The 'Chomskyan Era'
Excerpted from The Architecture of Language, 2000
QUESTION: With the growth of a 'Chomskyan era', linguistics has definitely become a discipline worth breaking heads over. At the same time it has become so esoteric that it has become restricted to people holding a job in linguistics only. How do you think this subject can become accessible to people beyond linguistics? What about its marketability?
CHOMSKY: I don't like this personalization. That is a wrong way to think about things. There is no personalization in rational inquiry, everybody is working on it. But I'll leave the question the way it sounds.
Well, first of all, a lot of linguistics is accessible. You could ask the same question about chemistry. An awful lot of chemistry is just unintelligible unless you have been through a pretty extensive education to know what people are talking about, understand the results, the background, the principles, and so on. But basic ideas can be made accessible to people quite readily. That is what popular science is about. Making the results of a technical inquiry accessible to people, at whatever level they want to understand it, is a very legitimate and socially valuable occupation. So if I am interested in learning something about quantum physics, I don't want to bother with all the details; I just want to understand roughly what is going on. There are people and books and so on that try to make it available at my level of interest. I think the same is true in linguistics.
What about marketability? Jobs are certainly a problem. When you get into any field that gets hard and complicated, there's always a question about where you are going to get a job. That is just as true in mathematics as it is in linguistics. Right now in the United States, there are, on an average, several hundred applicants for every available professional position in mathematics. That is a problem. It is not just a problem for linguistics; in fact it is, in many ways, less of a problem for linguistics.
In any case, it is a general problem. It does have to do with a social problem -- how much science should there be? Right now the answer to that question is given, in my view, extremely irrationally. It is not a big secret that wealth and power are very highly concentrated and the people in whose hands it is concentrated make the decisions. The way they make the decisions is largely by deciding what they want from the point of view of market value. That is an extremely irrational way for social decisions to be made. These decisions, like all decisions, should be popular decisions made on the basis of judgments as to where resources ought to go.
In my opinion, there ought to be a lot more science and everybody ought to be involved in it in some sense just like there ought to be a lot more literature and art. These are the enriching parts of human life; they should be made accessible to people. That means we should devote resources to them. But you don't make money for businesses that way and, since that is how jobs and resources are distributed, you get the results you have. I think that is pretty irrational but that has to do with lack of democracy in society in general.
QUESTION: What is in common between your science of language and your politics is the absence of any role of community and culture. The conscience of the community is what finds expression in justice as well as in language. In the study of language, don't you think better results will be obtained from giving positive values to the differences between languages, to relations of complementarity between two or more languages spoken simultaneously by the same community and by supposing the state of bilingualism to be normal to the species?
CHOMSKY: My political views are my own. Anything that one says about politics, of course, has to do with community and culture. How could it be otherwise? That is true not only of attempts to understand the world, but also to change it. In my own personal case, the point should be particularly obvious, if only because of my interest in and commitment to anarchism -- specifically, those tendencies within it that stressed the significance of community, association and culture.
The science of language is not mine. It is anybody's who is working on it; people don't own a science. So it is not Chomsky's science of language. The search for understanding of how the world works is a cooperative enterprise, and nothing that could be called 'X's science of Y' is even worth looking at. There's a field that is often called 'generative grammar', but it is not mine, or anyone else's.
This branch of the study of language is indeed marked by an absence of any role for community and culture ... There is nothing of any significance known, at least to me, about community and culture that relates to these questions about the nature of a certain biological system. If there is something known, I'll be glad to learn about it but I don't know about it. Therefore, as far as I am aware, there is no relationship.
But that is not to say that questions about community, culture and language are unimportant. They are extremely important but so is everything about human life; it is just that we have little scientific understanding of them. We ought to be very clear and explicit about what we understand, what we have technical knowledge of, and when we are in the same boat as everybody else. We just try to find our way through it as well as we can but without theoretical understanding of any depth. If that is wrong, I am happy to be instructed, but I don't know of any reason to believe that it is wrong.
Everyone working on language, myself included, focuses attention on 'differences between languages'. If we didn't, we would conclude that whatever language we happen to be looking at is innate -- which would certainly solve a lot of questions about the language faculty, language acquisition, etc. The first modern work on generative grammar happened to be on Hebrew. The first generative grammar published was on Hidatsa. So it continues. It is not a matter of 'better results' or 'worse results', any more than one could answer the question whether we get 'better results' or 'worse results' by studying just hydrogen or the differences between hydrogen and helium, or just fruit flies and not the difference between fruit flies and apes. At any moment, one concentrates on questions that look promising.
As to the positive value of differences between languages and bilingualism and so on, I really don't have any considered opinion on this. Obviously, you're a richer person if you have more diverse kinds of experience; that is certainly true. So exposure to various cultures and immersion in various cultures, languages and so on adds a certain richness to life and, yes, richness to life has a positive value; but I don't know what more there is to say about this.
Bilingualism is normal to the species in the trivial sense that the world is so complex that strict monolingualism is almost unimaginable. Even in the smallest, hunter-gatherer society with fifteen people in the tribe, there's going to be diversity. People aren't clones and as long as there is some diversity, you're going to have some small variety of multilingualism. It may be so small that you won't call it 'multilingualism' but there will be some variety. In that sense it is natural to the species but I don't see anything deep about that.
It is also well to bear in mind that 'multilingualism' is a vague intuitive notion; every person is multiply multilingual in a more technical sense. To say that people speak different languages is a bit like saying they live in different places or look different, notions that are perfectly useful for ordinary life, but are highly interest-relative. We say that a person speaks several languages, rather than several varieties of one, if the differences matter for some purpose or interest.
QUESTION: It is said that homo sapiens has the advantage of the faculty of language. Is it possible that actually the animals are better off than us because their system of communication is very sophisticated (saying more with less)?
CHOMSKY: I don't see any serious way to pose the question of who is 'better off' -- ants, birds, humans, whatever. There are no standards of comparison. Keeping just to communication systems, one finds all sorts in the organic world, including humans (gesture systems, etc.). Human language is used for communication too, as is virtually everything that people do, but here too, comparisons seem useless. Some animal communication systems could be regarded as in some (not very meaningful) sense even 'richer' than natural language -- continuous, as contrasted with the discrete infinity of human language, an unusual property of organisms. During the lively eighteenth-century debates on whether apes have language, one proposal was that they do, but are smart enough to realize that if they manifested this capacity, humans would put them to work as slaves; so they prefer to keep quiet when people are around. I always liked that one.
QUESTION: You said that it is in the overall architectural design of the human brain that the language acquisition device has a particular place with some kind of an interface but this interface is lacking in the case of primates. Do you mean to say that even animals have a language device but since they don't possess an appropriate interface capability they are not able to use a language?
CHOMSKY: I did say that but as a kind of joke. I said it is a possibility (it is a theoretical possibility); there is nothing we know about the natural world that tells us that it is false that apes actually have a language faculty but have no access to it. That is possible but there is no reason to believe it. So, yes, there is a possibility and, maybe, some day we will discover it to be true but nobody expects it; it is more likely that they don't have a language faculty. Either way it is kind of hard to explain. There is no known explanation for most of the complex properties of organisms. People talk about Darwinian evolution and that sort of thing, but that doesn't really give you the answers beyond simple questions. Not just in the case of things like language. Take biological organisms like viruses -- very simple organisms. They have certain structural properties like polyhedral shells. To attribute that to 'natural selection' would be missing the point.
Or, take the mathematical series called the 'Fibonacci series'. It shows up all over the place in nature; nobody knows exactly why. If you take a sunflower and you look at the flower, it has spirals that go in different directions. The number of parts that appear in adjacent spirals are related to one another as successive terms in the Fibonacci series. You find that kind of thing all over nature; it is not well understood why. There is something about the physical world that forces certain kinds of structures to emerge under particular conditions. If you can't explain what a sunflower looks like, you are not likely to be able to explain what natural language looks like; it is way more complicated. So, the fact that we do not know how to give serious evolutionary explanation of this is not surprising; that is not often possible beyond simple cases.
QUESTION: Would you please elaborate your views upon the statement that language is innate but it also has an overlaid function both at the articulatory and the representational levels?
CHOMSKY: Well, the issue of innateness of language is a curious one. There is a huge literature arguing against the innateness of language; there's nothing defending the thesis. So the debate is kind of funny in that it is one-sided. Lots of people reject the proposal that language is innate but nobody ever answers them. The reason why nobody answers is that the arguments make no sense. There's no way to answer them.
To say that language is not innate is to say that there is no difference between my granddaughter, a rock and a rabbit. In other words, if you take a rock, a rabbit and my granddaughter and put them in a community where people are talking English, they'll all learn English. If people believe that, then they believe that language is not innate.
If they believe that there is a difference between my granddaughter, a rabbit and a rock, then they believe that language is innate. So people who are proposing that there is something debatable about the assumption that language is innate are just confused. So deeply confused that there is no way of answering their arguments. There is no doubt that language is an innate faculty.
To say 'language is innate' is to express the belief that some crucial and relevant internal nature differentiates my granddaughter from rocks, bees, cats and chimpanzees. We want to find out what this internal nature is. On current understanding, it is an expression of genes, which somehow yields a language faculty (and, for example, a well-placed bone of the inner ear -- in this case for mice as well). How is unknown, but that is true for vastly simpler questions as well. The informal statement that language is innate to humans means something like this. Similarly, we say that arms are innate to humans and wings to birds.
Now a question that could be asked is whether whatever is innate about language is specific to the language faculty or whether it is just some combination of the other aspects of the mind. That is an empirical question and there is no reason to be dogmatic about it; you look and you see. What we seem to find is that it is specific. There are properties of the language faculty, which are not found elsewhere, not only in the human mind, but in other biological organisms as far as we know.
For example, the most elementary property of the language faculty is the property of discrete infinity; you have six-word sentences, seven-word sentences but you don't have six-and-a-half-word sentences. Furthermore, there is no limit; you can have ten-word sentences, twenty-word sentences and so on indefinitely. That is the property of discrete infinity. This property is virtually unknown in the biological world. There are plenty of continuous systems, plenty of finite systems but try to find a system of discrete infinity! The only other one that anybody knows is the arithmetical capacity, which could well be some offshoot of the language faculty. The more you go on the more it seems true.
When you get to questions of the kind we've been discussing here, there seems to be no analogue elsewhere in the biological world down to the level of, maybe, DNA or some level where you are talking about biochemistry really. So it looks as though language is not only innate but highly specific in rather crucial respects. I take it that that is what is meant by the question of 'overlay'. It is an overlay to other things, it is something inserted into a system that has other properties. That is where empirical inquiry leads you. If somebody can think of some other explanation of the facts, it'll be interesting to hear it. But there's no other proposal, so there's nothing to discuss.
The problem is to discover to what extent properties of language and its use are specific to this system. Thus, we may ask whether the tongue and teeth are specifically adapted for language use in some way, or did they evolve independently of language. Opinions vary, though on some matters (say, the migration of the reptilian jawbone to the inner ear), the answers seem pretty clear. Some of the most respected scientists studying speech analysis and perception doubt that there are any specific adaptations of sensorimotor systems to language; others disagree.
As for the far harder problem of representational levels, there are also varying opinions and interesting ideas, but naturally far less is understood. Suppose, for example, that one believes that an expression of natural language is mapped to the 'language of thought' (LOT). Some properties of the expression must determine to which expression of LOT the linguistic expression is mapped. Which aspects of the interpretation of the expression are part of the language faculty, and which belong to the 'semantics of LOT'? There are speculations, but not much more.
QUESTION: What are the latest trends in semantics? Is it likely to develop into a science some day with its own units?
CHOMSKY: That is a really interesting question. ... We have to ask what semantics is. If semantics is what is meant by the tradition (say, Peirce or Frege or somebody like that), that is, if semantics is the relation between sound and thing, it may not exist. If semantics is the study of relations like agency, thematization, tense, event-structures and the place of arguments in them and so on and so forth, that is a rich subject but that is syntax; that is, it is all part of mental representations. It goes on independently of whether there is a world at all just like the study of phonological representations. This is mislabeled 'semantics'. It would be like taking phonology and deluding yourself into thinking that phonology is the study of the relation between phonetic units and the motion of molecules; it isn't, that is a separate study. Phonology is the study of mental representations that one assumes are close to those parts of the processing system that ultimately moves molecules around. Most of what's called 'semantics' is, in my opinion, syntax. It is the part of syntax that is presumably close to the interface system that involves the use of language. So there is that part of syntax and there certainly is pragmatics in some general sense of what you do with words and so on. But whether there is semantics in the more technical sense is an open question. I don't think there's any reason to believe that there is.
I think it goes back to the old and probably false assumption that there is a relation between words and things independently of circumstances of use.
QUESTION: By virtue of knowing the concept 'climb', does the child know that the concept needs an agent and a theme for its realization? Does the child learn that the concept of 'die' is alternatively realized in English as 'die' and 'kick the bucket'? The innate conceptual and computational components are presumably different modules; does linguistic experience trigger some kind of interaction between them with the result that a predicate-argument structure is generated which is then converted into familiar lexically-filled syntactic representation?
CHOMSKY: These questions may be referring to a book of mine of about ten years ago in which I said that the child has a repertoire of concepts as part of its biological endowment and simply has to learn that a particular concept is realized in a particular way in the language. So the child has a concept, say, 'climb' in some abstract sense with all its weird properties and has to learn that it is pronounced 'climb', not some other thing. Jerry Fodor's important work for many years is relevant here, along with Ray Jackendoff's and much else. These are all perfectly reasonable questions. You can have various ideas about them; there isn't a lot of understanding. I could tell you what my own suspicion is about these questions but they are research topics.
There is overwhelming reason to believe that concepts like, say, 'climb', 'chase', 'run', 'tree' and 'book' and so on are fundamentally fixed. They have extremely complex properties when you look at them. This is not recognized in traditional lexicography. When you read the huge Oxford English Dictionary (the one you read with a magnifying glass), you may think that you are getting the definition of a word but you're not. All you are getting is a few hints and then your innate knowledge is filling in all the details and you end up knowing what the word means. As soon as you try to spell out what's taken for granted in the lexicon, you find that these concepts are incredibly complex. Actually that was understood a couple of hundred years ago. There is a tradition roughly from Hobbes through Hume which investigated questions like these with some sophistication. I think it was the tradition which should be expanded; it had Aristotelian origins in fact and interesting parallels to seventeenth-century neo-Platonism. But when you work these things out, it turns out that the concepts are very complex, which means that they've got to basically be there and then they get kind of triggered and you find out what sounds are associated with them. But then come these questions: how much of this is variable? How much is fixed? Is the agent-theme property fixed or is it variable? That is a research topic. In some of the cases we know; for example, for 'die' and 'kick the bucket' obviously that is just artificially imposed. But for the other questions one doesn't really know. Are the computational and conceptual components different modules? Really, that is not known very well either. That is the traditional question: can you have thought without language? If you ask how much we know about that topic, the answer is 'not much'. What we know is by introspection.
Now what seems to me obvious by introspection is that I can think without language. In fact, very often, I seem to be thinking and finding it hard to articulate what I am thinking. It is a very common experience at least for me and I suppose for everybody to try to express something, to say it and to realize that is not what I meant and then to try to say it some other way and maybe come closer to what you meant; then somebody helps you out and you say it in yet another way. That is a fairly common experience and it is pretty hard to make sense of that experience without assuming that you think without language. You think and then you try to find a way to articulate what you think and sometimes you can't do it at all; you just can't explain to somebody what you think. Sometimes you make judgments about things very fast, unconsciously. If somebody asks you how you made the judgment, it is often extremely hard to explain. Experiences like that seem to indicate that we can and do think without language and, if you are thinking, then presumably there's some kind of conceptual structure there. The question of how this is related to language is just another research topic which, at this point, can barely be touched; but it is potentially important and interesting.
QUESTION: Left-to-right ordering of syntactic constituents has been accorded a more central and integrated role in the Minimalist Program than earlier. Is it inherently central to the architecture of the language faculty or is it more of an interface-constraint imposed by the sensorimotor and conceptual-intentional ordering considerations?
CHOMSKY: This is an interesting research topic. My own feeling is that there is no left-to-right ordering. If you look at the structure of the generative system (the system that takes lexical items and puts them together into bigger ones, performs operations on them and ends up giving semantic representations), if you take a look at these operations all the way down to the interface of the conceptual-intentional systems, it seems to have no left-to-right ordering. In fact it may be that it has no ordering at all; it just has hierarchical relations.
On the other hand, the sound has a left-to-right ordering. My assumption is that that is imposed by the sensorimotor systems. Our sensorimotor systems are limited; they are forced to produce things from left to right, through time. So, somewhere along the line, this unordered system which has only hierarchy (and no order), gets an order imposed on it in order to meet conditions of the sensorimotor interface.
Notice that they are by no means necessary. In fact there are other organisms around that don't have that property. Take dolphins which have huge brains relative to size, not unlike humans. Dolphins have a complicated communication system. It comes out of their noses (dolphins make all kinds of funny noises); partly it is kind of sonar (they have to know where they are, if they are running into something), but partly it is apparently communication. Some sub-species of dolphins apparently can do it simultaneously through both sides of their nose. That means they have a richer form of communication than we have; they can produce sounds in parallel -- two-dimensional production. That is certainly feasible; then these sounds don't have left-to-right order. They have parallel outputs, maybe left to right within each but not all left to right. We don't have that; we have a single channel.
Incidentally, if you look at sign language, it doesn't have a single channel. It has multiple channels, but articulated language does have a single channel. That is a limitation of our sensorimotor apparatus and it forces things to be ordered. If we had the ability to communicate by telepathy, let's say (so that we didn't have to make sounds), there might be no word ordering in language at all. [Inaudible intervention from the audience] Oh, sure; that is absolutely true but that is a different question. Remember that the generation of an expression is an abstract operation; it is not the same as the production of an expression -- that is a totally different thing. When you produce an expression, of course, it is temporal because you begin at a certain point and then you do the next thing and you do the next thing. You may make all kinds of changes along the way, but that is not the same question. When you make changes you are just regenerating some new thing; but generation and production are completely different operations. They are obviously related in that the performance systems have to access the knowledge system; so they are of course related but they are different operations. To say that generation has no order is independent of the fact that production has an order because we do things through time. That is not under discussion. The question is, is there order in the abstract expression that provides the information? I think the answer is 'no' except close to the point of sensorimotor interface. But these are research questions; you can't be dogmatic about them.