Mathematician Timothy Gowers thinks groups are often smarter than individuals. In that light, last January he tried to solve a math problem no one had solved—a proof of the density Hales-Jewett theorem when k=3—in an online community. He posted the problem on his blog and welcomed people around the world to come together to solve it.
He said that many minds might solve a problem more quickly than a single one for the following three reasons (for his full explanation, see his blog post here):
1. Creativity is often fuelled by luck. If lots of people are working at once, it’s statistically more likely that one of them will get lucky.
2. Different people have different pools of information, and one person’s knowledge could add a missing piece to someone else's. An example: Consider a novelist who wants to write a book about the Iraq War. She might know a lot about constructing a good novel, but not much about Iraqi culture. If she hires a few Iraqi sociologists and reporters to fill her in, that's quicker and probably goes deeper than spending 30 years of her own studying the country before writing.
3. Different people have different styles for building creative ideas. Some people are great at brainstorming; others are good at lending organization to the brainstorm; others are good at critical editing. When we get these minds together, each can productively contribute from her own style.
Gowers posted some rules for contributors to follow--like being as clear as humanly possible--and then posted the math problem online. Within three months, after about 1,000 posts by a range of contributors--not all of them specialists in the same type of math--the problem was solved.
This idea--that many brains are quicker than one--might seem obvious. After all, we’re used to seeing collaboration fuel fields like math and science. It is precisely through group efforts like NASA, Apple, and research institutes that we’ve brought ideas to practical fruition in science.
But to a huge extent, we seem less willing to invest in the power of group work in other fields, like painting, music, and literature. And this reluctance might not exist for the most logical reasons. After all, remember that our first big stories were constructed by groups: Tribes made myths and wrote songs that evolved with group input through time. The Bible is a great collaborative book. But since we’ve moved through the Renaissance into a modern culture that values a division between private and public life, we’ve increasingly fostered the idea that genius comes from the individuality or independence of one person’s vision. We’ve particularly protected the cult of personality in fiction: We love to think of our brilliant writers as holed up in their rooms, working alone.
Some innovators have tried to shake us out of the cult of individuality in literature. For instance, Penguin launched A Million Penguins in 2007, in which they asked anyone online to contribute to a group novel. Anyone could help write and edit this novel for about three months, and then the site was locked, and the novel was pronounced finished. Other sites like Glypho.com and Protagonize.com similarly allow strangers across the world to come together to write joint novels. But in the Penguin experiment, at least, the book didn’t end up thrilling the people who started it, at least not on the level of aesthetics. “No, a community probably can't write a novel,” lamented editor Jon Elek on his blog when the Penguin book was done. He felt that the novel had a schizophrenic character: bright in spots but stitched together with a silly logic.
But I’m not convinced that a good online novel could not still be made.
Maybe online collaborative fiction has so far seemed loose or sloppy because--unlike when we’re doing online math--there's no single necessary endpoint for a novel, so when a contributor goes online to be an author, she is tempted to use her performance by drawing attention to her voice rather than doing the more subtle work of skillfully advancing plot and character with psychological accuracy. There is, after all, a psychological logic to what happens in novels, but it's not easily publicly identified.
Chekov said something like this: If a character in a story walks into a room and sees a gun on the table, there better be some logical follow-up, some psychological or structural implication that takes the other elements in the story into account. In a novel, it’s certainly not true that “anything goes.” A novelist needs to remember the events that have already happened and produce subsequent events that answer in some psychological or formal way to deepen interest in the scene.
So it’s hard to advance the “logic” in a novel, because it means spending a lot of time thinking about the psychological implications of what has come before. In contrast, contributors can focus on the purely logical implications of what’s come before in math. That’s an easier task for a group to approve or to vote down. That is, if someone contributes a false step to solving a math problem online, her work will be voted off for being counterproductive. It's comparatively hard to rate a creative writer. She can "contribute" without being called out (or even calling herself out) for a contribution that doesn’t deepen the scene.
Or maybe it’s this: Maybe math moves to a non-personal truth, and in contrast, literature moves to illuminate a particular, personal viewpoint. If this is true, then a billion viewpoints can average each other out in a productive way to help us reach an objective reality in math, but a billion viewpoints would just water each other down in fiction, where we need a quirky subjective perspective to come to life in all its coloring.
A different way to say a similar thing is to say that group work needs clear structure. In math, the math problem structures the work. In literature or art, we would go all over the place if we didn’t have a personal vision setting the agenda. There are examples showing this online. Kevan Davis, a British Web developer, tried to see how the visual arts might evolve through group online work by giving his readers two tasks. One was to develop a new typescript. People could participate in the activity by voting for what pixels in a small square to illuminate or darken, and through millions of hits, the letters came into formation. See them here, in various stages of formation and in their final edits. That worked out in a way, because we all have a paradigm for what an “A” or a “T” needs to look like. But the project didn’t work nearly as well when the goal became more subject to perspective. In one task, he asked people to draw a television (see here), but it didn’t really work. Neither did the face (here) or the goat (here). The goat got close to goatiness at 4,000 votes (here) but lost its form at 7,000 votes (here).
I wonder what you all think. Under what circumstances could group work foster something unique and beautiful in the arts or in literature? Right now I’m thinking of teams like the Cohen brothers, brothers who make movies together like Fargo and A Serious Man, or the former married team of filmmakers Charles Shyer and Nancy Meyers, who made hits like Private Benjamin and Father of the Bride. Maybe those couplings worked because the teammates really did share an almost singular subjective vision.
Does literature need a singular vision, or could many people come together to create one?