home     authors     titles     dates     links     about

the stuff of thought

21 may 2008

Take a pickup truck and some gravel. You can load gravel into the truck. You can load the truck with gravel. You can pour gravel into the truck, or you can fill the truck with gravel. But in the English language, at any rate, you can't pour the truck with gravel. And you can't fill gravel into the truck.

What's up with that? Is the impossibility of pouring a truck with gravel, or filling gravel into the truck, a feature of gravel, trucks, or the syntax of the English language? In The Stuff of Thought, Steven Pinker begins with this thought experiment in the syntax of weekend chores in order to argue that the human mind employs an innate physics with which it makes sense of the world.

Language is born, not made. We learn a particular language from our parents and peers, but we do not learn language itself, in many important ways. Language comes bundled inside us, and develops or unfolds as we grow. The syntax of truck-packing verbs is one of those innate features that draws the appropriate verbs to itself as we acquire our native language.

I wish I could say that the rest of The Stuff of Thought was as interesting as this opening look at "microclasses" of verbs that represent the natural world. Instead of pursuing the question of how innate language represents a world that the newborn hasn't seen yet, Pinker basically goes off on a wide detour through elementary philosophy, basic game theory, social relations theory of the Erving Goffman and Deborah Tannen varieties, and other mild exercises in why the way we talk about the world makes sense. But few of these topics give strong insights into how language encodes human nature, or even encodes an anthropic sense of non-human nature.

Part of the problem is that Pinker, no doubt in order to sell more books to a monolingual American audience, draws almost all his examples from English, occasionally pausing to reassure us that all other languages work this way. But without lots of examples from other languages, he can never really lay to rest the Sapir-Whorf problem. Edward Sapir and Benjamin Whorf theorized that the arbitrary workings of a language condition us to see the world in arbitrary ways (instead of the intuitive, "realist" other way around that Pinker prefers: that language strongly mirrors a working, rational reality). But if all your examples come from English, you can hardly show that other languages might not carve up the world into radically different segments.

For instance, Pinker's scorn for the notion that language can determine the way we see the world is largely directed here toward an example from an op-ed column, suggesting that "taxes" be renamed "membership fees" so that people would be more willing to pay them. Yes, that's a stupid idea. You can't fool even some of the people by calling a tax a fee, just as you can't persuade them that major combat is over when thousands of soldiers die over the next five years.

But can language work in far subtler ways to determine the way we see the universe? I'm thinking of the concept of grammatical gender. In French, pens and aunts are feminine, desks and uncles are masculine. The former are arbitrary; the latter are as naturally gendered as aunts and uncles in any language. French people do not think of pens as feminine in some way or desks as manly; it's just a difference in the sound and shape of the words.

But subjects of sentences are also gendered in French, depending on who's speaking them. When you answer roll call in a French classroom, you say "present" if you're a boy and "presente" if you're a girl. Does that influence how you think about gender in deeper and more pervasive ways? To say that someone is "sweet" or "cute" in French, you say "il est charmant" if he's male and "elle est charmante" when she's not. Are those different words, different concepts, different "prison-houses of language"?

Speakers of grammatically gendered languages tend to say that it doesn't make a difference, but without having equal access to both types of language as a native speaker, it's impossible to say whether this is true, or how much of what kind of impact grammatical gender may have. And in French, for instance, the gendering of predicate adjectives may be balanced by the lack of words for "his" and "her." A man's wife is "sa femme," and a woman's husband is "son mari." They aren't unclear on the sex of the possessor, but the pronoun takes the gender of the possessed. What prison-house are English speakers in, with their insistence on the gender of the possessor?

Now, I'm pretty sure Pinker is right in many fundamental ways. My critique is not of his theory, but of his book. It's one of the perils of writing about the wide world of language for a popular audience that would run screaming from the sight of a German declension. Just as popular books on economics must limit the math, Pinker must limit his linguistics.

I'm also puzzled not to see any mention of Jean Piaget in a book devoted to the unfolding of the mind's concepts of the world. Pinker cites tons of psycholinguistic research on children, devoted to showing how ideas and utterances develop. Jean Piaget invented this field, observing children with great care and elegant experiments. Perhaps his work has been discredited or is outmoded, but it seems odd that Pinker wouldn't at least touch on it.

For the rest of the book, it suffers at times from Pinker's fatal Cleopatra: the snappy one-liner. He has never met a wisecrack he didn't like, and many of them are recycled here from his earlier books. He's still a kind of adolescent right-libertarian at heart, addicted to word play and finding blue material hilarious in often-uncomfortably misogynist ways. He still thinks that campus speech codes are a clear and present danger, and he still can't understand why Anita Hill didn't just lighten up and have a chuckle over the unhygienic properties of Clarence Thomas's can of Coke.

The Stuff of Thought is at least far less strident and pugilistic than Pinker's previous book, The Blank Slate (2002). But much as I disagree with about 3/4 of The Blank Slate, I think it's a better book: more energetic, more provocative by far. The Stuff of Thought is not exactly phoned in, but it's a bit of a telecommute.

Pinker, Steven. The Stuff of Thought: Language as a window into human nature. New York: Viking, 2007.