Kit Fineinterviewed by Richard Marshall.


Kit Fine is a groovy metaphysician. He keeps asking questions about the fundamental questions of reality. He has the cool title of Silver Professor of Philosophy and Mathematicsat New York University. He has written the books Modality and Tense: Philosophical Papers, Semantic Relationism, The Limits of Abstraction, and Reasoning With Arbitrary Objects. He has brooded on Aristotle's account of logic and modality. He often writes out of curiosity rather than commitment. This curiosity led him to his theory of abstraction. You can get his flavour by looking at this lecture at the INPC 14 conferenceat Boise, Idaho, in April last year where he proved it is possible to discuss the determinate/determinable distinction under the influence of an indeterminate number of glasses of wine. He thinks we should do our metaphysics clean and clear of empiricism. He doesn’t think we are currently in a position to answer many key philosophical questions, such as whether freewill is possible, or what consciousness is. He doesn’t burn his armchair.

3:AM:You are a leading metaphysician and take a distinctive approach to this branch of philosophy. So what do you think metaphysics is and how do you approach it? You tend to approach it as an a priori subject and this raises the interesting question as to how is it possible for you to find things out about the world from your armchair? And what can it do that maths can’t? Your book The Limits of Abstractiondevelops out of curiosity about Frege’s work on the foundations of mathematics and abstraction, and this suggests that metaphysics and mathematics are related in some degree? But are they really, beyond their a priori status?

Kit Fine:Metaphysics is the philosophical study of the general nature of reality. It asks questions like: what is the nature of space and time?; what is the relation between mind and body?; do abstract objects exist or is everything concrete? I believe that metaphysics is like mathematics in being a priori; it comes to its conclusions without the benefit of any particular experience but, in contrast to mathematics, it is concerned with the general categorical features of reality and not with the purely mathematical or 'structural' features. It is a mystery how we can acquire knowledge of this sort from the armchair and perhaps even a mystery how we can acquire mathematical knowledge from the armchair, but I am in little doubt that we can in fact obtain such knowledge.

At the very least, for many of the questions of interest to metaphysics - such as the existence of abstract objects - it is unclear how empirical inquiry of the sort with which we are familiar in science could be even remotely relevant. In my monograph The Limits of Abstraction, I did employ mathematics. But the mathematics there is in the service of metaphysics and of philosophy more generally. The questions ultimately of interest to me are philosophical, such as: what is the nature of numbers?; how do we refer to them?; how do we know they exist?.

3:AM:You grapple with metaphysical questions ordinary folk like to think about too. So you have written about ontology, about what is real. We ask whether numbers exist, or chairs, or atoms and you suggest that there is an inherent confusion that haunts answers to this question. The confusions involve mistaking ontology with quantification. And this is due in part to Quinewho thinks the question is scientific when it isn’t. You say, ‘He asks the wrong question, by asking a scientific rather than a philosophical question, and he answers the question he asks in the wrong way, by appealing to philosophical considerations in addition to ordinary scientific considerations.’ I think this is a really helpful and illuminating approach to the issue that as I said is one that captures everyone’s imagination whether or not one is a professional philosopher. Can you say something about these confusions?

KF:This is a big topic, but let me expand a little. Even 'ordinary folk' may wonder whether there really are numbers or chairs or atoms or the like. Perhaps the world is entirely concrete or consists entirely of microscopic particles or is merely a construction of our minds. But what are we asking when we ask such questions? For surely we can all agree that there is an even prime (viz., 2) and an odd prime (say, 3) and so there are numbers. Or we can all agree that we are sitting on some chairs and so there are chairs. Or we can all agree that there are water molecules, each of which is made up of two hydrogen atoms, and so there are atoms. The answer to all of these questions appears to be obviously 'yes' and so do we even bother to ask them?

Quine thought that in asking such questions we were indeed asking 'quantificational' questions. For the case of each kind of object in question, we were asking whether there was an object of this kind. However, he thought that the answer to this question was not as obvious as one might have thought and that subtle philosophical considerations might be involved in attempting to answer it. In the case of chairs, for example, we might establish that there were no chairs by showing that every statement apparently about chairs could be paraphrased into one that was about particles.

I think that this is a mistake. In asking these ontological questions, we are not asking about what there is but about what is real. Are numbers real? Or chairs? Or atoms? The quantificational questions are relatively straightforward - they are to be answered by common sense or by science. Philosophy does not come into it. But the questions about reality are deeply philosophical and it is only through having a conception of reality, a philosophical Weltanschauung, that they can be answered.

3:AM:So you work to draw a distinction between existence as a quantifier and existence as a predicate. You reject that the question requires us to think of existence as a quantifier and instead argue that it is in these settings used as a predicate meaning something like ‘real’. Is that about right? Can you spell out this for us so we can grasp why this is helps us answer our questions about what exists? Do we end up being naturalists?

KF:Part of the problem here is that terms like 'exists' and 'real' are used in many different ways, especially in philosophy. For me, the fundamental question of ontology is not 'what is there?' but 'what is real?'. But what do I mean by 'real'?

As a first stab, I would say that what is real is what one must make reference to in giving a description of reality. Thus suppose you were a platonist and thought that reality included facts about numbers, such as that 2 + 2 = 4. Then numbers for you would be real. This is to explain the real in terms of reality. But what is reality? Here perhaps the best I can do is to explain the role of reality in metaphysical thinking.

Reality is what accounts, in the most perfect way possible, for everything that is the case. Suppose, for example, that I thought that the existence of a chair consisted in nothing more than certain atoms arranged in the shape of a chair. Then this would strongly suggest that reality did not include the existence of a chairs, since I could account for their existence in other terms. It is not at all clear, if one were to pursue this method of inquiry, that one would end up being a naturalist. For it is not at all clear that one can account for everything else in terms of the naturalistic facts.

3:AM:This is part of all sorts of debates, reductionism vs eliminitivism, and realist vs anti-realist arguments. To some outside mainstream philosophy these debates can seem a bit forbidding and technical but they are actually about issues that very often as children we start to ask and continue being puzzled by aren’t they? Can you say something about how philosophers are currently wrestling with these deep questions and perhaps explain why sometimes the discussions can become difficult for outsiders to grasp?

KF:There is a great deal of emphasis right now on just what the questions are. What are we asking when we ask 'do numbers exist?'. This branch of metaphysics even has a name - 'meta-ontology'. And so it is no surprise, given that there is no agreement on what the questions are, that there is no agreement on how they are to be answered.

But one approach to these questions, which I myself favor, is to ask 'what grounds what?'. In what does the existence of a chair consist? Or in virtue of what is an action right or wrong? These questions are meant to get at the ontological ground of the phenomena in question and it is thought that by working out what grounds what we can get at what is 'real' or 'fundamental'.

I suspect that the discussions are difficult for outsiders for a number of different reasons. One is that the discussions are sometimes quite technical - they might involve consideration of the logical role of the quantifiers, for example. Another is that they can involve subtle philosophical distinctions - between what there is, for example, and what exists or what is real. But the main reason is probably not peculiar to ontology - it is just that analytic philosophy as a whole has become highly specialized, with its own concepts and presuppositions, which can be very difficult to understand without some formal training in the subject.

3:AM:A groovy metaphysical question that philosophers have been grappling with for some time is the question of mereology, of the relationship of wholes to parts, both in space and time. Again, this is one of those questions that as a child you worry about. It raises all sorts of questions that I think are basic, like how can something that changes yet be the same. Persistence of identity is one of those loaded issues that never goes away. You’re a key player in contemporary arguments about this. So can you say something about how you think we should go about thinking about the problem. Perhaps you can use the example that you have written about where we are to ‘consider a piece of alloy and a statue that always coincide. Are they the same (the monist position) or not the same (the pluralist position)?’

KF:Yes, the concept of part is an important concept both in ordinary life and in metaphysics. My own view is that contemporary philosophers have had an unduly restricted conception of part. They have thought that the parts of an object are like the divisions one might make in space. Suppose, for example, that I have a rectangular surface. Then I might divide it into the upper half and the lower half, or the left half and the right half, or the diagonal half from top left to bottom right and the diagonal half from bottom right to top left, and so on. For these philosophers, all parts are like that.

But consider a sentence such as 'John likes the movie'. Then this has two main parts: the noun phrase 'John'; and the verb phrase 'likes the movie'. The verb phrase, in its turn, has two parts: the verb 'likes'; and the noun phrase 'the movie'. And, finally, the noun phrase 'the movie' itself has two parts: the determiner 'the'; and the noun 'movie'. Thus the sentence has a complicated hierarchical structure (which linguists set out in a tree). But there is nothing in the spatial model that could account for this hierarchical structure - everything is flat. In my view, the hierarchical part-whole structure that we find in a sentence is everywhere about us - in atoms, molecules, chairs, trees, cities and the like - and philosophers have come to an erroneous conception of the nature of these entities through having an erroneous conception of their parts.

Consider the example of the statue and the alloy of which it is made. They are not the same since the statue will cease if squashed into a ball while the alloy will still exist. But what then is the relationship between them? On the flat model of part, it is hard to say since there is just one region there, equally occupied by the statue and the alloy, and so how can they be different? But on the hierarchical model there is no difficulty, since the alloy might constitute a whole, the statue, at a higher level.

It is as if we were to say that the noun 'fire' properly constituted the one-word sentence 'Fire' with the sentence itself at a higher grammatical level than the noun itself or if we were to say that Socrateswas a part of singleton Socrates, the set whose sole member is Socrates, even though there were no other 'material' by which the member and the set might be distinguished. So just as the word 'fire' can constitute the sentence 'Fire', which is not identical to the word, or the individual Socrates can constitute the singleton Socrates, which is not identical to the individual, so the alloy can constitute the statue.

3:AM:Another terrific argument philosophers are currently engaged in thinking about is the idea of possible worlds. This is a logical tool for handle counterfactuals, but it has metaphysical consequences as well doesn’t it? For example, you argue that Aristotle was close to being committed to some sort of Megarian view that every possible world is one where the possible and the actual coincide. However, you make a distinction between understanding ‘world’ as ‘witness’ and ‘locus’. Although formidably technical in detail and proof, I think the argument is drawing a picture that is again mind expanding. Can you tell us about this argument as I think it broadens the scope of how we might think about issues about modality that again are pretty primitive and universal, like, ‘could I have been someone else?’ or ‘what if I had been born a lizard?’, standard fare for many a bored teen and for anyone with a scintilla of wonder in them?

KF:Yes, there has been a heavy emphasis on possible worlds in the philosophy of language and metaphysics. I think that to a large extent this emphasis is misplaced, that the work done by possible worlds would be better done by other means.

So instead of saying that necessarily, Socrates is a man, it would be more illuminating to say that Socrates is by his nature a man, putting the emphasis on the nature of Socrates rather than what is necessary. Or again, instead of understanding the counterfactual 'if the match were struck it would light' in terms of what would happen in the closest worlds in which the match is struck, it would be more illuminating to talk about the consequences of a situation (not a whole world) in which the match is struck.

The distinction between worlds as 'locus' and as 'witness' concerns a subtle point in the interpretation of Aristotle. Let me briefly explain what I had in mind. I can say 'possibly, you might have been a lizard'. But why the 'might'? For the possibility in question is not that you mightbe a lizard but that you area lizard. I suggest that the first modal locution, 'possibly', locates the world in which you are a lizard and that the second modal locution, 'might', bears witness to its being such a world, one which realizes a possibility. I then appeal to this distinction in explaining some puzzling things Aristotle says about the logic of modality.


3:AM:Your book Semantic Relationismbegins with a puzzle that you note has been puzzlingly not been noted as a puzzle by many. The puzzle is how are we able to say the same thing from one occasion to the next. The puzzle becomes clearly important when we consider a population who can’t say the same thing from one occasion to another. And if extended to thought, so we wonder how we can think the same thought on different occasions, then again, not being able to do so would put paid to memory. So it’s a capacity that is unremarked upon but remarkable, and you develop your theory of ‘semantic relationism’ to explain it. This theory is designed to replace a view that resemblance is the key to understanding how we can repeat ourselves. Can you tell us what your theory is and why it is far better than the resemblance view?

KF:Take two lines and suppose they are parallel. Then in virtue of what, if anything, are they parallel? One possible answer is that they are parallel in virtue of having the same direction. In other words, there is a certain feature of lines, their direction, and two lines are parallel in virtue of having this feature in common. This is the resemblance view. Now it is not at all clear that it is correct.

One might well have thought that lines do not possess an 'intrinsic' direction and that far from understanding parallelism between lines in terms of their direction one should understand direction in terms of parallelism, that direction is what is common to parallel lines.

Now consider the analogous case of synonymy, or saying the same thing. You say 'Obama is president' and I say 'Obama is president' and we are thereby same-sayers. But in virtue of what, if anything, are we same-sayers? Almost every philosopher who has considered this question has wanted to adopt the resemblance view. They have wanted to say that we are same-sayers in virtue of saying the same thing. In other words, there is something I say and something you say and it is in virtue of this thing being the same that we are same-sayers.

The basic idea behind semantic relationism is that this tempting thought is mistaken, that, just as in the case of direction, what is said is to be understood in terms of same-saying rather than the other way round. One advantage of the relationist view over the resemblance view concerns the question of why the sentence 'Hesperus is Phosphorus' is informative while the sentence 'Hesperus is Hesperus' is not, given that they appear to say the same thing. viz. that a given planet is the same as itself. The relationist can admit that they say the same thing in the sense of having the same intrinsic meaning, while denying that they stand in the same-saying relation. Just as two lines intrinsically the same in terms of length can be relationally different in terms of parallelism, so two sentences intrinsically the same in meaning can be relationally different in terms of same-saying.

3:AM:Some journalists and non-philosophers have been rather carping of certain types of philosophy, saying that its trivial, nit-picking and no longer doing what philosophers of old did. They also complain that they can’t read it anymore because it is technical and difficult to grasp what is at stake. Reading your work, what can seem philosophical abstruse issues and technically dense arguments open up startlingly fresh and new imaginative vistas don’t they? How would you characterize the point of doing philosophy?

KF:Philosophy is no more interesting than its questions. What is the nature of space of time? What is the relationship between mind and body? How do we know it's not all a dream? If you're not interested in the questions then you're not going to be interested in the subject and you might as well turn to something else. It might appear surprising that these questions, which we can all appreciate, should lead to work that is so difficult to understand. But consider Fermat's Last Theorem. All of us with a modicum of mathematics can understand the theorem but very few of us can follow the proof. I am not suggesting that the disconnect between the questions and the answers is as great in philosophy as it is in mathematics, but I do not think one should be surprised if seemingly innocent questions should lead, step by step, into ever greater levels of complexity and difficulty.

3:AM:Eric Schwitzgebelargues that metaphysical theories at some point must part company with common sense. He labels this position ‘crazyism’. Are you a crazyist in this sense?

KF:No, I'm a non-crazyist. I'm firmly of the opinion that real progress in philosophy can only come from taking common sense seriously. A departure from common sense is usually an indication that a mistake has been made. If you like, common sense is the data of philosophy and a philosopher should no more ignore common sense than a scientist should ignore the results of observation. A good example concerns ontology. Many philosophers have wanted to deny that there are chairs or numbers of the like. This strikes me as crazy and is an indication that they have not had a proper understanding of what is at issue. By recognizing that these things are crazy we can then come to a better understanding of what is at issue and of how the questions of ontology are to be resolved.

3:AMExperimental philosophy asks that philosophers leave their armchairs. So they would ask perhaps whether maths really is a priori, say, or whether principles of abstraction do work like Frege</strong > said they did, and then they’d experiment on the folk to find out. So what is your view about the role of non-a priori approaches to metaphysics?

KF:I am not especially enamored of my armchair and would be happy to leave it if I thought that it would be of help in answering the questions of interest to me. But I fail to see how it could be. Consider the question of whether mathematics is a priori or whether principles of abstraction of the sort proposed by Frege might provide a foundation for a significant part of mathematics. How could asking the folk possibly be of any help in answering these questions? Physicists don't ask the folk to look down telescopes and mathematicians don't ask folk to assess the plausibility of their axiom. And so why should it be any different for philosophy? Or, take another analogy. We don't ask the folk to read X-rays since it takes skill and training to know what to make of them - to understand whether a particular blotch, for example, has any real significance. It is no different, it seems to me, in regard to the intuitions of philosophers. One needs skill and training to know what to make of them and it would be a terrible retrograde step to rely instead on the untutored judgments of ordinary folk.

3:AM:You’re also concerned to pursue questions that you think we’ve got the right tools for. You give as an example the case of economics where they started to use the tools of physics in order to get precision and went disastrously wrong until Keith Arrow</strong > sorted out the mess. So are there areas in philosophy at the moment which you think are pretty futile to pursue, or examples from the past to illustrate what you think?

KF:Yes, you have to know when to ask a question. Until recently, there was no point in trying to prove Fermat's Last Theorem. The mathematical tools were not available. And similarly in philosophy. But great judgment is required in order to know when is a good time to ask a question. You have to match the tools available against the difficulty of the problem; and it is very easy to over-estimate the power of the tools and to under-estimate the difficulty of the problem. In my opinion, many of the questions philosophers currently consider are not worth pursuing - not because they are of no interest but because we do not yet possess the tools to make any real progress with them. These include the mind-body problem, the problem of free will, and the problem of skepticism. I feel that contemporary philosophers asking these questions is like the ancient Greeks asking about the constitution of matter or the nature of the cosmos. Good questions, to be sure but no point, at the current stage of inquiry, in trying to answer them.

3:AM:Have you ever changed your mind on a significant point through thinking hard about something or listening to someone else’s argument? It seems rare in philosophy and this is rather peculiar isn’t it if philosophical enquiry is about being open minded?

KF:Yes, I have changed my mind on a number of issues and one thing that has surprised me is that I have become more Kantian in my thinking. I used to hate Kantand everything he stood for, while recognizing that he was one of the greatest of philosophers. But I have now come round to some of his ideas, especially in the philosophy of mathematics.

Yes, it is relatively rare for philosophers to change their minds - and perhaps somewhat disturbing. Perhaps one reason is that decisive refutation is so unusual in philosophy and most of us have the tendency to stick to our opinions unless we are given some pretty compelling reasons to change them. A related reason is that good ideas in philosophy are like good wine; it needs to mature to be at its best. So we spend a lot of time letting our ideas mature and, before you know it, it's too late to do anything else!

3:AM:After the low period of the Positivist attack on metaphysics, the subject seems to be in the ascendency again. Is this your feeling about the subject and how do you account for its resurgence?

KF:Yes, it is in the ascendent. Why? Perhaps the main reason is that you cannot keep a good question down. You can be told that certain questions are meaningless and even be given apparently compelling arguments for thinking that they are, but if they strike you as meaningful then you are not going to be convinced.

Suppose that a friend of yours has been in a serious accident and you wonder whether they have died. A philosopher then tells you that it is a meaningless question and even gives you apparently compelling arguments (where do you draw the line between life and death etc etc). Are you going to stop worrying? Of course not. And the same is true for many of the questions of metaphysics. I personally could not care a toss for the many arguments that philosophers have presented for thinking that certain metaphysical questions are meaningless, even if I could not see what was wrong with them. If, on a careful reflection, a question just struckme as meaningful, then this intuitive evidence of meaningfulness would far outweigh any philosophical argument to the contrary.

3:AM:Having said that, there seems to be a kind of philosophy envy coming from the physicists these days. So Hawkingrecently said that philosophy is dead because physics explains it all which I took as an act of desperation on behalf of a scientific community that doesn’t do well in drawing very sophisticated inferences from their data. What do you think about this?

KF:I haven't read Hawking on this but I assume (or, at least, I hope) he just had cosmology in mind and not the whole of philosophy. Perhaps he is right that there are certain questions in cosmology that were once thought to be belong to philosophy but should now be thought to belong to physics. But even here, I suspect that philosophy can be of great help - not, of course, in doing physics - but in interpreting physical theories and relating them to the questions of interest to us.

3:AM:Are there any books or films outside of philosophy that you have found illuminating for your work?


3:AM:And finally are there any books on metaphysics that our smart but not necessarily philosophically trained readers at 3:AMwould benefit from reading, on top of yours of course?

KF:Saul Kripke</strong >, Naming and Necessity, though it is as much on the philosophy of language as metaphysics.


Richard Marshallis still biding his time.