JAC Volume 13 Issue 1

Q. In your essays, you seem intentionally to cultivate a kind of conversational and even colloquial tone. How would you describe your writing style?

A. A friend of mine, Arnold Isenberg, once told me—I think before I had written anything—that the way to write a paper in philosophy was to begin by asking a question that anybody could understand or by posing a problem in such a way that anyone would see that it was a problem. I followed that advice for a long time and began most of my papers with either a problem or a question, so I have attempted always to write my essays in such a way that the reader does not require any special background in philosophy in order to understand my meaning. I'm often told that my papers are difficult, but that was certainly never my aim. My aim was quite in the other direction. I'm gratified by the fact that people who don't have any technical background in philosophy do seem to make something of my ideas. I think the only other thing I can say about my style is that I sometimes find it incredibly hard to start writing. I often imagine the first sentence and then ask myself, "Wait! What comes next?" Pretty soon, I'm writing the whole paper in my head, and any problem in the composition or organization of the text stops me from even writing the first sentence for fear I would be somehow trapped. When I do finally write something, I often find that the first couple of pages, which usually sort of ease me into the subject, are better left out. So, I'll throw away these painfully constructed early pages completely.

Q. You've already anticipated my second question. How would you describe your writing process? Do your essays undergo several revisions?

A. I hope you won't hold this against me, but I don't do a great deal of revising. I always believe that I have a pretty clear idea about how a paper is going to go together before I start writing. However, in the throes of composing a paper, I find that I regularly think about the paper. When 'm trying to go to sleep or when I'm half asleep, ways of putting things often occur to me, or when I'm not in the midst of writing, a new idea or a solution for some problem of organization sometimes will come to me. I find that these relaxed moments—when I am not actually writing—are absolutely essential in my composing process.

Q. Do editors help you very much when you write an essay?

A. I don't think I've ever had any substantial help from an editor. I've often wished I had. The kind of thing I usually get from editors is advice on how to punctuate. For example, I've had several battles with the Journal of Philosophy. Concerning the first piece of mine that they published, the editors insisted that periods and commas go inside quotation marks and semi-colons and colons go outside. They said that this method of punctuation is absolutely standard practice in the United States. I wrote back and said, "Not so: the New Yorker does it the way I do it." Whereupon, they backed down. Aside from the fixation that the editor of the Journal of Philosophy had about punctuation and "whiches" and "thats," I can't remember much guidance from an editor. I should point out, however, that I often circulate a paper to colleagues and friends and ask them to make suggestions or comments, but they never do. The suggestions usually come years later after the paper has been published. Where I do receive a lot of help is from reading my papers aloud to audiences. In these presentations, I almost always get good ideas about what is difficult, what could be phrased in a better way, and so forth. Such feedback is extremely valuable because I see what it is that troubles people, and I realize there are difficulties I hadn't noticed.

Q. How would you describe your philosophical genealogy? That is, who are the thinkers who loom large in your own history?

A. As an undergraduate, I was first an English major and then went into comparative literature. I was interested in the history of ideas, and I did a lot of philosophy. By the time I finished my undergraduate work, I was in a combined field of classics and philosophy. So, I covered a great many different topics in those four years. For example, in my second year, Whitehead was teaching for the last time at Harvard, so I took both his undergraduate course and his graduate seminar, thinking I mustn't miss the great man. I was inspired both by his personality and by his writing, but he also encouraged me personally. He encouraged me in a way of thinking which was very congenial to me. Putting it very roughly, he taught me that I could learn from great figures of all sorts—both literary and philosophical. I learned, too, that the question of agreeing or disagreeing is subordinate to the question of getting an insight into various ways of looking at things; so, early on, I didn't think of philosophy as a subject where you tried to get it right; rather, I thought of it as a lot of interesting ideas, all of which you should take in as far as you could. So I'd say, Whitehead had quite an influence on me, not one that I now think of as altogether good.

The person who aroused me from my undogmatic slumbers was Quine. He was just starting his teaching career, and I felt as if I had discovered a new subject when I came under his influence. I was fascinated by mathematical logic and logical positivism which he was discussing and criticizing. Then I went into the Navy and there was a big hiatus. When I came back, I did my dissertation on something I had already started, namely Plato's Philebus, because I felt that I didn't know enough of the sort of thing that Quine was doing to write on his work. In 1951, I was hired by Stanford, and during the time I was there, I taught almost every course that appears in the catalog of a large philosophy department. So, I worked up one subject after another—philosophy of language, ancient philosophy, medieval philosophy, modern philosophy, epistemology, ethics—and I started formulating ideas in all these areas. In the philosophy of language, however, I found myself going back and reading Quine with a new eye, and as I started teaching, he slowly became more and more a guiding figure in my thinking. So, he entered my life twice over, so to speak, and has never departed; we've always been friends. My intellectual debt to him is greater than to anybody else.

Q. One of the intellectual developments that has directly influenced the study of writing is what has come to be called anti-foundationalism. In your survey of the passing philosophical scene, how would you assess the importance and the impact of anti-foundationalism? Has anti-foundationalism won the field in the philosophy of language?

A. The answer to the last question is no. It hasn't won the field. Foundationalism is alive. I'd like to think it isn't well, but it is certainly alive. There is a question as to exactly what one means by foundationalism. The British empiricists were foundationalists of one kind; according to them, everything comes from the senses in one way or another and this provides a foundation for epistemology. Another version is the Cartesian idea that what we know for certain is what is in our own minds and anything else we know we have to construct from that. Those are two quite different sorts of foundationalism, though they often appear together for various reasons. For example, Quine is a foundationalist in the general sense of the British empiricists, although his brand of foundationalism is clearly not the same as the positivists or indeed anybody really who went before him. I have departed from foundationalism completely, and this departure is one of the biggest differences between my views and Quine's. I think I never went for foundationalism; that is, I was never taken by any version of it. It always seemed to me to be a mistake, and I can remember, in fact, arguing with Quine—in my amateurish way—about this even back when I was a first-year graduate student. It's been a topic of pleasant debate between us through the years.

Q. In terms of large conceptual categories, your work has been tagged with the label "externalism" in opposition to "internalism." I know that a response to this next question requires more discussion than you can give it here, but can you describe the primary differences between an externalist conception of language and an internalist conception, and what hangs on these differences?

A. The internalist says that the contents of our thoughts—our beliefs, our desires, our intentions, and what we mean by what we say—are determined wholly by what is in the head. Generally speaking, this is a Cartesian position, and there are lots of internalists around. The externalist, however, maintains that there are factors external to the person which are determinants of the contents of our thoughts, and not just causal determinants—because that's obvious—but, so to speak, logical determinants, too. For example, from an externalist perspective, you can't have a thought about an apple if you haven't had at some point in your life some contact—indirect or direct—with apples. So, externalism has to do with your history and things that exist outside of you that make a difference to what you can think or what you are thinking at a given moment. Now, beyond this description, externalism takes a number of forms, but unlike Saul Kripke, Hilary Putnam, or Tyler Burge, I don't limit the extent to which the contents of our thoughts are fixed by external objects. I think externalism applies universally; there are connections everywhere between the world and the contents of our thoughts. It's not limited to a few words but is true of a very large number of them. So, I am an all-out externalist.

Q. Some of your terminology is now beginning to appear in both the areas of rhetorical and literary theory. I would like to ask you to discuss a few of these concepts. Would you explain first what you mean by "radical interpretation?"

A. Radical interpretation is a way of studying interpretation by purifying the situation in an artificial way. Imagine trying to understand somebody else when you have no head start: there is no translator around; there's no dictionary available; you have to work it out from scratch. It would beg the question, in trying to study the nature of interpretation, to assume that you know in advance what a person's intentions, beliefs, and desires are. I hold that you never could get a detailed picture of any of those things unless you could communicate with the person first. There is no master key or framework theory that you can have prior to a communicative interaction or situation. You've got to work your way into the whole system at the same time.

Q. In your formulation of communicative interaction, you employ what you've termed the principle of charity. What do you mean by this idea? Is it correct to say that radical interpretation relies on the principle of charity?

A. The principle of charity says that in interpreting others you've got to make their thoughts hold together to a certain extent if you're going to see them as thoughts at all, because that's what thoughts are like. They have logical relations to one another. Although people can certainly be irrational—they can have thoughts that don't go together—we can only recognize them as irrational because their thoughts lack rational coherence. You can't make sense of total irrationality. For example, if you believe everything is green, then you have to believe that this table is green, for if you don't, you temporarily depart from standards that you have to have if you're to have the thought that this is green or that everything is green. You can't have those thoughts without those thoughts being related to others, and those relations are logical. The principle of charity really just formulates the recognition of the necessary element of rationality in thought. Thoughts have propositional contents, and propositional contents are in part identified by their relations to each other.

Q. I think I wanted to make more out of the principle of charity. I thought the principle of charity in relation to radical interpretation constituted the very hinge on which swings the notion of comprehensibility. That is, in order to understand that we don't understand something, we already must share a great body of common experience about language and about our being in the world.

A. That's not wrong. An account of anybody's thoughts will no doubt uncover some inconsistencies, but in order to uncover inconsistencies, you must first identify the thoughts that are inconsistent to each other, and to identify them, you have to embed them in a complex of thoughts with which they fit. Otherwise, you have no way of identifying them as what they are. So, I'm not disagreeing with you about that. I also didn't mention the other aspect of charity, which is how thoughts are related to the world. Externalism says there's a connection between the contents of people's thoughts and their causal relations with the world itself. In fact, I would say if it weren't for that, we wouldn't be able to interpret anyone else. It's only because we share a world with others that we can get the hang of what they're talking about. So again, the word charity is a misnomer because it's not a matter of being kind to people; it's the condition for understanding them at all. Thus, charity has two features: one is that you can't understand people if you don't see them as sharing a world with you; the other is that you can't understand people if you don't see them as logical in the way that you are—up to a point, of course.

Q. In several of your papers, you talk about triangulation. I think of triangulation as a way around the problem of correspondence and language as mediation. Would you discuss triangulation? What do language users do when they triangulate?

A. Well, the idea of triangulation is partly metaphorical, but not wholly. The basic idea is that our concept of objectivity—our idea that our thoughts may or may not correspond to the truth—is an idea that we would not have if it weren't for interpersonal relations. In other words, the source of objectivity is intersubjectivity: the triangle consists of two people and the world. Part of the idea is this: if you were alone in the world—that is, not in communication with anybody else—things would be impinging on you, coming in through your senses, and you would react in differential ways. Now, here's where the metaphor comes in. If you were to ask, "Well, when you're reacting a certain way, let's say to some pleasant taste, what is it that pleases you?" We would say, "It's the peach." However, in the case of the person who has no one with whom to share his thoughts, on what grounds could you say, "It's the peach that pleases" rather than the taste of the peach, or the stimulation of the taste buds, or, for that matter, something that happened a thousand years ago which set all these forces in motion which eventually impinged on the taste buds. How far out are the objects that he is responding to? There would be no answer to that question at all: nothing for him to check up on, no way to raise the question, much less to answer it. So, the idea of triangulation is this: if you have two people both reacting to stimuli in the world and to each other—that is, to each other's reactions to the stimuli—you've completed a triangle which locates the common stimulus. It doesn't locate it in one person's mouth; it doesn't locate it in one person's eyes; it doesn't locate it five thousand years ago. It locates it just at the distance of the shared stimulus which, in turn, causes each of the two creatures to react to each other's reactions. It's a way of saying why it is that communication is essential to the concept of an objective world.

Q. Speaking of communication and how communication is "essential to the concept of an objective world," would you distinguish between what you call a prior theory and a passing theory?

A. The distinction between prior and passing theories is just the difference between what one anticipates that somebody will mean by something he or she says and what one decides was meant after one is exposed to an utterance. Whenever you talk to somebody, you have an unformulated theory of what that person would mean if he or she were to utter certain words. For example, you would know roughly what you yourself would mean if you were to utter these same words. However, plenty of things may tip you off that your interpretation is not the right interpretation. On occasion, someones words don't mean what you would have meant by those words. They don't necessarily mean what they've meant in the past. You discover that this might be a slip of the tongue, or it might be a clever invention on the spur of the moment; it might be a joke; there are a thousand possibilities many of which we are so good at catching that we don't even notice were doing it.

Q. So, theories don't actually need to match?

A. No, no. That's right. They just are nearer to or further away from being correct. Luckily, we can make lots of adjustments as we go along. Communication is always incomplete. It's not as though anybody ever gets everything right; it's a matter of degree.

Q. Some rhetoricians and composition specialists balk at the idea that communicative interaction cannot be codified in some way. They want what I think of as a "big theory" that can do some serious work for us and that actually can model successful communicative interaction. The notion of a passing theory denies this possibility, doesn't it? A passing theory can't be predicted in advance of a communicative situation, and in ordinary communicative interaction—in day-to-day utterances—the passing theory is still always there. It doesn't disappear. It's just that we have become so habituated to using linguistic crutches—conventions of language—that passing theories become, to some extent, transparent.

A. We share an enormous amount of linguistic lore. In fact, we don't ever share it exactly. Each of us has our own ideas, and each of us adjusts how we speak according to the audience. We have ideas about what words somebody else is apt to understand, or what concepts they're apt to control. Now, the passing theory, on the part of the speaker, is what the speaker thinks that somebody else will make of what he or she is saying. The passing theory on the part of the interpreter is that person's best bet as to how to understand what is being said.

Q. Let me tell you why I am pressing this point. Some rhetoricians and linguists claim that language mastery is enough to ensure successful communication. Once you master something—language conventions, rules of syntax, a grammar, a schema, a script, a model of cognitive processes—you have all you need to communicate. This notion disturbs me because I wonder what it is that one masters when one masters a language.

A. Obviously, if you get everything right, you've mastered something. However, what should be considered part of knowing a language? I myself think that it's quite easy to see that to understand even a single utterance you have to know something that cannot be a rule of language in any way at all. You have to know what language the person is speaking more or less. These sounds might mean omething completely different in another language, and the rules of the language can't tell you when somebody is speaking that language. So, that's something you have to bring in from outside. That's just a very simple case. I think it's a much broader point than that, but this example is enough to make the point; mastery of the language cannot be restricted in any way at all.

Q. If "mastery" means anything.

A. Exactly.

Q. This point ties in with my next question. In an oft-quoted passage from "A Nice Derangement of Epitaphs," you write that "linguistic ability is the ability to converge on a passing theory from time to time. If we do say this, then we should realize that we have abandoned not only the ordinary notion of a language, but we have erased the boundary between knowing a language and knowing our way around in the world generally. I conclude that there is no such thing as a language—not if a language is anything like what many philosophers and linguists have supposed." Now, this kind of talk strikes some rhetoricians as clearly counter-intuitive and others as downright crazy. Would you help us better understand your meaning when you say that there is no such thing as a language?

A. Of course, you've got to add the proviso, "if a language is anything like what many philosophers and linguists have supposed." In fact, elsewhere in that essay, I say in a fairly sharp way what those assumptions are that I think are wrong. They include such things as the idea that there is a single repertoire of expressions—with their meanings and their semantic interpretation—which everybody shares. That, I say, is just wrong. There is no such thing that's shared. People differ. Every person has got his or her own language in that sense. But to say that every person has his or her own language is to say that there is no such thing as what philosophers have supposed a language to be. Interestingly enough, this is a view that I apparently share with Chomsky, because he said exactly the same thing: everybody's got his or her own way of doing it. Of course, there is a tremendous amount of overlap from person to person in linguistic knowledge. If there weren't, talking would be much more a chore than it is. That the overlap is never complete is not a very shocking view if you understand it. What is much more shocking to some people is my view that it's not essential to linguistic communication that any two people who are talking to each other speak anything like the same language. In fact, it's inconvenient if they don't, but it can happen. I sometimes carry on conversations with people who are speaking French while I speak English. When I read papers in Europe, I'm constantly asked questions in German, French, and Spanish, and I answer them in English. They understand me, and I understand them. Of course, you might say that we have a background that allows us to translate back and forth; that's true. But, we could do it from scratch; it just would be very hard.

Q. And that's another way of saying that language is not convention-bound.

A. Yes, that's exactly right. This is something you probably know, but "A Nice Derangement of Epitaphs" was published along with two attacks, one by Ian Hacking and one by Michael Dummett. I recently attended a conference in Sicily with Dummett, and I finally produced my answer to his concerns. He is about to produce his answer to my answer, so this will go on, no doubt. He is strongly opposed to the idea that conventions are crutches. He thinks speaking as others do is not just a convenience but absolutely essential. He, like Tyler Burge, thinks that what we mean depends very much on what other people mean by the same words. There's a big issue here, for philosophers anyway, because were on the edge of Wittgenstein's private language discussion. You know Saul Kripke's book? Kripke attributes the following argument to Wittgenstein and seems to endorse it himself, although he doesn't absolutely commit himself. The idea is that a certain element of objectivity—which is essential to meaning something by what you say—is injected only by there being a social custom or a habit. The question whether you, for example, are meaning the same thing from moment to moment by some given word depends upon whether I use that word in a certain way. And, in fact, it's only that confluence or convergence on a usage, custom, or habit that provides an objective test of whether somebody is going on in the same way or not. This is not my account of objectivity; my doctrine of triangulation is an alternative account that doesn't depend upon people doing the same thing.

Q. Perhaps we might pursue this point. In the discipline of composition and rhetoric, we are fond of saying that knowledge is socially constructed. Following Kuhn, or what they imagine to be Kuhn's position, some rhetoricians desire to push this claim in order to argue the more forceful point that knowledge is communal in the sense that knowledge can be recognized as knowledge only within a community that shares a common discourse. It seems to me that this communitarian formulation buys into the idea that knowledge is relative to a conceptual scheme, a formulation that you roundly criticize. What are some of the problems we encounter when we insist that a framework of social conventions guides or structures what we can know or can say about the world?

A. Well, I shouldn't pretend any kind of expertise on what Kuhn has said, but if he says, as he seems to, that there is no objective standard against which to measure the correctness of a view except its acceptance, then this view seems to me to have all of the obvious flaws of various kinds of relativism about truth and the like. It sort of sounds attractive, but I don't know how one cashes it out.

Q. Let me approach this question from another angle. When social constructionists say that knowledge is relative to a discourse community, it seems to me that they posit a kind of conceptual scheme—made up of social norms or language conventions—to which propositional attitudes must be relative. Although social constructionists see their communitarianism as a break with cognitivism and the internalism inherent in it, I've been arguing that this position is not a break at all, but it's just simply a displacement of the same old Cartesian problem. Instead of meaning being located in mental categories, it now becomes located in a communal scheme of one kind or another, and you run into the old problems of skepticism and relativism.

A. It seems to me that it can be more complicated than that. I also think that there is an irreducibly social element in determining what it is that we mean. After all, the triangulation idea brings in a social element, but it doesn't say that something is right if it coheres with what everybody else is saying at the moment or something like that. It's not relativistic at all. It doesn't lead to a relativistic idea of truth. When people try to formulate this issue, they often get into the most obvious kind of trouble. You see, there are obviously some things that you can say that are true. As time goes on, what we mean by certain words changes, and, therefore, a sentence that might have been true at one time is no longer true because it doesn't mean what it meant before. That change in meaning is a social change and has to do with the development of theory, belief, and all sorts of things of that kind—but that's not relativism about truth. The truth of a sentence is relative to the circumstances of its use, but the intelligibility of this claim requires a non-relative concept of truth.

Q. You emphasize the importance of social change. How do you see your work in relation to social issues, such as institutional power, authority, and ethnocentrism? That is, do you see political or ideological implications embedded in your conceptions of belief and truth?

A. Well, it's not something I would want to push, but I do think that the rejection of certain kinds of relativism does make a difference about how we deal with people from different cultures, backgrounds, and periods. Instead of thinking of these thing as sort of blocks that are fixed one way or another, we might think of them as just variance which we understand in terms of what we share and see ourselves as sharing. Understanding other cultures is no different from understanding our next door neighbor, except in degree. It's not a difference of kind. In both cases—understanding a different culture or understanding a neighbor—the principle of charity is essential to yielding the best interpretation. Of course, the more different we perceive people as being, the greater the strain becomes on using the principle of charity. So, when we are faced with these differences among people, we may think what's required to understand them is an entirely different kind of act—namely, a leap of the imagination of some sort. That's a rather common idea. However, I think it's a dangerous idea, because it leads us away from simply doing our best to accommodate somebody else's view of the world. If we think of understanding as needing some magical leap of the imagination, we're no longer calling on ourselves to discover the common ground on which we can make whatever sense we can make of one another.

Q. Yes, I think of triangulation, for example, as a way around the problem of ethnocentricism, so we dont' have to throw up our hands and say, as Rorty seems to say, "Ah, yes, were ethnocentric; we might as well admit it, and learn to live with it."

A. There's a related issue that has always been a lively one in philosophy and that has taken on a new life in recent years. It has to do with the question of the objectivity of values—whether values are either relative and just plain subjective in some sense or whether some values are objective in some way. It is interesting in a way that this dispute is as alive now as it was in Plato's day. It's quite wonderful.

Q. More so, in some ways.

A. That's right. In this matter, I'm on the side of objectivity. I think we have no principled way of making a distinction between evaluative and other kinds of judgments. Intersubjectivity is the only source for understanding other people's values, just as it is for understanding their other beliefs. Certainly, there is a difference of degree in the sense that it may be harder to resolve disagreements—harder than it may be about some scientific matters, for example—but I think about such differences as matters of degree and not as matters of absolute distinction.

Q. Do you think, therefore, that there's no principled difference between our ability to resolve a problem in science and our ability to resolve a problem dealing with value?

A. That's right. People like Bernard Williams, with whom I argue about this issue, say, "Look, there's an absolute difference between moral and scientific views. In science, we have ways of resolving things, and even when we don't resolve them, we have methods of resolution, like performing an experiment. When it comes to ethical matters, there is no such methodology in important cases." I'm not convinced by this line of thinking. For one thing, if we do an experiment, the experiment doesn't tell us what to think. We decide what to think. It's foundationalism again. The foundationalist thinks that sense data are going to decide the thing and that certain of our beliefs are tied so directly to sensation that they are forced upon us. I don't think that's right. We interpret the experiment, just as we do anything else. In the same way, we experiment all the time in moral matters. Just by acting, we express our values and often find out how things work out.

Q. So the authorization of a value or a belief derives primarily from meaning holism, our appeal to other values and beliefs?

A. This is a pretty long story to get into in any depth. Let's just say that a very close connection exists between what we actually value and our value judgments, and our value judgments are not different in kind from our factual judgments. There's no principal distinction to be made. There's not a distinction that allows us to say that in one case truth has an application and in the other case it doesn't.

Q. Given these issues of value, how might a feminist employ your work? For example, how might your conception of language contribute to a feminist critique of political issues or social institutions?

A. I don't know. Its'not that I don't have any view about feminism, but I don't see that my philosophical views have any special bearing or application here.

Q. Would you say that language is so marked by gender that women think differently about the world than men do?

A. Well, I certainly think that some people have concepts that other people don't. This just means there are classifications that they have at their disposal and words that reflect these differences, just as a chemist has concepts that I don't have. So, I'm perfectly willing to believe that there are societies in which women have some concepts that are not widely shared or even shared at all by men. We know that there are cultures in which women speak a different language, but I take it that this is not really the basic question about whether or not a different language can or can't be translated. If you're asking me how widespread language differences are in our culture, I just have to say that such differences don't seem to me to be a major barrier to communication between the sexes. I think that the real barriers are economic and political questions of power

Q. You mentioned previously that you've taught a wide variety of philosophy courses during your long and distinguished career. How do you view writing in your courses? Do you see a relation between your philosophy of language—or your theory of communicative interaction—and your teaching?

A. Regarding your second question, I'm not quite sure, but there's a sense in which I don't think of myself primarily as instructing or, at least, passing along information. I tend to tell students what I think, and I try to tell them in such a way that they feel free to disagree with me. I welcome live exchanges, and I prize these exchanges over just straight lecturing. At the beginning of our discussion, you touched on the topic of writing, and it's something that I have some thoughts about. I do think there is a big difference between communication by writing and by speech. In what I think of as the best kind of teaching, this difference comes very much to the fore. As it seems to me, words which are extremely important to us—especially the big words that philosophers are fond of like truth, good, right, courage, sincerity—really do change in force according to the situation in which they're used. There are cases where we're apt to agree whether these words apply or don't, and then there's a huge shadowy area which often concerns cases that don't come up all the time or even cases that we just imagine for the sake of exploring the concepts. In the course of talking, we temporarily sharpen concepts through dialogue. 'That can't happen in writing, at least in the same way. So, I think that in conversation itself, we give words shapes, especially these big words which are important but vague. Perhaps these shapes will be ones that will remain with us after the discussion. A whole lot of philosophy is like that and teaching, too, if it's done right. It's a matter of people discovering what they think. It's not a matter of people bringing sharp ideas into conflict and then deciding who's right. That can happen, too, but much more important is the situation in which ideas are taking shape as the conversation goes on.

Q. I'd like to pursue this topic. How would writing fit into your triangulation metaphor? Let's distinguish between the production of discourse and the reception of it. Now, we can see pretty clearly how the reception of discourse would work; a text would be an object in the world, and we would triangulate with it and formulate different passing theories about its meaning. However, what about the production of discourse—the actual production of an object in the world? How''s that going to fit into your metaphor?

A. First of all, it seems to me that a lot of different cases exist. If you have two people who are dumb sitting next to each other and writing notes, then the fact that the note is written hardly makes any difference. It's the same thing as talking. If they're writing each other letters, well now a difference has occurred. For example, certain kinds of indexical gestures are lost. You can't point to something in a letter. Temporal references slip and so on. If you're writing a proclamation to a group, then further indexical elements drop out, because you're not sharing your ideas just with one person but with a lot of them. If you're writing a will, you're not going to be there to interpret it. So, I envision a whole continuum. A novel, for example, is an especially interesting case, because typically the author does not expect interaction with the readers. It doesn't mean that there is no feedback; it's just not relevant in any very important way, unless you happen to be Rushdie. This lack of interaction surely does make a big difference with the triangulation idea, and you have to ask how to apply the idea. I think the only way to do this is to say that something like novel writing absolutely depends upon the prior existence of conversational exchange. People have to have been in the triangular situation before they could make anything out of a novel.

Q. So, the principle of charity exists here in full force. The novelist must presuppose a large body of shared understanding about rationality and the world in general in order to be able to produce the text in the first place.

A. Absolutely. The normal way in which language gets related to the world is mostly lacking in a novel, or, let's say, it's deferred. It's established through indirect connections rather than direct ones. You can't learn what the proper names refer to from reading the book itself, and, of course, in a novel they normally don't refer to anything. You have to have learned it in some other situation. If someone were to talk about Paris and London or something like that, you can't learn from the book where they are, because the book can't point. The causal connections are lost to you; there's a lot of dependence upon the kind of case that I think of as basic. So, what is radical interpretation like when it comes to interpreting a book? I think that's a very tricky question and to tell you the truth, reading the essays in Dasenbrock's volume [Literary Theory after Davidson] has made me think about this. I'm surprised I didn't think of it before, since I began with a great interest in literature. I see that it's an extremely interesting question, and with all the discussion nowadays about the relevance of intention in interpretation, I need to think about some things I haven't thought about very hard. On the other hand, reading various things has made me realize that people who are talking about literary criticism need to make a lot of distinctions that they haven't been bothering about, at least the critics that I've been reading. They say either that intention matters or that it doesn't. Well, you really can't put it that way, because certain intentions on the part of the author must be known or assumed in order to make anything of the text at all. This is so outrageous an example that it may sound silly, but if we were to discover that the Iliad isn't in ancient Greek, that it's in an unknown language we just now learned to decipher and it doesn't mean anything of the sort that we thought it meant, would we go on reading it as being Greek? Some people say, "Yeah, read it any old way you please," but that seems wild. If we want to understand what's there, we need to know what the language is more or less. So, the intentions can't be totally irrelevant. There's no way that a book can say on its cover "understand this book as being in ancient Greek," because those very words may mean something else in the language it's written in.

Q. But clearly you can't go the other route either and say that it is only intention that authorizes interpretation.

A. No, no. All I'm saying is that you can't say that intention has nothing to do with it. On the other hand, it seems to me that when were reading something as literature, some of the authors intentions in writing it are indeed irrelevant. The thing is that any act at all has many intentions. There's not some one intention that either counts or it doesn't. Some of the intentions are relevant to our interests and some are not. The problem is to distinguish the various intentions, and it seems to me that here there are some very important and interesting things to argue about.

Q. How do you do that? How can you claim that you have discovered the relevant intentions as opposed to other kinds of intention?

A. Well, at the moment—and I may change my mind about this in the next couple of weeks—but at the moment, when it comes to aesthetic matters, I tend to say there's no point in telling other people what they should be interested in. Anything that you learn about a piece of literature may change its value to you—its interest, its informative content, whatever. Reading biographies of the author may be considered very low class, but, in fact, it may alert you to things that are in the work that you wouldn't have noticed before. I don't have some theory about what the right thing to look for is, but I could still make a lot of interesting distinctions. It's just that the discussions I've been reading in literary criticism about intention just seem to me to miss the fact that there are endless intentions involved in every single action.

Q. Let's go back for a second. Correct me if I'm mistaken here, but it seems that one of the implications of your triangulation metaphor concerns the relation of composition studies and literary studies. When we write or produce discourse, were interpreting, just as when we receive discourse, and in both cases there is nothing that helps us in advance to ensure that we have things right. Because writers and readers engage in a very similar activity, your conception of communicative interaction gives us a way of talking about a connection between discourse reception—what we mostly do in literature courses—as opposed to the production of discourse, which has been traditionally segregated in composition courses.

A. I completely agree with you. Anybody who is writing or speaking is constantly, if not consciously, asking him or herself, how will a reader understand this? What will the reader make of it?

Q. Your conception of communicative interaction, which we've been discussing, and your causal theory of action have created a great deal of controversy. What do you take to be the primary criticism leveled against your ideas? I think especially of someone like Charles Taylor, who has complained about some aspects of your work.

A. That's an interesting question. There are a lot of things that I've maintained that people have criticized a great deal. I find it hard to pick out just one issue. I could think of five or six theses that have attracted hundreds of articles and books, almost always critical. My causal theory of action, for example, rather won the day for a decade or so, and now it's under attack increasingly. In fact, it's absolutely true that in my concentration on what seemed to me to be the bigger issues, I neglected lots of fine points which have come to the fore, and I have to recognize as perfectly good complaints peoples objections to this fact. If I think of it chronologically, my theory of causal theory of action was the very first place where I stuck my neck out; next came my claim that a theory of truth for a language, more or less in the style of Tarski, could serve as a theory of meaning. Over the years this position has been criticized from many points of view; in fact, in light of the criticisms, I had to change the doctrine. I've made numerous adjustments so that now people are accusing me of being inconsistent, which is certainly true. Compare what I wrote ten years ago with what I write now;it's clearly different.

Concerning Charles Taylor, I've been the target of Chuck's attacks a number of times. I remember two cases when I was present: one was in London, when he gave a talk on philosophy and language, and then two summers ago in Santa Cruz, when there was a six-week institute on Heidegger and me. The idea of the institute was that Heidegger and I were both anti-Cartesians but obviously came to that position by very different routes. The central question had to do with the way these two such different ways of thinking about things could lead to a common conclusion. The ecumenical position was sort of the official one, but Taylor, who was there for a week to give lectures on the subject, said, "Look, Davidson and Heidegger couldn't be more completely different." For me, the curious part of Taylor's position was his claim that I was the one who was supposed to think that language is something that you can isolate and treat by itself apart from social context. This seems to me altogether wrong. Consider the passage where I say that there is no distinction between knowing a language—being an interpreter of a language—and knowing your way around in the world, a view which I take very seriously indeed. I think I would say that Taylor has not understood my work very well, but lots of people, of course, get something wrong here or there. However, in general, I don't complain that I haven't been understood. I would like to say this: one has one's relatively limited points, and then one also has a big picture. I would say nobody seems to me to have done as well in seeing what my overall picture is than Rorty. Even though Rorty and I think of things in quite a different way, he has made an effort to follow what I was up to, and it seems to me he gets the main thing right. It's extremely reassuring and pleasing.

Q. So, you validate Rorty's readings of your work?

A. He usually gets the biggest things. He sees what I'm up to and sees things I haven't seen or said as plainly. Sometimes he changes his mind, too. At the Santa Cruz institute, for example, Rorty had just read my Dewey lectures, which are three lectures on the concept of truth, and at that time he was rather scolding me for taking the notion of truth too seriously. He thought I'd gone off the track that I should be on. But then we met again about a month ago at a conference at Santa Clara, and he said that he's been rereading those lectures, and he now sees what I'm up to and thinks it's all right.

Q. How would you distinguish your work from Rorty's? In one way—and again correct me if I'm wrong—you seem to suggest that answers exist to enduring philosophical problems like skepticism. Rorty maintains that we should just change the topic since we can't answer the skeptic without accepting the skeptic's presuppositions. Is this a significant difference between you and Rorty? Or is it a difference at all?

A. I don't know. We certainly have a somewhat different approach to philosophy, a fact which I think we both appreciate. To say, as Rorty does, that the skeptic should get lost, or to try to answer the skeptic as I do, seems to me to be more a difference in style and attitude than a sharp distinction. I think each of us has teased the other about this. I say to him, "You just want to tell the skeptic to change the subject; why then do you refer to my arguments?" And, at least on one occasion, I think he said, "I'm not really interested in your arguments, it's your conclusion I like." I see this as a temperamental difference rather than anything else. My view about doing philosophy has always been that people should do it in whatever way they most enjoy doing it, and that there isn't one correct way. Rorty has given philosophy a big boost because so many people who are not in philosophy are interested in his work and are reading it. I think that this is all to the good, although not everyone shares my view. On the other hand, I don't think I could do what he does, though he might be able to do what I do.

Q. Yet, your influence on Rorty has been profound, as he acknowledges. In fact, most of his current views about language may be traced directly to your writings.

A. I think I'm right in saying that the piece that marked the point at which Rorty changed was something that was called "The World Well Lost." He wrote it after he read two lectures that I gave at the University of London attacking the idea of conceptual relativism, and he kindly gives me full credit. At that point, I hadn't published this material. I did subsequently, but I knew him through that formative period. We were colleagues at Princeton and, in fact, during the year or two that I was chairman there, he was the assistant chairman. We talked a lot, and we always got along extremely well.

Q. That you two teamed up to direct a department suggests a question about the role you see philosophy playing within the structure of the university. Rorty seems to think that philosophy has done its work. Do you?

A. Well, there are two questions there. One is just sort of social history—what has actually happened—and the other one is what do I think happened? Rorty's idea that philosophy has talked itself out of a job seems to me wrong. I think what he has in mind mainly is a certain way of doing epistemology which was a central part of philosophy through a period of a few centuries. I don't think of analytic philosophy quite in the same way that Rorty does. He thinks of analytic philosophy as being almost entirely concerned with epistemology, with producing one or another response to skepticism; however, I think of analytic philosophy as a method. The method is one which tries to state problems and arguments as clearly as possible and respects the pursuit of truth. So far as the place of philosophy in the university is concerned, my own views are not very sharp. I don't think there is some one thing that philosophy ought to do. Philosophy usefully overlaps a lot of other fields; literary criticism is one; artificial intelligence—cognitive psychology—quite generally is another; mathematics is another; the law is yet another. This is all to the good. Historically, philosophy has been like that. In fact, Plato and Aristotle made no distinction between psychology, mathematics, biology, philosophy; they were all part of the same thing. Some subjects like physics, when they became serious, went their own way, though there's plenty of overlap still. I prize those philosophers who really are competent in other fields. There are a handful of people who really know physics or really know certain branches of mathematics, or really know the theory of evolution. So, I'm against trying to draw lines somewhere and say, "This is philosophy and this isn't."

Q. It seems that these disciplinary boundaries within institutions are growing indistinct, if they're not being erased entirely. I believe that attempts to blur these distinctions are worthwhile.

A. I do, too, but it's probably not something that in general you can do much directly to promote. I think it just happens. Cognitive psychology was given a huge artificial boost, but the most successful cooperation I know about in which philosophy was involved really started before there was the injection of money. Linguistics and philosophy came into serious contact largely as a result of Chomsky's work which changed the way linguistics looked to philosophers. That's an area of cooperation that continues in a big way. There are a lot of people who are in both camps; you don't have to decide whether they are linguists or philosophers: they are both.

Q. However, some people get unnerved by this kind of talk. They don't desire to relinquish the master narratives that have held us together—or kept us apart—for so long. What do you say to calm those people?

A. Well, I guess I don't know how to calm them until I know exactly what it is that they are worried about. As I said earlier, there are two branches to foundationalism. One depends on the certainty that we're supposed to have about the contents of our own mind and the idea that this certainty is the basis of knowledge. I don't question that we know what's in our own mind. There's a presumption that we're right about what we think we think; first person authority is not something I would call into question. What I do call into question is the view that we can construct a world solely on the basis of what we know about the contents of our own minds. I substitute for that idea the triangulation situation which gives objectivity three legs to stand on. The other element in foundationalism stresses the dependence of knowledge on the senses. The senses also play a role for me because the triangulation argument has us reacting to the world and to other people. So, I don't see that anything is lost, but if someone has a reason to think that I'm undermining something valuable without replacing it, I need to be told what it is.

Q. I was thinking more along the lines of social philosophers or even theologians who think that something outside of our own beliefs must exist in order to authorize those beliefs.

A. If the argument is that I don't see anything to correspond with the truth, then that is certainly right, but no one has ever been able to make good sense of correspondence theories to begin with or to show how they would be any use. On the other hand, I've never said anything against God, but if He ever spoke to me, I guess I would have to raise the question on my own whether He was right.

Q. So, you would have to triangulate?

A. [Laughter] Yes.