The cyborg on the page

In an excellent anthology of his short stories, the author Joe Haldeman describes an exercise that he used to give to his students at M.I.T., where he taught a course on science fiction for many years. Reading it, I found myself wishing—for just about the first time ever—that I could have taken that class. Here’s what Haldeman says:

For this assignment, I gave each student a random number between 8 and 188, which corresponded to page numbers in the excellent sourcebook The Science in Science Fiction, by Peter Nicholls, with David Langford and Brian Stableford. They had to come up with a story using that scientific device or principle. I further restricted them by saying they had to use a story structure from one of the stories in our textbook The Science Fiction Hall of Fame, edited by Robert Silverberg. The point of the assignment was partly to demonstrate that art thrives under restrictions. (It was also to give them a starting point; many had never written fiction before, and a blank page or screen is a terrible thing.)

Haldeman notes that he always does his own assignments, at least to demonstrate the concept for a couple of pages, and that in this case, he was given the word “cyborg” and the structure of Daniel Keyes’s “Flowers for Algernon.” The result was a solid short story, “More Than the Sum of His Parts,” which was later published in Playboy.

Not surprisingly, I love this idea, for reasons that longtime readers of this blog will probably be able to guess. Constraints, as Haldeman observes, are where fiction flourishes. This partially because of the aforementioned tyranny of the blank page: any starting point, even a totally random one, is better than nothing at all, and a premise that is generated by chance can be more stimulating than one of great personal significance. (When you’re trying to write about something important to you, you’re often too intimidated by the possibilities to start, while it’s easy to get started on a premise that has been handed to you for free. As Trump might put it, what have you got to lose?) There’s also the fact that a kind of synergy results when you pair a story structure with a concept: the dialogue between form and content yields ideas that neither one could have generated in isolation. Nearly every story I’ve ever written has resulted from a pairing of two or more notions, and I’ve developed a fairly reliable intuition about which combinations will be the most fruitful. But I haven’t really experimented with structure in the same way, which is why this exercise is so useful. When I brought it up with Haldeman, he said that the assignment is designed to make students think of form as a tool—or a toy—that can be explored and enjoyed independently of plot, which is a point subtle enough that a lot of writers, including me, never get around to playing with it. But when I take my scheduled break in a couple of months to work out a new story, I’m going to give it a try.

It’s revealing, too, that the story that Haldeman uses to illustrate his point is about a cyborg, since that’s what we’re really talking about here—a mixture of artificial and organic parts that theoretically forms a single viable organism. (In the actual story, it doesn’t turn out well.) Sometimes you start with a few components from off the shelf, or an assortment of discrete pieces of information, and once you start to combine them, they knit themselves together with newly grown tissue. In other cases, you begin with something more natural, like the chain of logical events that follow from a dramatic situation, and then add parts as needed. And incorporating a bit of randomness at an early stage results in solutions that never would have occurred to you otherwise. There’s a famous design exercise in which students are told to draw the human body in a state of movement, and then to construct an apparatus that will support the body in that position. At the end, the teacher points out that they’ve been designing furniture. That’s how writing works, too. Writers are frequently drawn to metaphors from carpentry, as when Gabriel García Marquez compares writing to making a table, or when José Saramago says that any chair he makes has to have four stable feet. But the result is more interesting when you don’t think in terms of making a table or a chair, but of creating a support system that will hold up the bodies you’ve set in motion. A cyborg carries his essential furniture with him at all times, stripped down to its purest functional form. And that’s also true of a story.

If every story is a cyborg, there’s also a range of approaches to how visible the parts should be. Some wear their artificial components openly, like Locutus of Borg, so that the result is a style in itself, while others keep their enhancements hidden. A book like Joyce’s Ulysses, with its endless experiments and pastiches in form, looks like a manufacturer’s catalog, or a fashion spread in which the same handful of models show off various possible outfits. I don’t recall offhand if Joyce assigned the various epic episodes, literary styles, and symbols to the chapters of Ulysses at random, but I’d like to believe that he did, simply because it’s such a pragmatic tool: “Let the bridge blow up,” Joyce once said, “provided I have got my troops across.” Sometimes the writer takes pleasure in making the joints between the pieces as invisible as possible, and sometimes it’s more fun to play up the artifice, or even to encourage the reader to spot the references—although a little of this goes a long way. It’s a matter of taste, which is another reason why the use of randomness at an early stage can be a good thing: the more detached you are from the big conceptual blocks of the plot, the more likely you are to make the right decisions when it comes to the details. If you’re the kind of writer who wants to crank out a story a week for a year, as Ray Bradbury once advised, Haldeman’s exercise is invaluable. (As Bradbury says: “I dare any young writer to write fifty-two stories that are all bad.”) I wouldn’t want to take the same approach for every story, since there comes a point at which the author himself starts to resemble a machine. But when used wisely, it’s a nice reminder that every story is more than the sum of its parts.

Disrupting the printed page

Robert Louis Stevenson says somewhere—although I haven’t been able to find the exact reference—that all the words on a well-written page should look more or less the same. Stevenson’s advice is generally taken as a warning against the use of ornate vocabulary that doesn’t fit the style of the rest of the line, but in my own work, I’ve also applied it to the level of paragraphs and chapters. Not every chapter should read the same way, of course: a climactic moment should feel different from a chapter primarily devoted to setting up information for a coming run of scenes, and a novel that was written in the same tone throughout would soon grow dull. When you glance quickly over the text without reading it, though, every page of my fiction looks pretty much like any other. Along with the many other arbitrary rules I follow, I’ve never used narrative devices like found documents or diagrams, I stick to one typeface, and I’ve done what I can to make the surface of the book look as seamless as possible, presumably on the theory that any visual device that calls attention to itself can only distract the reader from the story.

This may seem like something other than a matter of style, since it’s primarily visual, but I don’t know what else to call it: it affects the balance between dialogue and description, helps determine paragraph length, and has a subtle but very real influence on the narrative register of my stories. A book that alternates between many different tones often reflects this on the page: the stylistic shifts in a novel like Ulysses are visible at a glance. This is also true of popular fiction, which can alternate between long passages of rapid dialogue, extended sections of description, and strings of short paragraphs and sentence fragments for action scenes. Part of the reason I’ve tried to keep my novels visually consistent is a desire to see if I can get the same effect through the writing alone. In a way, it’s another constraint I’ve laid down for myself: I try to make the story’s events as colorful and interesting as I can while remaining within the same narrow visual range. It limits my range of options while forcing me to develop other skills to compensate, and thus far, I’ve been pleased by the result.

All the same, I sometimes get a little jealous of novelists who seem comfortable with radical typographical or visual experimentation. I’ve never managed to get through all of William H. Gass’s The Tunnel, for example, but it still occupies a treasured place in my home library: every few months, I’ll leaf through it, my eye caught by its oddly sinister flags, shifting fonts, and stretches of comic strip narrative, each of which stands like an island in the middle of the sea of Gass’s prose. The same is true of the works of such authors as John Barth and Georges Perec, not to mention House of Leaves. When I flip through a novel in a bookstore and come across a diagram or unexpected illustration, I’m always a little tickled, as if I’ve stumbled on a bonbon for browsers. Indeed, a striking typographic trick will often make me more likely to buy a book, or at least remember it: they’re like advertisements within the text for the author’s ingenuity, or cleverness, which may be one reason why I resist them in my own work, at least in the absence of any overwhelming reason to the contrary.

And while I wouldn’t rule out using graphic elements in my fiction in the future, I have a feeling that their presence would be as systematic as their absence has been so far. I’m most comfortable when operating within clearly defined rules, even if they’re only obvious to me, so any attempt at formal experimentation I’d make would probably be closer to something like Dictionary of the Khazars, my favorite novel of this kind, which embeds considerable typographic and visual invention within an attractively uniform surface. It’s a choice that can have unexpected consequences these days, when it’s likely that many of my books will be read on Kindle or a similar format over which I have less control: few, if any, of the novels I’ve mentioned above would survive that transition. When all of your sentences look more or less the same, you don’t need to worry about how they’ll appear in print, and I’ve been glad to leave that aspect of my novels to professionals who know what they’re doing. That way, I can focus on trying to put variety into the story itself, regardless of how it’s laid out on the page—which is more than hard enough as it is.

Thinking in pictures

Last weekend, at the Printers Row Lit Fest in Chicago, I picked up a copy of Life: A User’s Manual by Georges Perec, a novel I’d been meaning to read for a long time. I’d been interested in Perec ever since reading about his work in Douglas Hofstadter’s Le Ton Beau de Marot, and while I’ve only begun dipping into Life, I’m already intrigued by the riches on display. As described in greater detail here, Life is an ambitious experimental novel, centered on a fictional apartment block in Paris, that Perec constructed using a system designed to generate a random list of items (an activity, a position of the body, a writer, even the number of pages) for each chapter, which he then had to incorporate into the narrative. The result, as Perec put it, is a “machine for inspiring stories.” Even apart from the merits of the novel itself, I find this premise tremendously exciting.

Regular readers of this blog know that one of my ongoing obsessions is finding new ways to insert randomness and constraints into the writing process. Writing a novel, at least as I tend to approach it, is such a left-brained activity that it’s necessary to create opportunities for the right brain to participate. Sometimes this happens by accident—while shaving, for example. But there are also ways of approaching randomness more deliberately. I’ve published stories based on juxtapositions of two unrelated articles from science magazines, used random selections from Shakespeare and the I Ching to guide chapters (although I’ve mostly dropped the latter, despite the fun of throwing the coins), and used mind maps to bind all these elements together. And I’m looking forward to applying some of Perec’s techniques to my own work, although probably in a much more limited sense.

Recently, I’ve also discovered another approach that might prove useful. In Origins of Genius (which, in case you haven’t noticed already, is one of the most stimulating books on creativity I’ve read in a long time), Dean Simonton describes a fascinating experiment by psychiatrist Albert Rothenberg:

[Rothenberg] and a colleague began by making up a set of visual stimuli that involved the superimposition of visual images. For example, one contained a photograph of an empty French four-poster bed placed in a period room superimposed over a group of soldiers in combat who were taking cover behind a tank. These highly incongruous homospatial images were then shown to writers and to artists, the latter including individuals selected in a national competition by faculty at the Yale School of Art. The writers were instructed to create new metaphors inspired by the stimuli, while the artists were instructed to make pastel drawings. In comparison with the control group (e.g., subjects who saw the images only separately), individuals exposed to these visual juxtapositions of unrelated images generated more creative products, as judged by independent raters.

In other words, juxtapositions of two unrelated concepts often result in ideas that would not have arisen from considering the two concepts separately, which only confirms one of my most basic convictions about the creative process.

What I find particularly interesting about Rothenberg’s experiment, though, is that the stimuli consisted of images, rather than words, which seems like an especially promising way of encouraging nonverbal, creative thought. With that in mind, I’ve started to incorporate a similar method into my own work, using images randomly chosen from three books that seem ideally suited for such an approach: Phaidon’s chaming little volumes The Art Book, The Photo Book, and The 20th Century Art Book. Each book consists of representative works by five hundred artists, one work to a page, arranged in alphabetical order—an arbitrary system that already lends itself to startling juxtapositions. For instance, in The Photo Book, by an accident of the alphabet, “A Sea of Steps” by Frederick H. Evans appears across from “Washroom in the Dog Run” by Walker Evans, exposing their haunting visual similarities. Two images, taken together, yielding a meaning that neither would have apart—that’s what art is all about, and why I’m looking forward to thinking more with pictures.