The following analysis addresses several puzzling features of e-mail correspondence, in particular, the emotional volatility and intimate disclosure in these distinctively prolix texts. A "chatty" (Shapiro and Anderson 23), "conversational" (Duranti 65), quality about electronic correspondence encourages e-mailers, as many observe themselves doing, to "ramble on." This prolixity can be understood partly by acknowledging the primitive word processing built into many e-mail systems. Revision is difficult. E-messages are first drafts, usually verbose. Moreover, without the physical boundaries of paper (page length, weight, and shape), these "flickering characters,. . . evanescent as fireflies," (Shapiro and Anderson v) seem to spill over effortlessly from screen to screen.

However, more is going on in e-mail than the lifting of logistical constraints. Inhibitions, too, are circumvented. E-mailers value spontaneity. They do not want to revise. For instance, the Pine mailing system used at the University of California at Irvine and elsewhere provides an easy-to-use spellchecker that writers usually ignore, more so than with ordinary word-processed prose. The result is associative, open-ended discourse that prompts self-disclosure. Moreover, e-mailers' messages are often emotionally engaging, full of playing with personae, puns, humor, self-deprecation, ingratiation, or of the opposites--flames, insult, obscenity, vituperation, rhetoric that is "stylistically closer to Beavis and Butthead than to Pericles" (Chapman 15).

These features are counter-intuitive. The absence of online embodied social cues and clues should create in e-mailers a sense of vulnerability and make them pause. Online human behavior, as elsewhere in such uncertain social situations, should play it safe, shrink back before risking exposure. Diffidence and modest decorum would be the more appropriate, expected response, as in this listserv posting: "P.S. This is my first tentative step . . . and I'm a little nervous. It has a very first-date feel to it. I'll get over it, right?" (Pickett). So, too, when entering a crowded room at a party or arriving on a friend's doorsep for a visit, people usually check things out before insinuating themselves into conversations or into ongoing family life. When they do join, their initial behavior is usually quite conventional, conforming strongly to social norms. Indeed, social scientists claim that "Empirically, the assumption that CMC [computer-mediated communication] is characterized by a weakening of social norms seems to have little direct or independent support. In fact, it could be argued that an absence of social cues from other interacting individuals, together with the resulting uncertainty, forces people to resort to default social norms to guide their behavior" (Spears and Lea, quoted in Mantovani 56). However, e-mailers do not default to conventional behavior. Exactly the opposite happens in cyberspace, "A zone of spontaneous combustion, where users are invited to betray the smoldering world stifled beneath the conventions of writing and speech" (Wittig 21). What is going on?

This discursive online behavior can be seen in one way as "'warm-up,' off-topic chatter" (Friedman et al. in press), that is, hedges and other gestures of preventive maintenance or repair of the conversational site, for example, as in this listserv posting: "Blame it on the late hour (late for those of us who teach 8 am classes) -- or maybe the coyotes were howling a bit closer to my window than usual . . . . Mostly I'm just sounding off because I have dissertation writer's block. I don't think I'm usually this hostile sounding" (Wilkins). In this sense, as philosopher Richard Rorty explains, "the point . . . is to keep the conversation going rather than to find objective truth" (377) or to manage other tasks. But this understanding alone does not account for e-mailers' continuing willing participation in their electronic conversations with individual correspondents or on group listservs and on other public forums. Why, for them, are these conversations worth maintaining in the first place?

Another, reductive, perspective reveals this behavior as addictive. A taste of a little pleasure lures e-mailers to increase their online activity, thereby developing an electronic habit. Indeed, a sizable literature is emerging about electronic addiction paralleling the numbers of psychotherapists specializing in this field. Gary Andrew Poole, writing in open computing, goes so far as to claim blatantly that "What drugs were to the sixties, computers are to nineties" (112), and this view certainly does explain users' otherwise puzzling and fanatic loyalty to mechanical features of equipment and software, an allegiance which can lead, according to John C. Dvorak's analysis of "Interface Addiction" in P C Magazine, to "War." Uncertain about its source, no one disputes the power of the medium and the intense gratification it gives.

This dynamic and its power--the focus of this chapter--can also be explained by two other mutually reinforcing and convergent perceptions of the electronic exchange of online conversation. The first explanatory model considers electronic conversation as literary genre: "conversation can be seen to exhibit features that have been identified as quintessentially literary" (Tannen 154). If e-mail is a genre, then which one is it? Whatever else e-mail may be, it is mail, which Jacques Derrida, fascinated by mail, claims in one of his post cards "is no longer a simple metaphor, and is even, as the site of all transferences and all correspondences, the 'proper' possibility of every possible rhetoric" (65). This first model accordingly places the electronic conversation of e-mail squarely in the genre of epistolary art and accounts for its appeal that way: "successful conversation is an aethetic experience" (Tannen 152). It may not be particulary good art, but as "art" nonetheless, at least potentially, e-mail is a gift exchange deriving its power from "the erotic life of the imagination," to modify Lewis Hyde's subtitle slightly.

The second explanatory model considers e-mail as medium. With McLuhan it acknowledges that "Any technology tends to create a new human environment" (7). Cyberspace, created by the new computer technology is, above all, a new social space. This second model recognizes generative energy in the medium's destabilizing hybridity between print and orality; in a corresponding textual hybridity that this medium shapes according to a distinctive and powerful logic of conductive, dream-like association; and in hybrid, liminal subjects--writers and readers--who endow this gift exchange with meaning. Various literary and critical theorists corroborate these perceptions as does, indirectly, the work of several social scientists, notably Marcel Mauss, Mary Douglas, and Victor Turner.