the new republic the new republic
the new republic
This Week
Privacy Policy
Contact TNR
the new republic Subscribe to TNR
the new republic
Subscribe to TNR the new republic
Get 4 Free Issues!
the new republic

Automatic Writing
by Keith Gessen

Only at TNR Online | Post date 05.22.01    

Just a few years before he became a merchant of schlock, Steven Spielberg directed the magnificent Close Encounters of the Third Kind, about a man, played by Richard Dreyfuss, who, after experiencing a run-in with aliens, begins to build volcano-shaped structures in his suburban living room out of soil and garbage. No one can understand what he is doing; his wife leaves him, taking the kids and the station wagon. Like Picasso, he persists. In the final scene, Dreyfuss is redeemed: As human scientists play a giant synthesizer to communicate with the UFO that's arrived in the vicinity of, as predicted, a dormant volcano, Dreyfuss is allowed to come aboard the alien spaceship.

Many of the pioneering digital writers gathered at the Digital Arts and Culture 2001 conference last month at Brown University had similar stories, minus the aliens: In the late 1980s, Deena Larsen attempted to construct a spatial epic in her room by linking poems, which were glued to model railroad houses, with thread and train tracks. Around the same time, the poet Robert Kendall was supporting himself by writing reviews of new presentation software; with the software already on his computer, he began to experiment with turning his poems into visual presentations, lugging the (desktop) computer to his readings to show animated text synchronized with music. Something was in the air. Computer poet and programmer Jim Rosenberg had begun writing "polylinear" poetry in the mid-1960s but had abandoned the format because it was meant, as it turned out, for the computer. "It's a little hard to explain," he told me. "But I knew this Internet thing would happen."

DAC2001, where over one hundred of the world's leading e-literati gathered to map and theorize about the future of hypertext (a text whose pages are connected via multiple links) and of electronic literature more generally, was a strange affair--all the academics were poststructuralists, all the poets computer programmers. Thomas Swiss, a 1978 graduate of the Iowa Writers' Workshop, seemed an emissary from another world with his dark blue collarless shirt, black suit, and easy verbal authority--the very model of a mainstream poet. "I did the poet thing," he said. "Got the grants. Published a couple of books with university presses. But then I looked around and asked myself: Is this where the cultural heat is?" He has been working on digital pieces for three years, and will begin an appointment in the Iowa English Department, teaching digital writing and English, in the fall. "I run into my old friends from that world, and they ask me have I published a book? I say no, but I've done six Web-based pieces. They're like, what? They certainly don't ask me where they can find them."

The poets, as any real estate agent will tell you, always precede the venture capitalists, and so too with the Internet. But whether literature has thus discovered a new neighborhood or just refurbished an old one is, as yet, unclear. It would be presumptuous, as the Internet boom recedes into memory, to announce that e-literature will go with it; but there is nevertheless reason to wonder if digital culture is ever going to produce a kind of writing to live up to its many manifestos. Or, if it does follow through on its promises, whether the result will be literature at all.


ecause the computer and the Internet are such social and economic phenomena, any discussion of electronic literature is fated to founder amid a confusion of terms. In the realm of distribution, for example, the promise of the Internet was at least as important to literature as any purely conceptual innovation: Whatever hidden threats it might pose to our immortal souls, the Web was potentially the most powerful samizdat tool in history--and just in time, too, as the 1990s witnessed an incredible consolidation of the publishing and communications industries. Unlike radio, film, and television--the great twentieth-century scourges of literature--the Internet was propelled by text, particularly in the early going. What is more, it was cool, literature finally plugging itself back into the zeitgeist: An entire apparatus of hype accompanied the Web's every move, each software upgrade like a call to arms; people were composing e-mails, coming alive to the possibilities of written English; text itself, which had been on a much-publicized farewell tour, was mounting a stunning comeback, migrating to every possible surface. And if The Matrix, a movie starring Keanu Reeves, could gain intellectual respectability by commenting on virtual reality, anything was possible. As the authors of the popular hypertext, The Unknown, wrote, a bit optimistically, in 1998: "We had built a literature, crammed it into a van, and we were heading for the Rockies."

There were also compelling literary-historical reasons to hop on board. Like all the great revolutions (the French, the Industrial, the Einsteinian, and so on), the digital one was shifting people's relations to their traditional communities, to space, to themselves in space. It was, even more potently, changing the way people read and wrote. As for what a computer literature ought to look like, this seemed clear enough: The Internet was capable--this was, in fact, its essence--of linking one document to another. According to such early visionaries as Ted Nelson, who coined the term "hypertext" in the 1960s, such linking could more accurately reflect the nature of consciousness than any previous human technology. The link could even replace the page as the standard unit of reading, and so when hypertext literature began to be produced in the '80s, its stated purpose was to extend and, indeed, to implement the insights of nonlinear fiction. Every nonsequential narrative in history was rallied to the cause--Tristram Shandy, Ulysses, Pale Fire, most of Borges, not to mention the metafictionists of the 1970s (John Barth, Donald Barthelme, Robert Coover). "All text is hypertext," the author and theorist Michael Joyce announced. "Never mind the bollocks," declared another author/theorist, Stuart Moulthrop: "Hypertext now." In 1992, Coover made the case for hypertext in an article in the The New York Times. Claiming, two years before the Mosaic browser flung open the doors of the World Wide Web, that hypertext and the computer would soon replace the book as we knew it, he argued that the link would overturn the (dictatorial) relationship between reader and writer. Hypertext, he promised, would be "interactive and polyvocal, favoring a plurality of discourses over definitive utterance and freeing the reader from domination by the author."

If this sounds familiar, it is: One of the most startling qualities of nearly any e-literature discussion is the preponderance of poststructuralist theory. After years of (self-)abuse, this academic language has found a home in cyberspace, where its esoteric terms seem, at times, to be literally true. Roland Barthes, especially, with his theory of a "writerly" text demanding the reader's input, appeared to be just what the early champions of e-literature needed. "The goal of literary work," Barthes had written, "is to make the reader no longer a consumer, but a producer of the text." Coover cited the "pre-hypertext but prescient Barthes" several times in his landmark Times article--the revolution had software, it had many megs of RAM, and now in Barthes it had a theorist-prophet. Coover began teaching a workshop on hypertext at Brown. All they had to do now was create a new way of writing.

This has proven difficult. In fact, nearly fifteen years after the first HyperCard program allowed pioneer hypertextualists to link different pages in their texts, the field of electronic literature is still inchoate, still uncertain of its direction, and still searching for legitimacy. Much of the work fits no genre category. What began as a textual form has sprouted audio accompaniments, photos, and video, so that, as Coover recently lamented, the text is often there merely as a caption. Many writers suffer from what one wit has styled the "anxiety of obsolescence," fearing that the software platform they painstakingly deploy for their texts will soon become antiquated. An even more significant problem is the persistent primitivism of computer usability: typefaces are ugly, and sustained attention to a computer screen is physically uncomfortable. "Cut these words," Emerson once said of Montaigne, "and they would bleed." Read too many words online, your eyes will bleed.

The vaunted link, meanwhile, for years the great hope for a computerized literature, is coming under increasing fire. The sentence quoted above from The Unknown--a perfectly nice sentence--is loaded down with two links. Here it is again: "We had built a literature, crammed it into a van, and we were heading for the Rockies." These links have the effect of destabilizing the sentence, collapsing its surface, and making it difficult to finish. And yet the raison d'Ítre of the Web, both in its utopian and capitalist manifestations, is the click; to resist the click is to resist the Web itself. And who would want to do a thing like that?

The link works; it's the theory of the link that now seems questionable. For if, in recruiting Barthes for his campaign against what he called the "tyranny of the line," Coover was not exactly misreading the slippery Gaul (hadn't Barthes spoken mock-ominously of "agents of the Sentence," as if they were Nicaraguan Contra assassins?), he was certainly proceeding with a literalness that Barthes would have found disconcerting. The problem with Coover's proposed application of Barthes' critical metaphor is that the reader, who can always leave a book, like an umbrella or the groceries, on a bus, has never been subject to the tyranny of anything except authorial whim; it was the author, wishing to scream and shout and cry but compelled, instead, to follow grammatical constructions, who felt imprisoned by the line. Freed from its pressures, many hypertext writers fall into a sort of recursive prose that is also recognizable from recent academic writing, circling around each object, repeating everything from different angles as if this could lend substance to a consciously insubstantial medium. And always the link lurks, a blue-fonted exit sign in every paragraph. It is an illusory freedom, and for the wrong person at that: it merely traps the reader in the author's associations instead of his own.


ut if the tyranny of the link is no better than the tyranny of the line, must e-fiction be abandoned and the vast frontier of cyberspace ceded to the vulgar profiteers? Not necessarily. Though the most popular examples of digital literature thus far, like The Grammatron and The Unknown, make promiscuous use of the link, the more promising work is linkless, or nearly so. The Jew's Daughter, written by Judd Morrissey and designed by Morrissey and Lori Talley, consists of a page on which one word is highlighted. If the cursor is passed over this word, a paragraph of text is replaced, but the rest of the page remains unmoved and can still be read sequentially. The meaning shifts, however, as sentences change their angles to one another and words find new contexts. The Jew's Daughter employs a cousin of the link, but its fluid page (which on certain browsers actually takes over your entire screen) manages what the linked page has always failed to do: It insists on its own urgency. Many other writers have taken to using one link per page--at the very end of the page. Another possibility is to display links only on a meta-page: one writer has posted her resume, with the links leading to brief episodes from the job or school in question. Robert Kendall, who teaches an online workshop in hypertext at the New School, uses a simple detective game format, with links, to lead the reader from poem to poem. Still others have discarded the link entirely: Rob Wittig is currently using email to serialize his novel, Blue Company, which novel consists of a series of shy emails to a female acquaintance from a marketing manager stuck in the 14th century. The novel, says Wittig, "is entirely and draconianly linear."

Wittig's experiment, though technically simple, is a perfect example of a distribution mechanism taking on the properties of literary innovation. A novel that reaches the reader via e-mail, on a daily basis, will be read differently, written differently, and evolve into a different animal. It could potentially enter the lives of its readers in the manner of the serialized novels of the nineteenth century (or of a television show), and could be discussed, even influenced, by those readers. It could free the novel from its unhappy existence as the least performative (and therefore the most product-like) of art forms. It could, in short, if adopted by more than a handful of writers, alter the nature of literary production. Considering the direction literary production has been tending (death), it will be difficult to argue with such a change.

That said, most of the work being done is an extension, though often an interesting one, of experiments begun in print. There are a number of poets working in the "concrete" tradition, which stresses the materiality of text. The computer, and especially Macromedia's new Flash animation software, allows letters in a line of poetry to morph into other letters, and into still other letters, demonstrating (at least in theory) the arbitrary nature of linguistic signs. Jim Rosenberg writes poetry inspired by John Cage's concept of a "note cluster"--the lines are piled atop one another and only emerge separately when you pass your cursor over them. These poets emerge from a tradition of highly theoretical poetry, to which the computer has only added further depth. DAC2001 presented the interesting experience of watching Rosenberg deliver a theoretical paper ("The Interactive Word Object") in the morning and then give an impassioned reading of his difficult poetry in the afternoon.

But if computer literature is often just encoding and extending innovations already made on paper, other developments actually bring to the fore certain changes in the nature of writing. There is the disconcerting fact, for example, that despite the Web's ability to transmit audio, it is the poets focusing on visual rather than sonic effects who are at the forefront. What this suggests is a remarkable shift, picked up by the subculture, away from verbal communication: Every reading at the DAC2001 conference had the work projected onto a large screen behind the author--placing the computer, like a marriage counselor, permanently in the middle of the literary exchange.

In this way, and especially as the narrow version of link-manic hypertext is abandoned in favor of experimental work that would, for example, generate its own texts according to certain prescribed rules, or develop a complex form or artificial intelligence to respond to the reader's movements through the text, electronic literature is beginning to function the way a true avant-garde should: less read than discussed, less important for its own achievements than for the ideas it might give more accessible writers, utterly uninterested in mainstream recognition or even readerly comfort (who, for example, actually wants to read Gertrude Stein?). Although the winner in the fiction category for the Electronic Literature Organization's first-ever awards ceremony, held on May 18 in New York, was a traditional hypertext, four of the other five nominees are not "written" in any traditional sense; one of them, "The Impermanence Agent," opens a window in the corner of your browser and over the course of a week incorporates into itself elements from Web sites you've visited--and begins to invade the text and images in your main browser as well.


n the early 1960s, the pre-hypertext but prescient Saul Bellow wrote a book about a bereft literature professor, Moses Herzog, who begins writing mental letters--first to friends and colleagues, then to the dead. "He wrote to Spinoza, Thoughts not causally connected were said by you to cause pain. I find that is indeed the case. Random association, when the intellect is passive, is a form of bondage. Or rather, every form of bondage is possible then. It may interest you to know that in the twentieth century random association is believed to yield up the deepest secrets of the psyche."

Why is it funny, and sad, that Herzog would compose letters to the dead? Because we know that, letters being so cumbersome to write, he would never actually write them; that if he wrote them, they would go unreceived. Herzog was lonely in a way that we, who have e-mail, are no longer lonely. There are no Herzogian letters anymore, there are only e-mails; it would seem that an entire literature of broken communication, from The Odyssey to Kafka's fables of messengers stuck in crowds, is rendered as obsolete as DOS.

This apparent loss has caused some consternation, and faulty logic, among defenders of the status quo. Valiant apologias for the traditional book have taken on the character of loyalty oaths for those, like postmodern elders and online book reviewers, who might otherwise seem suspect. The most eloquent and intelligent among them--and Sven Birkerts' Gutenberg Elegies remains the best plea for technoskepticism--are forced to fall back on the art-inducing loneliness of a Herzog as an argument against the superficial "connectedness" of the present. But Bellow was lamenting loneliness! It's like defending the modern workplace because it produced "Bartleby the Scrivener."

The truth is that Birkerts has nothing to fear. Whatever advances the Web has made in the sharing of information and the democratization of everyday life, it has not reduced by one centimeter the great mass of loneliness, and stupidity, and cruelty, against which literature registers its protest. The Web may have banished the peculiar Herzogian form of loneliness characteristic of the modernist novel, but this merely means we are now lonely in a different way. Electronic literature could begin to mine this change--all these changes--though first it may have to wait until the hypertext pioneers are taken, like Dreyfuss in the Spielberg film, up to the heavens.

And once they are gone it will have to decide whether it wants to be literature or television or a video game. An editor at the electronic publisher Eastgate recently responded to the latest print-world dismissal of e-literature by claiming that the new form, far from being the site of amateurs and charlatans, required real skill in "montage, juxtaposition, and other techniques and sensibilities derived not only from writing, but also from film, music, and visual art." If these sensibilities are meant to be harnessed in the work of forging new sentences, of responding to the computer screen and the millions of people with whom it connects us, then we had better start paying attention. But if the idea, as the market will certainly dictate, is to have a bigger, faster, brighter text, then it is only a matter of time before literature and the computer spit at one another and part. For we already have technicians adept at combining the many media. They work for a company called Dreamworks, and their movies are so powerful, their effects so dazzling, that people are willing to pay ten dollars and sit stupefied, entranced, for hours in their seats.

KEITH GESSEN writes for Dissent and Hermenaut.


TNR Online's Cyberspace Newsletter
Free news, links, updates, and recommendations.

Classic Reviews: The Movies and Reality
Virginia Woolf on Anna Karenina and other literary excursions into the new medium of film. (1926)

Online Music: Danses Russes
Adam Baer on two twenty-first-century stars.

Irony Scare
Benjamin Anastas asks, How did a literary device become a public enemy?

Online Films: Bringing out the Dead
Kenneth Branagh survives the Great Plague.

the new republic


the new republic
Home | Politics | Books & the Arts | Cyberspace
the new republic
Privacy Policy | Contact TNR | Subscriber Services
Copyright 2001, The New Republic
the new republic

the new republic