It’s been a while since I read any of Chomsky’s writings on linguistics. As is well known, he blew apart the old behaviourist model of language acquisition with his review of Skinner’s book “Verbal Behaviour” back in 1959. There must, he argued, be an innate linguistic competence, as the poverty of a child’s experience of language is in marked contrast to the complexity of the acquired linguistic ability. The battle was won and the Chomskian model has been the dominant paradigm in linguistics ever since.

Chomsky has never shown much interest in the evolution of linguistic competence. It was a reasonable approach for him to take in the sense that origins were not what he was interested in and wouldn’t help the business of untangling the actual inner workings of language, but it seems like what was originally a methodological decision to take innate linguistic competence as a given became a fact about the world, such that linguistic competence was beyond the reach of the usual Darwinian evolutionary logic. [I tried to argue here that this switch from methodology to ontology was something to be found in his political writings as well: an appropriate methodology for an American of concentrating solely on US sins turned into a world where the only sins were American sins.]

So where are we now? Carl Zimmer has posted the first of two articles on language on his website, and it gives those of us who aren’t prepared (or equipped) to wade through the professional journals as good an overview as we’re likely to get. On one side we have Steven Pinker (The Language Instinct) and Paul Bloom (Descartes Baby) arguing for the importance of evolutionary pressures for the development of language: against them there’s the Chomsky faction who argue that language may just be some sort of evolutionary side-effect (a “spandrel” in Stephen Jay Gould’s terms), or that nearly all the elements for linguistic competence are present in other animals, and all it needed in humans was a little extra something, like recursion. The latest article from Pinker, with Ray Jackendoff, is according to Zimmer an all-out attack on Chomsky’s recent writings, which, they hint, have a creationist ring to them.

Zimmer promises some sort of resolution in his second post. Should be interesting.

Posted in

11 responses to “Linguistics”

  1. denawardah Avatar
    denawardah

    Nowadays linguistics degree courses treat Chomsky’s work, if they bother with it at all, as naive and outmoded. There has been a huge change now that most academic perspectives on linguistics are grounded in real data and lots of it. Chomsky wasn’t interested in data collection and his theories are based on examples he made up. No way is his model now the dominant paradigm. It’s been totally superseded and much evidence has shown he was simply wrong.

    Like

  2. Bob-B Avatar
    Bob-B

    It is simply not true that ‘linguistics degree courses treat Chomsky’s work, if they bother with it at all, as naive and outmoded’. A look at linguistics degree schemes in the UK and in also in the States will show that he continues to enjoy considerable influence. It is probably true, however, that his influence is less than ten years ago. One aspect of this is the way his ideas are coming under criticism from researchers who were once very close to him such as Pinker and Jackendoff.

    Like

  3. Dom Avatar
    Dom

    It’s great seeing Skinner’s Verbal Behavior mentioned again. Is there any chance that this book will be rediscovered? I always felt that Chomsky was the new Lysenko, forbidding others to read Skinner for completely hairbrained reasons.
    I think it is safe to say the Chomsky has only a legacy value in Linguistics. Important for historical reasons, bu no one takes him very seriously.

    Like

  4. Bob-B Avatar
    Bob-B

    It is not true that ‘no one takes him [Chomsky] very seriously’. I could name numerous linguists who take his work very seriously. The most that can be said is that his influence is less strong than 10 or 15 years ago. One might wish it were oterwise, but there is no point in ignoring reality.

    Like

  5. Jim Avatar
    Jim

    There is a good, well detailed at least, account of the infighting over Chomsky in the 70’s and the 80’s in “The Linguistics Wars” http://www.amazon.com/exec/obidos/search-handle-form/002-4432485-7212830.
    Chomsky may still have adherents on university staffs, but he is a laughingstock among other linguists (Source: gossip from atrain buddy who is mid-way up in Microsoft.) Linguistics is to software engineering what physics is to civil
    engineering – you can only be so wrong so many
    times before everyone starts to ignore you.
    Denawardah sums it up – why waste time reading the Chomskey’s philosophical speculations in mathemathical-equation drag when you can read real research? Case grammar is a prime example. There is a lot of new work with actually occurring systems if that field interests you. Why bother with Chomsky’s solipsisms or his disciples?

    Like

  6. Bob-B Avatar
    Bob-B

    The term “Linguistic Wars” is applied to debates in the late 1960s and early 1970s in which Chomsky prevailed. He expanded his influence in the 1980s (with something called Government-Binding theory) but since the early 1990s (when he began to advocate something called Minimalism) his influence has declined somewhat. It is not true that ‘Linguistics is to software engineering what physics is to civil engineering’. Most linguistics is of no relevance at all to software engineering and none the worse for that. It is true, however, that Chomsky’s work has no influence in computational linguistics, which is an important branch of linguistics. Case Grammar was an approach developed in the late 1960s which enjoyed some influence until the mid 1970s. It is a thing of the past.

    Like

  7. Jim Avatar
    Jim

    “Most linguistics is of no relevance at all to software engineering and none the worse for that.” And most physics is of no relevance to civil engineering. The dependency runs the other way. The analogy holds. I remember software types in my linguistics classes back in the 70’s. I understand that software design owes a lot to linguistics.

    Like

  8. Jim Avatar
    Jim

    Bob-b,
    You are right about Case Grammar being out-dated, but I really meant that whole area of syntax, not a specific theory. Case still seemsot be a prety active area of interest.
    http://www.csa.com/csa/e_products/bacontent/LLB000452.html

    Click to access Bisang5.PDF

    http://listserv.brown.edu/archives/cgi-bin/wa?A2=ind0001c&L=conlang&F=&S=&P=25667
    One of Chomsky’s blind spots was his near total reliance on “data” from English – not real observed dtata, but intuitional data from his own non-native knowledge of English. That is not to blame him for using English; he have used Hebrew, but Yiddish examples might have been a little obscure.

    Like

  9. Bob-B Avatar
    Bob-B

    In 1950s Chomsky did work in a branch of mathematics called formal language theory. That is relevant to software engineering. Pure linguistics isn’t.

    Like

  10. Jim Avatar
    Jim

    Bob-B,
    Thanks for the update. I had always wondered what interest the software people had in pronominal systems and such. Not much, apparently.

    Like

  11. Erg Avatar

    Context-free grammars are used in programming language design and implementation (see a very cool tool called Yacc), which is only a small corner of software engineering. Working in the field, I can’t recall having heard of any other ideas from linguistics being applied to programming.
    Yacc: http://en.wikipedia.org/wiki/Yacc

    Like

Leave a reply to Jim Cancel reply