The Alphaville Herald

Herald Literary Suppository: Interview with Noam Chomsky on Mind, Representation, and Reality

by Alphaville Herald on 06/11/05 at 7:50 pm

Say what? An interview with Noam Chomsky? In the Herald? WTF??? Yes, dear readers, it’s true: for too long I have worried that the steady diet of griefer news, Linden screw-ups (known as “brownies” in the post-Katrina era) and nekkid avatars was turning all of you into drooling brain-dead pixel-zombies, seeking your daily dose of eye candy and scandal. But now I plan a series of uplifting, mind exercising posts and works of virtual literature, so that you can all become, if not exactly like me, then a shade bit better than you are. Think of this as The Herald Literary Suppository.

This is not the usual fare for Heraldites, but it does go to some issues that are near and dear to us: representation, reality, meaning, cool stuff. And the guy asking questions is one Peter Ludlow — the Phurst Philosopher of our little corner of the Metaverse. Enjoy.

Chomsky on Mind, Representation, and Reality
Interviewed by Peter Ludlow,
Interview conducted at SUNY Stony Brook
March 2003

Peter Ludow: Noam Chomsky is the principal architect of generative linguistics and has been the driving force in recent evolutions of the theory, including in the development of the principals and parameters framework, which one commentator has described as, quote, “the first really novel approach to language in the last two and a half thousand years.” Professor Chomsky has also been a key figure in development of the cognitive sciences generally. Indeed, the most important figure in the development according to Harvard’s Howard Gardner [The Mind’s New Science]. He is perhaps most famous for his writings and lectures on international politics and in media theory. It is less well known, however, that professor Chomsky has been an important figure in Anglo-American philosophy over the last fifty years and that he has been a key interlocutor with all the leading philosophers of language and mind within that tradition. He’s had lively exchanges with, for example, Willard Van Orman Quine, Hilary Putnam, Donald Davidson, Saul Kripke, John Searle, Michael Dummett, and now a new generation of philosophers. And I guess the first question has to be the following: Given everything that’s on your plate, why do you bother with the philosophers?

Noam Chomsky: Oh, that’s the most fascinating topic of all. It’s what I grew up with and still am obsessed with.

PL: Why do you think other linguists are not, or, in general other scientists don’t seem to be as engaged with the philosophical community?

NC: Well, scientists tend, I think, to be involved in their own technical problems and often don’t think much about what it’s about. So, for many years, I’ve taught courses in these questions and would have students read, say, Quine’s papers in which he argues that everything they are doing is completely called folly. He gives arguments that you can’t possibly do it this way. The students will read the papers, and say, “kind of interesting,” and then go back to doing exactly the same work, but what they have just read is folly and can’t be done and so on and so forth. And I think that is not unusual. If you read the scientific literature, there’s a lot of expressed contempt for what philosophers have to say about this. So you’ll read somebody working on the neurophysiology of consciousness—a hot topic—and there will be an obligatory first couple of paragraphs saying, “well, the philosophers have mulled this up for centuries, but now we’ll show how it’s done,” but with no engagement in the arguments that have been given or the thinking behind it.

PL: Does it trouble you that, say, most other linguists don’t engage philosophers on these issues?

NC: It troubles me, not so much, I mean it’s connected to a deeper problem. I think it should trouble us that we’re not thinking about what we’re up to, and those questions happen to be the domain of what philosophers pay attention to.

PL: Right. Let’s turn to some of these philosophers and some of the things that they have said. And the first one I want to look at is the Harvard philosopher, Hilary Putnam. In the mid-seventies, he wrote a paper called “The Meaning of ‘Meaning’,” which is very influential, and I guess the key slogan that came out of that is that “meanings just ain’t in the head.” Agree? Disagree? Or is that just incoherent?

NC: It’s incoherent without further explanation. First, you have to tell us what meanings are. You can’t decide whether they are in the head are not until we know what they are. If by meanings, we have in mind what people have in mind when they are using the word in English, then, sure, they’re not in the head. They’re all over the place. Talk about the meaning of life, it’s not in the head. If you have some more technical notion in mind, you got to explain what it is. If the technical … and there it’s a matter of choice. You can define the technical notions so it’s in the head or so that it’s not in the head. When you define technical notions, you have a choice. The way it’s done in the sciences, and the way it ought to be done, is you define a technical notion in the context of an explanatory theory. You don’t just define a technical notion out in space. So, let’s ask what’s the explanatory theory in which we’re going to give a technical notion, which will pronounce meaning or pronounce it some other way. And then we’ll ask whether in that theory is a sensible theory and does it place this technical concept in the head or not. But all that extra work hadn’t been done.

PL: Why can’t we say there’s just a pre-theoretical notion of meaning and what we want to do is try and elucidate or precisify it in some way or other?

NC: Yeah, you could do that, but there’s a million ways of doing it. So what about the meaning of life, for example? That’s part of the pre-theoretical notion.

PL: Is it?

NC: Is it? Sure it is. I mean, I thought people talk about what’s the meaning of life all the time.

PL: And you think that’s the same notion of meaning as when you say what’s the meaning of the word?

NC: No, there’s thousands … it’s a huge range of things. What did he mean by saying so and so usually has to do with what was his intention or something like that. So, part of the pre-theoretical meaning has to do with people’s intentions. In fact, when you talk about the meaning of the word, that’s a rather English-specific locution. And we could focus on that if we like, and if you want, you can try to clarify it, but again to try and clarify a pre-theoretical notion makes sense within the framework of an effort to understand something, an explanatory account.

PL: Well, suppose we go with the following attempt to clarify, or at least introduction of a piece of technical terminology. Philosophers will talk about the content of a mental state or the content of some sort of expression. Is that helpful?

NC: No, because “content” is a technical notion. It’s not being used in the sense of the ordinary concept, “content,” and it’s used by now in the last sense, but not in fact in Kripke and others. It’s typically used to refer to something out there. So, the content of our expression is not in the head by definition because we’ve defined it to be not in the head. Well, then the question is, well, does that have anything to do with our notions of… does that have anything to do with our understanding of the way language works and is used. Ok, that’s the question. You can certainly think it has very little to do with it.

PL: But, suppose we thought of the enterprise being sketched in the following way. There are two ways in which you might study primates, for example. You might study primate anatomy, in which case you’re just concerned about bones and muscles and tendons and so forth. And then you might also study primate ecology, in which case you’re studying the relation between the primate and its environment. And I take it that what these philosophers are talking about here when they talk about external content is that they’re saying a part of talk about meaning has to be analyzed in terms of thinking about the human organism in relation to its environment, right? So…

NC: It has to if you want to study the relation between the human organism and its environment. That’s fine. There’s nothing wrong with studying the human organism in its environment. That’s what sociologists are doing all the time. But if you want to do it seriously, you ask… first of all, let’s go back to primates. When you study primates and their ecology, you don’t just look at the anatomy of primates. You look at the physiology of primates, what you might call mental processes of primates. The way they seem to be interpreting the world, the perceptual abilities of primates. In fact, what you try to do is exploit to the extent that you can what we know about an individual ape […]. You want to know what we understand about that ape, and that information we will bring to bear in the study of how the ape interacts with other apes and with the rest of the environment and so on. That’s the way it’s done with primates, or with ants, or anyone else, and I think that’s the way it ought to be done with humans. So we should ask, ok, what do we understand about… the internal nature of the creature that’s doing all these things, and to the extent that we understand something about that, we can ask sensible questions about how it interacts with the external world, with other people and so on and so forth.

PL: Do you think that could potentially inform the sort of internalist investigation as well?

NC: Sure, it goes the other way. I mean, there is no order. You can study the sociology of ape communities, and that’ll tell you something, could tell you something important about the internal thinking processes of apes, or even their internal anatomy.

PL: And vice versa, presumably.

NC: Yeah, sure, science goes all ways.

PL: Here’s another technical term that has been used: “reference.” You consider that to be incoherent as well? Or…

NC: Well, until it’s explained, it’s incoherent. Once it’s explained, then it becomes non-incoherent.

PL: Well, you see one thing that I’ve never quite understood is exactly what the problem is with reference because I can say look, ok, now I’m referring to this coffee cup.

NC: No, that’s different. That’s quite different. You’re talking about an action of referring. That’s a common sense notion. It’s part of English and every language I know of. There’s some way to talk about such actions, but what philosophers have introduced is a different notion, a notion that’s supposed to hold between a linguistic entity and something in the world. Now, that’s not referring. Referring is an act that people do. People stressed and pointed this out fifty years ago…

PL: Yeah, well, this is a point that goes back to Peter F. Strawson, Right?

NC: … but I mean this goes way back, and it’s just a fact! And if you want to make up a technical term that you pronounce the same way, first of all, it’s questionable that you should pronounce it the same way because it’s misleading. But if you do, let’s at least keep clear that it’s a technical term, kind of like physicists use “energy.” They don’t mean it the way we do in ordinary language. They mean it the way they say they mean it. So let’s take the technical term, “refer,” or “denote,” or whatever you want to call it and tell us what it’s supposed to… tell us what you mean by it and tell us what explanatory theory it enters into. So when a physicist defines “energy,” you’re not interested in a definition. You’re interested in the set of principles and assumptions and problems and so on within which that technical concept is introduced, and then you look and you see how good the theory is. Well, here you don’t have any theory. And we don’t even have an explanation about what the technical notion is, so it’s impossible to talk about it.

PL: Yeah, I think it might be a mistake to think of it as a technical notion. I mean, I think it’s just a sort of precisified way of describing certain core level facts, right?

NC: Right.

PL: So for example, I might say, well, once you understand that people refer to things and so forth, it’s not a hard step to say, well, ok […] certain names are canonically used to refer to certain things and certain individuals. So the name ‘Noam Chomsky’, I can ask the audience here what person in this room does that term refer to…

NC: Well, you could ask what do you use that term for when you refer to something in the world. That you can ask. When you ask what the term refers to, you are assuming that there is a relation between terms and things, and they have to explain what that relation is because that’s a new one. We don’t have that in ordinary language. In fact, I think it’s leading us off into a wrong picture, as Wittgenstein would’ve called it. It’s leading us to a picture of language, which is… we know what it’s based on. I mean, it’s based on Frege’s theory of arithmetic, and that’s just the wrong picture for language.

PL: Let me… I want to change directions briefly. We may actually end up being sucked back into this issue. Donald Davidson in a paper called “A Nice Derangement of Epitaphs” said, or basically argued that there is no such thing as language, and I’ve always sort of thought that you probably—you may not want to admit it—but on some level or another, you probably agree with that, right?

NC: No. In fact, I think he ended up contradicting himself. I think there’s some discussion of it in here somewhere. If you look at the end of the paper, it turns out that he’s presupposing that there isn’t a notion of language in the technical sense, the technical sense of an internal generative procedure that relates sounds and meanings and so on. He says there is no notion of language in another sense, the sense of some community, property, or whatever. Well, ok, first of all I don’t even think that’s true. It’s not a scientifically usable sense, but I think the paper is just rifled with confusions that I’ve written about.

PL: So you think there are such things as languages?

NC: Yeah, yeah…

PL: In an informal sense?

NC: …like there’s such things as the meaning of life. And I understand it when people ask what’s the meaning of life. So yeah, there’s such a thing as the meaning of life. There’s such a thing as the financial crisis in Argentina, and there are all kinds of things in the world. But if you want to proceed to understand what you and I are doing, those notions just don’t help. You’ve got to look at it differently, the way we look at primates with other primates, in fact.

PL: Well, let me read a quote from something you wrote recently, ok. You say, “I doubt that people think that among the constituents of the world are entities that are simultaneously abstract and concrete like books and banks, or that have the amalgam of properties we discover when we explore the meanings of even the simplest words like river, person, city, etc.”

NC: Yeah, that’s a technical question. It’s a question if you try to figure out what a person’s folk science is, how people think the world is actually constituted of entities, which is not, “do I talk about books?” Of course we talk about books, we talk about the meaning of life, and so on. But if you ask people, well, how do you think the world really works, that’s a problem with ethno-science. Like you go to some other community and you try to figure out what’s their idea about how the world works. Like maybe the classical Greeks thought that Apollo pulls the sun through the sky or something. That’s their folk scientific picture of how the world works. That’s a hard topic. You can’t just do armchair philosophy about it. That’s why ethno-scientists have to work. And when they work, what they find… I think if they worked on people like us instead of just talking about us in the common room, they would discover that our folk science, yours and mine, does not include entities that are simultaneously abstract and concrete and does not include entities like the meaning of life. That doesn’t mean we can’t talk about them. Sure, we talk about them all the time, but we don’t—at least I don’t—and I presume other people don’t think of them as constituents of the way the world operates. We don’t do that when we are talking to each other informally.

PL: Now, that sounds a little bit more moderate than what you’ve said elsewhere. Here’s a passage from, I believe this is from New Horizons, where you say, “In the domain where questions of realism arise in a serious way, in the context of the search for laws of nature, objects are not conceived from the peculiar perspectives provided by the concepts of common sense.”

NC: That’s absolutely right. There are several different enterprises that you have to distinguish here. I don’t think it’s more or less moderate. It’s about a different topic. When you’re trying to understand something about the nature of the world, you and I—anybody—you start with some kind of what’s called folk science. Almost every society we know has some picture of the way the world works which is more or less commonly shared. If you try to do this more reflectively and carefully and bringing in other criteria and probably bringing in other cognitive faculties—we don’t know that for sure, but I suspect it—then it becomes the enterprise of science. Which is a different enterprise, and a peculiar one. It’s not folk science. It’s science. That works in other ways. This comment has to do with our culture in which the enterprise of science is understood…

PL: Right.

NC: … our intellectual culture, and in that, when we try to find out how the world works, we discard the concepts of common sense very quickly.

PL: But it sounds to me here like what you’re saying is that the only things that are real, right, are the things that science tells us are real. So…

NC: Well…

PL: It sounds like what you’re saying here is that, well, this table isn’t real, but maybe quarks…

NC: “Real” is an honorific term. You can use it any way you like. I mean, to say, if I say something is true, and then I add, well, it’s the real truth, I’m not saying they are two different kinds of truths, the truth and the real truth. I’m just emphasizing what I said, and the term “real” is basically used honorifically. So yeah, you can use it honorifically in various ways. If you’re trying to find out the way the world works, to really understand it in the manner of the sciences, then we’d very quickly give up commonsense notions. If we’re carrying out folk science, less reflectively, probably using different cognitive faculties, we also give up commonsense notions, but in different ways.

PL: Well, you can say a lot of them… I mean, claims about something being honorific, like ‘real’ being honorific. Some philosophers have argued that terms like “rational” and “moral” are honorific.

NC: Well, I don’t quite agree with that. I think “real” is quite different.

PL: But, what’s the difference in these cases then?

NC: Because I think rationality is something that we can understand and morality is something real and it’s part of us, and we can try to figure out what it is. We can try to figure out what our moral faculties are. We understand something about what rational action is. But about reality, we have to ask what we are talking about.

PL: Right.

NC: But, if we’re talking about reality in the enterprise of trying to discover the way the world works in a physics department or a linguistics department or whatever, commonsense notions are irrelevant. If we’re trying to explore our intuitive understanding of the way the world works, commonsense notions are relevant, but we discard them. If you have… if you’re using it in a more informal way, like is the meaning of life real, yeah, sure, ok.

PL: Well, look, there’s a sort of space between ethno-science and science, right, commonsense and all of those. It’s then explored by philosophers for 2500 years, and it’s called metaphysics, right?

NC: No, that’s different.

PL: Ok, but that is a question about what’s real, right?

NC: Ethno-science is a branch of science.

PL: Right.

NC: Ethno-science is the branch of science that tries to figure out what people’s beliefs are about the way the world works. Metaphysics is not that.

PL: I understand that.

NC: Ok.

PL: But do you think metaphysics is impossible?

NC: No, science is metaphysics.

PL: Ok, good.

NC: It’s talking about what the world is made of.

PL: Alright. So then the question is why do you think that science gets to claim what’s real? Now, let me give you an example. So, in The Scientific Image, Bas van Fraassen is a scientific anti-realist so he says the things posited by science, quarks, etc., are not real, but mid-size earth-bound objects are real. Now, you got the flipside of that.

NC: I don’t have any side because I don’t think the word “real” is sensible enough to use, although they’re all real in different senses. If you’re trying to understand the way the world actually works, whether you’re van Fraassen or you or me, we’re going to go to the scientists because they tell us how the world really works. If we’re interested in exploring people’s commonsense beliefs, we’ll go to the ethno-scientists and see what they discover

PL: What if we’re interested in something like whether there are events or whether there are properties or whether there are mereological sums or something?

NC: Well, let’s take events, which play a prominent role in modern semantics. So, here you can ask a lot of different questions. For one thing, you can ask whether in, say, Davidsonian semantics, where there’s a lot of… or anything that developed from an event based semantics, whether the things, whether what are called events are internal to the mind or outside the head. I think they are internal to the mind.

PL: Right, but can’t they be both?

NC: They could, but then we’re asking another question. If we’re asking, well, how do these things that are internal to the mind relate to something in the outside world, we’ll say, ok, let’s take a look at what you mean by an event. So, for example, is the American Revolution an event? Yeah, it was an important event in history. Does that event include the fact that the man who the indigenous population called the Town Destroyer took off a little time in the middle of the Revolution to destroy the Iroquois civilization? Is that part of the event called the American Revolution? Well, not when you study it in school. You want to find out about that event, you got to—probably the Iroquois remember, the one’s who are left—or you got to look at serious scholarly history. Then you find out that one part of what was going on in the event that we call the American Revolution was a side operation in [1779] to wipe out the Iroquois civilization so that the colonies could expand if they got rid of the British. Well, is that part of the event or isn’t it? Well, here comes hard questions about what we’re really going to call events in the outside world, and those questions don’t have answers because they’re highly dependent on our interests, our perspective, our goals; all kinds of factors. So, I don’t think we’re going to find external events in any sense worth pursuing for investigation for what the world is like.

PL: Why can’t these external events just be complicated objects?

NC: They could be anything you like, but is the Town Destroyer’s exploit part of the American Revolution or isn’t it? That event? It’s your choice, but that’s my answer.

PL: Well, now it sounds like you’re saying that, well, I have representations of events, right?

NC: No, I have representations.

PL: You have representations.

NC: I have representations, some of which we informally call events.

PL: Right, but now one might ask what on earth is a representation if it is not a representation of something.

NC: That’s a mistake that comes from a philosophical tradition. The way the term ‘represent’ is used in the philosophical tradition, it’s a relation between an internal object and an external object. It’s not the way it’s used either in ordinary speech or in the sciences. So, when a perceptual psychologist, say, talks about an internal representation of a cube or something, there doesn’t have to be any cube there. They’re talking about something that’s going on in the head. In fact, what they may be studying, and usually are studying, is the relation between things like tatistiscopic presentations and internal events. There’s no cube. But, nevertheless, they talk about it in internal representation. The concept, internal representation… There’s a long discussion of that in here. The concept, internal representation, is used in the sciences—and I think that’s ordinary speech too—in ways which don’t involve a relation between an internal thing and an external thing. And that derives from a particular interpretation of the theory of ideas, which said, well, ideas represent something out there.

PL: I should interject here.

NC: Incidentally, I should say that that’s not the interpretation of the theory of ideas that were given by the people that used it. So let’s take Hume for example. I quote him in there. He raises a serious empirical question. He’s says it’s about the nature of, the terms he uses, “the identity that we ascribe to things,” meaning how do we individuate things. And he asks the question, well, is this a peculiar nature common to the thing, or is it, what he calls “fictitious,” a construction of the mind? And he says, “fictitious.” There is no entity. There is no common nature. There is no nature common to the thing. There is a construction of the mind, which we use to talk about the world.

PL: But he’s an idealist.

NC: He was not an idealist.

PL: He was not an idealist?

NC: No, not here. He’s saying we interact. He believes there is an external world out there. There’s a coffee cup on the table, and so on. But he’s talking about the individuation of things, how we organize things, how we construct our picture of the world, and that involves the way our minds work. And that doesn’t mean the world isn’t there. No. It’s just what his predecessors called our “cognoscitive powers,” which use the data of sense to construct an account of the world. And he’s saying, well, you want to look at the identity of things, the identity that we ascribe to things, like what makes us call something a book or an event and so on. He’s saying, well, it’s fictitious in the sense that it’s a construction of the mind based on the data of sense. That’s not an idealist position. In fact, that’s the position of modern science.

PL: … because people will call you a crypto-idealist.

NC: Well, then they’re misunderstanding what idealism is.

PL: Ok. Let me… There’s an issue that I want to get to here, and this involves the thing we mentioned about representations and whether representation requires there being something that it is a representation of. Now, in a very important and somewhat influential book by Saul Kripke, there’s a revival of the sort of Wittgensteinian argument about rule following. Let me just read the relevant passage here. So in that book, Kripke says, “If statements attributing rule following are neither to be regarded as stating facts nor to be thought of as explaining a behavior, it would seem that the use of the idea of rules and competences in linguistics needs serious reconsideration even if these notions are not rendered meaningless.” I know that you’ve written on that in the Knowledge of Language.

NC: The crucial word is “if.” In the sense in which the term “rule” is used for thousand of years, in fact, in the study of language, it is not the kind of rule he had in mind. So if you read a book, you studied Latin lets say, or you studied it a thousand years ago; they would have a rule that would tells you when to use the appellative case or something. That’s not a rule in Wittgenstein’s sense. It’s a description of a part of the language. So the questions about rule following just don’t arise.

PL: But, we don’t need to get hung up on rules and so forth.

NC: But that’s what he’s talking about.

PL: I understand that, but in a certain sense he’s talking about any sort of computational state. So, take just a computer. Forget about human beings for a second.

NC: See, computers are a different story.

PL: Ok

NC: Let’s take an insect. Why aren’t these questions asked about insects? When you study insects, you attribute to them computational states. Is that a problem? I mean, is it not real? Like, if you say that an insect is determining the position of the sun as a function of the time of year and time of day and here’s the computation it’s using, why isn’t that science?

PL: That would be, but the argument would be that the reason you can get away with that is because you’re talking about what it’s…the representations that you’re attributing to the insect are externalistically anchored…

NC: No.

PL: …and that is you couldn’t do it unless what you had is an embedded system.

NC: That’s not true. You could do it in an experimental situation in which you have a light, and, in fact, if you knew how to do it, you could do it by stimulating the external sensory organs of the insect. It would all be interior. There doesn’t have to be a sun there. It’s just that, yeah, you’re talking about the way it happens actually in the real world, but you would say the same thing in an experimental setting where you don’t have an external world…

PL: Well…

NC: …because you’re talking about the internal computations of the insect on the occasion of sense. It doesn’t matter what’s out there.

PL: Notice the shift there though, because you went from saying you don’t need the sun in the experimental setting to saying that you could run the experiment in a world that didn’t have the sun, and that’s a different story. Right?

NC: No, no, it’s not a different story. The point is if you look at what insect scientists are studying, they are studying what seventeenth century philosophers used to call the “constructions of the mind on the occasion of sense.” Now, it happens that in the world that they’re looking at, the occasions of sense happen to be related to the fact that there’s something 93 million miles away, but the study could go on as if it’s what Hilary called a brain-in-a-vat. The studies are internalist because we don’t know anything else to study.

PL: But this is disputed, right? There is this dispute about David Marr. I mean, there are two stories for this. Tyler Burge and Martin Davis, for example, argue that Marr is sort an externalist sorts of…

NC: Yeah, but they are just misreading him. I mean, in fact, I happened to know Marr personally, but I’m sure if he was here he would say this. If you look at the informal exposition in Marr, in Marr’s Vision, let’s say. The book, Vision. If you look at the informal exposition, in order to motivate what he’s doing, he says, well, imagine an elephant, or anything—like a stick figure—and we want to know how that thing out there is interpreted by the visual system as some three-dimensional object. However, if you look at the experimental procedures of Marr, they didn’t have elephants out there. In fact, what they were using was tatistiscopes. So, they were having dots on screens, and if they had known how to stimulate the optic nerve, they would’ve done that. When you go from the informal exposition to the actual science, you see that like everything else, it’s a study of the internal nature of the beast. And in fact, they would’ve loved to get to the point where they could tell you something about how you identify an elephant, but they never got anywhere near that. However, even if they did, it wouldn’t matter whether the elephant is there or not, it wouldn’t matter what’s the occasion of sense. Again, the seventeenth century formulation of this was, I think, quite appropriate. On the occasion of sense, the cognoscitive—sounds archaic—but the cognoscitive powers of the mind construct complicated internal structures, which have all sorts of properties, gestalt properties, what Hume later called the identity that we ascribe to things, and so on. And that looks correct, and that’s the way modern science looks at it. The fact that the informal expositions talk about—sort of motivating what you’re doing—talk about identifying objects on the outside, that’s fine, but you have to know how to distinguish informal expositions from the actual scientific program. And if you look at the actual program, then you’ll never look at the things outside, aside from tatiscopic images because they are as close as you can get to the actual occasion of sense.

PL: It’s time to move to our very eager studio audience here. So, let’s turn to them.

Audience 1: Hi, how are you. You mentioned event semantics and how that’s sort of been pulled into, or is a part of some linguistic research, and I‘d like to ask a question about that and just hear more about it. The way Pietroski uses it in the article in the book that’s sitting on your table. He will take a sentence…

PL: We’ve been referring to this book, and I should point out it’s called Chomsky and his Critics which just came out from Blackwells.

A1: Right, and the article is by Paul Pietroski, I think. So, he’ll take the sentence, ‘John boiled the water’—I’m going to ignore the larger context—and then turn that into something that’s like there was an event of which John was the agent and it was concerning a boiling and the theme was water—terminated in a boiling, excuse me—and the theme was water, something like that, right? And then Pietroski tries to fit that together with a syntactic account of how that formal representation can turn into the sentence that you have here. So, you see, you recognize in your reply that it’s an open question whether the subject matter of this whole analysis is in the language faculty or in the cognitive …

NC: It’s all in the cognitive system. It’s all internal though. It’s all syntax in the sense that it’s involved with internal computations.

A1: Right. Fine. So, the question then is—and you say it’s an empirical question whether it’s in the language faculty or in the cognitive system. So, first of all, what would be some ways to determine empirically what that is and also what would it even mean for it to be… I have a trouble understanding what it would mean for it to be in the language faculty because then it seems that our language faculty has certain event metaphysics built into it.

NC: Not event metaphysics. It’s a, well, first of all, there is no event metaphysics in the ordinary sense of event. Events are things out there. And all of this is formal manipulations inside. We’re now asking the question what is the architecture of the mind, which is like asking what’s the structure of some benzene molecule. You want to know the structure of the benzene molecule, you can’t just think about it. You have to have some theory that tells you how you look at such questions. Well, the study of the architecture… I think there are real answers to the question of what is the architecture of the mind, but they’re not going to be easy to find. You want to find out what the language faculty is, well you’re going to… when you talk about any subsystem of an organism—say, the circulatory system, or the immune system, or the digestive system, or whatever it might be—you’re kind of presupposing that it makes sense to look at a complex organism as if it has components, each component having kind of an integrated character with its own properties. It’s worth studying in itself, but it’s not separable. You can’t cut the immune system out of the body […]. It’s in every cell you know, but you’re saying, look, it has properties that you can study by themselves. You can learn things about them. But when you put the whole picture together, you hope you’ll understand something about the organism. Well, same when you’re studying the cognitive architecture, we want to see is there a component, call it the language faculty. We don’t know what it is. You have to discover and refine it and change it. Is there a component, which is critically involved in what you and I are doing in some fashion, which has some intrinsic properties? So, is it going to turn out, say, that it has particular interface conditions and some kind of internal recursive computational process? Well, if it does, then the more we understand about that and how it fits into the general cognitive architecture, the clearer these questions become about whether something is inside it or outside it, and those are very concrete questions. I mean, take, say, anaphora, what are called referential relations, relations of intended referential dependency, ok. Are they inside the language faculty or outside? Well, that’s a substantive issue. If you’d asked me ten years ago, I would’ve said inside. Now, I think there’s evidence that they’re outside, right on the edge. And the reason for that has to do with beliefs about how the language faculty works. So, if the language faculty does involve optimal computation that requires cyclic derivations, and if anaphora looks at global properties—condition C of the binding theory—well, it’s outside. But that’s a question that you can’t ask in advance. You have to ask what does it seem these faculties are like. The more you learn about them, the more you can formulate these questions clearly, and that’s the kind of question that comes up about the internal notion of event. Is it in the language faculty, the semantic component of the language faculty, or is it in some other faculty that’s linked to it? And you can’t, you just can’t speculate about that. You have to learn about it.

A1: And I guess that’s my question, I just, from ignorance, have a hard time imagining what kind of empirical evidence would count in adjudicating between those.

NC: If I’d known it, I would’ve said it. I mean, I think these are the kind of questions… What is the empirical evidence that bears on whether anaphora is on one or the other side of the border of the language faculty? Well, the empirical evidence—in my mind at least—turns on whether in fact there’s a single cyclic derivational process that goes by, what I call phases, stepwise. That’s an empirical question but all kinds of things bear on it coming from everywhere. And once that empirical question is sharpened, you can ask whether the global property of anaphora is inside or outside.

A1: Thank you.

NC: Actually, similar questions, I might say, arise on the phonetic side. So take, say, prosody. Prosody appears to have global properties, ok, so is it inside or outside the cyclic derivation, which is forming bigger and bigger units? Well, if the properties really are global, it’s outside. But you can’t answer that, and the kind of evidence that bears on it comes from everywhere. It might come from chemistry for all we know, when we try to figure out how these processes work.

A2: You’ve made clear that science plays a primary role in philosophy, and I was curious to know what role, if any, does philosophy play in science?

NC: Well, here I think it helps to look at the question in a little more historical depth. If you had asked Hume, “are you a scientist or a philosopher,” he couldn’t have answered because there was no distinction. If you had asked Kant, “are you a philosopher or a scientist,” he couldn’t have answered. In fact, until the latter part of the nineteenth, around the middle of the nineteenth century, there was no really clear distinction between science and philosophy. If you study science in Oxford and Cambridge, it’s the department of natural philosophy or moral philosophy because science was just one part of philosophy. They weren’t distinct. By the latter part of the nineteenth century, they sort of became separated, and after that it’s really a matter of choice. Disciplines don’t exist. We construct them. So, I mean, Deans have to have ways of organizing departments because it’s too much trouble to have everybody do everything. But there’s no boundary to what’s in chemistry. There’s no boundary to what’s in philosophy. It just depends on what philosophers want people to call themselves… who come out of this tradition want to study, and a lot of what they want to study is questions in the foundations of science or say the kind of things that van Fraassen is trying to clear up by using the results of science. These are all fine questions, and people tend to call them philosophy. You could call them—parts of them at least—you could call thoughtful science or reflective moral theory or you can call them anything you want. But there’s a range of questions which have come to be in the domain of philosophy, and they can extend all over the place. I mean, there are people in philosophy departments that are working on the foundation of quantum theory and making contributions to it.

A3: Hello. I’d like to go back to insect navigation once more, and I take it that we explore insect navigation because we want to find out how insects navigate in their natural environment. And, you’re right to point out that in an experiment, we can replace the sun by an artificial light and so on. But, it seems like we still, in the whole explanatory context, we cannot just do without the sun and the actual objects in the real world because otherwise an explanation of insect navigation becomes meaningless. And, so, it seems to me that we have to have the assumption that insects represent the actual sun and not just dots of light.

NC: We don’t have to have that. In fact, the way science works… of course, you’re interested in the outside world. But if you’re taking a physics course here, they don’t use videotapes of what’s happening outside the windows because that’s just useless. It doesn’t let you figure out how the world is working. So you disregard that beyond the earliest stages of science. Yeah, in the earliest stages, you see apples falling from trees and that kind of thing. But as soon as you get anywhere, you start designing artificial situations called experiments in which you try to refine the evidence so that it will shed light on principles, which you believe will ultimately bear on what’s going on outside the window. But that’s not what you’re looking at. When you’re studying an organism, say, an ant, you don’t necessarily start… yes, you start with noticing that the ant is figuring out where the sun is and has a very strange computation—we can’t do it—as a function of the time of day and time of year. But as you go beyond, you ask, well, what are the actual computations going on inside that. And if you get far enough, you would set up experimental situations in which you wouldn’t bother with the sun. You’d figure out what those internal computations are. You’d then find that they interact with all kinds of other things. They are not done in isolation, and out of that, you expect you’re going to shed some light on what’s going on when the occasion of sense that the insect is operating on happens to be connected to an external object, like the one 93 million miles away. But that’s just kind of like a consequence of the investigation of the ant.

PL: If I can follow up on this question. This seems to be inconsistent with what you say in this reply to Ludlow here where you’re quoting Gallistel. So you say, “Gallistel (1990) argues that the representations play a key role in animal behavior and cognition. Here, representation is understood as an isomorphism, a one [to] one relation between mind/brain processes and an aspect of the environment to which these processes adapt the animal’s behavior.”

NC: It’s not inconsistent. He’s making an empirical claim. It’s a very interesting one, and which is an introduction to a couple volumes on animal representation, and his conclusion, which I quote there, is that for animals, there is in fact a one to one relation, an isomorphism, between an internal event and some property of the external world. Like… an odor is the example he uses. So the property of the external world is a particular odor, which is out there in the world all right. But you could, if you knew how to do it, you could just use the sensory organs. And he says that’s how animal representation works, with this one to one correlation between things that are outside the animal, though you’re really studying them at the sensory boundary, and the internal representations. Well, if that’s correct, that’s an interesting fact about animals, and they are very different from humans in that case.

PL: So then maybe he’s [Audience 3] right about that.

NC: Yeah, he [Audience 3] could be. Take the example I mentioned there. It’s taken from him [Gallistel]. That an ant—some species of ants, at least—will identify a corpse of a conspecific on the basis of a particular odor. Of course, if you give that odor, and there’s no conspecific around, they’ll do exactly the same thing. Because, according to this picture, he’s saying, well, it’s just triggered by the odor, which usually has to come from a conspecific. But if you can control the odors the way you can control lights, you could get the same behavior according to him, there’s no conspecific.

A3: Yeah, but just one more clarification question. It’s seems to me that no scientist can claim he has explained any kind of animal behavior if what we take to be its natural environment doesn’t play an essential role in this explanation. So, I think, if the relation between an ant and the sun, for example, doesn’t show up in this theory, then we haven’t really explained what we wanted to explain

NC: Well, it depends on what you want to explain. Let’s say that the sciences… As modern sciences have developed since Galileo, this is something of an innovation. They really are not trying to account for the phenomena of experience. I mean, indirectly, that’s what motivates them. But what they’re trying to do is discover the principles that enter into the way the world functions, and if you ask the guys in the physics department here, “can you explain this videotape of what’s going [on] outside the window?” They won’t even bother answering. They can’t say anything about that. It’s way too complicated, and way too many factors, and it’s not even their topic. Their topic is to find out the principles of nature. Galileo had a lot of problems with this. You go back and look at the history. He had a lot of trouble convincing the funders, the rich aristocrats who were the funders in those days, convincing them that it’s worth studying something as ridiculous as a ball rolling down a frictionless plane. First of all, there’s no such thing. Who cares anyway, you know. It was a real… it was a conceptual breakthrough to get to the point where you began to understand that the phenomena of the world are of interest for the sciences in so far as they provide evidence for the principles of nature. And as soon as you proceed very far, you find out that the ordinary phenomena of nature is useless for this purpose, and you do what are called experiments. And sometimes thought experiments. Like, you look back at Galileo, there’s no reason to believe he ever dropped the balls from the top of the Tower of Pisa. If you look at the argument, he gives a purely conceptual argument, convincing one, to show that that’s what is going to happen and it didn’t matter if you observed it or not. In fact, you don’t observe. But his argument was convincing because the logic of it, and that’s true of a lot of his experiments if you read through them. Scholars are now determined he couldn’t have carried out those experiments. There was no way for him to do it. Some of them he probably did. Some of them he was thinking about. But the point is almost none of them had to do with observations of a casual look at the world. The same with the theory of perception. Take an example of Descartes, which I’ll adapt to this situation, but it’s a Cartesian example. He said if I look out there [pointing to the audience], what I see is a lot of people sitting in a room. Well, Descartes points out literally that’s not what’s hitting your eye. What’s hitting your eye is something that’s coming from that guy’s head and that guy’s foot. But there’s nothing about people sitting in a room. Nevertheless, what you see is people sitting in a room. And then he says, “how does this happen?” Then comes the whole story about on the occasion of sense. The cognoscitive powers make these complicated constructions, which include imposing the structure of people sitting in a room on these fragmentary sensations that are coming to me. But, in perceptual psychology, as it proceeds, you don’t study phenomena, like people sitting in a room. That’s just way too complicated.

PL: But this isn’t fair to the questioner though because he’s not arguing against idealization. I mean, you can have a perfectly controlled experiment and still talk about… For example, take an old sort of Skinner type experiment with a pigeon pressing a bar. You can describe that behavior as the pigeon going, moving it’s claw or whatever, or you can describe it as bar-pressing behavior, right? Now, this is not… If it’s not an issue about idealization, the questioner could easily say…

NC: No, no, but, yeah, we’re asking… at least I understood him to be asking something different. If the scientist cannot give an account of my seeing people in a room, they’re not giving an explanation. That’s far too strong. I mean, the scientists are giving explanations even they can’t get anywhere near describing real life situations, and in fact that’s about all of modern science. The bar-pressing is an interesting case because Skinner did impose on it the interpretation, “bar-pressing,” and that turns out to be wrong. It turns out with closer look, that the pigeon pecking, which is what it was doing, actually incorporates different instinctive behaviors; the behavior of pecking for a seed and pecking for water turn out to be different instinctual behaviors, which happen to converge in this experiment. You just mislead yourself if you call them the same thing.

PL: Well, isn’t that Skinner’s mistake is that he was an internalist, and if he’d been an externalist, then he would see that pressing for a seed and pressing for water are different behaviors.

NC: It has nothing to do with externalism and internalism. I mean, the motivation for distinguishing seed from water comes from observing pigeons. But if you want to really carry it out further, you’ll find out what’s going on in the pigeon’s head on the occasion of sense. And if you could carry out the programs far enough, you’d forget about the seed and the water. Just like David Marr does. He doesn’t talk about external things because he’s trying to really discover what the principles are. So it’s not anything to do with internalism and externalism. What it has to do with is giving the right idealization of a wrong idealization, and we’re always doing that. Every experiment involves all kinds of interpretation as to what you’re going to think about this thing and that thing. As the sciences get more refined, you try very hard to not to cut it out because you can’t, but at least be consciously aware of what you’re putting in. So you can sort of compensate for it if it’s the wrong thing to put in.

PL: Ok. Next question.

Audience 4: Hi. I think I have a much less sophisticated question than the previous one. You’ve characterized I-language generally as an interface of sound and meaning and at the same time here you’ve also been very critical of the various senses of the term “meaning” and “reference.” So, I just wanted to ask you if you could clarify what do you mean by “meaning.”

NC: Well, that’s again like David Marr saying I’m trying to figure out how you see an elephant. We start with the intuitive notions of sound and meaning, whatever they are, and you go back Aristotle and he describes language as a pairing of something like sound and meaning, saying it in Greek. And, yeah, that’s what we start with. But as we proceed, we’re going to have to refine both of these notions. So, sound, as it’s used here, doesn’t have to do with what you and I call sound in ordinary talk, and “meaning” will be something very special. So, like maybe it will be Davidsonian event semantics. Ok, maybe it will be something built on that and maybe “sound” will be something built on my colleague Morris Halle’s conception of instructions for articulatory gestures or it will be whatever it turns out to be as the sciences progress. And it will end up having some loose relation to what we call “sound” and “meaning.” But, no more so than “energy” or “work” or “life” or any of the other concepts that are dropped in the sciences. Although they often keep the sounds when they talk about their new concepts. So that’s to be answered, not to start with.

A4: Ok. Thank you.

PL: Ok. I guess that’s it. Thanks a lot. Appreciate it.