Today I've got an article in the New York Times about the report in Nature that starlings can recognize syntax-like patterns in songs, and what that might mean--if anything--for the evolution of language. The blogs have been buzzing about the study since it came out on Wednesday, with the Language Log logging in several complaints about bad science and bad reporting. (Fortunately, they gave me a pass, and I hope not merely because I'm the brother of one of the bloggers there!)
So why should you now turn to Tuesday's science section of the Times, days after the wires and the blogs have had their way with this material? I hope that with the luxury of a little extra time, I've been able to get the science relatively straight, and to offer a sense of why it is inspiring so much debate. Plus, we've got starling songs for you to hear! Check it out.
Update: Tuesday 9 am--article link fixed.
- Log in to post comments
fyi - the link to your NYT article seems broken :(
It's not your fault; the NYT is always a little behind. It took them four days to mention Colbert's media roast, apparently, and approximately 700 years (give or take) to cop to the Flying Spaghetti Monster phenomenon. Blogs are for keeping up to date, and the Times is for... um... getting paid.
That said, I thought your article put a nice cap on the starlings question, which I've been following with some interest but little skill (I'm only just reading Chomsky now).
Jess, why start reading Chomsky now? I mean, I assume you've done okay thus far without him...
I kid, I kid. Sort of. In my experience, if you really want (or need) to read Chomsky, do some research first and find one of his books that will give you a reasonably big picture of the topic. If, after finishing the one, you insist on reading more another, you'll see that whichever book you read second repeats an awful lot of what the first one said. Ditto for the third and fourth and ...
As you can probably tell, I'm not a big Chomskian with regard to language. Nonetheless, my response to hearing about this starling research was squarely on "his" side of the debate. The research may well be interesting and important, but since I haven't read the "primary literature" on the research, I don't know for sure what implications it has, if any, for language (nor do I know what claims the authors make regarding language).
The first think I thought of, though, was a paper by JD Trout on the Speech Is Special hypothesis and what to make of "auditorist" claims that non-human animals perceive speech (or language) in basically the same way we humans do (pdf linked here). If I'm remembering Trout's argument, the basic problem is that behavioral similarity is not enough to establish equivalence. Chinchillas can distinguish between phonemes, as can humans, but that doesn't mean that Chinchillas have, or use, the same cognitive apparatus that we do when carrying out a phoneme discrimination task.
The same problem applies to this starling research, if I'm not mistaken. 9 out of 11 of the birds learned to discriminate between two types of songs that, on some level, can be argued to share structural characteristics with (some of) human language. Without a rigorously tested model of how they, and how we, carry out such a task, though, it doesn't tell us much of anything about how we (or our ancestors) represent, process, or acquire language.
Of course, none of this means the research isn't interesting or worth doing, and it's maybe too much to ask for a well-tested model already. And, again, I don't know (yet) what claims the authors make about the implications of the study. I just wanted to point out (or argue) that, whatever the merits of this study, it is unlikely to tell us much about human language.
I highly doubt I'm much of a Chomskian either, but I figure I should know what I'm not agreeing with.
Unfortunately, I picked the book via the time-tested "the one that was on sale for a buck from the Graduate English Organization, and I know that dollar goes into the end-of-semester binge drinking fund." Practical economic considerations, rather than intellectual ones.
I think Chomsky should give birds a little more credit. I just read this book (www.whybirdssing.com) a week before this study came out and it seems pretty clear there's more than just imitation going on.