:: Article

a Wittgenstein Kripke vertigo disturbance

Arif Ahmed interviewed by Richard Marshall

Arif Ahmed is a seriously funky philosopher who has never stopped feeling the thrill. He thinks Wittgenstein refuses premises, refuses conclusions and never answers straight. He keeps rethinking the relation of inner to outer, has examples about martians and triangular prisms, shows that Brutus’s suicide might be a problem (but that a Jasrow duck/rabbit might have the answer) and is always brooding on Kripke’s thought that every time we use a word it is a jump in the dark. Philosophical thinking is a sinister vertigo for him, which makes him a Jimmy Stewart mega-dude of the philosophical jive.

3:AM: Why did you become a philosopher? Has it been good so far?

AA: I studied mathematics as an undergrad. My University ran an excellent final year program in which you could choose 3 from about 50 papers, on topics ranging from mathematical logic through group theory to financial mathematics. I chose two options in mathematical logic and so happened to read something by Quine (I think it was ‘On Frege’s Way Out‘, which is about Russell’s Paradox); this took me to Russell and then on to Wittegnstein’s Tractatus.

I had always been vaguely puzzled by the fact that logic was supposed to be at the bottom of all science, that it described not only what was true of our world but what had to be true in all possible worlds. In that case you would expect logical laws to have nothing arbitrary about them. But if you look in any standard logic textbook the first thing you notice is that the axioms of the calculus are complicated, non-symmetric and generally have an arbitrary look about them. How could this be? The first of many things that thrilled me about the *Tractatus* was that it answered this question (or at least stopped me worrying about it): the relevant discussion is at 5.511-5.5151.

From that point on I knew that I was going to become some sort of philosopher, however long it took me. And so far it’s been terrific. It’s very hard to explain to anyone who has never felt it the quite peculiar thrill of working out something that is (i) novel (ii) interesting (iii) true, even in some very narrow and abstruse field of enquiry – or even just thinking that you have, which is what usually happens to me.

In addition I work in a department where not only do I have very smart colleagues, but also my graduate and undergraduate students are generally extremely intelligent, industrious and counter-suggestible. I feel very lucky.

3:AM: You edited a collection of essays about Wittgenstein. Wittgenstein seems like a philosopher who never seems to get a settled interpretation. It’s not so much that philosophers aren’t sure what to make of the answers he gave to puzzles, but that they can’t decide what the puzzles are. You say that the Philosophical Investigations is probably the most disturbing work of philosophy since Hume’s treatise. So firstly, why is Wittgenstein so difficult to pin down, and do you think he’s still important as anything but an historical figure?

AA: Certainly some of the blame for his being hard to pin down must fall on Wittgenstein himself, who despite having a very beautiful style set new standards of unclarity in philosophical writing, at least for analytic philosophy. It’s not only that he refuses to give straight answers to straight questions (like ‘Are you a behaviourist?’). You’d expect a philosophical treatise to argue from definite premises to definite conclusions: but in the Tractatus he misses out the premises, and in Philosophical Investigations he misses out the conclusions. So you are quite right, that even on very basic matters it is much harder to get a sense of what Wittgenstein thought than it is to get a sense of what (say) Locke or Mill thought about matters that were similarly basic to their philosophies.

But I myself don’t mind that very much. Once we have got in view the range of possible things that he might have meant by (say) s201 of PI (probably the most contested section in his entire oeuvre), I don’t care so much about which of them he actually did mean as about which if any is true.

This brings me to Wittgenstein’s enduring importance: what he wrote and said is important not just because of its influence on others but because it can be taken in a way that makes it both true and interesting, I think. Certainly the Tractatus made definite and lasting contributions to the subject – for instance, I think he was one of the first to use truth tables as a test of tautologicality. But his later work is full of remarks that when taken in a certain (and maybe unintended) way point to fundamental insights, insights that are as valid today as they ever were.

For example, consider the discussion on pp. 2-3 of the Blue Book. There Wittgenstein is considering the idea that we associate some mental impression or idea with a word, say a colour word. One reason for thinking this is that it explains how we are able to respond to public uses of that word. For instance, it is a fact that if you say ‘Bring me a red flower’ then I will indeed be able to pick out a red flower from amongst the many differently coloured flowers in this garden. What is supposed to explain this is that when you say ‘Bring me a red flower’ I get a mental image of red, and then I go to the garden and get a flower that chromatically matches the mental image.

But now suppose that you order me to ‘Imagine a red flower’. How am I supposed to do this? ‘You are not tempted in this case to think that before obeying you must have imagined a red patch to serve you as a pattern for the red patch which you were ordered to imagine.’ And yet I do in fact obey the order. So it must be that in at least some cases I can obey an order without this mental intermediary to guide me. And suddenly the idea, that this ever has to happen – that some inner mental process must guide the outer behavioural one – loses all of its power. The example in itself is trivial but it sets us up for a radical rethinking of the relation between inner and outer.

To illustrate one development of this line of thought, I’ll take as my second example just one strand of the extraordinarily fertile and wide-ranging discussion at ss139-242 of PI (a discussion that also contains at least a dozen others of similar interest and importance). There are normally two ways by which you or other people can tell what you mean by a word: one of them is what occurs to your conscious awareness when you hear the word for the first time or at other times; the other is how you go on to *use* that word in the (not always public) contexts of everyday speech, writing, following of orders etc. Now one of the things that Wittgenstein points out at ss139-141 is that these two methods might give discordant results. Thus to take his own example: suppose that whenever a Martian heard the word ‘cube’ a picture of a cube came before his mind’s eye, but that he ‘projected’ this picture onto reality in (what seems to us to be) a deviant way, the net effect being that our Martian applies the word ‘cube’ to triangular prisms.

The example illustrates three points. First: This certainly could happen: should we say in that case that he understands ‘cube’ as we do or that he doesn’t? Answer: you can say what you like. Our own concept of what it is to understand a word simply doesn’t cover such cases; and this illustrates the second point (which he states at PI 142): that many of our concepts depend on a tissue of contingent regularities that might well have not obtained. Third: it simply isn’t true that the objects of conscious awareness – the ‘ideas’ as Locke called them – that one associates with a word have in themselves any authority over its use. Construed in a classical way – as mere objects of inner awareness – the ‘ideas’ associated with a word just stand there inertly, and no more tell one how to use a word than do squiggles in the written language of an alien culture.

All of this is I think true, and important, and interesting, and it justifies though it does not wholly explain the extensive contemporary interest in Wittgenstein’s work.

3:AM: One big divide in Wittgensteinian scholarship at the moment is between Baker and Hacker’s interpretation, on the one hand, and new-Wittgensteinianism of Diamond and Conant on the other. Before saying what you think about this dispute, could you explain what are the main points of disagreement and why it matters?

AA: For reasons that I’ve already sketched I don’t have a dog in this fight. As I understand it the issue is between a ‘traditional’ reading (which actually covers many views) according to which Wittgenstein was at least trying to do substantive philosophical work in some sense – and so the content of his theses and distinctions is sensitive to that aim – and the ‘resolute’ reading according to which philosophy is just nonsense, and that Wittgenstein is aiming to get people to stop doing it – to stop wanting to do it – by producing a sort of nonsense that has that effect. So on the ‘resolute’ view Wittgenstein is trying to achieve what a very delicate lobotomy might also achieve, but on the traditional readings he is trying to do something else.

But I’m probably not the best person to ask about the content of this dispute. I’m certainly not the best person to ask about why it matters.

3:AM: Ian Proops has done much to make the Diamond new Wittgensteinianism thesis seem unlikely, but what do you think about this? Is this just a manifestation of a wider dispute in philosophy between pragmaticist philosophers like Rorty and others who wish to escape from a Lockean/Cartesian mindset and replace it with a therapeutic approach?

AA: For what it’s worth – I do think it’s unlikely, and it was Peter Hacker’s work that convinced me of this: on the Tractatus, for instance, see his essay in the Crary and Read collection (‘Was he trying to whistle it?‘). On the question of what the dispute manifests: I don’t want to speculate on the causes of the ‘resolute’ view. But the idea that philosophy results from some kind of failure of the understanding rather than being its highest expression, and that the discipline should aim at abolishing or curbing itself – that idea is surely much older than Rorty-inspired pragmatism. We find one version of it in Berkeley and another version in Kant, and neither of them could be called pragmatists.

On the dispute between the ‘therapeutic’ approach and what you call the ‘Lockean / Cartesian mindset’: if the latter is just the view that (a) some things are true and some things are false (b) it sometimes matters which: well I think that it’s true, and I don’t know any good evidence against it. ‘Must causes always precede their effects?’ or ‘Is anything really right or wrong’ (for instance) are perfectly good philosophical questions that have yes or no answers (that empirical inquiry will help us to find).

3:AM: In your own contribution to that collection you look at the temptation the philosopher undergoes to think that we can deduce the rules for using a word from its meaning. Is that right? Can you explain what the problem is?

AA: That’s right. What troubled me then (and still does) was the fact that there are some valid one-step arguments, in which the conclusion follows from the premises, whose recognition as such seems to demand resources beyond your mere understanding of the premise and the conclusion. Thus consider the argument ‘Brutus’s wife hates Brutus and Brutus killed Brutus; therefore, Brutus’s wife hates somebody who killed Brutus’. The conclusion really does follow from the premise, but to see this you need to carve up the content of the premise in a way that was not at all necessary for grasping it.

For to understand the premise you need at most to see that it conjoins two claims: the first that attributes to Brutus the property of being hated by his own wife and the second that attributes to him the property of suicide, both properties that Cassius might also have possessed. But to see that the conclusion follows from the premise you need instead to think of the premise as attributing to Brutus the property of ‘Being a killer of Brutus whom Brutus’s wife hates’ a property that Cassius surely did not possess.

But if it really is valid then (you would think) this must be because of the meanings that we have given to the premise and the conclusion. So then we have the following situation: you can grasp the meaning of a sentence without seeing it from every angle, so to speak: the meaning of a sentence is something that can have hidden features that are relevant to its deductive connections.

But that runs against Wittgenstein’s picture of things. Remember earlier on when I discussed section 139 of Philosophical Investigations: I said that one of its lessons is that the meaning of a word or sentence is not some object the apprehension of which dictates its use to us. But that is precisely what seems to be going on here. In short, if we take Wittgenstein’s picture seriously here then we are unable to account for the phenomenon of logical insight as illustrated in the example.

3:AM: So what is your analysis of what Wittgenstein is doing here and why is it important?

AA: I’m not sure that Wittgenstein ever directly addressed this sort of case. What I tried to do in that paper was to see if we could use other resources from his work to address it. In particular I thought that some of his insights with regard to aspect shift were transferable. By ‘aspect shift’ I mean what Jastrow’s duck/rabbit illustrates: you can look at a picture-duck (say) and suddenly experience a ‘shift’ in which it looks different (it looks like a rabbit) and yet the same (the lines in the drawing are visibly unchanged). What I thought was that there is an illuminating parallel betwen this sort of case and the dawning of a ‘logical’ insight in cases like the ‘Brutus’ example. It’s illuminating because some of Wittgenstein’s insights about the former might resolve the tensions for his philosophy that the latter seems to generate. Maybe he thought that there is a connection too, because he says at one point that when an aspect shifts you perceive an internal (a logical) relation.

Why does it matter? It matters if you care about the nature of linguistic meaning and linguistic understanding, because it sheds light on what sort of things they could be. Does the phenomenon of logical insight place constraints on the correct theory of meaning, constraints that rule out the separation Wittgenstein envisaged between the object of apprehension and the pattern of use? I don’t think so but it isn’t straightforward.

3:AM: Another important contribution to interpreting Wittgenstein was made by Saul Kripke. Can you say what Kripke brought to the discussion and whether Kripke’s approach still considered a valid way of looking at Wittgenstein’s concerns, or is it better understood just an important misreading?

AA: Kripke’s book is a very focused and powerful interpretation of sections 138-242 of Philosophical Investigations. The basic problem is as follows. Suppose that two persons A and B, similar in all relevant respects, are taught to use the ‘+’ sign in basically the same way that we all get taught to use it. For instance, we train them to calculate sums like 2 + 4 = 6 by counting two fingers on one hand, four fingers on the other hand and then counting all of the fingers together. For larger sums we teach them techniques of manipulating numerals on paper, and so on. E.g. when I was a child I learnt to carry numbers across from the units column to the tens column where necessary and then add it on when adding the tens, and then if necessary carry another number across to the hundreds column etc. We may suppose them also to be trained in this technique – or whichever substitute gets taught nowadays – but the details don’t matter much beyond that point. What does matter is that there is a number such that neither A nor B (or anyone else) has ever done sums involving numbers bigger than it. There is no harm in supposing like Kripke that 56 is such a number.

Now suppose that you ask both A and B to calculate 57 + 68 whereupon A says ’125′ and B says ’5′. Of course you will think that B got it wrong, but suppose that when you try to explain it to B he refuses to recognize any mistake. For instance, you might say to B: ‘Look at these examples: 2 + 4 = 6, 5 + 7 = 12 etc. you agree with those don’t you?’ – ‘Yes’. ‘So then you are supposed to do the same with 57 and 68 when you add them as you did with 2 and 4 when you added them‘. – ‘But I am doing the same.’ What seems to you a bizarre deviation from the rule that you taught him seems to him to be the most natural continuation of his training.

So now: must there have been some prior difference between A and B, either in terms of their training, their experiences or their ‘inner’ lives that justifies this difference between their responses? The point of Chapter 2 of Kripke’s book is that no, there need not have been any such difference. God himself could not have seen any such difference. But this means that there is nothing in A – who is in all essential respects just like you and me – to which his actual answer is being faithful. Prior to his actually saying ’125′, neither he nor his teachers had done anything that invested the sign ‘+’ with such a meaning as to make ’125′ and not ’5′ the right answer to ’57 + 68 = ?’ The inescapable and incredible conclusion is that neither A nor any of us ever really meant addition by ‘+’ in the first place. And the argument generalizes: so we have an a priori refutation of meaning. Every time you use a word no prior grasp of its meaning guides you: it is, every time, a leap in the dark.

Putting Kripke’s point in a slightly wider context: we all thought that your understanding of a word imposes two sorts of normative constraint on you (i) it makes some of the things that you might say or write true, and other such things false; some things justified and others not (ii) your conscious engagement with your understanding guides you to use the word in one way rather than another. What Kripke showed was that given certain reasonable assumptions nothing could impose these obligations on you. So his argument had some resemblance to, though a wider scope than, Mackie’s roughly contemporary argument about ethical obligations.

It is hard to overstate the depth and importance of Kripke’s argument, though both he and Wittgenstein make it easy to appreciate its almost sinister power to disorient. What it has got to do with Wittgenstein’s actual intentions is another matter. The example at s143 and ss185ff of Philosophical Investigations looks like a very straightforward case of what Kripke has in mind; on the other hand there is plenty of evidence that Wittgenstein himself was not skeptical about meaning. There are very good discussions of this in McDowell’s paper (‘Wittgenstein on following a rule‘) and in Martin Kusch’s more recent book (‘Sceptical guide to meaning and rules: ch. 8).

3:AM: Your book on Saul Kripke discusses his contribution to Wittgenstein’s arguments about rule following and private language. But Kripke’s ‘Naming and Necessity‘ is equally significant isn’t it? He put metaphysics back on the table didn’t he – contra Wittgenstein and empiricists? Can you say something about how he managed to reverse the disrepute of metaphysics prevalent at the time?

AA: It’s probably more significant historically, though in my view the arguments are not nearly as plausible or interesting. In the 50s and early 60s things were from one perspective proceeding in a very satisfactory way, in the US if not the UK: the distinctively metaphysical questions (‘What things are there? How do they fit together?’) were still around, but (if you believed Quine) the answers were to be extracted from a scientific theory, not just plucked out of the air (or contrived to serve one’s religious or ethical purposes).

What Kripke did was to convince a generation of philosophers – I am still not sure quite how – that there was interesting a priori work to be done on questions of what there is, and how things might have been, and what is the essence of an object or of a kind. Indeed he probably did more than anyone in the last century to revive that long discredited Aristotelian notion. Nowadays many philosophers think that your intuition that President Obama might have been a Republican but could not have been a light switch is the answer to an interesting question, for it shows that being a Democrat is not an essential property of Obama but not being a light switch is an essential property of Obama.

And what are these ‘intuitions’? You don’t get them by thinking or observing things or experimenting upon them. As far as I can tell they are what you say *before* you have done any thinking or observation or experiment: you are supposed to get them by somehow communing with yourself. And yet Kripke makes the amazing claim that he finds it difficult to imagine what could be better evidence for anything than an intuition.

And a whole flourishing field of metaphysics has grown up around these intuitions, so you get philosophers engaging in (to my mind) often sterile disputations based upon their ‘intuitions’ about possible worlds and abstract objects and whether one property ‘grounds’ another property. All of this is conducted at such an abstract level and in such a freewheeling way that it is hard to see that anyone could learn anything interesting from it. Such is the disastrous legacy of Kripke’s ‘revival’ of metaphysics.

As I said I’m not sure how he did it. On this point Naming and Necessity seems to contain much more assertion than argument, so maybe it owed something to an atmosphere that was generally receptive to this sort of change. But what made philosophers receptive to it just then? Fodor once wrote a very good piece in the London Review of Books suggesting that Kripke was showing how philosophers could carry on with a priori work of conceptual analysis – which was then very popular – only now they could think of it as being about the world itself and not just footling with meanings of words. When the philosopher tells you that Obama could not have been a light switch she is telling you something about the man Obama, not the word ‘Obama’. Perhaps some people found that exciting.

3:AM: Kripke was also influential in disputing the idea that meaning was mediated by something in the mind of the speaker, an idea that might be associated with John Locke. How did he do this?

AA: That’s right. On one view – internalism – what you mean by a word depends only on what is going on in your head; on the other view – externalism – two people might have the same thing going on in their head and yet mean different things by the same word.

The version of internalism that he attacked most closely was ‘descriptivism’: the idea that the meaning you associate with a proper name (‘Columbus’) was the description that you associate with it (‘The first European to visit America’): the name referred to whatever the associated description picked out, and it did so via that description, so that the associated description gives not only what the name denotes but also its sense or connotation. This plainly is a sort of internalism: for if anything is internal then the description that you associate with a name is internal.

Kripke gave three interesting and influential arguments against descriptivism of which I’ll mention just one. It often happens that you have no definite description associated with a name that you none the less understand; and it also often happens that you do have such a description, but it does *not* pick out what the name picks out. (Thus in the example ‘Columbus’ picks out a 15th-century Italian, whereas ‘The first European…’ probably picks out some 10th-century Scandinavian.) So the descriptivist thesis cannot be true – at least that is Kripke’s view.

So what does make ‘Columbus’ refer to Columbus when spoken by somebody who is completely ignorant about him? Kripke’s view is roughly that there has to be the right sort of causal connection between the person’s use of ‘Columbus’ and the use of that name (or some ancestor of it) to christen the object itself. There are details to be filled in here, but for our purposes the crucial point is that whether or not my word ‘Columbus’ refers to Columbus depends only on that causal chain and not on my knowledge of it. I might be completely ignorant of any such connection and yet as long as the causal connection exists it is true that ‘Columbus’ as used by me is referring to that man. You might have two people with exactly the same ‘internal’ goings on such that one of them and not the other is referring to Columbus. (As you might have two paintings that happen to be exactly similar, but only one of them is ‘of’ Mona Lisa because she only posed for that one.)

How much does this show? I’m not convinced that internalism need retreat very far to accommodate cases like this: it may be for instance that Kripke’s examples underestimate the ‘internal’ data that we really do associate with individual names, and that when we take these into account descriptivism seems much more plausible. Moreover, we can distinguish two things: (a) the actual mental associations that explain your particular use of a word; (b) the causal background to that use. And we can distinguish between an ‘internal’ and ‘external’ notions of reference that are determined by (a) and (b) respectively. The point at issue is then: which of them is the ‘real’ reference? – and I’m not sure how to answer that, or what exactly would turn on the answer.

I should add that this only covers a very small part of the debate between externalists and internalists. There are many more things that might motivate externalism, not only about meaning but also about mental content itself (what is in your mind does not only depend on what is going on in your head). But Naming and Necessity came out at just the time that externalism of various sorts was starting to become popular, and that fact probably has something to do with its influence.

3:AM: Some things look like they’re necessary, things like maths truths eg 2+2=4 for example. But necessity itself doesn’t look on the face of it necessary. You’ve thought about this and what various philosophers have said. Before you tell us what you think, can you say something about why non-philosophers ought to understand the significance of this debate?

AA: It looks as though there are two completely different dimensions to our thought: one is thought about how things are (about how they actually are); and another is thought about how things might have been. You might wonder about whether Napoleon might have not invaded Russia in 1812 even if you know for a fact that he actually did. And it seems that the first dimension of thought might have existed without the second: that we should wonder and care about how things actually are, were or will be, without in addition wondering, or caring, or even having the resources to inhabit or speak about that second, modal dimension. (It is not so obviously possible to have the second dimension without the first. Maybe God thinks like that but it isn’t clear that humans could.)

But in fact we do have this second dimension, and it matters to us a lot. At an everyday level we wonder about how things would have gone had we acted otherwise, and might feel regret or relief based upon the answer. At a more abstract level it is natural to think that your responsibility for your actions depends on whether (and in what I think is the same sense) you might have done otherwise in the circumstances. At a more abstract level still you might wonder whether determinism entails that nothing could have been otherwise. And at the most abstract level you have philosophers saying that 2 + 2 = 4 is something that could not have been otherwise, and wondering whether Obama might possibly have been a light switch, and so on.

So an obvious question is: what explains our practice of having and responding to this second dimension of thought? Why does it matter to us? What role does it serve? Are our judgments based on insights into the nature of things, on limitations in our imagination, or on something else? And are they ever true or are they always false? These are in my view empirical questions to which philosophy can none the less contribute: they belong in or near a field known as ‘modal epistemology’.

Here is why it matters to me: I am strongly attracted to the view that individuals are the authors of their own knowledge through the operation of their own senses, i.e. empiricism. But empiricism has trouble with modal epistemology, because it is just not clear what sensory processes, or sensorily checkable theory, might help with this second dimension. The senses can tell you what is going on in the actual world: how can they tell you about other possible worlds? (Hume had a similar problem when he argued that you can’t see the necessity with which a cause is supposed to produce its effect; it survives even if you drop the reductionism for which Hume and his positivist successors have been criticised.)

So it looks as though the empiricist has to say that this whole second dimension of thought is an enormous and deeply rooted superstition. It is just a mistake to think that as well as facts about whether Napoleon invaded Russia there are also facts about whether or not he might have done otherwise. It is possible to go around the houses on this point, and there are many people that disagree with this, but my own view is that empiricism must in the end reject modality.

But then in much the same way as the atheist must explain religious belief, the empiricist must explain where this superstition comes from. It must have made some difference to our non-modal thought, or in some other way to our actions or welfare, if we are to explain its ubiquity and persistence without postulating any ‘must-detecting’ faculty. And so we are led to what John Divers calls the functional question about modality: what do these beliefs do for us?

3:AM: So what are the various arguments for the necessity of necessity and which is the one you think works?

AA: I’ll mention one sort of argument that has been going around for a while: there is a version in Crispin Wright’s 1980 book on Wittgenstein, and again in his 1986 paper; also in a posthumous paper of Ian McFetridge’s and finally Bob Hale’s synthesis of these versions in a 1999 paper in Mind. The drift of the argument is that if you are going to engage in any sort of reasoning at all about how things would have been in alternative possible scenarios, there must be some assumptions that you take to be true on all such possibilities – otherwise your reasoning would never get started – and these are the *necessary* truths.

I should say that this argument doesn’t establish the necessity of the modal dimension as such (nor was it meant to): at best it shows that if we are going to think about how things would have been if…, then we are going to have to take some truths as holding however things would have been. I should add that Hale is not the sort of empiricist that I am (if he is one at all), and he doesn’t share my skepticism about modality, so he is under no obligation to justify as much of modal thought as seems genuinely questionable to me.

But I don’t think that the argument works even on Hale’s terms. Very briefly, it involves a fallacy known as ‘quantifier shift’. It’s true that given that you are reasoning about alternative possibilities you must take some truth for granted. It doesn’t follow, nor does the Wright/McFetridge/Hale argument show, that these must be the same truths in all cases. So even if you think that there are such things as genuine alternative possibilities and that we can reason about these, you aren’t committed to thinking that there are real necessities too.

However I do have a little more optimism about another argument that tries to answer the functional question. What difference does it make that one believes that if one had acted otherwise then such and such would have been the case? After the event – well, it might make you regret it. But what difference does it make that you regret something? In itself, nothing. What matters is that you anticipate regretting it. For instance: suppose you offer me a choice between $10 now and $12 tomorrow. If I take the $10 now then I have to pay an additional cost tomorrow i.e. the additional regret that comes from thinking that if I had taken the $12 then I’d be better off now. This adds a (prospective) cost to the more short-sighted decision. So the effect of modal belief is to make us more prudent or (equivalently) to lower the rate at which we discount the future. It’s easy to see how this might matter and also what good it might do. Anyway, all of this is just the beginning of what is (at the moment) just a theory. I drone on about it at much greater length in my forthcoming book on decision theory.

3:AM: Is your attraction to Kripke driven by similar feelings as those that drove to Wittgenstein, that his work is profoundly disturbing?

AA: Pretty much. I don’t think it was the disturbance that first attracted me to Wittgenstein (see my answer to q1) but it is one of the things keeps me interested in him, and the same is true of Kripke. I never forgot the sense of vertigo that I felt on reading the discussions of rule-following in Philosophical Investigations and Wittgenstein on Rules and Private Language. It’s also true that they both have distinctive and attractive writing styles. One thing that I greatly admire about Kripke is that in spite of his technical brilliance he writes in plain English, i.e. without formalism, wherever possible.

3:AM: You recently argued against the existence of God in a Cambridge Union Society debate. What made you intervene in the rather bad-tempered contemporary debates about this issue? I might have thought that with your sympathies towards Wittgenstein and Kripke you might have been more quietistic?

AA: Of course if people want to believe in fairies that is their business, but you have to object when they start persecuting homosexuals or insisting that schoolchildren be taught rubbish. Monotheistic religion is in my view a greater threat to humanity than climate change or nuclear proliferation, and I’d hope that every citizen do his or her little bit to ‘ecraser l’infame’.

Of course there is an argument that it isn’t rational argument that stops people going in for this stuff, and I used to think that too. But experience has taught me otherwise: I know many people that are both Christian and otherwise intelligent (so susceptible to argument); I also know many recovering Christians whom rational argument has helped to turn away from its doctrines and practices.

For this reason I have taken part in a number of debates against believers of all faiths and I hope to do so again. In so far as W and K were or are sympathetic to religion this is not something about them that I admire.

3:AM: What are the reasons against God’s existence then for you? Is there anything new in this debate that couldn’t have been raised in medieval times?

AA: Pretty much same as those against Father Christmas’s existence. If God had existed you would expect to observe many things that we do not in fact observe. I include the problem of evil in this category of reasons.

Writers in the mediaeval period were generally more credulous on this matter, but of course they had the excuse of living before Hume, who took the decisive step in the debate in his Dialogues Concerning Natural Religion. There he demolished the main arguments for God’s existence. They have really never recovered from the serious beating that they got then. I know that there are professors working in Theology and Philosophy faculties who would disagree with that statement and that many of them have come out with new arguments for God’s existence. But I am yet to see one that is any good.

3:AM: And finally, when you’re not being disturbed by Wittgenstein and Kripke, are there books, films, music that you’re really into outside of your philosophical realm?

AA: Favourite books:
1. Henry James, Bostonians.
2. Nabokov, Pale Fire.
3. Kingsley Amis, Lucky Jim.

Favourite music:
1. Mozart, Piano quartet in g minor K478
2. Haydn, Piano sonata in c minor HXVI/20
3. Beethoven, Piano sonata in E major op. 109

Favourite films:
1. Any episode of Columbo except: the one where his niece gets kidnapped and the one where he goes undercover

3:AM: And finally finally, are there five books (other than your own which we’ll be dashing away to read straight after this) you could recommend to the disturbed readers here at 3ammagazine?

AA: D. Hume, Enquiry Concerning Human Understanding

An excellent introduction to empiricist thought. De Finetti thought his treatment of causation to be the highpoint of all philosophy.
Hume, Treatise of human nature Book I

More difficult than the Enquiry but even more disturbing. Part IV is spectacular.

Nietzsche, Human all too human

Pithy and often spot-on discussions of metaphysics and religion.

Wittgenstein, Blue Book

More accessible than PI and the Tractatus but still full of startling insights about mind and language.

W. V. Quine, Word and object

I don’t recommend this to beginners, and the notorious Ch. 2 is by no means the clearest treatment of its subject. But reading it still gives you a sense of the power and resourcefulness of the empiricist tradition, at least in the hands of somebody like Q.

photo-on-2011-08-09-at-15247
ABOUT THE INTERVIEWER
Richard Marshall is still biding his time.

First published in 3:AM Magazine: Monday, October 15th, 2012.