typically (initially) connect to these groups in order to make subsequent rendezvous. The objective is first and foremost the situational encounter, although this can lead to subsequent meetings if the participants so desire.

Although the medium is textual, exchanges in IRC are organized much like ordinary conversation. For example, we find recognizable turn-taking sequences in the scrolling dialogue. The contributions are often just threads or fragments which serve to elicit further contributions in a question/response schema. The different threads of text get picked up and continued in a not very urgent or systematic way, something like how different topics are developed or left alone at a cocktail party. In this way IRC functions in a similar manner to the Minitel messages. Michel de Fornel writes, "...Bien qu'il n'y ait pas coprésence physique des interactants, on se trouve dans une situation d'interaction diffuse (non focalisée) qui présente des traits structuraux assez.similaires à d'autres situations d'interaction diffuse telles les soirées ou les fêtes." (Fornel 1989). The conversational style leads Cherny to call them "page conversations" and they seem to follow the internal structure of a conversational mode, as participants seem to treat turntaking and organisation as if IRC were ordinary conversation.

A system of addressivity (Sack's Speaker select), which functions in a similar manner to gaze in face to face interaction, directs the turntaking either toward one particular participant or to the group in general. These turns are then followed by subsequent responses directed toward the initial sentence or to the group at large to open the discussion. "Cette organiszation des prises de tour est renforcée par l'emploi régulier, au niveau des séquences d'actions, de paires adjacentes, telles les couples questions/reponses." On the screen one sees multiple simultaneous conversations between the various participants which can be distinguished due to the nicknames directing the contribution to a specific interlocuter. (8) The messages are posted onto the page continually and as they are sent, a rhythm develops which permits the interactants to read and to manage their own participation in the exchange.

Yet, even as IRC approaches spoken form through its system of turntaking, it has developed unique codes of behavior and communication to harmonize the interplay of synchronic communication and textuality. These constraints lead to a variety of techniques to augment the speed and the capactiy of information transfer such as abbreviation, elipsis and a telegraphic style which reduce the quantity of words that need to be typed, sent and read in order to transmit a meaningful message and thus to participate in the conversation. This abridging functions in a similar way to word clipping and sentence reduction in spoken language, especially among interactants who are familiar with eachother or the jargon. (9) The technical constraints also offer an interpretive context for the behavior found in IRC, such as the sequential threads of dialogue and the management of multiple concurrent conversations. This hybrid written/verbal language constitutes a unique and complex sign system.

The construction of the IRC exchanges takes on a structurally conventinal form. Turns last generally no longer than a line or two to compensate for the technical lack of overlap and interruption mid sentence. (10) In addition, whereas a turn in spoken dialogue may necessititate a response, turns in IRC are often not responded to because either a person was too slow in sending his message, or the group does not want to pick up on the thread of information.(11) The consequence for not responding to a turn in IRC is far less drastic than in face to face interactions, or private messages exchanged between a pair of interactants. That is, there is usually no reaction. Ignoring a thread may also be a strategy to avoid certain topics or the inclusion of a certain participant. Formal greetings and farewells serve as "openings" and "closings".

 There is also a discursive technique in which participants describe their own or related actions from a third-person point of view. These "emotes" (as they are known in the MUD language which is similar to IRC) have been compared to performatives as they often use proclamative and action oriented verbs (cf. Cherny 1995).

The most important defining quality of IRC is that the text is governed by temporality and immediacy. IRC exchanges take place in real time. This situation brings cognitive reaction and processing time closer to that of face to face encounters. Exchanges are rapid and participants need to respond promtly in order that their contributions retain conversational relevance. Long pauses and lengthy reponses also cause general delays which are unacceptable in this conversational mode. Through its form and its unfolding text, we can see that IRC communication lies somewhere between written and spoken modes of communication. (12)

There is also presence involved in the dialogue aspect of the turn-taking sequences, presence because the interactants react and respond to things that are written on their screens at the moment which have not been prewritten and saved. Presence is marked by the contextual relation with the preceeding messages to create a thread and to continue the line of the conversation. The transmission of a message relies on a person decoding the message on the other end. For there to be a conversational mode, it is imperative that this recipient be physically present at his computer and to manifest this presence by his participation.

It is important to remember, however, that IRC is synchronic computer mediated commmunication between real people. That is, although the physical body is not part of the exchange itself, an IRC mediated interaction can only take place when there are people simultaneously connected to the same channel and participating in the on-going dialogue.So, there is requisite physical presence during the immediate turn taking sequences. The user must be at his computer reading and responding.

The participants are sensitive to the fact that there is some unseen userer out there typing and sending a message, or a response to one of their own messages. His message substantiates his existance. This could explain also why participants announce any absence away from the screen, for example, going to the fridge or answering the phone with <brb phone> (be right back) (even if this is not away from the screen the attention is not given to the page conversation) or when they quit the channel, <zap is away now...messages will be logged>.


In addition, presence is technically part of the IRC system as the server (computer) announces each time a participant joins or quits a channel. And, during the time that one is connceted to the channel, the nickname (and a given e-mail address, fictitious or not) can be seen in the userlist window which lists all of the participants of a given channel. Here, there seem to be different degrees of presence between those connected but who are not contributing (lurking or working on another window), and those who are actively involved in the immediate activity.


3 Establishing the structure of frame and face

 Frame and face are notions by Goffman used to describe and analyse interaction. He asserts that both function automatically each time two (or more) interactants engage in conversation. The original premises of face and situational framing according to Goffman, require physical co-presence as in face to face interaction which he considers the pinacle of communication. Goffman writes :

[...interaction is] the reciprocal influence of individuals upon one another's actions when in one another's immediate physical presence. An interaction may be defined as all the interaction which occurs throughout any one occasion when a given set of individuals are in one another's continuous presence. (also called an "encounter") (Goffman 1959: 15)


In addition, sociologists such as Schutz feel that all of our experience is interpreted through the means of our bodies and therefore it is essential to integrate the latter into our interactions in order to understand them. (14)

Goffman's analyses are based on the physical signs - the images given and given-off - to maintain and repair face and to establish a situational framework for communicative activity. Physical presence permits us to see the sub-conscious indications of sincerity and reliability through embarrassment, humiliation or the emotions which are transmitted visually and/or aurally during the interaction and to interpret the context and tone of the interaction. (15) Presence is needed on a physical and cognitive level to create interactional structures and for the construction of the self/identity within this interaction. But this becomes problematic when the encounter is not face to face.

There are various degrees of physicality in all modes of mediated communication, such as the telephone, CB radio, and teleconference. We can consider that mediated modes of communication are derivations of face to face interactions and each requires situational co-presence in order to sustain an interaction. So that, as perceptual elements are missing which would correspond to the face to face equivalent, other conversational and technical strategies fill the roles left empty. Interactive mediated conversations require some form of presence, but this presence will have to be classified and qualified according to new norms and functions. I assert that co-presence is required as un underlying cognitive stucture to all encounters, although physicality itself is supplanted in disembodied IRC exchanges.

Consequently, the participants and their interactions must create situational co-presence. Within a frame of a social exchange, the participants cooperate and mutually maintain face with very similar ground rules for investing themselves as in face to face interactions. Conversational strategies having recourse to imaginary forms and content fulfill the requirements of co-presence in a pseudo-physical environment so that factual co-presence is not necessary to sustain a viable interactive setting. More precicely, physical co-presence is conjured on an imaginative discursive level. This framework becomes a closed system in which participants can express themselves and through which their actions and words are qualified. The ACTUAL physical setting of each interactant is in front of his computer and the screen (etc, hardware), yet the FRAME allows for him to interpret the context of IRC as a social encounter in "cyberspace" or the chat room.

3.1 Framework

The participants of IRC use framing in order to extablish a sense of presence. It follows that a certain number of assumptions are required to establish the frame of a physical space within the data exchange network of IRC. There is a type of common knowledge when one connects to IRC which establishes certain necessary conditions for the framework or cognitive structure. These are:

1) The media restricts aspects of ordinary face to face conversations which has effects on the syntax of written sentences, construction of "utterances", on turn-taking sequencing, on discursive responsibility/reliability and verity. In addition, there are no physical attributes which restricts prosodic cues such as the expression and creation of intonation, gaze and proxemics. So all interaction must be condensed into text form. Yet, at the same time, this lack of physical details gives people a certain freedom to create new forms of expression. Condon writes, "Because participants in s-interaction (synchronic electronic exchanges) do not share the same physical environment, all understandings they achieve must be established in the linguistic forms they enter on their keyboards, together with the interpretive strategies that apply to those forms." (Condon/Cech 1996: 65f.)


 2) Upon connection, one is entering a social arena where there are firmly established or loose rules of behavior particular to each group (This is the notion of Netiquette). These rules exist for conversation behavior, for the limits of imaginative expression, for cognitive reconstructions or for described actions. They can be the same or quite different from those in the world at large. As in face to face interactions, the interactants of IRC negotiate the space and validate themselves and their presence through their discourse. During the exchange, they concentrate in particular on the words themselves because the discursive object is their unifying link and their mutual creation. Simeon Yates writes, "IRC is a semiotic field which is defined by a situation, as a social structure and as a physical location with discursively available material objects, and by the topic of the interview as the main discursive object." (Yates 1996)


The setting is a verbally described place. It can be either communal or individual, continually recreated or a predetermined fantasy world. In three dimensional communities, though the possibility for concurrent realities people usually meet "somewhere" already predetermined and then create their own contributions. (16) The users have accepted the imaginitive structure of an environment as a foundation to build their discursive symbolic system. This discursive community becomes a "place" to meet others.

3.2 Framing the metaphor of the virtual community

frame is through the use of metaphor (cf. Lakoff/Johnson 1980). Since the earliest days of computer mediated communication, users, theoreticians and journalists (among others) have superposed the metaphor of the virtual community onto Computer Mediated Communication, and particularly in Newsgroups and in IRC. The result is that now the public at large and IRC interactants organize their activity and their discourse around this common idea.

In an IRC, as in real life (17), people tend to define things in reference to their natural experience categories such as their bodies, their environment and their interactions with others.The first and most well known/common frame is the community. This is part of an extended metaphor found throughout CMC and Internet involvement. In IRC this metaphor is restricted and becomes the notion of a smaller, more intimate group, within the larger cybercommnunity.

The community metaphor has developed parly because of the natural tendancy to use familiar referents to define something less concrete, but also as a result of the adherance to the metaphor itself and its popularity. Several qualities are associated with being a community, such as, a kind of informal attitude that implies familiarity of the participants, sharing and working together to establish the common good and a common atmosphere. (18) By framing the exchange as a community, the participants have created a viable environment for interpersonal exchanges.

This metaphor is made up of a network of ideas and other metaphors which substantiate the Internet exchanges and IRC conversations as a virtual community. The result is a large and coherent network of entailments such as:

1. The metaphor highlights certain features while it surpresses others. For example, it highlights features such as the reinforcement of the "intimate group model". For example, participants often use greeting patterns (overt openings) to demark a conscious effort towards group adhesion. But, at the same time, the metaphor ignores certain behaviors of an intimate group and, therefore, the lack of participation from an inactive participant or "lurker" is considered acceptable.

2. It only entails certain aspects of the concept. While it makes reference to the cohesion of a group of "familiar faces", it does not highlight their responsibility to eachother.

3. It gives the concept a new meaning. By adhering to this metaphor, the idea of a community is altered from being something locally (geographically) based to being a loose union of geo-distant individuals with similar interests.

4. This metaphor also justifies certain actions. Establishing the group as a community also justifies certain actions such as asking for advice, whereas the participants are techinically strangers, using an informal tone.

5. The metaphor's meaning is partly cultural and partly subjective or individual. The metaphor of a virtual community is partly a cultural idea, reinforced by books such as those by Howard Rheingold. But, it manifests itself in individual ways through the contributions of each participant. The term becomes subjective and each interprets it in the way he wants. Nancy Baym writes : "One apparant truth drives the metaphor : computer-mediated communicators themselves see their activity as inherently social. (Baym 1996)


3.3 Creation of the Physical and Social Environment

Participants use and substantiate the metaphor of the virtual community by reconstructing a physical and social environment in which their actions and exchanges are played out. For example, we see an artificial reconstruction of the physical setting. The participants imbibe their discursive activity with contextual co-presence as the following examples demonstrate.

AnN96 bows to Anthony, her knight in shining computer bytes
Anthony huggles anna
_jessica looks around for someone to chat with
Chord wanders in, waves and climbs into her bb chair
Certain discourse behaviors such as the description of physicality or the use of verbs which evoke physicality are constructed around the metaphor of an intimate group. In the following example, the participant acts as if he has just walked into a room where he finds other members of his community, or his friends.
  1(s) Mamoo: wow everyones here
2(s) camel: 2later guys :) hmmm...and girls too :)
3(c) *** camel has left channel #c
4(s) Mamoo: bye all
5(s) Birdie: bye mamoo
6(a) Mamoo give birdie a catcher off birdseed
7(a) Mamoo: and hugs and kisses all the ladies in the room and shakes the hand of
the males

 The participant, <Mamoo>, greets the others as he enters and implies that they are all in the same physical space together, a constructed environment. (19) The second character, <camel>, is signing off at nearly the same time and bids farewell to the others in a typical opening and closing sequence (cf. Cherny 1995). (He also makes sure to use "politically correct" address by writing guys and girls.). Likewise, when <Mamoo> leaves, he concentrates on group cohesion by addressing all of the other participants, following a traditional (albiet archaic) protocol of departure, and by giving out presents, eg the catcher of birdseed. He responds as if he is in a discursive "place".

In their textual exchanges, participants are aware of their own recreated physical presence and sometimes exaggerate this physicality, to amplify their attributes or actions. There is an effect of hyper-compensation for the lack of physical reality and the simultaneous minimisation of a plausible reality, so that actions and things are described which would be impossible or inexistant in real life. This creates a discursive surface structure which diverges completely from a parallel reality and establishes a convention of fictionally based exchanges, while the underlying structure is still modeled after real life referents and protocol.

-Sunsword shimmers and vanishes
-SysOP smacks razz with a 9 GB SCSI-2 Hard drive. Nice to see ya again old buddy!!
-Juni passes around forms "Apply here ;)" -Small_fry EATS the forms.

 Physicality exists within the world of IRC as a frame on which the rules of interactive conduct and "reality" within the CMC are based. We see, then, that by passing into an imaginary level, the need for physical co-presence has been supplanted by this discursive reconstruction.


4 Face

The IRC Community is a place for situational interaction, essentially, a social environment where the interactants consciously choose to be. Like in the notions of face and face maintenance are important to ensure the overall success of the communicative encounter.We see that many of the usual implications for facework apply using similar ground rules as in the physical world. This is so because, first, the frame has been set as a valid environment and, secondly, there is a socialization process which takes place, in which all of the participants are influenced by the social norms and rituals of the IRC system.(20) The mediated interaction becomes ritualistic like all social exchanges and automatically appeals to face.

How do we see facework in IRC? According to Goffman:

"Face may be defined as the positive social value that a person effectively claims for himself by the line others assume he has taken during a particular contact. Face is an image of self delineated in terms of approved social attributes - albiet an image that others may share" (Goffman 1967: 5).

On the most basic level, people maintain face by accepting and recognizing other participants given in their nickname. According to Brown and Levinson (1987), all adult members of society have face and are aware of other's face. Here this is transferred onto the imaginary character.

Other facework characteristics in IRC are that participants often have a "line", that is, they have a pattern of textual contributions (discursive acts) by which they express their views of the situation and through this their evaluations of the participants, especially themselves. A certain front is maintained in a particular framework or setting. Keeping one's line involves ritualized symbolic action containing stigma symbols, prestige symbols and disidentifiers. For example, <SunSword>'s line is that of a magical character, and he reinforces this idea when he writes, <SunSword shimmers and vanishes> as he leaves.

4.1 The individual's identity: Nickname

As protocol, to join a channel one first chooses a nickname, which is not permanent and can be changed at any time before, during or after an IRC session. This nick can be anything from numbers and punctuation to a highly personal and/or evocative name. Here are some examples (21)

Mamoo so-hot camel Birdie _sunnyman dthangel Goof-BBL Lyn2 Tungzzz @\gizsleep dohcan @W. (22)

The nickname is the first sign of individuality when one encounters another participant. It serves as a first impression and shows the aspect of the face that the participant wants to present. Elizabeth Reid (1996) compares the first impressions and judgements of the nickname to the physical details gleaned upon meeting someone such as his haircut, clothing or palour. All manner of sociological cues can be surmised from the information packed into the nickname, for example, if the user is male or female, if he has read William Shakespeare or William Gibson, or what his general interests, class, or approximate age may be. Of course, these details are also superficial and can be completely eroneous. Therefore, there is an ongoing verification process in the dialogue to find out if the participant is what or who he says he is.(23) The nickname is the beginning of a person's line. The verification process attempts to discover holes in the consistancy of this line.

It is interesting to note that the implications of presenting a distorted or fictitious character are far less serious in CMCs than in the physical world. The truth value of an encounter and an exchange seems to be reduced, first, because of the material difficulties of verification and, secondly, because the participants seem to tolerate a larger expression of imagination and fantasy. Goffman notes that someone's line is often verified on a larger scale which covers more than the situational encounter. That is, the interactant judges the congruency of his line in view of possible future encounters. "An encounter with people who he will not have dealings again leaves him free to take a high line..."v (Goffman 1967: 7).

The criteria for the construction of the online identity diverges from that of the physical person. (24) We notice that the idea of psychological continuity of one's identity is questioned as the search for diversity and expression takes precedence. With each successive nickname one can assert a new individuality and recreate the limits of his "self", whereas in the physical world one is technically less able to change his identity with such maleability. A dimension of self reflection arises in CMCs as the participant technically can question his identity on a physical, superficial, social level each time he connects. In constrast, because the referent is less changeable, in physical world the connection between one's name and one's physical identity is much stronger. (25)

At first one is tempted to assume that with this system the participants are anonymous and can represent themselves in any shape or form. Technically, this is so, and some participants are quite creative with their choice of nickname and personality. However, a recent study by Diane Schiano (1997) finds that most people say that they represent themselves (with this or that characteristic slightly amplified) while on-line. Dibble (1998) speaks of how when a participant is new to online communities, he often experiments with his personality and different forms of himself. But, with "CMC maturity" this dies down and the perceived anonymity of early days develops into pseudonymy, where participants reciprically recognizes eachother. As in the physical world, one finds a strong relationship between one's self and the face/persona that one presents in IRC as well.

In his analysis of Minitel message services, Michel de Fornel (1989) asserts that the pseudonym is a symbolic locus for presence, which establishes presence of the individual and therefore allows an exchange to take place. He asserts that for the participants to create a non-physical co-presense, they must believe that the messages are being exchanged with someone who is copresent in front of his computer somewhere. The pseudonym (nickname) acts as a marker that individualizes this interactant and indicates that he is co-present at the moment of interaction (ibd.).

There is a fascination and attraction to be involved in an online conversation. Attachment to one's own face may explain why participants become committed to an IRC encounter (ibd.). Also, people feel engaged because connecting to a channel in IRC is a voluntary and conscious act to link onesself into an online social environment. Since it is known by all that the participants are geographically distant, then, the users are connecting to be social in the situational environment of the IRC channel. It is partly this desire that drives people to engage and to invest themselves. The more they invest of themselves the more complex and interesting the channel can be(or this could be the logic concerned). There is a degree of personal responsibility to enliven the channel, a kind of civic duty to their own community.

Participants also observe forms of politeness, such as the positive and negative politeness discussed in Brown and Levinson. They establish social norms adapted to the virtual community metaphor, they are considerate and they tend to work against face threatening acts (FTAs), for example, they concur on many things and try to avoid conflict. They are self regulating members of a socialized system and participate accordingly in social encounters. Nancy Baym writes that although "participants' communicative styles are oriented around common social practices before they even enter into CMC, practices that are unlikely to be supplanted by computer mediation (Contractor and Seibold 1993 - p141 cybersociety)" (Baym 1996: 141)


Research on the social uses of CMC has demonstrated, in a variety of CMC contexts, ways in which participants form group specific forms of expression, identities, relationships and normative conventions.

4.2 The cookie convention

On several chat channels that I observed, a conventionalized pattern of social behavior seemed to emerge which imitates the collective value of generosity in order to create a pleasant group atmosphere. In IRC conversations one remarks on the friendly behavior of the participants towards eachother as they give and share online. This behavior has been enlarged into codes of symbolic action, and in particular, that of the distributing imaginary cookies. In each example one participant initiates the "action" of giving the cookies to the others as an offer of goodwill and to set the tone of the interaction and the ambiance of the chat room. Other participants follow suit and contribute to develop the metaphor attributing the "cookies"such qualities as homemade, flavor, ingredients, and origin. The two following sequences originate from two different IRC channels and are dated six months apart, but revolve around the same activity (which has also been recorded in real life situation, cf personal letter from (26)


 In the first sequence, the participant <SunSword> announces that he has a headache - which would dampen the positive orientation of the interaction. So, <Aries^>, using the third person to describe her own "action", offers him some medicine (Aleve) which elicits a "thank you". But his mood seems to really improve when, <Aries^> offers cookies as an alternative to aspirin. <SunSword> reacts with punctuated exclamation.


1(a) RachelMcL offers homemade Easter cookies that her mom sent her in the mail!!!
2(a) Ripper sits in the corner
3(s) Ripper: *sigh*
4(a) Chord sits beside Ripper.. 'sup?
5(s) _jessica: ripper, why are you all alone?
6(s) c2it: ripper doesn't like homemade cookies?
7(s) Ripper: nothing
8(s) Ripper: no
9(s) Ripper: thn end of my last reason to continue
10(a) RachelMcL eats one herself and says they are good!!!
11(s) amsue: rachel-- i'll take you up on your offer of a cookie.... :)
12(s) c2it: any hints of flavors, rachel?
13(a) RachelMcL smiles and hands a bunny cookie to amsue!
14(s) amsue: awww thanx rachel :)
15(s) RachelMcL: Sugar cookies, and sugar cookies with raisins.
16(s) c2it: mmmmmmmmmmmmm!
17(s) amsue: yummy :)
18(s) c2it: Rachel, you're making me hungry, even if I prefer chcolate chip oatmeal raison the works
19(s) RachelMcL: Hmm, mom didn't put any chocolate chips in.
In the second sequence, the cookies again serve as a way to repair the dower ambiance of the chat channel due to the negative attitude of <Ripper>. Here, when <RachelMcL> hands out the cookies <Ripper> does not accept them. So, the cookies, and the offering in general terms, become the focus of the discussion among the whole group. This serves to solidify the group.

4.3 Facework and Social Norms

In the next example, we see how social norms are reinforced by the reprimand of the various users in the chat channel.

1(s) c: Macfisto you are always asking so many demanding questions
2(s) macfisto: who the fuck died and made you king
3(s) SunSword: macfisto, watch the language.
4(s) Aries: Nice mouth Macfisto
5(s) c: voila another question! (just kidding)
6(s) macfisto: well sorry for my outburst
 Here the participant, <macfisto> was offended by <c> who criticised his asking too many questions. He strikes out against this criticism with an explicitly aggressive response. He is subsequently reprimanded by several members of the group who seem to feel that he has overstepped the boundaries of polite speech and behavior on the channel. He capitulates and excuses himself, reestablishing a friendly atmosphere by adhering to group norms. However, he changes his tone also in response to <c>'s own repair, softening the initial offence in a paranthetical key that indicates that the comment was intended as jesting language. The final sentence in this exerpt serves as a double reparation. The first to reextablish the group norms and the second to reequilibriate the exchange with <c>. This is a clear example of Goffman's reparation model where both offending parties (<c> and <macfisto>) repare the ritual equilibrium of the interaction.

According to Goffman, the risk of losing face can have physical consequences such as blushing, stammering, losing public image and being excluded from the interaction, etc. Interactiants work to avoid such ruptures in the exchange. In IRC a participant who loses face also risks receiving a barrage of pitiless insulting comments which are used to drive the unwanted participant from the channel. He can also be techinically excluded or kicked off the channel, at least momentarily, which is the most drastic overt sanction in IRC because the person can no longer be part of the social fabric.

The motivation to follow face norms seems not to be generated by the fear of sanctions, as these have very little real consequence on the person himself. Instead the motivation is derived from the desire to continue to participate in the interactive/interactional situation.

The following example demonstrates FTA management in IRC. We can observe through the sanctions given what type of language and imagery is socially acceptable and what is not.

(c) *** SysOP ( has joined channel #Chat
(s) razz: take it easy people I have to get something to eat friends are probably
waiting....I am here a lot though so look for me....
(a) SysOP smacks razz with a 9 GB SCSI-2 Hard drive. Nice to see ya again old buddy!!
(s) cube: sysop stop being an idiot
(a) SysOP whips out his 4 foot schlong and presses it against cube's forehead
(c) *** SysOP has been kicked off channel #chat by cube (moron)
(c) *** SysOP ( has joined channel #chat
In this excerpt a face threatening act committed on <cube>. It is an imposition of one participant on another and an infraction of politeness because the act is considered vulgar. The sanction comes quickly by <cube> (who happens to be a channel operator), the recipient of the face challenge and SysOp is eliminated.

From this example, we note that face maintenance mechanisms in IRC are slightly modified from those in the real world, because there is a reduction of the reality level of discourse which is replaced by an increased use of the imaginary level of discourse. Here we find the framework of exaggerated physical characteristics and a physically impossible feat which is interpreted as a face threatening act (thinking litterally, the locus of the aggression is also on the "face" of the recipient). The sanction is again a type of exaggeration of the physical world, which has no practical face to face equivalent (there is no real transfer of behavior of this type in IRC/face to face). Yet within the frame this action is completely legitimate and not considered rude itself. It is treated like only a slight slap on the hand and the offender quickly rejoins, saving his face as an independent actor, and adjusts his subsequent behavior. We see, for example, that a FTA in the physical world would rarely by sanctioned in a similar way and be so inconsequential to the participants involved. A real life equivalent would amount to throwing a person physically out of a conversation or shouting "get out now!" which would most likely elicit an equally aggressive refusal or reaction.

The IRC program also allows for passive sanctioning of a FTA with the possibility to selectively eliminate the contibutions of another character. A user can select (through a technically simple process) not to receive any postings from a particular nickname. He can effectively ignore the other character. In the real life we can choose not to listen to someone, we do not usually choose not to hear the person. The "silenced" participant, however, can change his nickname and be "heard" again. Only a channel operator can eliminate another nickname and, so, only under very specific circumstances can the rank and file users of a chat group appeal to this higher source, and then perhaps the provider, to eliminate the nickname from the entire chat group or even to cut the IRC connection of the projenitor (cf. Diblle 1998).

Thus, the channel operators in IRC have more technical power than other participants and are there to regulate the group activity. This feature of IRC parallels some social hierarchies found in the real world. A channel operator is the person who created the channel and any other designated participant he grants these powers to. The special role of this monitor or channel operator is to maintain general politeness and the social norms for the group by using his few extra technical "powers" which set him above the other users. For example he can set the "topic" of the conversation, or he can bar impolite, aggressive or non-conforming behavior and eliminate certain users from the channel for having threatened the group face, the social norms. If a channel operator kicks a participant nickname off the channel it is usually only momentarily, although, as mentioned above, if the offense is serious, this could be a permanent ban on the nickname. But this doesn't stop the participant from changing his nickname and rejoining the channel. So, the politeness equasion proposed by Brown and Levinson (1985: 15, 74-83; relation of Power/Ranking of the imposition/Social Distance) is first weighted for power in favor of the channel operator, making a hierarchical situation for the control of social norms, etc. And, the surface manifestations of these sanctions are exaggerated forms of the base structures to control the group's "proper" behavior. But the medium also favors the distance factor as each participant can change his former identity and rejoin the social activity. (27)

5 Conclusion

 IRC is a unique medium because it strips communication down to a perceptually minimal mode. Its combination of textuality and temporality creates an environment in which text is structured in a conversational mode, permitting socio-cognitive structures such as one finds in face to face interaction. In addition, this mode also allows for an enlarged possibility for identity experimentation and fictive exaggeration of discursive action.

We have seen that this discursive environment is made possible through situational framing. Participants recreate co-presence through the use of metaphor and the interaction with other interlocutors (as established by their nicknames). To do this, the interactional presence is recreated by the imaginative discourse and the interpretive stuctures the particpants use to decode this discourse. They establish a text based interactive community, like an intimate group within the larger metaphor of the virtual community.

Conversational co-presence is again underlined by the fact that due to the technical synchronicity of IRC, there is real, albiet perhaps geodistant, co-presence among the interactants each attending the encounter from his personal computer terminal.

Through imagination and discourse, the participants create a discursive physical realm in which there are conventions of behavior and language. Face here is reinforced by the participants' investment in their own online personality and in the social environment that they have established. To a large extent, the interactions structurally parallel those found in the real world, while particpants exploit the linguistic possibilities unique to the hybrid text-conversation medium.

Here we see the establishment of social norms and face maintenance in which the interactants are careful to retain the ambiance set in the channel. This they do partly to avoid sanctionning, yet mostly out of a desire to continue the social, conversational ritual equilibrium they intended to achieve.

IRC allows us the opportunity to pare down the parameters of interpersonal interaction in order to try to see the underlying structures. Given the technical restrictions on physical expression and co-presence, we note that the participants readjust their criteria for a valid exchange and adapt their contributions and the medium to their desire to communicate.The key to this is the recreation of presence as the cognitive foundation upon which all other conversational strategies are based.




(2) This phenomenon has been compared to the continual flow of information found with the radio or the television. That is when one turns on these media, he comes in during some point in the continuity.

(3) Notes on how to read the IRC scritp: (1) Messages posted by the server computer are preceeded with three astericks. These messages concern the state of the channel, for example when the channel was created, as in line 1; the topic of the channel; mode changes; when a participant joins, is kicked off or quits a channel; nickname changes, etc. In the following transcriptions these are marked with (c). (2) Contributions understood as "speaking" (lines 2-6, 8-14) begin with the nickname of the participant who sent the message followed by a colon and the "spoken" text. In the following transcriptions these are marked with (s). (3) Messages to be read as "actions" are indented more than the "spoken" texts. These contributions are otherwise unmarked with punctuation, but are usually delivered in the third person singular, as in line 8. In the following transcriptions these are marked with (a). (4) Some common abbreviations and symbols used are the emoticons :) showing content, ;) a face winking for sarcasm or a joke, :=) a German version of the happy face. These emoticons can also be exaggerated, for example :)))) meaning very content.

(4) rofl is the acronym for "rolling on the floor laughing".
(5) lol is the acronym for "laughing out loud".

(6) In one recorded IRC exchange the topic was set to "Don't PING me". The interactants began to send auditory cues to eachother and notes to back these up. Finally one user wrote, <juni: pinging hurts, it's like poking someones bellybutton ;p>.

(7)The cues given by the time it takes to read, type and send a response or to have it posted are interpreted as urgency, involvement, system delays and/or lulls in the conversation.

(8) In addition, participants can also send private messages to eachother. One cannot see the private messages sent between other participants, one can only see his own and those directed at him. In some large channels, there is practically no interaction on the collective screen, and we can assume that there are several private conversations taking place between pairs of interactants.

(9) Here we can recall Grice's maxime of quantity to send a relative message in as few words as possible. IRC messages seem systematically to break the rules of Grice's maxim of quantity.

(10)Unlike in BBSs, participants do not generally formulate long arguments. These would be considered as a monologue and, therefore, impolite because it barrs the others from contributing and upsets the conversational mode

(11) Not responding to a thread is less consequential than not responding to a spoken sentence. Sentences are often left open, with trailing endings, cf Werry (1996: 51, footnote 12).
(12) For further examinations of the differences between IRC, written and spoken forms of communication, see Herring (1996).

(13) Julian Dibble writes that it is standard in MOOs (MUD-Object-Oriented), a similar form of synchronic mediated communication, for participants when they are not connected to have a "sleeping" character, which then wakes up when the real person connects.

(14) For a discussion of Schutz's theories in relation to IRC, see Velkovska (1997)..

(15) Hugh Miller (1998) writes, "It could be argued that electronic communication is not interaction in Goffman's sense at all."

(16) MUDs and MOOs, for example, which are based on the construction of a three dimensional physical space, provide a skeletal structure for participants to contribute their ideas and to add onto the already existant fictional physical environment.

(17) "irl" is a commonly used abbreviation in CMCs and means In Real Life, in contrast with the cyber-life.

(18) Rosello (1994) objects to the metaphor of the Virtual Community citing Heidigger and pointing out that Internet exchanges lack the "everdayness" of life. Cf. Zickmund (1997: 185).

(19) Ray Oldenburg discusses the neutral ground, or third space, on which communities are based. This concept stipulates a physical common ground which can be mirrored in social interaction. Quoted in Rheingold (1993: 25f.)..

(20) J. Dibble talks about losing his "newbieness" after gaining some experience in MOOs and this loss engenders a different attitude towards what is said and behavior in the exchanges.

(21) It is interesting to note that the protocol of CMC research now stipulates that nicknames and chat channel names should be changed in data in order to protect the privacy of the participants. This may seem redundant as nicknames are already masks which replace the real name of the individual, but as one user wrote in Diane Schiano's survey, "Pseudonymy is not anonymy".

(22) The arabesque @ preceding a nickname indicates that this user is a channel operator and has some additional technical "powers" over those who are just visitors. Discussion follows.

(23) It would constitute an entire sociological study (for which I do not have time in this paper) to examine the role of Habitus, the social baggage that we carry with us and that we expose whenever we open our mouths or write etc, in IRC, that is, to evaluate what image particular nicknames incur, what sort of social appartenance a nick suggests, which nicknames are socially acceptable and under what circumstances or contexts. For further discussion cf. Lawley (1998).

(24) For an analysis of Internet identity creation see Donath (1998).

(25) Through imagination, tranvestitism and masquerade people do experiment with their self-representations, but the physical form often reveals contraditions.

(26) The action is initiated by participants who have different nicknames, although this could technically be the same person.

(27) Analysis of this FTA equasion as concerns IRC deserves further analysis.



Baym, N. (1996): "The Emergence of Community in Computer-Mediated Communication". In: Jones, S. (1996) (ed.): Cybersociety: Computer Mediated Communication and Community. Thousand Oaks.

Brown, P./Levinson, S. (1987): Politeness: Some Universals in Language Usage. Cambridge.

Cherny, L. (1995): "The Modal Complexity of Speech Events in a Social MUD" to appear in Electronic Journal of Communication, Summer 1995, posted <>, (21 February, 1998).

Condon, Sh./Cech, C. (1996): "Functional Comparison of Face-to-Face and Computer-Mediated Decision Making Interactions". In: Herring, S. (1996) (ed.): Computer Mediated Communication: Linguistic, Social and Cross Cultural Perspectives., Philadelphia.

Dibble, J. (1998): "A Rape in Cyberspace. or How an Evil Clown, a Haitian Trickster Spirit, Two Wizards, and a Cast of Dozens Turned a Database Into a Society." <> (21 February 1998).

Donath, J. (1998): "Identity and Deception in the Virtual Community. <> (21 February, 1998)

Fornel, M. de (1989): "Une situation interactionnelle negligée". Reseauxx, N°38 (1989): 31-49.

Herring, S. (1996) (ed.): Computer Mediated Communication: Linguistic, Social and Cross Cultural Perspectives. Philadelphia.

Goffman, E. (1959): The Presentation of Self In Everyday Life. New York.

Lakoff, G./Johnson, M. (1980): Metaphors We Live By. Chicago.

Lane, E. (1998): "The Sociology of Culture in Computer-Mediated Communication: an Initial Exploration", <> (21 February, 1998).

Miller, H. (1998): "The Presentation of Self in Electronic Life". <> (21 February, 1998).

Reid, E. (1996): "Virtual Worlds: Culture and Imagination". In: Jones, S. (1996) (ed.): Cybersociety: Computer Mediated Communication and Community. Thousand Oaks.

Rheingold, H. (1993): The Virtual Community: Homesteading on the Electronic Frontier. New York.

Schiano, D. (1997): "Convergent Methodologies in Cyber-Psychology: A Case Study". Behavior Research Methods, Instruments and Computers 29 (2): 270-273.

Velkovska, J. (1997): Formes de sociabilité sur les réseaux électroniques, unpublished DEA thesis from the Ecole des Hautes Etudes en Sciences Sociales.

Werry, C. (1996): "Linguistic and Interactional Features of Internet Relay Chat". In: Herring, S. (1996) (ed.): Computer Mediated Communication: Linguistic, Social and Cross Cultural Perspectives. Philadelphia.

Yates, S. (1996): "Oral and Written Linguistic Aspects of Computer Conferencing". In: Herring, S. (1996) (ed.): Computer Mediated Communication: Linguistic, Social and Cross Cultural Perspectives. Philadelphia.

Zickmund, S. (1997):, "Approching the Radical Other: the Discursive Culture of Cyberhate". In: Jones, S. (1997) (ed.): Virtual Culture: Identity and Communication in Cybersociety. London.