[The Stuff of Thought: Language as a Window into Human Nature, Steven Pinker, Viking, 484 pages]
Speaking Your Mind
By John Derbyshire
BACK IN 1854, English mathematician George Boole published a book entitled An Investigation of the Laws of Thought. The objects of his inquiry, Boole tells us, were "the fundamental laws of those operations of the mind by which reasoning is performed." He sought to mathematize those laws and hoped, incidentally, to gather "some probable intimations concerning the nature and constitution of the human mind." Looking back on Boole's work a half-century later, Bertrand Russell sniffed, "If his book had really contained the laws of thought, it was curious that no-one should ever have thought in such a way before."
What Boole in fact succeeded in doing was creating symbolic logic, a branch of applied mathematics-the algebraization of deductive reasoning. True, there's much more to thought than just deductive reasoning, so Russell had a point. Still, the idea that our thoughts obey their own laws and that those laws can be worked out and expressed mathematically, like the laws of physics, is very appealing. It is more appealing now than ever before, as experimental neuroscience, fortified by new techniques for brain imaging and new understandings of the human genome (which has a construction template for the brain, as for every other organ), allows us to treat thought as a physiological process, like digestion, and observe it taking place and speculate about its evolutionary history.
Since we use language to express our thoughts, one obvious way to investigate the "laws of thought" is by studying language. This is Steven Tinker's approach in his new book, The Stuff of Thought: Language as a Window into Human Nature. Pinker reminds us, though, that this commonsensical point of view is controversial. Twentieth-century behavioral psychologists came close to asserting that thought does not exist and that only language, along with other forms of observable behavior, is worthy of study. Their spirit was carried forward by linguists Edward Sapir and Benjamin Whorf, who in 1956 famously hypothesized that the "laws of thought" are different for speakers of different languages. So pervasive were these ideas, Pinker tells us, that while writing this book, he had to stop telling people it was about "language and thought" because they all assumed it would be about how language shapes thought-"the only relation between the two that occurred to them."
Modern psycholinguistic theories can in fact be laid out in a spectrum. At the left end of the spectrum (using "left" here with Orwell's Newspeak in mind) is linguistic Determinism, the idea that if thoughts exist at all, they are entirely at the mercy of language. I don't think anyone believes the precisely opposite thing, that language has no influence on thinking at all, but Pinker's "conceptual semantics" is well to the right of center on the spectrum.
In The Stuff of Thought, he gives over a whole chapter to refuting three different current language-drives-thought theories: Extreme Nativism (nothing to do with immigration, Nativism is a term of art in cognitive science, referring to innate mental structures), Radical Pragmatism, and Linguistic Determinism. None of these can be fairly summarized in a sentence or two. Suffice it to say, Pinker is properly respectful of serious intellectual opponents, but succeeds in showing that each of these three theories has loaded onto a single true idea more weight than it can bear. At the end of this chapter-the most difficult but most rewarding in the book-Pinker nails his own theses to the church door. The parentheses are my own. Word meanings can vary across languages [the true idea over-loaded by Linguistic Determinism] because children assemble and fine-tune them from more elementary concepts. They can be precise [Extreme Nativism] because the concepts zero in on some aspects of reality and slough off the rest. …