Luddism, SF, and the Aesthetics of Electronic Fiction

The New York Review of Science Fiction May, 1994

I come from a generation of amphibians, born in Print but developing, early in life, the ability to breathe in the Electronic aether. So I can hear the current debate about the aesthetics of electronic fiction from both sides: The Linear who argue against the erosion of authorial voice, and younger avant readers, polysequentially perverse tadpoles digital from the egg, who ask why any one person has to write a fiction at all, and, come to think of it, why are there are so many words in these things.

Any attempt to talk about e-fiction has to deal squarely with this issue. And I am indebted to the fine groundwork laid by Sarah Smith ("Electronic Fictions: The State of the Art" (NYRSF #63) and Stuart Moulthrop (Electronic Fictions and "The Lost Game of Self" (NYRSF #64)). They are part of the group of writer/theorists who are shaping this medium both through their fiction and their analysis. In her essay, Smith elaborated the various metaphors current in e-fiction (electronic book, storyspace, architecture, game, film) and suggested some principles of any acceptable e-fiction environment (good graphical presentation, capability for reorganizing elements, link creation, and, most importantly, "a single e-fiction should be able to be written by both "author" and "reader.") Moulthrop, building out a scenario for e-fiction's development, argues that the future will see the corporate Information Superhighway commoditizing the contributions of "authors" to online e-fictions, while marginalizing (and charging for) "amateur" authors' participation.

Both Smith and Moulthrop agree on one thing: there will be a dichotomy between single authorship and the "consensual imagination" of the networked text. I believe that this (and the related issue of constructive versus exploratory fictions) is the central issue in electronic literature, in fact, these issues define electronic literature. And so, I'll try to approach these along several axes (technology, social construction, and literary theory) in an attempt to generate a first pass at what electronic text should be. If you're still around at that point, I'll say some things about how these texts should be, and what will make them worth reading and writing – particularly for critics and writers of science fiction.

GEnie meets ISD(jin)N
There is much that was inarguably true in Moulthrop's pesssimistic appraisal of the future of the medium. The "imagination industry" he described is indeed terrified at the prospect of "undermining the very idea of unified sequence." But this particular genie can't be stuffed back in the bottle; once you give people a taste of the interactive, the curve of rising expectations makes a read-write revolution inevitable.

In twenty years, channels will be extinct. The narrowband of online services like Genie and AOL will be subsumed into the larger datastream of the ISDN pipe. Or some other protocol; it's impossible to tell which particular technology is going to win, only that cable or the telcos will be delivering access to huge bandwidth, through which we will both access content and communicate.

Mitch Kapor argues, in a recent issue of Wired, for an open, Jeffersonian information highway, where,"one should be able to create and distribute programming, not just receive it." While such a result is far from certain, there are strong indications – including the one most important to corporate culture, market research – that people want interactivity. Kapor relates the cautionary tale of Prodigy, an online service based on a broadcast model, which was overwhelmed by the amount of message traffic among its subscribers. Even in the presence of "content," people will still make stuff up and share it with each other.

Many critics reacted negatively to Prodigy because of its implicit assumptions about information consumption. But, like remote controls and ATMs which have slowly taught whole hitherto unexposed segments of the American public what it means to interact, Prodigy should be regarded as the first derivative, the flexpoint of change, indicating what is to come. And coincidentally, like any good mind-expanding experience, it encourages people to try the harder stuff….

Telnetting to Cambodia
There is something paradoxical about the recent proliferation of books about the Internet, (including even an Internet "White Pages") as if fish could learn to walk, somehow, by reading about it. The Internet, that massive, messy, anarchic network which sprang organically from assumptions about decentralized communication after a nuclear exchange, is an ecology which must be experienced, in process, to be understood. (Perhaps even, I suspect, to be believed.)

In the shared worlds of the MUDS, and in the World Wide Web, a hypertext paradigm for tagging and accessing documents, we can see the future of reading and writing. And I find myself unable to believe that Moulthrop's predicted commoditization is possible, for while there is, for example, a "real" author doing pay-per-page humor (clari.feature.dave_barry) there is an equal traffic in unmoderated humor sharing, (rec.humor) edited slush humor (rec.humor.funny), and the over the top antics on, where people call dibs on postings from other boards and MIST them, inserting intertextual riffing á la Mystery Science Theatre 3000. And more to the point, though there are MUDS of a definite sharecrop character, based on many of the same series that dominate the monthly sf lists, there are also some genuine, weird, original realities.

Shoot Whatever Moves
I did a trade show gig recently, where I was unfortunately stuck next to some folks from a computer company demoing VR equipment: a big Silicon Graphics workstation, hand controller, and enormous video headgear. "It's like a ride," they would repeat, over and over, as the line of people ducked, acolytic, to have the helmet lowered over their face, "Shoot whatever moves." For me, working the spillover crowd and hearing this repeated for five hours straight, it became the mantra for VR. It is indeed a ride; it is the inevitable final extension of technology as entertainment, metastasizing out from the amusement parks that have kept it contained, trigger finger on the ready, on the prowl for helpless polygons.

The implications for the future of fiction are pretty bleak. Moulthrop, in a brilliant McLuhan-tinged analysis in the online journal Postmodern Culture, laid out his trademark paranoid take:

But what can this mean – talking books in homeboy jive? Street rap accompanied by Eliotic scholia? Nintendo with delusions of cinema? Or worse, could we be thinking of yet more industrial light and magic, the disneyverse of eyephones and datagloves where YOU (insert userName) are IN THE FANTASY?

—Moulthrop, Laws of Media

Well, yes, there's that. But J. Yellowlees Douglas has pointed out that the theoretical underpinning of modern fiction is "the genuine post-modern text rejecting the objective paradigm of reality as the great 'either/or' and embracing, instead, the 'and/and/and'." The postmodern, and sf in particular, drove the evolution of media in the direction of multilinear multivocality. What are technologies, after all, but the embodiment of our fundamental ideas about the universe? Can a century which has seen the death of absolute space and time really stuff people back into passive thrill-rides through the Matrix? Eventually, mustn't one run out of things to either kill or mate with (through the miracle of teledildonics!) surfacing issues from higher levels on the hierarchy of needs?

Almost Coming Home
There is, it can be argued, an unspoken, but nonethless dominant set of values underpinning much science fiction. Even though sf writers read and watch different instances, they still share a very real abstract sf mindspace. There may be disagreements about what is "good" in the field, but most will agree about "the field" itself, its general countours, its joys and terrors. Science fiction has achieved its hard-won respectability after a long adolescence in poverty and relative obscurity, publishing in paperback and circulating elaborate critical zines where writers alternately purr and hiss, (it sounds like they're killing each other, but they're just making more cats...) raising the stakes in a self-reinforcing definition of the field.

It should be no surprise that sf resists the electronic, coming as it does just as sf was beginning to enjoy some respectability in the mass market. Along comes this new form and threatens the whole painfully constructed edifice. When I was growing up, an sf title on the bestseller list was rare as proton decay; I can still remember being physically stunned the first time I saw Lucifer's Hammer on a rack in a supermarket.

And I appreciate the dues that were paid to make that happen, by generations of writers whose names are nowhere near as well known as the new kids making the big advances. But I'm arguing like a marketer: How should sf/f position itself to survive the inevitable coming of the digital? Should it adopt a reactionary stance, fighting to privilege "print" against digital fiction, or should it take a leap of faith into the matrix?

Within a generation, that's where the audience is going to be: out on the World Wide Web and its pay-per-page progeny. Not all of them, no. Not the casual reader who picks up an occasional bestseller from the airport racks, but the hardcore fans, the ones who read this stuff like it matters and who dream of being writers, they're all going to be off in the Net. Leaving them to the tender mercies of the hacks and hackers of the Hollywood-Silicon Valley axis seems like the worst mistake sf could ever make.

Because here is a ricorso any Golden Ager should recognize: suddenly, and again, science fiction has the opportunity to interact with and shape a radical new technology. In much the same way that the Space Age was made by a generation of kids who grew up reading about pulp rocketships, here is an opportunity for cyberspace to actually be shaped – not just indirectly, by anxious influence on kids in EE programs, but physically, textually and architecturally, by just those writers who have the most to give, and the skill to share that vision.

A retreat from this opportunity because it is challenging to notions of authorship, privilege, and individual craft is something for which history is unlikely to be forgiving. How many illuminated manuscripts do you have in your library?

We are indeed in another Incunabula, like the period of the early book where type designers struggled to reproduce the characteristics of handwriting in their letterforms. The phrase "expanded books," which Voyager uses to describe their product line of hypertext fiction and non-fiction, embodies precisely the kind of confused metaphor every new medium experiences in its cradle period. (How long did "horseless carriage" last?)

But the time has come to at least attempt some taxonomy, and perhaps a preliminary and provisional aesthetics, recognizing all the while that it is the nature of the medium to be protean, mercurial, and polymorphous.

Constructive or Exploratory
Michael Joyce, author of the e-fiction afternoon, and co-developer of the hypertext authoring package Storyspace, created, in a 1988 essay in Academic Computing, a widely quoted taxonomy of hypertext: exploratory and constructive. Exploratory hypertexts are typically adaptations of preexisting material (like encyclopedias and print novels) and are essentially objects of exploration. One can poke around in the nooks and crannies of the text, search, even make marginal notes, but these texts are closed systems. It has been argued that almost all material produced in hypertext thus far has this character. But Joyce went on to describe what he sees as the ideal form of the medium, constructive hypertext:

Constructive hypertext requires a capability to create, change, and recover particular encounters within a developing body of knowledge...constructive hypertexts are versions of what they are becoming, a structure for what does not yet exist.

Evolutionarily, we are naive realists and phenomenological pragmatists. Survival skills of the protohominid amounted to being able to parse experience symbolically with the highest degree of predictive power. Anticipating the return of absent tigers is a competitive advantage – and a pragmatically constructed a priori.

Writing is the extension and codification of this fundamental human need to forge symbolic connections. Until now, in order to render those connections apprehensible to the waking mind, we needed to tie them up in a narrative that "made sense," that "cohered with experience." Narrative is a tacit metaphor of causality, and our notions of causality have, until now, been determined by experiences of an almost completely linear character. And yet, we know that much of the universe behaves in ways which make our naive notions of causality appear, well, naive.

Electronic writing and hypertext systems introduce a fundamental change in our conceptualization of the text. We no longer have to deal with bound volumes comprised of linear sequences of expression, but with textual spaces which may contain a great number of alternative sequences. The literary imagination becomes spatialized, topographic.

—Jay David Bolter, Writing Space

I strongly suspect that the constructive hypertext, echoing as it does indeterminacy and superposition, represents the human creative attempt to enact, on the level of macroscopic reality, ideas of the quantum. And not just in the language of mathematics or ontology, but rather, in the visceral, storytelling way that separates us from the lower primates and sitcom writers.

And there is a real need for this; we live in a media ecology where the utility of print-based though has broken down. Gregory Bateson argues, in Steps Toward an Ecology of Mind, that linear thinking will always result in either the teleological fallacy or the myth of some supernatural controlling agency. Perhaps this is why we always seem to end up with gods or ghosts in the matrix; the linear text may snake unpredictably, but it always does end up meandering past Eve and Adam's place.

Authorship or Shared Text
Related to the issue of indeterminacy, but quite distinct from it, is the question of authorship. As an author of a hypermedia fiction, I recognize my personal vested interest in juggling this issue in my favor, but I think that there are larger, less blatantly personal arguments which have some merit.

The issue is this: why, in a network world where everyone is a theoretically equal participant, should there be a privileged position called "author." The MUDdwellers who spend a great deal of time in collaboratively created fictional environments wonder why they should have anything to do with archaic monolithic hypertexts. I have had experiences like the one Moulthrop describes (in fact, we were on the same panel once when it happened) where the digital radicals charge us with being counterrevolutionaries preserving the notion of authorship.

As an amphibian, I can reply, "Yeah, I am. Deal with it."

The spaces created by digital media are diverse. No one complains that digital video isn't a cellular phone. It's not an "all-one" kind of space yet – we haven't cranked the TeV sliders on the Accelerator of Babel up high enough to achieve Grand Reunification. The MUDS and the MOOs offer a paradigm for truly open, collaborative texts. And that's great. They're fun to write into, fun to hang out in, and some high, wonderful stuff gets done there. But to fail to see the limitations of such fiction – or to naively suppose that it has no limitations is to ignore reality.

Participatory writing, like jazz or improv, has a flow and logic, but it also has structural constraints. These do not derive from any weakness of the participants, or failure of vision, but from the medium itself. A jazz combo will not produce the Goldberg Variations; a team of actors will not improvise The Piano, a MUD will not collect, of an evening, and crank out Gravity's Rainbow.

Not to say that they should. That's not the point I'm arguing. Only that there are levels of complexity, linked to recursion and reflection upon emergent structure, that inhere in the process of developing fiction. And yes, there is something about having an individual author of these complex structures. There is a difference between closing the loop of author-reader feedback and submerging the notion of individual authorship into the socially constructed text. We've had such socially constructed texts for a while now. (How many screenwriters worked on The Flintstones?) And although the social space of the net will be unlike any we've ever seen before, this is not an unalloyed virtue: the difficulties of evolving a social system while trying to accomplish something (like collaborate on a fiction) will be thorny for a distributed team like the participants of a MUD.

On the other hand, only MUD fictions will be able to capture the essence of Net society, in the way that only a culture's native stories capture its inner truth. If we expect that this is the forerunner of mediated global culture, then perhaps we should crawl up on the shore and play in the MUD for a while, if only to understand what the sf market may look like in ten years.

Not many people know this, but all the characters in the Armchair, Daydrama, and Proscenium channels now write their own lines. It's a new gimmick, intended to promote formlessness, to combat sequentiality, and so on: the target-research gurus have established that this goes down a lot better with the homebound.

—Martin Amis, The Time Disease

Suggestions for a hypertext aesthetic

If we accept the rhizome as a metaphor for electronically mediated exchange, then hypertext is its apparent fulfillment, and Deleuze and Guattari's "approximate characteristics of the rhizome"-- principles of connection, heterogeneity, multiplicity, asignifying rupture, and cartography and decalcomania–may be seen as the principles of hypertextual design.

–Kathleen Burnett, Toward a Theory of Hypertextual Design Postmodern Culture, 1/93

In 1988 I had the opportunity to attend one of the first – if not the first – hypermedia writing workshops, at Humboldt State University. It was run by Rob Swigart (whose Portal is still one of the best sf e-fictions) and Vonda MacIntyre, and had the ambitious goal of producing a collaborative hypertext fiction in a little over a month. In the tremendously fertile and confusing weeks of the workshop, as the group struggled to learn software, adapt skills to the new medium, and formulate a plan for a large, shared-world story, we managed to articulate two main ideas, which, though I plead guilty to refining and elaborating, owe much to the pioneering writers of the Future Fiction workshop.

1) "This is not a game."

We thought this so important that we put it on the t-shirt, in real big letters. No parsers, no puzzles. (Anyone who has ever tried unsuccessfully to get the Babelfish in the Infocom Hitchhiker's Guide will probably agree…)

When Michael Joyce says, "games are the enemy," I take him to mean it in several senses. Pragmatically, writers can't compete with games. I could have stood next to that VR display for the rest of my natural life and not attracted as many people as they had lined up. And if fictions begin to exhibit gamelike characteristics, then they risk being judged by the standards of games ("How come there's all those words?") and dismissed as being "graphically inept," or worse yet, "old fashioned." So avoiding gamelike characteristics has a pragmatic component. But there's the larger, fictional issue as well.

Although one may view narrative as an "infinite game," in James Carse's sense of a game which we play in order to keep playing, rather than to win, this is not the way games are normally viewed. The payoff for "correct" play usually to win; to play "incorrectly" is to lose. This is very much at odds with what one might loosely call goals of fiction: exploration, insight, and the renewal of the perceived world through alterneity.

And though it is true that in my own fiction, Uncle Buddy's Phantom Funhouse, there is in fact a puzzle, this actually arises from adhering to our second workshop principle.

2) Hypertext fiction should be artifactual.

The best example of such artifactual hypertext is still Ursula LeGuin's Always Coming Home. I can recall many occasions at the workshop -- and in the time since -- where, when tempted to take the easy way out, to just hack out a piece of plot-moving narration, thinking about the elegant, embedded structure of LeGuin's world would still that impulse. LeGuin's wordcraft and her immense patience combine to grant the reader the authority to project a world from a rich variety of found objects: myth, stories, songs, maps, a dictionary. It is required reading for anyone who really wants to explore hypertext narrative.

Disembedded narrative is a print construct. In oral tales, there is always a narrator -- she's the one sitting there talking to you. When the audience vanishes into the virtual distance, the speaker's proxy, the narrating voice, takes over. Clever writers, from Sterne to Borges to Lem (now there's an infield for you) can problematize that authorial relationship, but it inheres in the medium. There has to be someone, a voice, who is telling the tale. Artifactual hypertext has no such requirement.

To be precise, in artifactual hypertext, the narrator disappears into the interface, with the logic of the hypertext becoming the "narration." Which is why, in cases where you are creating a fictional narrator who might be given to puzzles or games, such devices can be appropriate. But only within, and as aspects of, that narrating interface.

For those e-fictions which retain authorial narration, there is both a literal proxy and a programmatic double (in afternoon, the narrator speaks directly to the reader ("Do you want to hear about it?") but the fiction also skips and shunts, presumably under the author's control. And that's key -- we do presume control, and this control represents authorial craft. "Why did I get to this screen?" is one of the most common questions from first time readers of hypertexts. It has to be learned, in the same way that we learn to tell linear stories, that there is some connection, the character of which is an aspect of the fiction itself, not a stray programmatic artifact.

To those two main principles, I could add:

3) Be modally appropriate.

This is actually an extension of number two. But think about how whatever information you need to impart (character relationships, dialog, backstory) would actually be uncovered by a real person. Narrative that exists on the level of the character's reality is okay: stories that characters tell, letters containing their narratives, even print fiction that they read or refer to and hence is comprised within their reality is okay. Narrative that steps out on the veranda of colonial Text and bosses the characters around while sipping a mint julep had better start listening to the grumbling around the cook fire.

Yes, it means that the occasion of interaction itself can – and I believe should – be ingtegrated into the fiction. We should never sit down at the computer and look at POV shots walking down hallways; we are not walking down hallways. We are sitting at a computer. Use the medium. The computer interface can be amazingly plastic; so much of our information comes to us through screens that finding the right one is often a matter of relaxing and allowing the obvious to smack us, like a diamond bullet, in the forehead.

4) Be polysequential, parallel, polyauthorial.

In the same way that you can't get rid of linearity (We are, darn the luck, Beings-in-Time) you can't really get rid of authorship. Someone wrote this chunk of text, someone wrote this piece of code. What we are really trying to get rid of is the fetishization of the author. (Now wait a freaking minute, say the sf novelists in the audience. It took me long enought to goddamn get fetishized, and it doesn't pay that well as it is...) The best you can hope for, as an individual author, is to try and actualize a variety of voices in appropriate interfaces (and this is much tougher than just writing a multi-POV novel; one must think about how each of these people designs, what sorts of interfaces they are comfortable (or uncomfortable) with, how they compose for different media (a scanned in scrawled love note as opposed to a piece of e-mail to their lawyer), even what they look like. (A rich, robust interface should include modally appropriate images: not cartoons that we see, stalking though some polygonic dungeon, but images the way we find them in the world: blurry newsprint, somebody's driver's license, a family photo from the fourth of July.) One of the axioms of (screen)writing is "show, don't tell." In electronic ficiton, imagine communicating ideas about character and "plot" not through description, but through the normal modalities one employs in the Real world. You join a large corporation. How do you get a sense of the people from different departments? Maybe you see a memo under their signature. Check their names in the online directory and find out what projects they work on. Maybe you notice they post to a certain discussion board. You begin to develop a picture of who they are, based on the sorts of thigs they read – and write.

5) Traditional narrative modes are suspect.

Narrative arc and Freytag's triangle look distorted and dysfunctional in the non-Euclidean geometry of electronic space. Embedded narrative is pointillist, probabilist – dare I say it? – chaotic. Think of events as basins of attraction around which narrative units will accrete. (A plane crash, for example, will produce police reports, radio traffic, news photos, TV commentary, death certificates, manilla envelopes stuffed with watches, glasses, keys, wallets, small change, condoms, scraps of paper with phone numbers…well, you get the idea) Objects all full of meaning just waiting to be taken up into the web of authorial intent. A thing by itself, a thing-in-itself, has no meaning. But a thing taken as object by a perceiving intelligence….

Paging Captain Ludd

"Do it the way Balzac did it."

—Tom Wolfe arguing against fabulism with John Barth, Nightline, 12/15/89

Presumably as opposed to the way Roland Barthes un-did it in S/Z, dissecting Balzac like an episode of Quincy, hunting down the mechanism of textual infection. This is a challenge, particularly to this generation of writers: there is yet no payoff in writing e-fiction, and there won't be, for a while. There is no respect from the print community, and the truly digital wonder why one would bother with such practice at all. Perhaps the best justification comes from a larger perspective on the whole enterprise of writing:

Young writers take that most communal object, language, and perform on it that most individual act, creation. Years pass; and, doing much the same...older writers take an object now known to be, if not exactly private, certainly more idiosyncratic...and perform on it an action now known to involve so many communal facts...that the Romantic notion of 'individual artistic creation' becomes hugely shaky – if it has not, indeed, crumbled.

—Samuel R. Delaney, Neverýöna

Print writers, because they are "noticings" are always noticings of difference, and hence their texts are always, to a greater or lesser extent, Remembrances of Things Past. A truly new medium will require that the writer themself change; they must inhabit a medium which offers a new space for constructing fictions, but they must inhabit it with the power of observation and craft that comes from being, as science fiction writers consistently posit themselves, aliens in their own culture. Here is the intersect between water and aether, a niche which sf writers not only fit uniquely, but also one they helped to create.

So why the hesitation, on the part of the group which one would think to lead the charge on the shoreline? When I put on my new head, the one that lives full-time in the pure electronic aether, I wonder how they can be so blind. Of course, I know. There are still the colleges and the hierarchical corporations, ruled by the Invisible Hand and the payoff matrices of print, to be navigated before one can have sex and health insurance. We must, still, return to the Text to spawn.

But for all the difficulties and the risk, there is unique opportunity; for one must remember that we amphibians alone who can survive in both worlds; fish asphyxiate, and mammals can drown.

John McDaid is the author of the electronic sf novel Uncle Buddy's Phantom Funhouse. He can be reached at