In the New York Times, Timothy Egan worries we stifle creativity in the search for expert quantitative analysis. He writes, in “Creativity vs. Quants“:
Here’s how John Lennon wrote “Nowhere Man,” as he recalled it in an interview that ran just before he was murdered in 1980: After working five hours trying to craft a song, he had nothing to show for it. “Then, ‘Nowhere Man’ came, words and music, the whole damn thing as I lay down.” …
We’ve bottled lust. We’ve refined political analysis so that nearly every election can be accurately forecast. And we’ve compressed the sum of education for an average American 17-year-old into the bloodless numbers of standardized test scores. What still eludes the captors of knowledge is creativity, even though colleges are trying to teach it, corporations are trying to own it, and Apple has a “creativity app.”
But perhaps because creativity remains so unquantifiable, it’s still getting shortchanged by educators, new journalistic ventures, Hollywood and the company that aspires to be the earth’s largest retailer, Amazon.com.
An original work, an aha! product or a fresh insight is rarely the result of precise calculation at one end producing genius at the other. You need messiness and magic, serendipity and insanity. Creativity comes from time off, and time out. There is no recipe for “Nowhere Man,” other than showing up, and then, maybe lying down.
From here he wanders into political analysis, a field where the “quants” have done rather well lately:
In relaunching his data-driven FiveThirtyEight website this week, Nate Silver took a swipe at old-school commentators. He recalled the famously off prediction of Peggy Noonan, who criticized people “too busy looking at data on paper” to pick up on the “vibrations” of a Mitt Romney victory in 2012. “It’s time for us to start making the news nerdier,” Silver wrote in his manifesto.
Data journalism has certainly done much to clean up the guesswork in a profession still struggling to find its way in the digital age. On election eve, it’s far better to look at the aggregate of all scientific polls than to listen to a pundit’s hunch. But numbers, as Silver himself acknowledged, are not everything in the information game. Satire, journalism’s underappreciated sibling, belongs to the creative realm. And there are no quants on the planet who could write Jonathan Swift’s “Modest Proposal,” or a decent episode of “The Daily Show.”
Nor could they produce an original film. Sure, they’ve tried. Most of Hollywood’s big budget, so-called tent-pole openings are the net result of exhaustive crunching of the elements of a hit. A robot can write a screenplay — about robots fighting one another! — that is just as effective at the box office as the fart-joke formula of an Adam Sandler movie. Before a major release, audiences are tested and polled, and producers fix and calibrate.
I think the reason the article seems to wander this way and that is that Egan is missing a key element in the analysis: craft. It’s very true that no number crunching data analyst can come up with a song like Nowhere Man. But the song didn’t just come to anyone as he was laying down, it came to someone with years of experience listening to, and crafting, songs. Creativity is a wonderful thing, but successful songwriters, playwrights, poets, video game designers and chefs, know technique – they have to. It is great to encourage children to experiment and explore, to instill a love of creativity. But they won’t turn into adults that make genuinely interesting creative works until they have learned technique. “The Daily Show” sketches cannot be written by someone who only understands how to analyze data, Egan is correct. But neither can they be written by somebody with no experience or sense of how television comedy works.
And in that way I think he misunderstands the “quants” as well. Universities turn out each year hundreds, thousands of graduate students with training in quantitative analysis – how to run regressions, how to interpret a time-series, when to use instrumental variables, and so on. But only some will succeed in producing work with wide interest. And that’s because there is a craft that accompanies quantitative analysis, the craft of knowing what are the interesting questions, what constitutes an interesting answer, and how to present the results so that readers see that it is in fact interesting and, yes, innovative. This craft is developed through observation, mentoring, practice. When I was co-editor of the Journal of Cultural Economics I received many submissions that had impeccably correct quantitative work, but that were simply not very interesting – the question not important, and the answer unlikely to influence any future research by other scholars.
The need to teach “creativity” has achieved a lot of buzz lately, as Egan notes. But is it misplaced? Should the emphasis rather be placed on technique, know-how, rather than some generally vague notion of creativity? Misleading to characterize the issue as one between creativity and the quants.
UPDATE: From Paul Krugman.