The chicken-and-egg of language

Steven Pinker is an experimental psychologist involved in research into the human mind, but he is also an unabashed popularizer whose books are full of pop culture references (especially comic strips). Apart from a few tedious sections, The Stuff of Thought: Language As a Window into Human Nature (recommended to me by a fellow blogger who merely read an article about it) is one of his best books. It applies a scientific perspective to a favorite subject of mine, the relationship between language and thought. But it does it with style, exploring a range of Americana from the semantics of Bill Clinton's lies (a topic that has already received far more attention than it deserves) to the grammar of profanity. I find the following hard to read without smiling:

Woody Allen's joke about telling a driver to be fruitful and multiply but not in those words assumes that Fuck you is a second-person imperative.... But Quang makes short work of this theory. For one thing, in a second-person imperative the pronoun has to be yourself, not you--Madonna's hit song was titled "Express Yourself," not "Express You." For another, true imperatives like Close the door can be embedded in a number of other constructions:

I said to close the door.
Don't close the door.
Go close the door.
Close the door or I'll take away your cookies.
Close the door and turn off the light.
Close the door when you leave tonight.

But Fuck you cannot:

*I said to fuck you.
*Don't fuck you.
*Go fuck you.
*Fuck you or I'll take away your cookies.
*Fuck you and turn off the light.
*Fuck you when you leave tonight. (pp. 362-3)

The book's overarching theme is how the human mind influences the structure of language. Like most linguists, Pinker largely dismisses the notion that the influence goes the other way. That notion is the basis of the controversial Sapir-Whorf hypothesis, which predicts, for example, that if you grew up speaking a language like Hopi, which lacks verb tenses, you would end up with a different perception of time than if you grew up speaking a language like English.

Pinker discusses some of the alleged evidence for this hypothesis before disposing of it. For example, one Mayan language has no words for left and right. The speakers orient themselves using the mountain slope where they live, with the words "upslope" and "downslope" corresponding roughly with south and north, respectively. Researchers found that the speakers have trouble distinguishing left from right but can locate north and south after having been spun around blindfolded while indoors!

Pinker spoils the picture by revealing that another Mayan people with the same aptitudes does have words for left and right. Apparently, since both groups spend most of their lives outdoors, they have a stronger sense of north and south than we do but little use for the concept of left and right. The absence of those words from the language of one group is an effect, not a cause, of the group's traits.

Distinguishing cause and effect is the subject of the book's most fascinating chapter, where Pinker explains how the whole concept of causality, so central to our common experience, is tantalizingly hard to define. We perceive the flow of time as consisting of nothing but causes and effects, and this intuition is deeply entrenched in language. But "the world is not a line of dominoes in which each event causes exactly one event.... The world is a tissue of causes and effects that criss and cross in tangled patterns" (p. 215). The challenge of identifying which causes are most relevant and guessing what would have happened if not for certain events--effectively imagining an alternate universe--underlies everything from scientific knowledge to moral responsibility.

One of his examples is President Garfield's assassin, who argued that "The doctors killed him; I just shot him." The wound was potentially nonfatal, but the doctors were wildly incompetent even by the standards of their day. Did this get the assassin off the hook? The jury didn't think so, and they sent him to the gallows.

A more recent example came in the aftermath of 9/11. Insurance companies were pledged to reimburse for each destructive event. But was the destruction of the Twin Towers one event or two? This question held billions of dollars at stake.

Questions like these are almost unanswerable because the world, contrary to our perceptions, is a continuum without clear boundaries between things. This dichotomy can be seen in the two categories of nouns, count and mass. Count nouns are words like book, which you can count: you can talk about one book, two books, etc. Mass nouns are words like jello which lack that property. You can't talk about one jello or two jellos; there's just jello.

Curiously, some mass nouns, like furniture, refer to material that should be countable. (We get around this problem by talking about "pieces of furniture.") And many nouns can perform both roles: rock is a mass noun in the sentence "The ground is made of rock" and a count noun in the sentence "I'm holding two rocks."

Speakers will occasionally transform a count noun into a mass noun by imagining that something discrete is made up of an amorphous substance. Pinker's example is the distasteful statement "After he backed up, there was cat all over the driveway." His point is that the count/mass distinction doesn't force us into any particular way of thinking, because we can escape that thinking by manipulating the language. But the distinction does reveal how we choose whether to view matter as a collection of objects or as a lump of "stuff."

I've only mentioned a fraction of what the book covers. With each topic, Pinker builds on the thesis that language reflects more than affects our minds, which can see past the constraints it imposes on us. (You might think this undercuts the point I made in my post "The cage of language," but actually I think it reinforces it.) Identifying these constraints helps us understand how we perceive the world and thus provides a way for us to transcend those perceptions.