Stuart Moulthrop creates “stories” that consist of very many small text files on one or more computer, all linked to each other like pages on the internet, all telling part of some larger “story.” This new art form is called hypertext fiction. A work of hypertext fiction may be on the internet, on a CD-ROM or other digital media. The following interview was conducted by exchanging emails in the winter and spring of this year.
RT: I’d like to talk to you about hypertext fiction and how it relates to traditional fiction and other arts, but also more generally about new media and the internet.
First, in traditional, linear fashion, how about some background. You teach web design at the University of Baltimore, you have presented numerous papers on the relation of hypertext to literary theory, you have authored numerous hypertext fictions, notably The Color of Television and Heigirascope, and, with Adrienne Wortzel, you judged the recent New York University Press Prize for Hypertext (was that the first of its kind?). You are a literary person who works primarily with computers. How did you come to these particular coordinates? Was there a time of deciding you wanted to write? What was the first computer you owned? Does asking more than one question at a time help break down linearity?
SM: >How did you come to these particular coordinates?
It’s hard to answer this question because I feel I’m still maneuvering, so the coordinates are somewhat momentary or transitional. I’ve been interested in contemporary fiction and various kinds of experimental writing since I was a teenager (mid seventies). My graduate training is in literary criticism and theory, especially narrative theory. While I was in grad school, a couple of articles by Mark Edmundson (then a classmate) got me re reading McLuhan, which in turn made me think about media more generally. It was about this time that the Macintosh came along, with Storyspace and HyperCard hot on its heels. There’ve been a few moments of professional risk taking (e.g., deciding to write Victory Garden instead of an academic book), but things have worked out very nicely since the advent of the Web in the early nineties.
>Was there a time of deciding you wanted to write?
I’ve wanted to tell stories since I learned to read. When I was much younger I wanted to be Jack Kirby (who counts in my book as something more than a “writer”). Actually I’d still like to be Jack Kirby, but at my age there’s no point aspiring to genius.
I love science fiction and have sent a few miserable stories to editors who wisely slushed them. (Compare my colleague John McDaid, more serious in all things, who’s a prizewinning SF writer.) I wrote a bunch of stories as an undergraduate, one or two almost passable, scribbled notes toward a bunch of terrible novels, then was fortunately distracted by graduate school.
At the end of the eighties, shortly after HyperCard, I got the only sabbatical I’ll probably ever have, the result of which was a draft book on postmodern fiction and a fragmentary hypertext. After a year or two the hypertext seemed more interesting than the book, so I chucked my career prospects and wrote Victory Garden. The rest is work in progress.
>What was the first computer you owned?
Answering this question makes me feel like a total geek, but I’ll wear my pocket protector with pride. It was a Commodore VIC 20. I sold it after a few months to finance a Commodore 64, which is boxed up with a bunch of old Macs in our basement.
>Does asking more than one question at a time help break down linearity?
Not the way *I* answer questions…
RT: This also could have been my first question: what is hypertext and what’s it good for?
SM: >What is hypertext…
The canonical definition, given by Theodor H. (“Ted”) Nelson about 1964, is “non sequential writing,” which I like to understand as something more like “polysequential discourse.” Hypertext doesn’t abolish sequential relationships; on the contrary, it makes them all the more significant since sequence emerges through a system of articulations, or as Nelson later said, through “free and knowing user movement.” Also, while much hypertext is still word based, there have been interesting moves toward more graphic or “spatial” conceptions, and toward polysequential form in
time based media like sound and video. These days I’m more comfortable with the term “hypermedia.”
Let me also offer a critical definition: Hypertext is NOT identical with the World Wide Web or with most things produced using Hypertext Markup Language (HTML). I’ve been known to call HTML “Hypertext, More or Less.”
That’s not to say the Web isn’t a legitimate implementation of Nelson’s ideas, just that it’s not the only one, or even the most interesting one a point the inventor of the World Wide Web, Tim Berners Lee, makes very clear.
In a nutshell, the Web tends to enforce a limited conception of the link.
It basically has a single predicate: “go there.” Other hypertext systems
allow more sophisticated logics of transition and relationship. In Storyspace, for instance, a writer can change the behavior of a link depending on what the reader has read previously. Instead of just asking, “where do you want to go today?” we can add, “and where are you coming from?” Other, even more advanced systems complicate this process even further.
>and what’s [hypertext] good for?
There are two answers to this question, one concerning instrumentalities and the other concerning imagination, or art.
The instrumental answer is easy: Remember what life was like before Amazon.com. Today the World Wide Web confers small but substantial benefits on a growing number of people. Folks like Nelson and Douglas Engelbart (who built the first practical hypertext system in the late sixties) have argued for decades that these benefits will scale up into a broader literate franchise, maybe even a true commonwealth of information. People still dismiss this as hokum, but it gets harder and harder to do
that, even as the Fortune 500 open their “portals” all around the Web. Once you’re through those doors, after all, links can lead anywhere.
The case for hypertext in art or literature demands more subtlety. I’ve suggested in a few scholarly essays that hypertextual art is a complicating practice, disruptive and disorienting to be sure, but part of a larger cognitive program that represents a legitimate human response to information technology. For some reason, critics tend to overlook the second part of this argument. The quote comes out: “hypertext is a
complicating [and] disruptive practice” period. So I move to a simpler, more adversarial stance.
If you define art (speciously, in my view) as a filtering operation that separates dross or “noise” from worthy, definitive statements, then hypertext is a wrong turn, or to follow the etymology, a *perversion* Sven Birkerts refers to exclusive, linear sequence as “the missionary position of reading.” Things happen differently when you allow a range of movement. The business might not follow a predictable track. Hypertext (to return to the subject) may entail recursion and redundancies. As Mark
Bernstein says, “repetition in hypertext is not necessarily a vice.” Or as Michael Joyce opens his classic afternoon: “There is no simple way to say this.” Hypertext is, well… kinky.
Birkerts describes his critique of hypertext as part of a broader “war” in defense of an ink on paper literary tradition. This brings to mind the “culture war” we’ve been having since about 1980 and reminds me which side I fight for. Hypertextual art stands in opposition to any singular view of the world. To be sure, partisan interest can never substitute for talent, craft, and hard work, but there are plenty of hypertextual projects
works by Judy Malloy, Michael Joyce, Ed Falco, Jim Rosenberg, Shelley Jackson, among others that live up to those standards. These works present an important alternative to the narrow, print bound tradition of the Gutenberg elegists. A very smart Norwegian critic calls the sort of reading we do in hypertext “ergodic,” which is a Greekified way to say “pathfinding.” As the world gets more complicated, this can be a crucial skill.
What’s hypertext (fiction) good for? In a single proposition: it inscribes
RT: People unfamiliar with the medium might think of, in addition to the web, popular CD ROM games as similar to hypertext. Particularly strategy games, such as Gettysburg, or fanciful puzzle games, such as Myst, involve a sort of story which has options. Are these related to hypertext? Could they become more related?
SM: I carry in my hip pocket a column by the doyenne of American book critics, Michiko Kakutani, in which she predicts that hypertext writers can produce at best “Myst or Warcraft II as re imagined by Robbe Grillet.” You can see why they gave this woman the Pulitzer! This is absolutely correct: At the moment, and perhaps for the long haul, the most important prospects in polysequential art lie with things we call “computer games” and I consider the work of Rand and Robyn Miller, from The Manhole (1988) to the marvelous Riven (1997) particularly groundbreaking. The Miller
brothers are the Lumières of these dim days. They’re inventing a new way to tell stories. Cross Myst with In the Labyrinth or with Mumbo Jumbo or The Crying of Lot 49 or Always Coming Home and you might have something really significant.
Of course, La Michiko wouldn’t agree. Leading book reviewers still regard
anything that doesn’t look like a 19th century novel as suspicious, sub cultural, infra dig. You get the feeling they’re not at all sure about cinema and the telegraph. Why should this amaze anyone? It’s their business, after all, to award prizes and keep down hoi polloi. Once this made me angry, but not so much anymore as the poet Costello says, “I used to be disgusted. Now I try to be amused.” Much of this amusement is self directed, largely because someone asked me recently to judge a hypertext prize, which taught me a few things about practical criticism. To begin with, anything you say about something as complex as literature, let alone hypertext, will seem wrong to most people. Critics are the straight men and women in history’s variety act. What dolts! Even when we’re right, we’re wrong. Culture wouldn’t be half so funny
So, a rash prediction: Within 20 years (which means hopefully I’ll get to see it), someone who probably hasn’t yet been born will produce an enchanting, complex, spatial graphic text that weaves a deeply moving network of stories and becomes indispensable to millions of people, not as throwaway amusement but in the way of “great” films like Gone With the Wind, Casablanca, or (ahem) Star Wars, and also of “great” novels like Pilgrim’s Progress and The Sorrows of Young Werther, and other “great” examples of name your form. I have no idea what the medium or conventions
of this supreme fiction will be, but I hereby predict, putting aside my straight man role, that its cultural DNA will contain sequences derived both from today’s games and, to a lesser extent, from hypertext fiction.
Hoo ha! Tell us another one.
RT: How do you start on a piece of hypermedia? How, generally, does work progress with a technically simple piece that just links from one text segment to other text segments?
SM: >How do you start on a piece of hypermedia?
Once upon a time (for instance, with Victory Garden) I spent many weeks making plans, flow charts, schemata, mock ups, and prototypes, trying to come up with the best marriage of form, function, narrative, and purpose, and going seriously crazy in the process. I’m pretty sure I spent as much time planning Victory Garden as I did writing it. Eventually I threw out all of the early scaffolding, after I had convinced myself that I had a minimal sustainable structure that would let the work write itself. For Victory Garden this structure consisted of John McDaid’s War Night videotapes plus the plot line of “The Garden of Forking Paths” structures that now don’t seem particularly central, but which gave me the strength to jump.
These days I like to think I’m more spontaneous, but in fact I spend so much time tinkering with interfaces that I’m always at that stage of pointless experiment. Tinkering can be important.
>How, generally, does work progress w/ a technically simple piece that just links >from one text segment to other text segments?
As Robert Coover wisely says, every attempt to get off the Line will eventually bring us back to it. Writing is never more linear than when it’s hypertextual, because hypertext is all about *multiple* lines, or threads if you prefer. Victory Garden and Hegirascope both have lots of very linear threads, though I think the weaving has gotten more complex in the later work. The new project (“Reagan Library”) isn’t so linear, since it’s largely a work about space, or about losing the thread of discourse. But I still believe in lines, somewhat (to compare great things to small) the way jazz players do.
Polylinear writing has a simple but powerful charm: Lines can go across as well as along the flow of a reading. So as you start laying out a storyline moment by moment (in Victory Garden these “moments” were defined by the amount of text that would fit into the “classic” Macintosh 9 inch window) you find yourself sensing connections to other lines, and sometimes articulations that involve several lines at once. It’s in this way that the text begins to write itself.
RT: Three small things about readers, or “users” storming the hypermedia studio:
My main experience w/ CD ROM games has been with Douglas Adam’s loopy
Starship Titanic. I loved the elements of story and media, but being a bear of little brain, the puzzle elements were often too hard for me. So I went a web site with “spoilers” answers to the puzzles. The guy who’s site it was said he’d hacked the executable to the game and found a list of key words for prompting responses from the game’s robots (who talk back to players) which he printed. This enhanced my enjoyment of the game.
Last I looked, the Online Pynchon Pages were set up so that by chopping off all but the first part of the site’s address you could see a full index of all the pages on the site useful, since the organization of the entries and the way they are linked to each other is (I bet deliberately) arcane.
3)Not too drastic
Sometimes when your Hegirascope loads the next segment too fast (I bet deliberately), I hit my “back” button to finish what I was reading.
Any thoughts on hacking a work of art (not w/ a knife)?
SM: >a list of key words for prompting responses from the game’s robots (who
>talk back to players) which he printed. This enhanced my enjoyment of
I’ve had a similar experience (though less ambitious) with Riven. After spending a solid and joyful week poking around in the Miller Brothers’ amazing universe, I gave in to what had become an unbearable curiosity, opened up the Hints and Solutions book, and took the complete tour. This did put me in a position where I don’t want to re play the game, though I wouldn’t say it’s diminished my respect or admiration, or my sense of huge enjoyment.
>Last I looked, the Online Pynchon Pages were set up so that by chopping off all but >the first part of the site’s address you could see a full index of all the pages on the >site useful, since the organization of >the entries and the way they are linked to each >other is (I bet deliberately) arcane.
You can do this with most Web sites (though interestingly enough, not mine at the moment). The better servers can be configured to display a directory structure if there’s no appropriate head file. Security freaks find this horrifying, but I approve of the practice (and will implement it when we migrate to our new server).
>3)Not too drastic
>Sometimes when your Hegirascope loads the next segment too fast (I bet >deliberately), I hit my “back” button to finish what I was reading.
That’s okay with me. It’s part of the scheme.
>Any thoughts on hacking a work of art (not w/ a knife)?
I deeply approve of all the things you’ve mentioned here. Though I don’t insist that cybertexts have to be hackable, or that readers of these texts have to hack, nonetheless one possible argument for this artform is that it encourages people to “jump outside the game” or recognize the contingency of media constraints.