Over To You, Steve

A story that’s been making the rounds the past few days (thanks to the indefatigable JK for sending along this version of it) has to do with recent research that casts doubt on a cornerstone of contemporary thought about human language: namely that we all are born with a “language module” that constrains possible grammars to a few basic systems.

This idea has become more or less hegemonic amongst linguists due to the work of Noam Chomsky (and subsequent proponents such as Steven Pinker), and has also been taken as a model for work on the evolutionary psychology of morality by people like Marc Hauser. If it’s wrong, that’s a pretty big deal; it will be interesting to see how the research community responds.

Learn more here.


  1. Jesse Kaplan says

    The ideas here are interesting. If I’m understanding correctly, historically known phonemes track theories of human dispersal, but rules of grammatical construction do not.

    If a new word like “internet” comes along, it will be incorporated into various languages without affecting the grammatical rules of those languages. Words are not phonemes, but they consist of phonemes. Therefore, historical commonality/sharing of vocabulary will leave a residue of greater commonality of phonemes. This would also explain why more isolated languages will contain fewer phonemes.

    I’m not sure how convincing that graph of phoneme numerousness is, though. Also, just because more isolated languages contain fewer phonemes does not necessarily imply the opposite, that less isolated languages contain more. Within the scope of known history, the center of gravity for phoneme exchange would be more like the Middle East. Therefore, an alternative hypothesis (or conclusion) would be that more isolated languages will EITHER have fewer or more phonemes, and less isolated languages will have a middling number of them. One can see how that might be: maybe languages at the linguistic center of gravity will tend to utilize phonemes that are most adaptable among “diverse tongues.” Also, of course, theoretically an isolated language might reach EITHER toward more & more phonemes to expand its vocabulary OR toward greater word or grammatical complexity utilizing a small set of phonemes.

    Leaving that aside, I like the idea that grammar does not correlate with human dispersal because there is not a “language module.” I find it encouraging to think that rather there is a deeper level of abstract, linear cogitation that merely utilizes language. Then language would not “think us” as much as we think, which in turn maybe means we are less “mechanical” than we fear. This line of thought is really the theory-of-consciousness issue of whether there is a meta-language of consciousness. As language manifestly does “think us” often, the implication is that we have both an innate thinking tool and a slightly more external one that sometimes helps us reach results we might not otherwise. Maybe linearity and complexity of concept-manipulation is not language-dependent. If true, maybe animals can think more than we think. Also, here could be an argument against artificial intelligence, that it involves only syntactical manipulations but not the alternative meta-language-of-consciousness of living things. Also, we are used to the thought that in an “oral tradition” memory is more powerful, that fixing thoughts in writing transfers that function to paper. Now we have the possibility that fixing thoughts in language is the same thing at an earlier stage; the exact rules of grammar (and phonemes) used are not important. And what was lost with this trade-off? Perhaps fluidity of thought? Or is it extended linearity itself? Do grammatical and other language constraints inhibit intuition, idea-production? Does language permit more extensive linearity of thought, or merely encode it? If language is less reflective of “intelligence,” might animals be more nearly as mentally acute as we, but we are missing some of this because language limits and structures our recognition of non-linguistic thinking? Which came first, language or logic?

    My first points are very precise and my second ones are very unstructured and not novel. I also confess I rewrote more writing this than I have with almost anything.

    Posted April 23, 2011 at 11:47 am | Permalink
  2. Malcolm says

    Hi Jess,

    Plenty there to think about.

    One thing that I’ll say right now: I think maybe phonemes aren’t as flexible as you are assuming here. I agree that word exchange surely happens at a high rate in crossroads-type places like the Middle East (and the whole world is now becoming such a place) — but I think it’s important to keep in mind the distinction between words and phonemes. For example, Japanese, which has been one of the most isolated languages in the world, has incorporated a lot of English words since the War, and a great many more lately — but when it incorporates a new word, it does so by making a version of it that’s constructed out of the same old familiar Japanese phonemes. Creole is like that, Chinese is like that, and I think most languages are like that. So phoneme groups may be a lot more stable than you think, and are certainly far more stable than vocabulary.

    Also: I think that consciousness is probably not as important here as I think you think it is; I don’t see why consciousness has to have anything to do with language at all, really (other than that we are aware that our thoughts — the ones we are even aware of at all, that is — proceed in this or that language). But below that, we get into whether our unconscious cognitive processes are conditioned by our language,as you say, and all I can do about that is to speculate.

    I don’t think this research refutes the idea of a “language module” – it probably means that there is just more flexibility to it than it seemed. Learning language is such an important skill, and is so complicated a cognitive task, and we do it so well right out of the box, that it seems crazy to imagine that we don’t come with some factory-installed hardware for it. I’m curious to see what the boffins, in particular Pinker, have to say about it.

    Posted April 23, 2011 at 10:25 pm | Permalink
  3. Jesse Kaplan says

    Bad my.

    Posted April 24, 2011 at 11:53 am | Permalink
  4. Hi Jess, I was wondering at what point a new word is allowed into the dictionary, does it have to be used a number of times to appear. What about words like ‘init’ which is short of isn’t it, obviously this is not a word, but would it become one if used enough and at what point would it be put into the dictionary. I am intregued.

    Posted September 4, 2011 at 7:48 am | Permalink

Post a Comment

Your email is never shared. Required fields are marked *