Writing, Play and Performance on Internet Relay Chat


  • Brenda Danet,

    1. Brenda Danet is Professor of Sociology and Communication at the Hebrew University of Jerusalem. She has been studying aspects of language, communication, and culture on the Internet since 1991. Her interests include the nature and implications of playfulness in digital communication; consequences of the loss of the text as material object in computerized communication; the new media and the future of performativity; email and the history of letter-writing; gender, play and performance on the Internet; and virtual theatre. During the 1996-97 academic year she is on sabbatical at the Center for Folklife Programs & Cultural Studies, The Smithsonian Institution, Washington, DC, where she is working on a book Keybo@rd K@perz: Play, Culture and Communication in the Early Digital Age. Email: msdanet@pluto.mscc.huji.ac.il, 1996-97: bdanet@erols.com or cfpcs.danet@ic.si.edu.
    Search for more papers by this author
    • 1

       We would like to thank Barbara Kirshenblatt-Gimblett, Tamar Katriel, Irit Katriel, and Hagai Katriel, for help and encouragement, and Fay Sudweeks, Sheizaf Rafaeli, and Margaret McLaughlin for feedback on earlier versions of this paper, and for the opportunity to be honorary members of ProjectH. Nachman Ben-Yehuda and Noit Meshorer provided information on terms associated with the culture of smoking marihuana. We owe a special debt to <Thunder>, the central figure at the “virtual party” analyzed in this chapter, for facilitating Ruedenberg's entrance into the world of IRC and for making this virtual party possible.

  • Lucia Ruedenberg-Wright,

    1. Lucia Ruedenerg-Wright became interested in online communications while completing her Ph.D. in Performance Studies at New York University. After a postdoctoral year at the Dept. of Behavioral Sciences, Ben Gurion University in BeerSheva, Israel, she began to teach Internet skills and issues in the Dept. of Education, Ben Gurion University and in the Dept. of Communication & Journalism, The Hebrew University of Jerusalem. In addition, she participated in the founding of the Ramat-Negev Freenet in the northern Negev Desert and has researched its impact on residents of the region. Currently she is Webmaster for the Graduate School of Arts and Science at New York University, and teaches an Internet course remotely, via the Internet, at Ben Gurion University, Department of Education. Email: lucia@bgumail.bgu.edu or lucia@ramat-negev.org.il or ruednbrg@acf2.nyu.edu.
    Search for more papers by this author
  • Yehudit Rosenbaum-Tamari

    1. Yehudit Rosenbaum-Tamari is a doctoral student in sociology at the Hebrew University of Jerusalem. The subject of her doctoral thesis is Play, Language and Culture in Computer-Mediated-Communication. She is the director of the Research Department at the Israel Ministry of Immigrant Absorption. Rosenbaum-Tamari's research interests include the sociology of computer-mediated-communication; play theory and research; the sociology of language; and migration and immigration and acculturation. Since 1990, she has been conducting an ongoing survey on processes of absorption of the current wave of immigrants from the former Soviet Union in Israel. Email: msrosen@pluto.huji.ac.il
    Search for more papers by this author


Digital writing is strikingly playful. This playfulness flourishes particularly in synchronous chat modes on the Internet. This paper is a study of writing, play and performance on IRC (Internet Relay Chat). We analyze a “virtual party” on IRC, whose highlight was a typed simulation of smoking marihuana. Three interrelated, yet analytically distinct types of play are discussed: 1) play with identity; 2) play with frames of interaction; and 3) play with typographic symbols. We adopt a qualitative, textual, and micro-sociolinguistic approach, drawing on work in discourse analysis, the study of orality and literacy, and the anthropology of play and performance. In all play there is reduced accountability for action. In the material world, masks and costumes at carnival time liberate participants; here, the ephemeral, non-material medium, the typed text, and the use of nicknames provide the mask. Although the improvisation analyzed here is typed and occurs between geographically dispersed strangers, it has fascinating affinities with “live” interactional forms such as jazz, charades, and carnivals.


Computer-mediated communication (CMC) is strikingly playful. Millions of people are playing with their computer keyboards in ways they probably never anticipated, even performing feats of virtuosity with such humble materials as commas, colons, and backslashes. Not only hackers, computer “addicts,” adolescents and children, but even ostensibly “serious” adults are learning to play in new ways. This paper is a study of writing, play and performance on Internet Relay Chat, known for short as IRC, a network program that allows thousands of users all around the globe, at any hour of the day or night, to “talk” to each other in real time by typing lines of text. We adopt a textual, micro-sociolinguistic approach, informed by recent work in discourse analysis, the study of orality and literacy, and the anthropology of play and performance.

To borrow Turner's (1967) metaphor, written interaction on IRC is a veritable “forest of symbols”–typographic symbols. Phenomena which we have observed on IRC partially resemble real life (RL) genres of play such as charades, a carnival or a masked ball, having a party, and putting on a show. Yet most of what happens consists only of text–letters and typographic symbols dancing on a computer screen. Events on IRC are a form of disembodied “virtual play” (Aycock, 1993), floating freely in cyberspace.

The computer keyboard is something like a piano keyboard. On the piano, in addition to playing individual notes to create a melody, one can also produce chords by playing several notes together, much expanding the expressive possibilities of the instrument. On the computer keyboard, creative individuals sometimes produce amazing effects merely with the mundane options of upper and lower case, numbers, and typographic symbols.

“Emoticons” on computer screens are icons for the expression of emotion, or for marking one's intent as non-serious. They are composed of clusters of typographic symbols, and are popularly known as “smiley” icons or “smileys” (Raymond, 1991: 142-143; Godin, 1993; Sanderson, 1993; Witmer and Katzman, 1997). The best known ones are a smile, a wink, and a frown, respectively:

:-) ;-) :-(

To view them, tilt the head toward the left shoulder. Some initiates to digital writing reject emoticons as “in poor taste,” or in conflict with standards of good writing associated with literate culture; for others, using them reflects or signals socialization to the new culture evolving in cyberspace. Although we produce these symbols sequentially, they are experienced almost simultaneously, as a gestalt, both when producing and when reading them. Thus, the effect is somewhat similar to that of a chord in music. In the exploitation of these and other possibilities of the computer keyboard, playful digital messages have fascinating affinities with graffiti, comics, the language of advertising, jazz, and improvisational theater.

Pioneering researchers on CMC in the late 1970's and early 1980's were slow to notice playfulness in the new medium, and to consider it worth investigating. Primarily interested in the instrumental, rather than the expressive aspects of communication, early research was concerned with the effects of the new medium on organizational functioning. The terms “teleconferencing” and “conferences” were frequently used (Short, et al., 1976; Hiltz and Turoff, 1978, 1993; Kiesler, et al., 1984), and are still quite common today among researchers on organizational communication, developers of the technologies, and in popular discourse about CMC. Ordinarily, the term “conference” refers to a work-related meeting; thus, many people may have expected the general frame of messages exchanged to be “serious.” The perception of the medium itself as cold, anonymous, and lacking in “social presence” because of the absence of non-verbal cues such as facial expression also contributed to this expectation (Short et al., 1976; Kiesler, et al., 1984; Rice and Love, 1987; Walther, 1992).

Persons who use computers for communication only sporadically may not necessarily have noticed or even been exposed to playful phenomena of the kind analyzed in this chapter. There is, in fact, an extraordinary amount of playfulness on the “net.” Children and adults log on to read and post to bulletin boards (BBSs), participate in recreational and work-related discussion groups, and “talk” to each other through chat programs such as IRC or text-based virtual realities known as MUDs (Multi-User Domains or Dungeons) and MOOs (MUDs–Object-Oriented; Curtis, 1992; Marvin, 1995; Turkle, 1995). In the last five years or so, researchers have begun to pay close attention to linguistic and other features that appear in these modes, including those that pertain to playfulness and expressivity.

The classic theorists of play, Huizinga (1955) and Caillois (1961), saw play as activity set apart in time and space from ordinary life. Digital play shares with RL play many important characteristics: it is voluntary, intensely absorbing, done for its own sake, and, as we shall see, more or less rule-governed. However, in cyberspace it makes little sense to ask “What time is it here?” or “Where on the globe are we?” Cyberspace provides perfect insulation to maintain a play frame. Even when physically “at work,” individuals seated at their computers can be engrossed in deep play.

Like other forms of playfulness, that in digital writing takes place in the subjunctive mode of possibility and experimentation. It is a liminal activity which engenders communitas (Turner, 1974; Turner, 1986a; Handelman, 1976) and many have suggested that the medium has a democratizing influence (e.g., Hiltz and Turoff, 1978, 1993; Kiesler, et al., 1984; Poster, 1990). Like other forms of play, those to be examined in this chapter are also amoral; they communicate “what can be” (Handelman 1976: 186). By conventional criteria, playfulness in CMC can easily be viewed as threatening to the social order:

[Play] is defined as “unserious,”“untrue,”“pretend,”“make-believe,”“unreal,” and so forth, precisely because it is a fount of unorder against which the social order must be buffered. By definition it cannot be permitted to define a moral community, since…the directions of its transformative capacity are uncharted (Handelman, 1976: 189).

Thus, the breaking down of social barriers in digitized communication can, in certain circumstances, have a negative effect. People can “go out of control,” lose their tempers, get involved in sudden incidents of “flaming” (Raymond, 1991; Lea et al., 1992; Danet, in press a). Communication is also volatile: the “play” frame can break down at any moment if participants become disturbed or offended by what is happening. When dramatic RL events such as death, war, or natural disasters like earthquakes impinge on players' consciousness, the medium is mobilized for the dissemination of late-breaking news, online expressions of solidarity, and so forth.

Cyberspace as a New Frontier

Another factor fostering playfulness is the frontier-like quality of this new world, which is highlighted in the title of Rheingold's (1993) book, The Virtual Community: Homesteading on the Electronic Frontier. Hackers are sometimes called “computer cowboys” (Hafner and Markham, 1991: 10), or “digital explorers” (Levy, 1984, Preface). Barlow suggests that:

Cyberspace, in its present condition, has a lot in common with the 19th Century West. It is vast, unmapped, culturally and legally ambiguous, verbally terse…, hard to get around in, and up for grabs. Large institutions already claim to own the place, but most of the actual natives are solitary and independent, sometimes to the point of sociopathy. It is, of course, a perfect breeding ground for both outlaws and new ideas about liberty. (Barlow, 1990).

Users of IRC treat the medium as a “frontier world:” a virtual reality of virtual freedom, in which participants feel free to act out their fantasies, to challenge social norms, and exercise aspects of their personality that would under normal circumstances be inhibited (Reid, 1991).

Melbin's (1987, chap. 3) analysis of night and night culture is also highly suggestive. Like the American West, he finds, night is characterized by uneven stages of advance; organized sponsorship rather than solo activity of individuals; sparse and homogeneous population; chances for escape and opportunity; a wider range of tolerated behavior than in daytime life; fewer status distinctions; novel hardships; decentralization of authority; lawlessness and peril. The similarities are striking, indeed!

An Inherently Playful Medium

Digital writing is inherently playful, first of all, because the medium, the computer, invites participants to “fiddle”, and to invoke the frame of “make-believe” (Bateson, 1972; Goffman, 1974; Handelman, 1976). When this frame is operating, participants understand and accept the meta-message “this is play” (Bateson, 1972; Handelman, 1976).

Four interrelated features of CMC foster playfulness: ephemerality, speed, interactivity, and freedom from the tyranny of materials. The prominence of playfulness grows as we move from basic word-processing of author-absent texts (Heim, 1987; Bolter, 1991), to hypertext and interactive fiction (Delany and Landow, 1991; Bolter, 1991, chap. 8; Lanham, 1993) and e-mail and discussion groups (Myers, 1987a, 1987b; Ruedenberg, et al., 1995; Aycock, 1993, 1995; Kozar, 1995; Baym, 1993, 1995a, 1995b), to synchronous communication (Reid, 1991, 1995; Bruckman, 1992, 1993; Curtis, 1992; Danet, 1995; Jacobson, 1996; Turkle, 1995). It is in the latter genres that writing is most intensively experienced as “talking,” and the distinction between process and product breaks down (Kirshenblatt-Gimblett, 1976, Introduction). Thinking primarily of word-processing, Heim (1987) notes:

My stream of consciousness can be paralleled by the running flow of the electric element. Words dance on the screen. Sentences slide smoothly into place, make way for one another, while paragraphs ripple down the screen. Words become highlighted, vanish at the push of a button, then reappear instantly at will. Verbal life is fast-paced, easier, with something of the exhilaration of video games (Heim, 1987: 152).

Thus, playfulness is present even when there is no apparent partner to play with. One is playing with the program and the machine, not a person. Similarly, with electronically composed literature–hypertext–in mind, argues that:

Playfulness is a defining quality of this new medium. Electronic literature will remain a game, just as all computer programming is a game. [Hypertext]…grows out of … computer games….the impermanence of electronic literature cuts both ways: as there is no lasting success, there is also no failure that needs to last. By contrast, there is a solemnity at the center of printed literature–even comedy, romance and satire–because of the immutability of the printed page. (Bolter, 1991: 130).


Synchronous modes invite playfulness because they offer an engrossing “flow experience” in which action and awareness are fused (Csikszentmihalyi, 1977). Interaction with computers is often felt to be totally absorbing; computers are experienced as a “second self” (Turkle, 1984). An important factor which fosters this sense of flow is the “magical” quality of instant efficaciousness in interaction with the computer–even when no human partner is involved. We receive instant feedback to our own feedback. This characteristic has come to be called interactivity (Rafaeli, 1988; Laurel, 1991). The sense of flow may be even greater when participating in synchronous modes than when interacting only with the computer, or even when reading and composing e-mail. People often lose all sense of time, suddenly discovering that hours have passed.

Release from the Tyranny of Materials

Yet another aspect of digital writing which fosters playfulness is the release from dependence on physical materials. From the dawn of human efforts to make marks on the world, to encode information in graphic form, there has been a struggle with materials, both to prepare materials to write on and to do the writing itself. Writing was invented in the ancient Near East some 5000 years ago in Mesopotamia and Egypt (Gaur, 1984; Gelb, 1963). For thousands of years, the dominant materials were clay and stone –not very easy materials to work with! (Gaur, 1984; Walker, 1990). Eventually, materials requiring somewhat less physical effort came to be used, e.g., papyrus in Egypt and parchment in medieval Europe. However, even these lighter, somewhat more flexible materials still required endless toil (Gaur, 1984; Troll, 1990; Avrin, 1991; Olmert, 1992; de Hamel, 1986, 1992).

Many of us tend to read-and-delete e-mail, and to send messages without making either an electronic record or hard copy of them. Similarly, while it is often possible to log a chat session, as in IRC, we suspect that individuals do not generally take advantage of this feature. The game's the thing–not the outcome: like speech, typed interaction is fast-paced and as ephemeral as the wind.

Playfulness in Hacker Culture

Playfulness is not only inherent to the medium of computers but has come to be cultivated and valued by the pioneers of its creation and use–the hackers (Raymond, 1991; Barlow, 1990; Meyer and Thomas, 1990). Hackers love to play with words and symbols (Raymond, 1991; Barlow, 1990). “They often make rhymes or puns in order to convert an ordinary phrase into something more interesting,” as in “Boston Glob” for “Boston Globe” (Raymond, 1991: 9), or “snail-mail” for ordinary mail services (Raymond, 1991: 325-6). “Dry humor, irony, puns, and a mildly flippant attitude are highly valued–but an underlying seriousness and intelligence are essential” (Raymond, 1991: 20).

Many of the emergent conventions of digital writing originated in hacker usage (Raymond, 1991: 15-16). Thus, entire messages in CAPITAL LETTERS are interpreted as shouting. There is a preference for writing all in lower-case, even for the beginning of sentences and names. While this practice may have originated out of considerations of speed, it has spread from hacker subculture to the rest of us, and is frequently found both in informal email style and in typed real-time chatting. Another common feature is the use of asterisks to emphasize words denoting actions (*grins*) or sounds (*bang*).

Some forms of playfulness are familiar from earlier genres of writing. The use of capital letters to emphasize a word or phrase, as in “I REALLY LIKE THAT!” is familiar from the comics (Abbott, 1986; Inge, 1990) and from personal letter-writing. Children and young people are naturally quite richly expressive in handwritten personal letters; this expressivity tends to be suppressed by the teaching of literacy in the schools. Other forms of playfulness, however, such as the use of nicknames, virtually de rigueur on IRC (Bechar-Israeli, 1995), and of evocative names for one's character on MUDs and MOOs (Reid, 1991; Curtis, 1992; Bruckman, 1992; 1993; Turkle, 1995) – have more in common with certain oral, not written forms of communication, such as Citizens' Band radio (Powell, 1983; Kalcik, 1985).

The Masking of Identity

In all play there is reduced accountability for action (Handelman, 1976; Honigmann, 1977; Turner, 1986b, 1986c). In the material world, the releasing effect of masks and costumes at carnival time is the paradigmatic example (Turner, 1986b). In the present textual form of CMC, it is the ephemeral, non-material medium and the typed text which provide the mask (Danet, in press b). The absence of non-verbal and other social or material cues to identity frees participants to be other than “themselves,” or more of themselves than they normally express. This is especially so when participants adopt nicknames or “handles” (Reid, 1991; Bruckman, 1992, 1993; Rheingold, 1993, chap. 5; Bechar-Israeli, 1995; Turkle, 1995; Danet, 1996).

Stylization, “Orality” and Performance

Linguistic features previously associated with oral communication are strikingly in evidence in this new form of writing (cf., e.g., Maynor, 1994; Ferrara, et al., 1991; Murray, 1991; Yates, 1992a, 1992b; Bolter, 1991; Collot and Belmore, 1992; Leslie, 1994). “Electronic text is, like an oral text, dynamic” (Bolter, 1991: 59). Although simplistic notions of the differences between (pre-CMC) speech and writing have generally been dismissed in the last decade (e.g., Tannen, 1982a, 1982b), we have tended, in literate culture of the last 500 years, to think in terms of two contrasting models of communication: the contextualized, ephemeral oral conversation, which may be deeply personalized and concrete and is in varying degrees a performance of some kind, versus the decontextualized, author-absent written text–frozen in print, an entity of physical and symbolic integrity (Ong, 1982).

Just as performance tends to be highly stylized in oral cultures (Bauman, 1975, 1977; Finnegan, 1977; Ong, 1982; Edwards and Sienkewicz, 1990), textual and typographic art are prominent in CMC (Reid, 1991; Danet, et al., 1995), both by incorporating ASCII art originally composed offline, and in spontaneous online improvisations. Participants are conscious of their audience and pay special attention to the display of communicative competence, to how their messages are packaged (Bauman, 1975, 1977). The poetic function of communication is dominant; formal aspects of language are foregrounded (Jakobson, 1960). The need to say in writing what we have been used to saying in speech calls attention to the communicative means employed in formulating the message. The reduced transparency of language heightens meta-linguistic awareness, and leads us to treat words as objects and to play with them (Cazden, 1976).


We turn now to an analysis of a “virtual party” which took place on IRC in December 1991. The analysis focuses primarily on three interrelated, yet analytically distinct types of play–play with identity, play with frames of interaction, and play with typographic symbols. The party began when Lucia Ruedenberg logged on from Jerusalem, about 1:30 a.m. local time, and made contact with a system operator (sysop) nicknamed <Thunder> with whom she had developed on ongoing electronic relationship.

During the one and one-half hours or so that Ruedenberg was logged on and chatting with <Thunder>, six other persons joined them for varying amounts of time. Except for one player with whom <Thunder> had a RL acquaintance, and his evolving relationship with Ruedenberg, all the others were to the best of our knowledge strangers, not only in RL but on IRC as well. We learned that except for Ruedenberg, an American in Israel at the time, and one person in Finland, all the participants were in the United States, though quite dispersed geographically.

Play with Identity: “Nicks”

The nicks (IRC-ese for “nicknames’) of the eight persons who appear in our log are listed in Table 1. Note that all nicks are displayed in angle brackets, as they appear on IRC.

Table 1.  The players at a virtual party on IRC.
NickUserid and AddressAdditional Material
<Thunder>root@xxxxxxxxxxxxxxxxxx(-: Raam / Chundeung :-)
<Kang>GENGHISCON@xxxxxxxx(<Drax the D>)
<Lizardo>lizardo@xxxxxxxxxxxxxxxxx(Doctor Lizardo)

At <Thunder>'s request, we have deleted details about the mainframe computers through which all participants except Ruedenberg were logged on. In addition, we refrain from using <Thunder>'s real name.

On many grounds there is no apparent need for researchers to disguise the identity of participants any more than participants have done so themselves. IRC is by definition a “public” space. Anyone with an electronic address and access to a server can log on. Moreover, all have the option of making a log of interaction. In addition, a distinction between public and private channels is built into the program, and a channel operator can make any channel private at any moment, and anyone objecting to being a part of ongoing proceedings in a public channel can instantly leave the channel. While participating in public channels, individuals have the option of exchanging private messages. Participants can change or add information about themselves, other than the address from which they are logged on. This information is viewable on screen by anyone who wants to check someone's address.

These considerations argue for treating a log as “public” and therefore not ethically problematic for researchers. However, what is perceived by participants as private or public may be quite another matter. Therefore we sought and obtained permission to publish this analysis from <Thunder> who plays such a prominent role in the log.

As for the nicks themselves, we feel there is no ethical need to disguise these disguises. There are two great principles of design in nature, the principle of camouflage and the principle of conspicuous marking (Gombrich, 1984: 6). The use of masks and costumes at carnivals and masked balls brings these two principles together (Turner, 1986b). Masks are meant not only to hide a player's real identity but also to call attention to the person and to the mask, its expressive power, imaginativeness, capacity to instill fear, evoke humor, and so on. Similarly, textual masks–online nicknames–are not only a means to disguise RL identities but a form of online “plumage.”

There is a certain amount of risk in the interpretations of the nicks we develop below since we did not interview the players, except for <Thunder> in this respect. However, we are actually in the same position as the players themselves, who must develop their own interpretations of the textual mask presented by any given player.

The Masking of Gender

Nicks on IRC often appeal to fantasy and the fictitious–to nature, mythology, the occult, comics, children's literature, science fiction, films, etc. (Bechar-Israeli, 1995). Some of these themes from popular culture or student culture find expression here too.

Perhaps the most striking thing about players' nicks is that it is impossible to know if the players are male or female. Retaining her RL name, Lucia is the only obviously gendered individual in the group, though only we could know that this was also her real name and that her RL gender was indeed female. Those knowing that Kang is a male character in Star Trek or that Genghis Khan was a notorious male, medieval conqueror could at least assign a gender to the nicks <Kang> and <Genghiscon>, if not to the player behind them. It is known that more men than women participate in modes like IRC; therefore we believe that most of the other players were male. <Thunder> once telephoned Ruedenberg while she was in Jerusalem, so we could confirm from his male-sounding voice that he is a man.

There is a tendency to play with gender identity in synchronous chat modes: either players “cross-dress”, impersonating a person of the opposite gender (not obvious for any of the players in this particular case), or they choose a gender-neutral strategy, at least in nickname or persona, if not in actual behavior (Bruckman, 1992; Turkle, 1995; Dickel, 1995; Cherny, 1995; Reid, 1995; Danet, in press b). Until now, we could contemplate the notion of gender-free personhood only in science fiction, e.g., in Ursula LeGuin's The Left Hand of Darkness (LeGuin, 1969; Dickel, 1995; Danet, in press b). A fifth of over 7,000 characters registered on LambdaMOO in February 1996 were assigned unconventional genders, and about a third of the 1000 or so on MediaMOO (Danet, in press b).

Analysis of Individual Nicks

<Thunder> root@xxxxxxxxxxxxxxxxxx (-: Raam / Chundeung :-)

The nick <Thunder> evokes power, loud noise, even perhaps an intention to instill fear in one's interlocutors–certainly control–but according to <Thunder> himself, the name was to commemorate his RL dog by that name, who had died. <Thunder> logged in with the userid “root” in his electronic address. The term “root” traditionally refers to something in nature; however on a computer, this is the standard superuser userid for a system administrator on a UNIX mainframe, and thus it is a technical term. <Thunder> appends a sequence that includes two “smiley” icons, facing each other, within which are two other apparent nicks: “Raam” and “Chundeung.” We soon realized that ra'am is Hebrew for “thunder,” and <Thunder> confirmed that chundeung is “thunder” in Korean. He was playful with his “calling card” in still another way. The insertion of the “smiley” icons is a visual or graphic pun: the end-brackets serve both as the smile on the smiling face and as conventional brackets for the text enclosed within them! Thus, he recycles the “smiley” back to its originally functional character as an abstract typographic symbol, all the while retaining its playful use as emoticon.

<Kang> GENGHISCON@xxxxxxxx (<Drax the D>)

For <Kang>, we have no less than three nicknames. <Kang> is the name of a Klingon character in the television series Star Trek. The userid GENGHISCON is obviously a play on “Genghis Khan,” the Mongol conqueror. In American pronunciation, “Khan” and “con” are homonyms, words that are pronounced the same though they are spelled differently. Thus <Kang> is playing with the relation between spoken and written language. Constraints imposed by the number of characters allowed in one's userid might also have in fluenced his choice–“con” having fewer letters than “khan.” Nevertheless, the choice of “con” rather than “kon” or “kan” suggests that playfulness is present here too.

“Con” invites other readings into the nickname. It is reminiscent of “con” as in “convict”. Is <Kang> pretending to “con” us into thinking that he is Genghis Khan? In computerese “con” stands for “console”–the computer keyboard–as in the “copy con” command in DOS that copies whatever is on the console to whatever destination you specify. The term “con” also is used to refer to “conventions” for science fiction fans as well as computer culture enthusiasts. <Kang> must have had a predilection for certain sounds: /g/ and /k/ recur in <Kang> and in “Genghiscon.” <Drax the D>, which is appended to his identity, may be a reference to a character in Dragons and Dungeons. Dragon imagery is very popular in interactive genres, especially on MUDs and MOOs, which were developed from the original offline game Dragons and Dungeons.

<Rikitiki> rpa3@xxxxxxxxxxxxxxxxxx

<Rikitiki> is <Thunder>'s fellow college student. This nick also plays with sound. It contains reduplication: two components which rhyme. There are at least three possible associations to this nick. One is to a character in Rudyard Kipling's Jungle Books (Kipling, 1961), a second to Tom Lehrer's song, “Rickity-Tickity-Tin,” popular with college students in the late 1950's and early 1960's, and a third to George Bernard Shaw's play Man and Superman, in which the heroine, Ann, disparagingly calls a character called Octavius “Ricky Ticky Tavy.” Today's college students may no longer be familiar with Tom Lehrer's cynical songs; Kipling, on the other hand, may still be favorite childhood reading. Kipling's Rikki-tikki-tavi is a little mongoose whose war-cry was “rikk-tikk-tikki-tikki-tchk!” (Kipling, 1961).

<BlueAdept> dlahti@xxxxxxxxxxxxxxxxx

<BlueAdept> is the name of a novel by the science fiction writer Piers Anthony. The convention of eliding words that begin with capital letters stems from the way “names” are created in object oriented programming. This plays with typogaphic conventions, eliding two words which are normally written a space apart, and capitalizing them where normally they would not be capitalized. A nick on IRC may have up to nine characters; thus eliding “Blue” and “Adept” allows this person to have both words in his nick. This nick also invites associations to characters in the comics, such as “The Green Hornet.”

<Jah> miksma3@xxxxxxxxxxxxxx (Baba)

<Jah> has chosen a one-syllable nick. We have no way of knowing if it has substantive associations for him, or whether he just liked the sound of it. This might be a playful spelling of “ya” as in the German or Finnish for “yes.” <Jah> has another nickname, “Baba”, inscribed in his/her electronic address. Here too we find reduplication, of the syllable /ba/. <Jah> seems to like the sound of /a/, since it appears in both nicknames.

<Lizardo> lizardo@xxxxxxxxxxxxxxxxx (Doctor Lizardo)

<Lizardo> is a version of “Doctor Lizardo”, which appears in the full address. Both play with the English word “lizard,” adding an /o/ to make it sound foreign–say, Spanish or Italian. Thus <Lizardo> is playing with the codes of language, mixing the grammar of one with the lexicon of another. The use of a mock title–“Doctor”–is, of course, also playful and ironic, especially in combination with the name of an animal, rather than a person.

<Teevie> ssac@xxxxxxxxxxxxxxxxxxxx

<Teevie> appears to be a playful spelling out of “TV”, that household word in contemporary America. It is also playful to adopt the name of an inanimate object as a nick, something rarely done in RL, but quite common on IRC.2

Frames of Reality and Multitasking

One of the most fascinating aspects of encounters on IRC is the nature and complexity of the frames of experience which are invoked and sustained by the participants. The activation of frames in real life is related to the notion of “multitasking” in computer operating systems. In computer jargon, multitasking is “loading and running several applications at the same time” (Sheldon, 1992: 13). When using a multitasking operating system, one opens up one or more windows to be kept running simultaneously, but not necessarily visible on screen at the same time. Technically, only one frame at a time is “foregrounded” while all others are “running in the background.” As humans, we do this without thinking: we may write while listening to music, or cook a meal while carrying on a conversation with someone.

Five Frames and the Relations Among Them

We identified five frames in the log we captured from IRC: (1) REAL LIFE; (2) the IRC GAME; (3) the PARTY frame; (4) the PRETEND frame; and (5) the PERFORMANCE frame. Influenced by Victor Turner's (1986c) analysis of the frames invoked when he and his students experimented with performing ethnography, we too see the various frames as nested within one another (Figure 1).

Figure 1.

 Five nested frames at a virtual party on IRC.

The PARTY frame is nested within the basic IRC frame; within the PARTY frame lies the PRETEND frame, and within that lies the PERFORMANCE FRAME. From within any of the four inner frames, participants can step “out of frame,” and into REAL LIFE, momentaril y, while still logged on and interacting with the others. Of course, they return to REAL LIFE when they log off.

Frame #1: REAL LIFE

By dint of being alive and functioning, the IRC players are already in the REAL LIFE mode, which in Western culture is dominated by the expectation of “serious” activity in the workaday world (Goffman, 1974; Schechner, 1988). Adopting Bateson's (1972) and Handelman's (1976) notion of meta-communicational frames, we suggest that the meta-message of this frame is, “Everything inside this frame is everyday life, grounded in physical space and time; actors are accountable for their physical and verbal actions, for the well-being of their bodies, and for their social commitments.” This frame continues to be activated but is moved to the background when participants log onto IRC, thereby activating frame #2, the basic IRC GAME. Within frame #2, <Thunder> initiated the PARTY frame–frame #3, when he invited Ruedenberg to join him in a channel called #weed.

In multitasking terms, REAL LIFE was kept hidden in the text most of the time during this sequence–“running” but “not visible on screen.” Only rarely did it pop into the interaction overtly. Participants stepped out of the PARTY frame, or out of the nest of frames #2-5 shown in Figure 1, to talk about things that are happening or that they have to do in REAL LIFE. For instance, at the beginning of the log, <Thunder> sends <Lucia> a private message, using the /msg command:3

Line 9 *Thunder* I am gonna shower soon

In a return private message, she comments, “first laundrey (sic), then shower…” [you're going to be a] clean boy.”

Frame #2: LET'S PLAY IRC

To log on to IRC is to activate the second frame, that of “LET'S PLAY IRC,” or, simply, “Let's talk.” There is no pretense or “fooling around” at this level, just as there is no necessary pretense or “fooling around” when people play chess. “Chess” is what they are doing. Any interaction made possible by the constitutive rules, the commands of IRC, occurs by definition within this frame. The content of the interaction may be playful, or it may be serious, as when a group of scientists discuss their research on IRC. Therefore the meta-message of this frame is “Anything may be said in this frame; participants enjoy reduced accountability if they choose to communicate in a playful mode.”

If the “talk” on IRC can be “serious,” can we still say that IRC is a game? In the broadest sense, the answer is yes. Wittgenstein's famous notion of language as a tool kit is very a propos:

Think of the tools in a tool box, there is a hammer, pliers, a saw, a screw-driver, a rule, a glue-pot, glue, nails and screws–The functions of words are as diverse as the functions of these objects (Wittgenstein, 1968: para. 11)

Just as language provides the tools for a series of spoken (and, secondarily, written) language games, IRC provides a set of tools for many different kinds of writing-based language or symbolic games. The meta-message of the IRC GAME frame is, then, “Anything may be communicated in this frame.”


The third frame in the log is one whose meta-message is, “Let's have a party; let's have fun”; participants enjoy reduced accountability; action and utterances are primarily in a playful mode. This is the first of three frames or windows in which three different genres of culturally constituted, playful activity are activated–PARTY, PRETEND-PLAY, and PERFORMANCE. All messages communicated within the PARTY frame enjoy reduced accountability; “anything goes.” The players get acquainted, chat, fool around, “drink,” flirt, and generally have a good time.

The “party” got started when <Thunder> created the channel in which it could “take place.” He proposed different whimsical names or themes, finally settled on one, and set a playful topic to go with it:

Line 53 *** Thunder invites you to channel 14
    *** Thunder invites you to channel 2
  55 *** Thunder invites you to channel 3
    *** Thunder invites you to channel 4
    *** Thunder invites you to channel 5
    *** Thunder invites you to channel 6
    *** Thunder invites you to channel 0
  60 *** Thunder invites you to channel hmmm this is confusing
    /msg thunder hmm so much to choose from ! hehe
    *** Thunder invites you to channel -)
    *** Thunder invites you to channel \
    /msg thunder hahahaha
  65 *** Thunder invites you to channel -)
    *** Thunder invites you to channel \
    /msg thunder no way!
    *** Thunder invites you to channel ok. I will stop now
    *** Thunder invites you to channel +bagelnosh
  70 *** Thunder invites you to channel +noshbagel
    /msg thunder oh yeah?
    *** Thunder invites you to channel +hsonlegab
    ***Thunder invites you to chanel +kinky sex with
    riding crops and handcuffs
  75/msg thunder I'll think about it.
    *** Thunder invites you to channel +weed

<Thunder> starts the party by playing with different nonsense names for his channel, first numbers, then typographic symbols which hint at the “smiley” collections circulating on the Net (Godin, 1993; Sanderson, 1993), or which may indicate that he was planning to perform for Ruedenberg in some way typographically. He changes the channel names so quickly that she cannot join any of them.

“Bagelnosh” was a channel that Ruedenberg and Madeline Slovenz-Low, a fellow graduate student at New York University, had often created on IRC. It had become a “hang-out” where regulars chatted daily. Play with the word “bagelnosh” continues to signal PARTY. First, <Thunder> reverses the two components as “noshbagel” (line 71), and then he types the word backwards, yielding “hsonlegab” (line 73). The next channel name proposed is “kinky sex with riding crops and handcuffs” (line 74). Suddenly the PARTY may be turning into “dark play” (Schechner, 1988). <Lucia> coyly says just, “I'll think about it.” Finally, in line 76, <Thunder> settles on the channel name #weed, issuing the invitation to this channel five times–repetition being in itself, of course, playful. <Lucia> joins the channel. For anyone familiar with American student or youth culture, “weed” immediately signals marijuana, relaxing, having a good time together. <Thunder> further elaborates the PARTY frame by setting a topic:

Line 90 *** The topic is: sssssssssssssss hmmm wheres all that smoke from?+weed

Play with symbols now becomes more sophisticated. <Thunder> anticipates the form of play which eventually becomes the highlight of this sequence. From the juxtaposition of sssssssss…and the phrase, “where's all that smoke from?” we realize that sssss… is meant to simulate undulating smoke, as well as–on further reflection–the sound made by the person smoking the weed, and /s/even happens to be the first letter of the word “smoke” in English! At first, Ruedenberg doesn't “get it”, thinking “weed” refers to “garden.” Finally, <Thunder> makes his intentions explicit:

<Thunder> I was thinking more of grass, herb, pot, marijuana :-)

As we mentioned earlier, another “activity” that signals PARTY in modern urban culture is alcoholic drinking. Soon after <Lucia> and <Thunder> finally settle into their “channel” and its “topic,” <Kang> joins them (line 123). Later on, <Rikitiki> and <BlueAdept> arrive. <Kang> takes on the role of party “host” when he offers <BlueAdept> a drink: Line 336 <Kang> icewater or rum blue? Not receiving any clear acknowledgement of the offer, he expands the range of choices to include wine and coca cola (line 342). <BlueAdept> asks for beer (line 344), but then changes his mind and asks instead for “peppermint schnapps” line 336), saying that it has a better taste (line 350). Eventually even “herbal tea” is added to the list of options.

Another range of content which signals PARTY is flirtation and sexual innuendo. <Kang> is very curious about <Lucia>:

Line 592 <Kang> lucia=female as i suspect?
637 <Kang> so lucia single long?

Not getting any answer, he says it a little “louder”, but <Lucia> plays coy:

   >oh my
  667 <Kang> how long has your life been thus far.

Not getting anywhere, he tries again, addressing his question to the group:

Line 695 <Kang> how old are y'all?

<Jah> quickly answers “18,” and <Kang> follows this with the query “lucia?” (line 698). But no reply is forthcoming. Much later, when <Kang> leaves, he asks:

Line 810 <Kang> are we engaged?
   <Jah> IRC-engaged?
   >I don't think so
   <Kang> just kidding never did get your age lucia

In addition to overt flirting there is subtle sexual innuendo in the log, as in:

Line 538 <Kang> here lucia *hands bong* long reach (over ocean)
  540 >I don't smoke
    >I just watch
    >and enjoy your games
    <Kang> must be pretty funny
  545 >can't handle the stuff myself
    >yes it is cute
    <Kang> who can?
    >well you haven't seen me
    >I don't go up
  550 >I go down
    <Kang> eh?
    >I curl up into a little ball and don't speak
    <Kang> I'll be kind and not razz about that one…
  555 >no fun at all
    >razz if you want
    <Kang> good thing we're not on +hottub!
    >up and down?
  560 >haha


The highlight of the party was a virtuoso simulation of smoking marijuana, that took place within frame #4: “LET'S PRETEND,” and culminates in a performance in Frame #5 (Figure 1). All symbolic activity in the “pretend” frame contains the meta-message “Let us make-believe; let us suspend belief” (Handelman, 1976).

In RL, extended pretend-play of this type is more common among children than adults. Just as children improvise with props created from objects around them in the physical world, the participants in our log improvise with the only means available to them on IRC–those on the computer keyboard. They simulate the various stages of smoking marijuana via words which depict or represent action and experience, and via pictorial means–“smiley” icons that “act out” these actions, in a kind of typographic pantomime. The full explication of how this is done is presented in the next section on the PERFORMANCE frame.


The meta-message of the PERFORMANCE frame is “Let's show each other what we can do with the keyboard.” Many features which we have known in oral performance in the past are present in this new variety of written interactive communication. Bauman's (1975; 1977) formulation of the nature of oral performance fits online interactive writing as well:

Fundamentally, performance…consists in the assumption of responsibility for a display of communicative competence…. Performance involves on the part of the performer an assumption of accountability to an audience for the way in which communication is carried out, above and beyond its referential content….the act of expression…is thus marked as subject to evaluation for the way it is done, for the relative skill and effectiveness of the performer's display of competence. (Bauman, 1977:11).

The PRETEND and PERFORMANCE frames are not necessarily both activated simultaneously, as they happen to be in the present case. Children can pretend to “play doctor” without any concern with showing off to each other. Here, the players show keen interest in showing off, and express admiration for their own and each others' skills.

The PERFORMANCE frame was activated from the very beginning of the log. <Thunder> knew that <Lucia> usually logged her chat sessions. The simulation and the performance actually begin when <Thunder> types:

Line 107 <Thunder> sssssssssssssss *passes joint to lucia*

Once again, he draws on the sequence of s's to simulate the undulating smoke and the sound made when smoking the joint. Notice the use of the third person verb “passes”, to describe his own action, marked with a pair of asterisks. This is a striking and very common feature of communication in synchronous and even some non-synchronous modes (Godin, 1993; Sanderson, 1993). This convention resembles the representation of action in the comics.

When <Kang> joins the interaction, he also mobilizes third-person descriptions of his own actions (*puff*, *hold,**exhale*), to simulate the acts of inhaling the smoke and keeping it inside one's body. <Thunder> keeps the simulation going by commenting whimsically that smoke is “filling the channel” (line 145). In the next line he sets the topic to “ssssssssssssssssss.” At first, this seems to be just a repetition of line 90. However, there are more s's than before–an invitation for all to become even more deeply involved in the game, or else a way to express the meta-comment that they are “really getting into it.” <Kang> responds appreciatively with a midget “smiley” (the nose is omitted– :)), and he remarks that they had better be careful–the smoke will “spread to other channels”. <Kang> further simulates the effect of the drug on him by intentionally misspelling “suspicious”:

Line 156 <Kang> they might be sushpishus, huh? :-) …(italics added)

When <Rikitiki> and <BlueAdept> join the party, the verbal simulation of smoking marijuana continues. Sometimes the participants use third person verbs, as in

Line 204 <Rikitiki> *inhales deeply*
  214 <Rikitiki> *exhales slowly*
Line 291 <Thunder> *exhales* ssssssssssssssssss

<Thunder> continues to “inhale” and “exhale,” representing the gradually dissipating smoke by reducing, line by line, the number of “s's”. Pleased with himself, he comments “wow”.

   <Thunder> sssssssssssssssss
Line 370 <Thunder> sssssssssssssssss <Thunder> ssss <Thunder> ss <Thunder> s <Thunder>
  375 <Thunder> wow

<BlueAdept> (line 386) “tosses 1/4oz of red sense on the end table.” <Thunder> “throws seeds into channel +hottub” and says “look they are floting (sic) in the tub hehe” (line 391); “maybe they will root” (line 393). Then <Thunder> contributes an unexpected, serendipitous pun:

Line 397 <Thunder> need to pack a new bowel heheheheh hehehehehehe…

<Thunder> has had a slip-of-the-finger: instead of “bowl” he typed “bowel.”“What a typo!” he chortles, in line 400. Kang is quick to pick up on his mistake. Thunder corrects himself, “laughing” at his mistake. Kang “packs” the bowl for him:

Line 401 <Kang> not bowel!!!! bowel?
Line 403 <Thunder> need to pack a new bowl hehhehehehe
Line 408 <Thunder> that is shitty pot I would say if it was packing a bowel

This word play is all the cleverer because, unlike other examples discussed so far, it is the spontaneous byproduct of a “typo”–a typing mistake.

Although <Lucia> declines to “smoke dope” at the beginning of the log, she shows appreciation for <Thunder>'s whimsical suggestions for channel names. She laughs at his silly suggestions (lines 61, 64, cited above in the section on the PARTY frame). And when <Kang> joins them and immediately begins simulating smoking dope, <Lucia> “smiles” appreciatively at his performance.

Line 126 <Thunder> ssssssssssss *passes joint to kang*… <Kang> thanx dude *puff**hold* >:-)

<Kang> comments on his own action:

Line 138 <Thunder> kang exhale. you will die :-) <Kang> *exhale* <Kang>;)

A new stage of the simulation begins at line 419 when <Kang> and <Thunder> start playing with icons as a means to simulate and represent the various stages of inhaling, exhaling, and so on.

This clever improvisation begins when <Kang> draws from his digital repertoire an icon which is not widely used: :|. He performs it twice, then adds another icon.

Line 419 <Kang> :|
   <Kang> :|
   <Kang> :\
   <Thunder> heheheh
   <Thunder> heheheheh
   <Thunder> that was great

Neither <Kang> nor <Thunder> need to say explicitly what the icons mean or why <Thunder> is laughing and complimenting him. The sequence means something like “puff, puff, hold the smoke inside.” Inspired to continue, <Kang> improvises further:

Line 425 <Kang> :/
   <Kang> :)
   <Thunder> hehehehehehe
   <Kang> *exhale*
   <Kang> :0

Again, one can follow quite easily what is happening: in lines 425-6, <Kang> smiles after inhaling. He reverts momentarily to the verbal *exhale*, but then, once again zips out from his repertoire of icons the one using the figure for “zero.” So now the entire sequence has been rendered pictorially. This inspires <Thunder> to improve on <Kang>'s improvisation. Beginning in line 430, he tries to put the whole sequence together in one line:

Line 430 <Thunder> :| :| :\sssss :)

<Kang> laughs appreciatively, and <Thunder> continues:

Line 432 <Thunder> :-Q :| :| :\sssss :)

Here, like <Kang>, <Thunder> has retrieved a seldom-used icon :-Q whose meaning is usually given as “man smoking a cigarette,” the cigarette being the tail in the letter Q. Now the sequence is perfect, and one can “read it” as “I put a joint into my mouth; I inhale twice, exhale, let the smoke out, and then experience pleasure.” <Thunder> laughs with delight at his own performance, and <Lucia> compliments him too (line 435: “:-) cute”). The improvisation concludes when <Thunder> adopts the sequence as the “topic” of his channel (line 438).

Line 441 *** Thunder sets the topic to: \:-Q :-| :-| :\sssss :-)


The textual party analyzed here revealed a remarkable degree of structure and coherence. Like REAL-WORLD parties, it had a clear beginning, middle and end; when the last guests “left,” the party was over. No doubt, it owed its structure in part to the previous contact between <Lucia> and <Thunder> and to his awareness that she had an ongoing interest in his ability to perform on IRC. However, except for <Rikitiki>, whom <Thunder> knew both in REAL LIFE and on IRC, we believe that the players had not interacted with each other before on IRC.

Like many REAL-WORLD parties, this virtual one had a featured activity which engrossed participants intensely, not dancing or drinking or flirting, but a textual form of virtuoso collective mime–not smoking pot but improvising a typographic simulation of it. The double agenda of play and performance was evident in the meta-comments that participants made about the simulation–they gave themselves and others compliments for their performance.

IRC and the Theory of Play

Playful interaction on IRC contains many important components of Caillois's (1961) classic typology of types of play and games. In its moment-to-moment quality as a textual “happening,” our material has clear affinities with the phenomena of tumult, agitation, immoderate laughter, improvisation and joy that Caillois associates with PAIDIA–“letting oneself go” through the spontaneous manifestations of the play instinct (Caillois, 1961: 28).

At the same time, LUDUS, “a taste for gratuitous difficulty,” as expressed in games of all kinds (Caillois, 1961: 27), is also strongly present, not only at the level of the constitutive rules of the basic IRC game, but at the level of emergent regulative but optional conventions of expression, such as those for the depiction of physical action. Perhaps even more important, LUDUS is dominant when the participants are conscious of performing for one another, because it is here that skill comes to the fore. As we saw, this self-consciousness was most evident during the simulation of smoking marijuana.

Caillois distinguishes between four categories of play, each of which has a PAIDIA-dominant set of play forms, on the one hand, and a LUDUS-dominant one, on the other. Caillois's famous typology is presented in Figure 2. Types of play shown in the first row are closest, in his view, to the PAIDIA end of the continuum. Those in the second and third rows are closer to the LUDUS end. For agon or contest, Caillois contrasted the turbulent activity of racing and wrestling with football and chess, both heavily constrained, indeed constituted by rules. To illustrate alea, or chance, his second type of play, he distinguished between counting-out rhymes at the PAIDIA end and betting and roulette at the LUDUS end. For mimicry, or simulation, children's tag and the use of masks are contrasted with theater and spectacles. Finally, ilinx, or vertigo, characterizes horseback riding and waltzing at the PAIDIA end, and skiing and tightrope walking at the LUDUS end.

Figure 2.

 Caillois's typology of types of play and games.

Where among all these types of play does interaction on IRC fall? At the level of the text, two are prominent, as we have seen–agon and mimicry. Ilinx and alea, on the other hand, are probably not usually discernible in the text itself, but rather are part of the experience of the players.

Contest is prominent in participants' tendency to show off, to perform for one another, as in sports and other RL forms of competitive play such as verbal dueling (Huizinga, 1955; Caillois, 1961; Labov, 1972; Gossen, 1976) and playful modes of “flaming” online (Danet, in press a). In this respect, interaction on IRC seems to fall midway on the continuum between PAIDIA and LUDUS: contest is neither totally unregulated nor totally stylized or dictated by the rules. Second, mimicry is rampant in IRC, not only in specific instances of the representation of action (as in *puff**puff**hold*), but in more extended sequences like the simulation of smoking marijuana in the present case.

Third, we suggest that there is also a component of ilinx, of losing oneself, in IRC. While this is not directly evident in the text, we know from our own experience that sitting for hours and hours in front of the keyboard while on IRC can indeed be addictive. This aspect of IRC is discussed further in the next section.

As for alea, it is a matter of luck who is already logged on when an individual joins the crowd in a particular channel, especially when visiting an established channel for the first time, or when creating a new one which others eventually join. Similarly, if a channel is left “open” or public, participants have no prior knowledge or control over who will join them (though a user can type “/whois channelname” to see who is on a channel before joining it). Finally, in any given encounter, it is also a matter of chance how skilled the other participants are in manipulating symbols in cyberspace–thus contest and alea are linked, here as elsewhere.

Virtual Conviviality: A New Form of Leisure for the Young and Computer Literate

Stylized online communication in real time is rapidly becoming a new form of leisure activity for the young, educated, and computer-literate. In the six years since this virtual party took place, thousands of channels on IRC have become stable gathering places for online conviviality and typed wordplay–each with its own “regulars.” These virtual pubs and cafes have become a new kind of “third place”–places lying somewhere between “home” and “work” where people can engage in relaxed conviviality (Oldenburg, 1989; Coate, 1992; Rheingold,1993; Kirshenblatt-Gimblett, 1996).

Many channels now have regular RL gatherings, which may be annual, biannual, monthly, and so on. For instance, regulars on #gb, a channel mainly populated by persons living in Great Britain, gather once a month in a London pub. #TahitiBar, a channel for Finns, holds RL beach parties; other festivities include an October Fest and a Winter Sports Day. Dozens of channels now have Home Pages on the World Wide Web, complete with photographs of regulars, reports of past group gatherings along with photographs of those present, news about upcoming RL social events, and so on. These developments can be viewed as a partial return to artful communication patterns characteristic of cultures without writing (Ong, 1982; Bolter, 1991; Lanham, 1993).

Expanding the Expressive Possibilities of IRC: the Hamnet Players

More or less unplanned online improvisation is the typical pattern on IRC. A group called the “Hamnet Players” took the expressive possibilities of the IRC software and the computer keyboard a step further, in an attempt to create virtual theater, or an online performance art forum. Founded by Stuart Harris, an Englishman now living in California and a computer professional with a good deal of theater experience, the Hamnet Players have scripted and performed three parodies thus far: “Hamnet”, “PCbeth–An IBM Clone,” and “An IRC Channel Named #Desire,” spoofs of Hamlet, Macbeth, and A Streetcar Named Desire.

The main source of humor in these partly scripted, partly improvised productions is the incongruous, outrageous juxtaposition of canonical plot and language with IRC and other computer jargon and various types of content from contemporary culture. Many of the phenomena discussed in this paper recur in even richer form than in the material analyzed here (Danet, 1994; Danet, et al., 1995).

The Light and Dark Sides of IRC

We should not be misled by the charmingly playful tenor of the log analyzed in this chapter. Cyberspace is by no means wholly benign, and IRC is no exception (cf. Reid, 1991; Dibbell, 1993). The synchronous modes of CMC can release aggressive, even shockingly malicious behavior, including sexual harassment and racism. Moreover, people can get themselves into fairly unpleasant RL trouble. One striking example, by now well known on the Net, is the case of the New York male psychiatrist who misled a number of co-participants on the CB channel of CompuServe into believing he was a woman (cf. van Gelder, 1986). Perhaps the most dramatic instance, to date, is a textual rape on LambdaMOO and the genuine shock and furor it generated (Dibbell, 1993; MacKinnon, this issue).

In short, along with the exhilaration, sheer fun, zany inventiveness, and sense of “flow” that participants enjoy on IRC, there is a fair amount of offensive aggressiveness, none of which is evident in the log we studied. In part the potential for it is built into the game, which includes such commands as /kill and /kick, which offer, respectively, the possibility of removing another person from the channel or from IRC altogether (one can use the /ignore command to prevent contributions from a certain person from appearing on one's own screen). The channel operator can use the command /kick to remove another person from a channel, but only IRC operators can actually/kill a person–“bumping” them altogether from IRC (participants can, if they want, immediately log back on, with another nick). Ironically, although these commands were originally created to control the behavior of aggressive persons, they are often used, in fact, as a means of aggression.

Another aspect of the “dark” side of IRC and other synchronous modes is its “addictiveness,” mentioned above. This is a special case of the more general addictiveness of interaction with computers (Turkle, 1984). Strictly speaking, one should be wary of labeling it as a psychological addiction in the full sense. Nevertheless, participants on IRC often speak of addiction not just metaphorically–admitting that they spend long hours at the computer at the expense of their physical and mental health, social life, studies or occupation.

It is not hard to argue that chat modes can be “counter-productive” for individuals, given that the majority of participants are, apparently students, many of whom may be neglecting their course work. The potential for subversiveness on IRC and other chat modes is similar to that of comics and graffiti, both in the diversion of the medium to other uses, and in the actual content produced (Estren, 1974; (Sabin, 1993; Abel and Buckley, 1977; Castleman, 1982). This was well illustrated by the log analyzed in this chapter, whose highlight was the simulation of an illegal activity. Players' delight could only have been enhanced by the knowledge that they were simulating something illegal.

We view “regulars” on IRC as pioneers experimenting with new forms of human expression. Contrary to the pessimism about the alienating effects of computers so often heard in the media, and to the position of technological determinism adopted by academics like Ong (1982) and Heim (1987), among others, our work identifies some of the ways that people domesticate, transform, and subvert these new technologies to make them their own. Meyer and Thomas (1990) argue that it is “this style of playful rebellion, irreverent subversion, and juxtaposition of fantasy with high-tech reality that impels us to interpret the computer underground as a postmodern culture” (Meyer and Thomas, 1990). The log we have analyzed in this chapter illustrates the postmodern nature of digital writing and speech play in chat-forms like IRC (Poster, 1990: 128; Meyer and Thomas, 1990; Reid, 1991).


  • 1

     We would like to thank Barbara Kirshenblatt-Gimblett, Tamar Katriel, Irit Katriel, and Hagai Katriel, for help and encouragement, and Fay Sudweeks, Sheizaf Rafaeli, and Margaret McLaughlin for feedback on earlier versions of this paper, and for the opportunity to be honorary members of ProjectH. Nachman Ben-Yehuda and Noit Meshorer provided information on terms associated with the culture of smoking marihuana. We owe a special debt to <Thunder>, the central figure at the “virtual party” analyzed in this chapter, for facilitating Ruedenberg's entrance into the world of IRC and for making this virtual party possible.

  • 2

     See Bechar-Israeli (1995), the corpus of 260 nicks.

  • 3

     The asterisks around Thunder's message are an indication from the IRC software that this was sent as a private message; that is, he has typed /msg lucia I am gonna shower soon. The asterisks around his nick would show up only on the screens of all other s logged onto the channel.

  • 4

     Lines marked with three asterisks are acknowledgements by the software of execution of commands; thus <Thunder> had typed, /invite lucia #weed.