While Noam Chomsky’s importance and influence on contemporary philosophy is undeniable, Georges Rey, professor of philosophy at the University of Maryland, is troubled by Chomsky’s is unwillingness to accept all the implications of his own theories of language and mind. Specifically, Rey argues that Chomsky doesn’t appreciate his theories’ commitment to intentionality, or the way in which mental states are “about” things – specifically syllables, words, and sentences with elaborate syntactic structure.
Rey is currently on a Fulbright fellowship from the United States, at the Center for the Study of Mind in Nature (CSMN), writing his book Chomsky and Philosophy – Sifting the Insights from the Excess. In this interview Rey explains why he thinks Chomsky’s dismissal of various “mind-body” problems, particularly those surrounding intentionality, is an unnecessary excess, but that philosophers who are understandably put off by this excess shouldn’t let it get in the way of an appreciation of the deep insights a Chomskyan approach to linguistics has to offer.
Why did you come to Oslo to write a book on Chomsky?
Mostly because I have colleagues here that I like talking to a lot – Nicholas Allott and Carsten Hansen in particular. I have visited here in the past and it has been very productive. It has been fantastic this year, partly because Allott has himself been finishing a book on Chomsky. Allott and I disagree, but very fruitfully, and so we have been bouncing off each other. I have been giving criticisms on his manuscript and he has given criticisms of mine. He is wonderfully open-minded. While he has a lot of admiration for Chomsky, he is not as rigidly ideological as many Chomskyan can be.
Would you say that you approach language from a philosophy of mind perspective?
I am interested in linguistics in a way that Chomsky himself is, namely as a part of psychology. He thinks of linguistics as being a theory of our underlying psychological competence, and that is really an interesting case for the study of the philosophy of mind.
You often argue in favor of the Computational-Representational Theory of Thought (CRTT) as being compatible with Chomskyan views on language and the mind. How does that theory relate to psychology?
CRTT is the view that mental processes should be understood as computations, like those in a modern computer, over representations of material in the domain of the mental process. Thus, a computational-representational theory of linguistic competence would characterize that competence in terms of computations over representations of words of natural language and the syntactic structures in which they figure. The CRTT is really a strategy that emerged from the work of Alan Turing and the growth of computer science. It seems to afford a general and intelligible way of answering Descartes’ problem of how it is that a material thing can think. Well here is how: It can perform computations on representations, computations defined à la Turing, representations that are to be explained by a theory of intentionality.
Is intentionality meant to equal representation? What is the relationship between the two?
Well, intentionality is just the theory of the content of representations, specifically the content of the representations that figure in the “computational/representational theory” in which Chomsky couches his theory. Suppose you try to understand how people are able to see objects. You might pursue the course that was originally prescribed by David Marr in the 80s in his book Vision, in which you imagine there are representations of very simple excitations on the retina, intensity gradients on the retina, and there is a process of computation which takes you to the detection of edges, detection of surfaces, and finally to the detection of objects in two-and-a-half-dimensional space, and it is built up step by step until you get to the recognition of, say, a cube or an animal or a face. And the hope is that if we can figure out the right computations on what is represented in different stages we can get a successful explanation of visual experience, e.g. why we see things accurately when we do, and why certain illusions appear under certain conditions.
Does this apply to both language and perception?
Yes. These are the two areas that have been the most promising, and there has been deep research done in both. There is also work in other areas. For example, I am very impressed by the work of Daniel Kahneman’s systematic efforts to understand human reasoning, which also requires a CRTT in principle, although it can be a while before we work out the details. CRTT can also be applied to further areas such as: general audition, in theories of children’s understanding of objects, causality, morals and other minds, logic and decision making, e.g., issues surrounding attitudes towards risk, discounting of future satisfaction, revising preferences in light of frustration, and so on.
Chomsky is in general dismissive of any talk about intentionality. Why do you think that is?
This is a very vexed and complicated topic, and I am not sure what to say quickly except that he makes frequent pronouncements against intentionality. He says in New Horizons in the Study of Language and Mind (2000) that intentionality has no place in serious science. And he claims his own theory is not committed to it. He is reasonably opposed to various theories of intentionality that philosophers have proposed, and thinks the whole topic is not really amenable to scientific treatment. He doesn’t appreciate the full range of theories philosophers have proposed. I think this is likely mistaken, and certainly confused. I think he doesn’t think about this very seriously or nearly carefully enough. He is impatient; he wants to go on with his own work. He does his fantastic linguistics and his politics, but he should not make these throwaway remarks about intentionality in his theory, because of the misleading – indeed, I think, pernicious – influence they will have, encouraging, ironically enough, the very “methodological dualists” (who also don’t think intentionality belongs in science) that he elsewhere rightly opposes!
One specific way in which he is confused is historically odd and unfortunate. In the 19th century, Franz Brentano was the philosopher who in a modern way called attention to what medieval philosophers had called “intentionality.” This is a very unfortunate word, which has little to do with the ordinary use to mean something like “deliberateness.” Rather it has to do with the fact the thought are “about” or “directed upon” their objects, e.g., tables, chairs, people, words, even non-existent things like ghosts! And that is really weird, because how can they be about something that does not exist? If I can’t a kick a non-existent thing such as a ghost; then how can I think about such a “thing”? This was a real puzzle Brentano struggled with, as have philosophers since then. It is a traditional philosophical puzzle. You can trace aspects of this problem all the way back to Plato. There are other problems as well, which I set out in chapter two of my book, Contemporary Philosophy of Mind (1997).
A common response to this problem that has been influential for the last century is to say we should really focus on representation when it is about real things, and that representations of unreal things should be somehow derivative, “secondary,” or maybe not representations at all. This move, which is customarily referred to as “externalism,” began with traditional empiricists, such as Locke and Hume, was resurrected by Bertrand Russell, and has been energetically pursued in recent years by philosophers such as Saul Kripke, Hilary Putnam and theorists of so-called “direct reference” and has become very influential.
Now what is odd is that Chomsky notices this in the 80s and 90s, and gets impatient with it, because what he thinks is interesting in linguistics and semantics are issues more purely “internal” to a mind. He is interested in a kind of semantics that has to do with what he therefore calls “I-language” and with purely internal psychological states, not with relations to the external world, which he thinks are irrelevant to linguistic theory. “I-language” is the system internal to a person that is responsible for their linguistic competence, as opposed to the “E(xternal)-languages,” like English and French, or their dialects (or even idiolects). Chomskyans are skeptical to whether much scientific sense can be made of these latter notions. (Note also that they think of language as manifested in what every child speaks, not in what only a minority of the world can read or write.)
I think he is right about his internalist views, or I am at least maybe 95% on board with that. (I am what I call a weak internalist, who allows that some mental states may have their content by virtue of some or other relation to some or other external phenomena. It makes a difference in only about 5% of the issues.) But Chomsky read externalism as if it were the last word on the issue of intentionality. He sees intentionality itself as being committed to relations between a brain and an external object. And since he doesn’t think those relations obtain for the linguistics he is pursuing, intentionality generally has no place in linguistics. Because he thinks intentionality must be externalist, he dismisses it entirely, in my mind throwing out the baby with the bathwater.
I want to say: Hold on! I don’t buy into the theories of external representations either. I am a Brentanian; I think that we often think about non-existent things, indeed, that it might be the default case (Brentano thought the “inexistent” cases were paradigmatic). So I am with Chomsky entirely here, but I think that it is still intentionality. But this is where things get confused and hard to sort out.
So he dismisses intentionality on the ground that he takes it to be tied to extensionalist views?
He thinks that trying to give a theory of the relation between our brain and the external world is a fool’s errand. It is too complicated and it is a mess. We interact with the world in any variety of ways, and he doesn’t think there is going to be a clear theory of that relationship. I think he is right about that. And, since he thinks that intentionality is committed to externalism, there is no place for it in serious science. I come in and say: “No, but we should give up this notion of intentionality that has to have to do with external objects.” In fact I agree with Brentano (as Chomsky should agree too) that it is a characteristic of intentionality that it can be about objects that might not exist, and that is the default case. Unlike Brentano, I think we sometimes do refer to real objects (Brentano got into some problems about this). Sometimes we actually think about real objects and refer to them, and that is fine, but sometimes we don’t, and I think that Chomsky is right that the cases where we don’t are the theoretically deeper cases, and that we will explain the times we succeed in referring to real objects as being based upon this deeper ability.
Could you just give a small indication of what such an explanation would look like?
So we might start with the idea that we have an internally generated concept of triangles, which may in fact never be actually realized in the world (as every high school student learns, there couldn’t possibly be genuine Euclidean triangles). This concept is evoked by certain patterns of stimuli, e.g., roughly triangular shapes in the environment, to which we then roughly refer with the concept. That is a controversial claim; there are many people who disagree with it, such as Tyler Burge for example. It is an interesting philosophical issue. I am willing to take Chomsky’s side on this, oddly enough; the default case is the internalist case when you don’t have an external object you are thinking about.
What prompted you to write a book on Chomsky at this point?
In the 90s I was going to write a different book on the cognitive science revolution and the computational theory of mind, which seemed to me then (as it does now) the best thing since sliced bread, and I was going to use Chomsky’s work as an example. But then I started reading that he was denying that kind of theory, and I thought, Good grief, what is happening!? I then got invited to write a piece in the Chomsky and his Critics volume. That was about 2001. In that article I tried to spell out a few of the things discussed here. And I had a long exchange with Chomsky about this, about a hundred pages of email. It was by and large fruitless because he simply got defensive and stone-walled. He didn’t begin to get the idea nor, really, even the problem. It was a worthless exchange. Much better defenders of Chomsky’s view are John Collins and David Adger, particularly in recent work they presented at a conference we had in Trondheim last month.
A surprising number of people are not interested in these discussions. This is a little odd; it has to do with an odd sociology. Chomskyans tend to be impatient with it, “don’t bother with it, let’s go on with our work,” and I understand that. On the other hand philosophers are exasperated with Chomsky; they are bored with him partly because they share my view that he is often careless and impatient, so they just don’t pay much attention to it. So I am in a funny state of being somewhat alone in this pursuit, but not entirely: I have these arguments with Collins, Adger and, again, Nick Allott and Carsten Hansen here in Oslo.
Would you agree that Chomsky puts too heavy constraints on philosophy as a form of science?
No, I agree with him on that point. I don’t respect disciplinary boundaries. I just think there are questions out there. There are really interesting questions about intentionality, the nature of the mind, the nature of thought, perception and so forth; may the best people, whatever discipline they may be, think about it and come up with insights! So I agree with Chomsky that philosophers should pay a lot of attention to science. I don’t think philosophy is an autonomous discipline, at least not in this area. There might be parts of philosophy that are. Philosophy is a huge mess of stuff, and I don’t see the point of making sweeping statements or generalizations. What I am interested in are general aspects of the theory of mind as very deeply puzzling in ways that has been called philosophical. There is the unfortunate fact that, for example, historically and sociologically, the only people who have noticed that there is a problem of intentionality and have studied it are philosophers. It is a fact that has to do with oddities in history that are hard to explain. So I can see why people say, “Only you philosophers are interested in intentionality,” but I think scientists should be equally interested. They just haven’t noticed intentionality because it doesn’t make a difference immediately to their experiments. But it will in time.
There seems to be a connection between Chomsky’s views of intentionality and his defense of what he calls ”methodological naturalism.” Could you explain this?
Methodological naturalism I am all in favor of. This is roughly the view that the investigation of the mind should be conducted in the same way as the investigation of any other part of nature. What is exasperating is Chomsky’s exclusion of intentionality from that investigation. He is responding to the externalist development that is not the only development in the history of this subject, and so he tends to dismiss it as not a part of science. Ironically enough, he does this in much the same way as the very philosophers, such as Quine and Davidson, the very methodological dualists he opposes. So he turns out to be a methodological dualist at the end of the day, he just does not notice it. I can’t begin to tell you how exasperating it is. I think of myself as trying to save Chomsky from his own foolishness, from his rashness. He just hasn’t thought out what his view should be, so he says things that are incoherent. For example: The phonologists talk about a feature of speech called voicing. And Chomsky wrote series of articles about this at the same time. He says that +voice, or voicing, i.e. the property of voicing a phoneme, such as /gɑ/, both refers to something in your brain and it is a feature of a sound. But, wait a minute − it can’t be both of those things. A feature in your brain is one thing, and a feature of a sound is something different.
So there are several candidates for what a word is. How about the auditory way in which it is perceived by the ear and brain? Or perhaps it is a feature of the articulation.
Exactly. One could also say that. But the articulatory thing and the sound are connected in a way that one can unify. That is at least what I think, and phonologists do too. The idea that +voice is something that is in your brain is what philosophers call a “use-mention confusion. A confusion of, say, a property, like +voice, with a representation of that property, e.g., the expression “+voice.” It’s like confusing dogs with the word “dogs.” I think what is in your brain is a representation of +voice, in the same way as when you see an illusory Kanizsa triangle (just google the term for examples). There is no triangle there, and there isn’t one in your brain either. There is no triangle anywhere, it just looks like there is, and I would say the same thing with words and so forth. Words are a kind of illusion we enjoy as a result of acoustic [blasts] causing us to classify them in terms of standard linguistic categories, nouns, verbs etc. Words are heard as sounds made by someone in the external world, not as features of anyone’s brain. What is in my brain is a representation of that sound as a word. The word itself isn’t there. Anyway, for various accidental reasons, the use/mention confusion is particularly easy to get away with in linguistics. Saying a word is in the brain doesn’t sound so silly as it would if you were to say there’s a white triangle there. It doesn’t sound so silly when you commit this confusion in linguistics, and so it can go by unnoticed. A lot of my work is put into trying to show that it is there. If you read carefully you will see linguists are often making the same mistake, but it is hard to point out for various reasons.
A keyword here seems to be interdisciplinarity, doesn’t it?
Well, not really for me. I mean, it is true; it is not that what you say is false. It is just that I don’t like to think of it that way because I don’t like to think of the relevant disciplines here as being in any important way different from one another.
You would not like to have them delineated?
Right, I think that is completely artificial. The delineation between disciplines is largely an historical and sociological fact you have to live with. Universities are organized this way and it is not going to change very quickly. What I like particularly about the place I work, at Maryland, and it is also true of CSMN, is that you have a lot of people who are interested in philosophy who are equally interested in science. I can’t distinguish when I do the one or the other. It is funny to call it interdisciplinary because I don’t really think of the disciplines as being separate. I’m interested in the nature of the mind, in the very general nature of mental processes. Some philosophers have drawn attention to very significant problems about understanding them, and psychologists have drawn attention to lots of other problems. Let us put all of what they say together and try to figure it all out. For example, I find talking to Nick Allot tremendously useful. Now, he was not trained as a philosopher, but he has an incredible native philosophical talent. But his training is entirely in linguistics and pragmatics. He is a good example. I don’t care whether he is a professional philosopher, we discuss the relevant issues together, and that is what I care about.
But, are philosophers not trying to come up with necessary truths whereas scientists are trying to come up with contingent ones?
Well, that has been disputed I think quite well since Kripke and Putnam who have pointed out that scientists come up with necessary truths as well, such as “water is H2O” and that “polio is the activation of a certain virus,” and those are necessary truths. I don’t see why necessary truths should be the domain of philosophers alone.
The way I am inclined to think of necessary truths, is as an extension of a Quinean conception. Ironically, Quine himself wouldn’t have bought it, but I think it is right. Modality, thinking of what might have happened even though it didn’t, thinking of possible worlds, is just another way of describing this world. It just turns out to be essential to our understanding of the world. Thus, it’s essential to chemistry to know what would have happened if some salt had been placed in water. Such “counterfactuals” are part and parcel of serious scientific understanding. So modality is here to stay and with it probably some necessary truths: that is just part of science. At least the necessary truths that interest me (e.g., analyses of chemical elements, diseases or mental states) indicate necessary structures in the world.
Returning to where we were, you said that words are not features of sounds, and neither are they in our brain. What we have are representations of them. Could you say something about how such an internalist conception can work?
I think along the lines of the example of the Kanizsa triangle, and what phonologists say about acoustics, and the relations between acoustics and phonology. I think Kanizsa triangles do not exist, and words do not exist. I do not utter any words right now, and that is OK, because it is enough that I intend to utter a word and produce a lot of acoustic waves. And it just happens that we are nicely and exquisitely designed, so that when we get those waves to our ears we are able to construct what was intended to be uttered, at least on the whole. However, you won’t find anything in those acoustic waves that corresponds to the words we take ourselves to utter and hear.
It is enough that we believe in words?
That is roughly a part of the view I am defending like a Brentanian: that words don’t exist. They are, as it were, projections onto the world of the content of our internal mental states.
What about words as existing in the brain – you wouldn’t call them words? Concepts?
They are not words of natural language. There is a language of thought. Words in that computational language are representations in the brain, but I think the language of thought is a separate thing. Something very much like concepts, yes.
I should perhaps mention that here is a point at which I get maybe a bit outrageous. I don’t only have this view about Kanizsa triangles, words, and linguistic entities. I also think it is true of sensations such as pain, tickles, and “qualia,” like the “taste of beer.” I don’t think they exist. All you have are representations of them in your brain. Your brain is one very complicated, intricate representational system. The difference between beliefs, desires, perceptions and sensations are differences in the way different representations in these different domains are computed.
So if we took away our representations we would be left with nothing?
Well, you’d be left with your brain and body and the rest of the external world. But, right, so far as our mental life goes, there’d be none left. I don’t think qualia exist at all. There are representations of qualia, but – and this is the crucial thing – representations of qualia are just as bad as the qualia themselves. The representations of pain can be just as horrible in your experience, you will hate the state every bit as much as the pain itself. I think that qualia are idle; they do not do any work. That is controversial. That is pushing the view further than most. Denying the existence of words and sentences is one thing – here Chomsky and I agree. Denying the existence of pains and qualia? I suspect even he would blanch at that. But fortunately that’s one of the few issues he doesn’t address, and I won’t, at least not in my book about him.
Finally, when will we have the pleasure to read your book on Chomsky?
I am hoping to finish it by June or July this year. I am working hard at it as we speak.