Linguistic Weirdness: Brazil's Piraha

The Whorf-Sapir hypothesis is one of those bits of pseudo-science that just won't die. Almost every time a language with strange properties is discovered, some popularizer somewhere will jump to the conclusion that the relevant bit of weirdness proves that Whorf was right.

Well, he wasn't.

Still, this stuff about the Piraha of Brazil sounds pretty interesting. As with Hopi (or Navaho? Dang...which was it?) I expect that the language of the Piraha will turn out to be far less interesting than it seems once more people can understand it. And, of course, mere linguistic weirdness in now way lends even the slightest support to the Whorf-Sapir hypothesis.

As in the case of the Hopi (Navaho? Crap.), some people are apprently asserting that the Piraha "have no concept of time." As with the Hopi (Navaho...whatever), this will almost certainly turn out to be false. It'd be fairly easy to design some experiments to test this claim, though. (Interestingly, this article does contain descriptions of the results of some experiments on the Piraha's math skills.) Of course that's going to force us to explain at least roughly what it means for someone to have a concept of time...

Of course some people have access to concepts that others lack. There's nothing weird about that. (Not everyone, for example, has the concepts reductio ad absurdum or mathematical limit or representative democracy.) But, contra Whorf and his epigones, this in no way indicates that language determines thought. What's most likely is that thought and language interact, influencing each other. They may even be something like two sides of the same coin. No one would claim that language has no effect on thought--and it's rather bizarre that many people think that language is the only thing that influences thought. You probably think I'm making those people up, but I'm not.

Among the many other reasons for rejecting the claim that language perfectly determines thought is that this entails that thought is not possible without language. Among the many reasons for thinking that that latter claim is false is that we know of some people who had thoughts before they understood language. Steven Pinker discusses (in The Language Instinct) deaf-mutes who had very complex thoughts before they acquired sign language. One such person was a locksmith before learning to sign--and that's not something you can learn to do if you can't think. He also reports a normal mental life before he learned to sign--including very complex philosophical thoughts, about, for example, the existence of God.

The Chomskians also end up getting mixed up in these debates, and I'm always left wondering why, exactly, they think that all languages have combinatorial syntax. I take it that's a kind of constitutive claim? That a symbol system without CS is not really a language? I've got nothing to say about that. But doesn't that just push the question back to "does any human group use a primary symbol system that's not a language?" I mean, it's unlikely that recursive grammar sprung up as soon as humans started using sounds to refer to things. There's probably a fairly smooth transition from animal warning noises to full-blown language. So at some point in the evolution in at least some languages, it seems like we'd predict the existence of a fairly complex but not fully-developed language or quasi-language. Something with a relatively impoverished bank of concepts, and a relatively unsophisticated grammar.

I don't see why anyone would think that such primitive (quasi-?)languages are impossible, but I also don't see why anyone would think that these possibilities tell us anything terribly important about thought.