An interesting debate flared up in the blogosphere in late March between Nobel-prize winning economist and New York Times columnist Paul Krugman and data-guru and political prognosticator Nate Silver. The impetus was the launch of Silver’s new website, fivethirtyeight.com, which he describes as a “data journalism organization” and vastly expands the scope of his work from politics to sports, health, entertainment, the environment and just about anything else you can imagine. Krugman found that Silver’s reach exceeded his grasp, at least in the early going, and the two had some testy back-and-forth, with various observers lining up on one side or the other.
I find it heartening that such media heavyweights are debating the proper role of data in public life. It is a testament to the growing interest in big data and its uses and abuses. But I am troubled as well.
Take a look at fivethirtyeight.com and note the fox logo. In the opening post, SIlver explained the logo thus: Our logo depicts a fox (we call him Fox No. 9) as an allusion to a phrase originally attributed to the Greek poet Archilochus: “The fox knows many things, but the hedgehog knows one big thing.” We take a pluralistic approach and we hope to contribute to your understanding of the news in a variety of ways.”
The link from the quote takes you to a excerpt of an essay by the scholar Isaiah Berlin, by far the most thorough discussion of the hedgehog and the fox. The whole essay is worth reading. Berlin uses the metaphor as a way into understanding history, particularly as seen through the work of Leo Tolstoy. Tolstoy, says Berlin, was a fox who desperately wanted to be a hedgehog; he saw the world in its countless details but sought, and in his old age believed he had found, a way to integrate them in some sort of vast, spiritual calculus.
Silver obviously knows Berlin’s argument, but what worries me is that he has not taken it seriously. I think he suffers from the same contradictions that Berlin explores in Tolstoy, but without acknowledging them. Silver clearly implies a value judgement; he picked the fox for his logo because that is preferable to being a hedgehog (though he did himself no favors when the first post on climate change misrepresented the data and required a quick rebuttal.) What he misses, as Krugman points out, is that data never tell their own story. The accumulation of disparate facts explains little. They need a model or a narrative or some other explanatory context. It seems likely, in fact, that Silver, like Tolstoy, is at odds with himself. All of Tolstoy’s heroes, writes Berlin, grope for some framework for understanding nature, but Tolstoy himself, “has not, do what he might, a vision of the whole; he is not, he is remote from being, a hedgehog; and what he sees is not the one, but always, with an ever growing minuteness, in all its teeming individuality, with an obsessive, inescapable, incorruptible, all penetrating lucidity which maddens him, the many.”
“An ever growing minuteness” seems like a cogent warning to the pursuit of data for its own sake. Conservation science is not yet at that point, but it would be an ultimately self-defeating strategy.