That said, I don’t agree with the premise of Joshua Rothman’s piece at The New Yorker.
Should literary criticism be an art or a science? A surprising amount depends on the answer to that question. If you’re an English major, what should you study: the idiosyncratic group of writers who happen to interest you (art), or literary history and theory (science)? If you’re an English professor, how should you spend your time: producing “readings” of the literary works that you care about (art), or looking for the patterns that shape whole literary forms or periods (science)? Faced with this question, most people try to split the difference: if you relate to criticism as an art, you take a few theory classes; if you relate to it as a science, you put on bravura close readings. (Louis Menand’s article about Paul de Man, in this week’s magazine, quotes the critic Peter Brooks, who recalls how de Man could “sit in front of a text and just pluck magical things out of it.”) Almost no one, meanwhile, wants to answer the question definitively, because, for a critic, alternating between one’s artistic and scientific temperaments is fun—it’s like switching between the ocean and the sun at the beach.
This to me sounds like a typical “New Yorker” frame: start out with a short, provocative-sounding question, and then try to unravel the complications behind it. That’s fine, except, I don’t see the point of asking whether literary criticism should be an art or a science. It’s obviously not a science and never will be. There’s simply no way as a critic that you can form hypotheses and test them, that being the heart of the scientific method. Yes, sure, you can try to determine the structures beneath texts, movements, etc, but I’ve never seen a literary critic make a single falsifiable prediction, not even in the sense of how it’s done in social sciences like economics and political science. And of course theorists like Paul de Man did nothing of the sort . . . not even Roland Barthes, who’s probably much closer to a “literary critic/scientist” than de Man, got even close to science.
My sense is that people who care about books are much more interested in other questions, and I would even count Moretti himself in that group. Yes, Moretti obviously uses ostensibly scientific methods to look at texts, but so have a lot of critics. Structuralism came out of the science of linguistics, but I don’t think any of the structuralists were interested in being scientists. And nor to I get the sense that science as such is important to Moretti. My impression is more that computation is the best method for thinking about books in ways that fascinate him. There are so few (if any) mathematical equations in Moretti’s books. There aren’t even that many numbers. The man is a success because he knows how to use words. That makes him a writer, not a scientist.
Also, a bit of a nitpicky point, but this is a pretty ridiculous misreading of one of Moretti’s essays:
Yet Moretti has critics. They point out that, so far, the results of his investigations have been either wrong or underwhelming. (A typical Moretti finding is that, in eighteenth-century Britain, for instance, the titles of novels grew shorter as the market for novels grew larger—a fact that is “interesting” only in quotes.)
The essay that Rothman is alluding to is actually collected in Distant Reading, so I guess he must have read it, and if he did he must know that this is just the starting point of Moretti’s findings. (It’s actually one of my favorite essays in the book.) Saying that the above is the whole of Moretti’s finding would be like saying that Newton discovered that apples fall from trees. There is actually some very intelligent, criticism of Moretti’s work (some of it available right in Distant Reading, for instance), and Rothman, who’s clearly smarter than this, should have found some of it for his piece.