Natural language and the language of thought

I found Thom’s book extremely illuminating, insightful and enjoyable. I learned a great deal from it, and look forward to this online discussion, from which I’m sure I’ll learn a lot more.

One point where I was left feeling rather frustrated was in the brief discussion of Chomsky’s views on language and adaptation (section 6.2). I had been hoping to get some guidance on how to think about the increasingly acrimonious debates between Chomsky and others on the existence or non-existence of a dedicated language faculty or Universal Grammar, but Thom remains officially neutral on this. As he says in the Précis,

“What might be a natural object of study is an innate cognitive mechanism – sometimes called a Universal Grammar – without which we would not be able to acquire and use languages. I say that this only “might be” a natural object of study simply because whether such a mechanism actually exists is a disputed and much vexed issue, on which I am personally agnostic.”

Although I would have liked to hear more on the pros as well as the cons of Universal Grammar, what mainly frustrates me is the possibility that what Chomsky means by language is not the same as what Thom means, so the discussion may be at least partly at cross-purposes.

Thom makes a convincing case that language evolved to make ostensive communication expressively powerful, whereas Chomsky repeatedly denies that language has a primarily communicative function. Thom defines language as “The suite of cognitive traits that allow us to acquire and use languages”(i.e. public languages like French or English). For Chomsky, though, language seems to be more like a language of thought, and this has become increasingly obvious in his recent writings. Here are a couple of extracts from a recent informal talk by Chomsky on ‘Language and the Cognitive Sciences’ at Carleton University (my italics):

“It appears overwhelmingly clear that a generative process suddenly emerged at some pretty recent point … Well it emerged in an individual, mutations don’t take place in groups, so some individual was fortunate or unfortunate enough to get this generative capacity… Furthermore, there was no selectional pressure at that time. There couldn’t be. It’s just something that happened to an individual. So you’d presumably expect what appeared at that point to be just determined by natural law, there’s no other pressure, something kind of like a snowflake. And the same would be true as this capacity of this property is transmitted to offspring. Notice that the capacity itself HAS selective advantage, the person who had this capacity could think, it could plan, it could interpret, you know, could construct, internally of course, complex thoughts. That, you would expect, would have advantage transmitted to offspring in some small hunter gatherer group, maybe a couple of hundred people. It could take over most of the group after some period, and at that point there would be a reason to externalise it, to make it available to others so that it’s not just in your own head. And that seems to be the way language works, with externalisation being an ancillary process.” (Chomsky 2011)

And here is a longer extract from the questions after that talk:

“Question: One of the things you argued in support of is Language’s productivity, and Fodor uses the same argument for a language of thought. I wondered if you think language plays a greater role, like Fodor does, and is there such a thing as a Universal Grammar of thought?

Noam Chomsky: Well, I DO think so, and, in fact, what I was describing was a language of thought. What actually seems to have happened, as far as we can piece anything together, and as far as the empirical evidence shows, is that at some point—maybe seventy-five thousand years ago—some small neural rewiring took place, of course, in some individual, because it’s the only possibility, and that individual had a computational process which was somehow linked to pre-existing conceptual structures. Now, what those pre-existing conceptual structures are, we haven’t a clue about that. That’s the problem I ended with. Nobody has a clue what that could be or where it could have come from. But it’s there, and it’s totally different from anything in the animal world. And if this generative procedure could link to it, you do have thought. So, that’s a language of thought. And then somewhere down the line it got externalised and you get interactions among individuals. So, yeah, I think it is a language of thought. But I don’t see any reason to think it’s separate from language, I think it just is language.”

Chomsky goes on to add, rather cryptically,

“Noam Chomsky: In fact, if you look at Fodor’s work, and you ask the question ‘What do we know about the language of thought?’, well it turns out to be English.

[Audience murmurs]

Question: You are saying that everybody thinks in English.

Noam Chomsky: I’m not saying that we think in English, but the reason it turns out to be English is that’s the language everybody’s using. Whatever the language of thought — this internal, our own internal language, yours and mine — whatever it is, is inaccessible to introspection. Okay, if you introspect, you can’t go one minute without talking to yourself. It takes a tremendous act of will not to talk to yourself. In fact you do it all night, it keeps you up all night. It’s just impossible not to do it, but what you’re introspecting is the externalised language. So, you can tell when you’re talking to yourself, you can tell whether two sentences rhyme, okay. Or you can tell how long they are, or something like that. And actually, if you really pay attention, you’re not really talking to yourself in sentences, just kind of odd little fragments. Something is going on deeper which we CAN’T introspect into, any more than you introspect into the mechanism of vision, and that’s the language of thought. And it’s probably universal. It’s hard to imagine how it could be anything else. There’s no evidence for acquiring it.”

Now if this is what Chomsky means by ‘language’, it’s easy to see why he denies that language has a primarily communicative function and that it emerged through natural selection. On the other hand, as long as the capacity for ostensive communication was present before “externalisation” took place, Chomsky’s account might well be compatible with the view that the externalisation process was driven by selective pressures, along just the lines Thom suggests. (I don’t know if this makes sense – it’s a long time since I thought of myself as a linguist – and am quite happy to be put right.)

2 Comments

  • comment-avatar
    Thom Scott-Phillips 2 July 2015 (19:42)

    Deirdre asks about Chomsky’s views on language and adaptation. Before I answer, let me take this opportunity to thank Deirdre for her work in developing Relevance Theory. As I expressed in SOM, I believe that Relevance Theory provides a genuine scientific paradigm for pragmatics, and for this work alone, all of us interested in cognition and culture owe her a huge intellectual debt.

    Deirdre observes that I made very few comments about Chomsky’s views on language evolution in SOM. What I did argue (see p.134-136) was that if there is a cognitive mechanism worthy of the name Universal Grammar, then it would have to have been selected in order to aid the acquisition and use of (what I call) languages i.e. sets of conventions created in order to enhance the expressive capacity of ostensive communication.

    As Deirdre observes, Chomsky would be unlikely to endorse this terminology. In line with the quotations that Deirdre picks up on, Chomsky defines language as “a computational cognitive mechanism that has hierarchical syntactic structure at its core” (see p.1 of the paper that Deirdre links to). The presently preferred hypothesis about what this mechanism consists of is a very simple one: “human language syntax can be characterized via a single operation that takes exactly two (syntactic) elements a and b and puts them together to form the set {a, b}” (ibid.). This operation is called ‘merge’.

    Let us grant Chomsky all of the above. Let us accept for the sake of argument that this “generative procedure” is “a language of thought”; that, rather than being separate from language, this procedure “just is language”; and that all that communication brings to the table is externalisation. Deirdre suggests that this view could be compatible with the one I put forward in SOM: “as long as the capacity for ostensive communication was present before “externalisation” took place, Chomsky’s account might well be compatible with the view that the externalisation process was driven by selective pressures, along just the lines Thom suggests”. And she is right: the two views are compatible in just this way.

    This acknowledgement of compatibility should not, however, be taken as an endorsement of the Chomskyan approach to evolution. Chomsky believes language, thus defined, is uniquely human. To the extent that I understand what ‘merge’ is, this seems unlikely to me. ‘Merge’ is not computationally difficult, and the computation of putting things together seems necessary for all sorts of animal cognition. Perhaps I have misunderstood the idea, but if I have not, it seems to me that much more elaboration is required to support the claim that ‘merge’ is uniquely human. (It is possible, however, that it was only in humans that there was selective pressure for such an ability to be externalised in communication. Indeed, this is exactly the point I made about Universal Grammar in SOM (see above).)

    A second issue is adaptation. If merge is uniquely human, then that invites the question: Why? Why should such a trait have evolved only in one species, and why this species? Adaptationism can provide candidate hypotheses to such questions, but Chomsky does not offer any. On the contrary, he is famously hostile to adaptationism, and hence refuses to even propose such an explanation. He instead suggests that the emergence of language is, effectively, just a lucky accident (“what appeared at that point to be just determined by natural law, there’s no other pressure, something kind of like a snowflake”). To be blunt, this seems to me an abdication of intellectual responsibility. I agree with Steve Pinker (among others) that if one is committed, as Chomsky is, to studying cognitive computation like one would any other biological trait, then adaptationism is a critical part of the toolkit.

  • comment-avatar
    Nicholas Allott 3 July 2015 (23:30)

    I hope it’s alright if I join in with this very interesting discussion, even though I haven’t yet read Thom’s book. A few brief points about Chomsky’s views on the evolution of the language faculty. First, Tim Wharton in his comment elsewhere in this discussion makes a crucial point: it’s not at all clear that languages are sets of conventions. Obviously there are different theories about what conventions are, but I can’t see that any of them would make conventional (e.g.) the putative fact that grammatical relations are mediated by C-command. So Chomsky certainly wouldn’t – and shouldn’t – agree with Thom that “if there is a cognitive mechanism worthy of the name Universal Grammar, then it would have to have been selected in order to aid the acquisition and use of (what I call) languages i.e. sets of conventions created in order to enhance the expressive capacity of ostensive communication.” And so, as Deirdre says, there is a real risk that Thom and Chomsky are simply talking past each other. Thom’s suggestion in his reply to Tim that all the syntactic differences between languages are conventional even if shared properties of languages are not, is interesting, but not much more plausible, I think, especially given the extra commitment that they be “created in order to enhance the expressive capacity of ostensive communication.” At any rate, one wants to know how the idea would go in detail. I also agree with Deirdre that disagreement about ‘adaptationism’ is inessential to Thom and Chomsky’s disagreement. Chomsky’s proposals with colleagues about the evolution of core of the language faculty can be stated in a way that is neutral about such arguments among evolutionary biologists, I think. Chomsky, Hauser and Fitch make the bold conjecture that FLN (the ‘narrow faculty of language’), which is defined as what is specific both to humans and to language, is just the operation Merge, which puts together any two linguistic constituents, plus the requirement that the structures generated are legible by the two systems that need to use them, the Conceptual-Intentional (i.e. thought) system and the Sensori-Motor (phonetic) system. As a slogan: “Interfaces + recursion = (the core of) language”. Part of the motivation is to explain what Chomsky and co-authors take to be the narrow time-frame between the appearance of natural language in the species and its dispersal over the planet, since which time it apparently hasn’t evolved (given that infants from everywhere can acquire any language with apparently equal facility). A related motivation is to dispense with what earlier syntactic theory (Government and Binding-era generativism) seemed to require: the simultaneous evolution of several distinct ‘modules’ of the grammar – theta theory, Case theory etc. –which would be largely useless without each other. Anyway, as Deirdre says, the proposal is that a single evolutionary event in an individual gave him/her the ability to put together atomic thought-components recursively. This ability spread to a group. I think that it’s inessential to the account whether it was selected for, or just not selected out, or was a side-effect of some other change that was selected for. After this there was another event that allowed it to be externalised, and this has not been selected out – again, for whatever reason. If adaptationists like the overall story, then they are free to claim that selection pressure was crucial – following the initial mutation, of course – at both stages. Thom doubts that Merge is uniquely human, saying that “Merge is not computationally difficult, and the computation of putting things together seems necessary for all sorts of animal cognition.” Maybe. But do we have good evidence of unboundedly recursive cognition in other animals? That seems doubtful. Since that is what Merge provides (given that it can operate on its own output) then they either don’t have Merge or some other limitation prevents them from using it in the powerful way that we do. One final note: I see no reason to suppose that “all that communication brings to the table is externalisation” and I doubt that Chomsky thinks this. After all, he often suggests that we have pragmatic competence, which we draw on along with our linguistic competence in (some of) our use of language. Equally, selection pressures connected to communication might have been important for the development of parts of FLB (the ‘broad faculty of language’), that is, the pre-existing systems and abilities that became part of a full language capacity with the emergence of FLN. Certain properties of the articulatory apparatus are pretty good candidates, for example.