Tuesday, June 1, 2010

So maybe there is no universal grammar...?

Here's an interesting article I got from Marian at the Analog forum, talking about linguistic diversity and possible explanations for patterns of human language, including the weakness of so-called language universals. As a person who always maintained a degree of skepticism for Chomsky's theories, I find these ideas compelling:



  1. I think part of the problem is we don't understand in any deep way how we learn to do some things, especially things we (as a species) are good at. There seem to be modules in the brain that allow us to do these things, but we don't understand them. They probably aren't as rigid as classic Chomskian Universal Grammar may seem to imply.

    For example, humans seem to be better at throwing things accurately, like a baseball or a rock, or hammering in a nail, than our close relatives such as chimpanzees and gorillas. (Although to be honest, I don't know how much effort has been spent on teaching a chimp to pitch, or a gorilla to hammer a nail.) The neurophysiologist William Calvin has made a big deal of this, and has tied together this human ability to that of speech. It's a bit of a just-so story, but not completely insane: throwing objects accurate and hammering nails require a complex and carefully timed sequence of muscle actions (letting go of the ball a hundredth of a second too early or late and you get a ball instead of a strike). And speech is, well, a set of complex sequences of phonemes.

    Not unrelated: we, and many animals, are very good at categorizing the natural world. For example, although we often joke about our pet dogs as thinking themselves as people, it's clear that dogs perceive people, dogs, and other animals as quite separate categories. We had found the shelter a dog that was fearful and aggressive towards other dogs. But it was always friendly and gentle with people, and despite our fears never once confused a small child with a dog. He clearly knew the difference, somehow. Other dogs loved to chase ground dwelling animals--lizards, squirrels--but completely ignored birds, even if the bird was on the ground. How did they know? Clearly there is something in the brain, something pretty fundamental, that allows brains to categorize the world. And we have no idea how that works, at the neural level. If we did, we might understand language and language "universals" much better.

  2. Calvin,
    I entirely agree. Categorization is a really important skill and I don't think one we can easily tamper with. In the area of grammar, though, I think it's about time we started working with what's measurably there and seeing if we can understand it without relying on UG as an intermediate construct.

  3. But I wonder, without understanding the neurobiology (of "modules") beneath our skill sets, be it categorization or language, how deeply can we understand? I'm not saying we shouldn't try; after all, Mendel deduced the laws of inheritance without knowing about DNA. But once we understood the machinery of DNA, we could understand in great detail and subtlety and predictability inheritance.

    (So that seems like an SF idea there--the discovery, a la Watson and Crick, of neural modules that allow us to go forward in understanding language...)