A colleague alerted me to this interesting article by Malcolm Gladwell (author of The Tipping Point) that looks at a bunch of boffins who think they’ve come up with a mathematical means of identifying hit movies. It contrasts their belief that there are rules that can identify potential hit movies with William Goldman’s famous dictum that “nobody knows anything,” and suggests there are two basic approaches to the idea of “rules” in art:
What Goldman was saying was a version of something that has long been argued about art: that there is no way of getting beyond one’s own impressions to arrive at some larger, objective truth. There are no rules to art, only the infinite variety of subjective experience. “Beauty is no quality in things themselves,” the eighteenth-century Scottish philosopher David Hume wrote. “It exists merely in the mind which contemplates them; and each mind perceives a different beauty.” Hume might as well have said that nobody knows anything.
But Hume had a Scottish counterpart, Lord Kames, and Lord Kames was equally convinced that traits like beauty, sublimity, and grandeur were indeed reducible to a rational system of rules and precepts. He devised principles of congruity, propriety, and perspicuity: an elevated subject, for instance, must be expressed in elevated language; sound and signification should be in concordance; a woman was most attractive when in distress; depicted misfortunes must never occur by chance. He genuinely thought that the superiority of Virgil’s hexameters to Horace’s could be demonstrated with Euclidean precision, and for every Hume, it seems, there has always been a Kames – someone arguing that if nobody knows anything it is only because nobody’s looking hard enough.
I’m inclined to think that Gladwell’s boffins are barking up the wrong tree, with way too many variables in their system to ever allow reliable calculations. Gladwell starts by talking about music, an area where it seems more persuasive that their approach might work: I can see how there might be particular patterns of beat or melody that just “sound right,” and which could be mathematically described. For example, Gladwell mentions that Gnarls Barkley’s “Crazy” scored super-high on their scoring system, and this seems plausible: if ever there was a song that sounds like it would max out a computer’s hit-single algorithm, it’s that one.
But I just don’t think it could ever work for movies, and Gladwell’s rather vague description of the buffs’ trial run of the system does nothing to change my view. Gladwell makes it sound impressive, talking of the application of the system (called Epagogix) to nine unreleased movies:
On three of the films – two of which were low-budget – the Epagogix estimates were way off. On the remaining six – including two of the studio’s biggest-budget productions – they correctly identified whether the film would make or lose money. On one film, the studio thought it had a picture that would make a good deal more than $100 million. Epagogix said $49 million. The movie made less than $40 million. On another, a big-budget picture, the team’s estimate came within $1.2 million of the final gross. On a number of films, they were surprisingly close. “They were basically within a few million,” a senior executive at the studio said. “It was shocking. It was kind of weird.”
Gladwell puts a reasonably positive spin on this, in the circumstances, but break this down and what do we have? A third were off completely. Of the remaining six, we’re told it only predicted whether the film made or lost money: but that allows a huge spread in terms of what counts as a successful prediction. (If a $100 million movie is predicted by Epagogix to make $250 million but actually makes $110 million, then the system can still be said to have correctly predicted it would break even). We have one out of the nine that we know the system got within $1.2 million, which is very close, but hardly compelling with eight other guesses that were obviously further off. Even for the film correctly predicted to underperform, Epagogix was still more than 20% off.
So I’m a sceptic about the ability to predict the gross of movies. But I’m not having a go at Gladwell, who it eventually becomes clear has much the same reservations as I do about the system. And I think that the idea that art is designed by rules that can be objectively analysed is true to a point. It should be noted that Gladwell conflates Goldman’s dictum, which is about predicting success, with the Hume / Kames debate, which is much more about formal aesthetic principles that underly art. There is some degree of overlap between a discussion about what is good and what will make money, but there’s also a a fair bit of difference too, and that difference is crucial. Predicting the latter involves, effectively, mastering an incredibly complex system of multiple variables (relating to both the qualities of artwork itself and the response of the wider population) and the even more impossible task of factoring in essentially random outside stimulus. How, for example, can a computer program intuit that its values for Tom Cruise’s box-office value are worthless because he jumped on Oprah Winfrey’s couch?
Yet describing the formal properties that make a piece of art good is a much more respectable exercise. We don’t baulk at all at the idea that there are formal principles that matter in visual arts like painting, photography, or architecture. We can instantly tell the difference between a well-composed image and a badly composed one, and some very simple principles (like the rule of thirds) can be persuasively demonstrated. And there are any number of similar rules of thumb for extended narrative forms such as the cinema: that a story should build to a climax, for example, or that the lead character should be taught a lesson during the course of the plot. These basic principles are so ingrained we generally don’t even think about them.
What is much more difficult, though, is in describing that extra something that makes a film great. You can tick all the boxes of the formulae that are taught in screenwriting manuals, and what you’ll get will be servicable, but it tends to be the intangibles that make for real greatness. This might be particular actors, sparkling dialogue, aspects of the directors technique, the innovative departure from the usual narrative rules, or something even more indefinable. A careful analysis of a completed film should be able to tease out both the formal elements that work, and also the more unpredictable elements that lifted it above the mundane. But before it’s made, you have only the formal stuff – that the plot is well constructed, say, or that the protagonist has a clear motivation – and that isn’t enough to tell whether the final product has that extra something.
Which might be the problem with the Epagogix exercise: only mediocrity can be scientifically described. This comes out in grimly humorous latter part of Gladwell’s article, in which the boffins give their suggestions for improving Sydney Pollack’s The Interpreter: essentially they decide that to maximise profits the film should be turned into The Bodyguard, a conclusion that upsets even those who devised the system. It’s a point that reminds me of Roald Dahl’s story “The Great Automatic Grammatizator,” which contemplates a machine that generates short stories and novels according to mathematical principles. In Dahl’s vision, the hand-crafted stories of human authors can’t compete with the mass-produced versions, and gradually authors sign contracts to turn their own names over to a literary agency for use on machine-made stories.
So we should probably hope that the Epagogix formula never really takes off. Certainly the prospect of studios lured to the sure cash of endless machine-tweaked Bodyguard clones brings to mind the final refrain of Dahl’s story, as the author tries to summon up the courage to not pursue profitable option: “Give us strength, Oh Lord, to let our children starve.”