Linguist - Author - Historian: Noam Chomsky
Posted: Fri Mar 24, 2006 12:07 am
Andrew L. wrote:si-maro wrote:I will assert this: I'm no academic linguist badass, but everything I've read from Chomsky and everyone else's analysis of Chomsky suggests that the above quote from Andrew Knight is complete horseshit.
Yeah, that's what I thought. Chomsky is solely interested in syntax, and lexicography is presumably epiphenomenal to Universal Grammar in his view(?)
Yet, Chomsky does suggest some weird stuff.
Here's Knight's response to specific criticisms of his characterization of Chomsky (Chomsky aside, Knight's polemic re the disciplinary stratification of knowledge production is of increasing interest to me). I still think Knight is misrepresenting Chomsky, but the matter seems more vexed than I'd thought.Al Cohen: I have to say that Chomsky's project has been to change the field of linguistics from a soft to a hard science.
Chris Knight: Yes, as everyone knows, this has always been Chomsky’s stated project. By ‘hard science’, Chomsky means ‘natural science’. By contrast, any kind of social science is deemed ‘soft’. This is my point. It is precisely his acceptance of this institutional/disciplinary divide that compels Chomsky to be so ruthless in defining language as a natural object. The slightest concession on this score would lead to the implication that linguistics could be scientific while also being ‘soft’. Chomsky wants linguistics to be completely ‘hard’. This compels him to redefine language as completely ‘natural’, with the logical corollary (logical in Chomsky’s mind) that it is neither ‘social’ nor ‘cultural’. Once this novel definition has been established, it can be ruled legitimate to exclude all social science
perspectives. This is precisely what Chomsky does.
Al Cohen: Knight is completely wrong to say that Chomsky's project has been to say that 'the meanings of words are not socially negotiated but wired into the brain.' Chomsky never makes claims about the lexical semantics of a language, but rather of its syntax and grammar.
Chris Knight: I am afraid this comment is not well informed. Chomsky makes no such distinction between syntax/grammar on the one hand, semantics on the other. Instead he conflates the two. I agree that this is very strange. But although it is strange, it is nevertheless true.
Rather than argue in a vacuum about whether it is or isn’t true, I think it
best to accept Chomsky’s own word on such matters. Recourse to Chomsky is a valid procedure because our dispute just now is over what Chomsky does or doesn’t say. So here, just for the record, is Chomsky himself:
‘In my view, most of what’s called semantics is syntax. I just call it
syntax; other people call the same thing semantics. Syntactic Structures, in my view, is pure syntax, but the questions dealt with there are what other people call semantics. I was interested in the question, “Why does ‘John is easy to please’ have a different meaning from ‘John is eager to please’?” I wanted to find a theory of language structure that would explain that fact. Most people call that semantics; I call it syntax because I think it has to do with mental representations. Take a point we discussed earlier: the word house, the concept “house,” and the use of the word house in real situations to refer to things. There are two relations there, and I don’t think you can turn them into one as is commonly done. The common idea is that there’s one relation, the relation of reference, and I don’t believe that. I think there’s a relation that holds between the word house and a very rich concept that doesn’t only hold of house but of all sorts of other things. That relation most people would call semantics. I call it syntax because it has to do with mental representations and the structure of mental representations.’1
As for whether Chomsky says that ‘the meanings of words are not socially
negotiated but wired into the brain’, I am again perplexed. Can anyone
familiar with Chomsky’s writings possibly doubt this? Chomsky himself makes his position abundantly clear. The lexical concept ‘house’, he explains, is wired into the brain. Here is Chomsky in his own words:
'There’s a fixed and quite rich structure of understanding associated with the concept “house” and that’s going to be cross-linguistic and it’s going to
arise independently of any evidence because it’s just part of our nature.' 2
Of course, Chomsky is not stupid. He knows perfectly well that there are
different kinds of houses – mud huts, country mansions, men’s houses, public houses etc. etc. – and that for this and other reasons people in different cultural situations will be using the word ‘house’ (in whatever language) in different ways, for different literal and/or metaphorical purposes and so on. But he is not interested in all this. For him, such matters concern linguistic usage, not the nature of language as such. As we have just seen, Chomsky treats semantics as syntax. Lexical concepts can be used in various ways, he says, but in themselves they are fixed features of the human Language Faculty, in this respect no different from syntactic features. His fundamental point is that variations in semantic usage are superficial. Beneath all such variation, he asserts, the lexical concept ‘house’ – a component of the human language faculty – is universal and hard-wired. It is a pre-installed internal feature of the ‘semantic component’ of the distinctively human ‘language organ’.
Al Cohen: No responsible linguist can state that lexical semantics can
represent universal truths about onotological objects.
Chris Knight: I did not say that. I am well aware that for Chomsky, the whole of language including ‘the semantic component’ is ‘internal’ to the Language Faculty rather than ‘external’ in the sense of accurately mirroring the environment.
Having said that, Chomsky’s position is very strange. Why should the lexical concept ‘house’ be part of the human genome? Are we sure that palaeolithic hunter-gatherers had houses? In the recent ethnographic record, most Aboriginal Australians didn’t really have houses. Very often, they slept in the open around a camp-fire. But even assuming our paleolithic ancestors did have houses, it still doesn’t solve the wider problem of where lexical concepts can have come from.
It seems extremely unlikely that the lexical concept ‘carburetor’ could have become installed during the evolutionary emergence of Homo sapiens. So what about this concept? Is it an exception? Do we adjudicate that certain concepts (such as ‘house’) are part of the genome while others (such as ‘carburettor’) are just socially or culturally determined? As a scientific theory, that would look a bit messy and inconsistent.
In a sense to his credit, Chomsky opts for consistency. To this end, he
decides that even in the case of ‘bureaucrat’ and ‘carburettor’, the concepts are hard-wired. Our stone age ancestors already had these concepts, even though as yet they didn’t need to use them.
Yes, I agree, this is strange. I fully appreciate that few supporters of
Chomsky are likely to believe me unless I quote Chomsky himself. So let me quote Chomsky himself.
After defending his idea in a general way, Chomsky elaborates:
‘Furthermore, there is good reason to suppose that the argument is at least in substantial measure correct even for such words as carburettor and bureaucrat, which, in fact, pose the familiar problem of poverty of stimulus if we attend carefully to the enormous gap between what we know and the evidence on the basis of which we know it. The same is often true of technical terms of science and mathematics, and it surely appears to be the in substantial measure correct even for such words as carburettor and
bureaucrat, which, in fact, pose the familiar problem of poverty of stimulus
if we attend carefully to the enormous gap between what we know and the
evidence on the basis of which we know it. The same is often true of
technical terms of science and mathematics, and it surely appears to be the case for the terms of ordinary discourse. However surprising the conclusion may be that nature has provided us with an innate stock of concepts, and that the child’s task is to discover their labels, the empirical facts appear to leave open few other possibilities.’3
‘Thus Aristotle had the concept of an airplane in his brain, and also the
concept of a bicycle – he just never had occasion to use them!’, comments Dan Dennett, adding that he and his colleagues find it hard not to burst out
laughing at this point.4
So there we have it. Chomsky views word meanings – the concepts to which words are attached – as mental representations drawn from an ‘innate stock’ of such representations provided by ‘nature’. The child who acquires a particular language, according to Chomsky, doesn’t have to learn the concept. The task is simply to rummage through its innate stock of concepts and then discover the relevant label.
This is what I said in the first place. I very much thank Al Cohen for giving
me this opportunity to explain. I hopeI have made myself clear.
References.
1. Gary A Olson and Lester Faigley, 1991. Politics and composition: a
conversation with Noam Chomsky. In Journal of Advanced Composition 11.1, pp.
1-36.
2. The same.
3. N. Chomsky (2000) New Horizons in the Study of Language and Mind
(Cambridge: Cambridge University Press), pp. 64-66.
4. D. Dennett (1991) Consciousness Explained (London, Penguin), pp.
192n-193n.
The study of human language can never be a "hard science" because any "science" that involves human volition is de facto hard to make predictions and lawlike generalizations about. Soft science.....ok....but never a science like geology or oncology.
The hybris of modern scientists is that they have FAITH in conquering all that is when in fact they nor has anyone else seen all that is.
Linguistics is interesting insofar as it is useful. Pure theorising about the ultimate origin of language is stupid, much like any theorising about ultimate origins.
ssssssssss
Unless.........................you accept a purely deterministic view of reality.....a view that is continually, empirically disproved by every voluntary human action.