December 7, 2012
Noam Chomsky turns eighty-four today, more than a half century after he exploded onto the scene of linguistics, in in the late nineteen-fifties, as a young professor at M.I.T. His career began perhaps most notably with a book review that helped launch an entire field of linguistics (known as generative grammar) and laid waste to another (the behaviorist view of B. F. Skinner that then dominated psychology). From that moment forward, linguistics truly has never been the same.
He remains as influential, and divisive, as he was when Larissa MacFarquhar wrote a Profile of Chomsky in The New Yorker nearly a decade ago (“The Devil’s Accountant”). He has at least three new essays on linguistics coming out soon, and if time has slowed him down, it’s not by very much. A few months ago, I sent him a manuscript and he replied, with comments, in less than half an hour.
I can’t speak to his politics, for which he is equally well known. But since his earliest days, Chomsky’s scientific concerns have been as much about philosophy as linguistics. For most of us, words and sentences are tools for communicating. But for Chomsky, words and sentences are tools for understanding the nature and origins of knowledge. Chomsky sees himself, correctly, as continuing a conversation that goes back to Plato, especially the Meno dialogue, in which a slave boy is revealed by Socrates to know truths about geometry that he hadn’t realized he knew. Plato’s question was whether any of what we know about the world is innate as opposed to acquired through experience. For Chomsky, the interest in linguistics isn’t so much whether one language uses infinitives and another uses subjunctives but whether all languages are, at some level, deeply related and constrained by what Chomsky dubbed “universal grammar.”
That idea of universal grammar didn’t just change linguistics, it had repercussions for virtually every field that concerns the mind. In developmental psychology, for example, no idea has ever been as controversial, or as pivotal. How much do children know about language before they even begin to talk? Do they learn language simply by imitating their parents, or is there a built-in language-acquisition device (or what Steven Pinker called a “language instinct”)? Parallel questions soon arose in other aspects of cognitive development as well.
Part of Chomsky’s argument begins with the observation that language is infinite, with no end to the number of possible sentences that you can produce and comprehend. Take Chomsky’s sole entry in Bartlett’s Quotations, “colorless green ideas sleep furiously”; even if you don’t know what it means, you can still immediately apprehend that the sentence is grammatical, whereas “green furiously ideas colorless sleep” is not. Yet any given child’s experience is finite; we constantly encounter sentences that we have never heard or seen before. For Chomsky, the question of linguistics is how children bridge the gap between finite and infinite, from the finite input that they have heard to the infinity of what they can comprehend. In framing the problem in this way, Chomsky took an ancient and seemingly imponderable question—about nature versus nurture—and turned it into something that is actually testable.
One of Chomsky’s most remarkable traits is his willingness to change his own mind, like Bob Dylan suddenly going electric to the consternation of his early fans. Take for example the distinction that he once made between “deep structure” and “surface structure.” In its crudest form, the notion is that an active sentence (“John loved Mary”) and a passive sentence (“Mary was loved by John”) might seem superficially different; yet they have some important underlying commonality, both in meaning and in their representation in the brain. It’s a neat idea that makes certain very specific claims about how language is represented in our mind, and how sound relates to meaning; decades of linguistic work have been based on it. It also (somewhat uncharacteristically for Chomsky) makes for a perfect sound bite; there are plenty of people who know nothing about linguistics, but have the sense that what he was talking about was “deep versus shallow”; there was an even a Nobel Prize winner, Niels Jerne, who used the metaphor in his Nobel address about language and the immune system. Most people would have lived off a metaphor that good for the rest of their careers; Chomsky has spent the past twenty-five years arguing that he made a mistake. Although the basic metaphor is simple, the distinction between deep structure and surface structure required a great deal of behind-the-scenes technical examination in order to make it work with the complexities of different languages. In its place, Chomsky has recently been trying to develop a simpler, more elegant theory (known as the Minimalist Program) that encompasses the spirit of the original. (Not all of us are convinced about the success of that approach; my own view is that language is irreducibly messy, and that the elegance that Chomsky seeks will not be forthcoming.)
More recently, in a co-written 2002 paper, Chomsky seemed to open a door to a view that he’d long criticized: the idea that the “faculty of language,” as he called it, might draw on parts of the brain that weren’t specialized for language. Up to then, Chomsky had been known in part for idea called “the autonomy of syntax,” which, in crude terms, suggested that grammar was cognitively separate from other aspects of the mind (like our understanding of the world and our desire to eat pizza for dinner). I was so surprised by the dramatic shift that I wrote to him to ask. “A lot of people take [your new] paper to be a renouncing of your earlier arguments.” Was that really the case? His response, as immediate as ever, “As for my own views, they’ve of course evolved over the years. This conception of ‘renouncing beliefs’ is very odd, as if we’re in some kind of religious cult. I ‘renounce beliefs’ practically every time I think about the topics or find out what someone else is thinking.”
Nine academics out of ten never change their mind about anything; most (though there are salient exceptions, like Wittgenstein) lock into a position earlier in their careers and then defend it to the hilt. Chomsky, in contrast, has never stopped critiquing his own theories with the same vigor with which he has criticized others. For fifty years, his search for linguistic truth has been relentless.
At times, Chomsky can be maddening. He is not a particularly good listener, and he aims to win every argument, preferably by taking the most contrarian stance possible, like arguing that language evolved not for communication but for internal thought. The odder the arguments are, the more he seems to enjoy himself. A friend of mine was a student of Chomsky’s in the mid-seventies, and he recalls his strategy for meeting with Chomsky. Instead of the usual one-on-one meetings that most students have with their advisors, my friend would bring a classmate along. The two students would take turns debating Chomsky, one stepping in when the other ran out of things to say. Chomsky would win every argument (or at least never admit defeat), and the two would go home exhausted, but also elated. Each debate with Noam brought them a step closer to understanding the true nature of language and mind.
Chomsky can also be dismissive, in ways that still rankle—and stir people to action. He has, for example, never been kind to the field of semantics (which is about meaning) as opposed to the field of syntax (which is about grammar). He has also paid too little attention to the experimental data of psycholinguistics, which focusses on the dynamics of how people understand sentences as they unfold, millisecond by millisecond. Because Chomsky has a kind of first-mover advantage, his views in any area tend to dominate, even in areas, like the evolution of language, in which he is not an expert. When he makes a bad choice and unfairly dismisses an important question (like the way in which meaning is represented in the brain), linguists who aim to answer the questions that Chomsky neglects often face unnecessary resistance, as if the guru hasn’t sanctioned their mission. A good way for a young linguistics graduate student to make a name is to develop an intriguing idea that Chomsky mentions in one of his footnotes; it’s a riskier move to study something that Chomsky doesn’t find to be important.
A year and a half ago, at a symposium at M.I.T., Chomsky said, somewhat flippantly, that he didn’t think that statistical data mining approaches to language had contributed significantly to our understanding of how language worked. A more polite person would have put that sentiment more gently; a less influential person would simply have been ignored. Instead, one of the world’s busiest and most influential software engineers—Peter Norvig, the director of research at Google—wrote an eight thousand four hundred word blog post, extensively footnoted, critiquing Chomsky’s remark. It was two hundred or so paragraphs in response to a handful of Chomsky’s sentences. In practically any other context, it would be unseemly for a leading researcher at one of the world’s largest companies to spend so much effort picking on an off-the-cuff remark made by a man in his eighties. But I see it in a different way: two titans facing off, with Chomsky, as ever, defining the contest.
When I was a second-year graduate student at M.I.T., Chomsky taught a class on philosophy, which I was lucky enough to sit in on. The class itself was an event, almost a circus; people came from all over Boston, not just M.I.T. Grad students in my department (brain and cognitive science) would walk over en masse, facetiously chanting, as if it was a mantra, “Noam… Noam… Noam.” We would poke fun at Chomsky’s body language, and his teaching habits, as if we were above it all. But the truth is that few of us have ever been to a course that stimulating, before or since. To my eternal chagrin, I became exceptionally busy that semester (working on a language acquisition project that had just taken off), and had no choice but to drop the class when I realized I wouldn’t have time to write the required paper.
Sometimes I think of my whole career since as a kind of penance, still trying to wend my way through the philosophical stage that Noam had set. None of the questions Chomsky has posed has yet been fully answered, to his satisfaction or to anybody else’s, but no scholar of the mind has ever been more influential. Chomsky may not always have the right answers. But he has always had the wisdom to pose the right questions.
Gary Marcus, a professor of psychology at N.Y.U. and the author of “Guitar Zero: The Science of Becoming Musical At Any Age,” has written for newyorker.com about the facts and fictions of neuroscience; moral machines; and Ray Kurzweil’s new book.
Photograph by Philip Jones Griffiths/Magnum.
Read more: http://www.newyorker.com/online/blogs/newsdesk/2012/12/the-legacy-of-noam-chomsky.html#ixzz2ERQZhipF
No comments:
Post a Comment