I think it was P.G. Wodehouse who observed that the English strike Americans as funny when they are just being English. Similarly, philosophers strike the laity as funny when they are just being philosophers, and that makes it hard to be as funny about them as they are when they’re left to their own devices. But Michael Frayn is among the honoured few who have succeeded. I fondly remember a piece of his from the 1960s (about fog) that purported to be a newly discovered fragment of Wittgenstein’s Philosophical Investigations. Wittgenstein generally writes with a transcendental pomposity that makes parody seem superfluous, not to say impossible. But Frayn pulled it off. For years Frayn’s Wittgenstein was to be found pinned to the bulletin boards of anglophone philosophy departments all round the world.
So the news that Frayn had done a whole book on philosophy was a cause of anticipatory glee. What’s more fun, after all, than seeing one’s colleagues skewered? But the skies darkened when a copy actually turned up in the mail. For one thing, it’s clear at a glance that this is no joke; it’s a book of philosophy, not a book on philosophy, and I can’t imagine an author who is more in earnest. It’s also clear that the thing is much too long. These days nobody writes philosophy in chunks of four hundred pages (plus notes). Partly that’s just fashion; partly it’s tenure politics; but mostly it’s because the problems philosophers work on have turned out to be much more subtle than we used to suppose them, and much more idiosyncratic. You have to do them one at a time, and the progress you make is generally inch by inch. For better or worse (I think, in fact, it’s much for the better), almost nobody has ‘a philosophy’ any more. What one has, if one is lucky, is a glimpse of an insight into (as it might be) the semantics of intentional contexts; or the behaviour of modals in obligation ascriptions; or the way natural laws support counterfactuals; or whether knowledge is warranted true belief – and what, while we’re at it, does ‘warranted’ mean? After what seems in retrospect to have been a very extended adolescence, philosophy has settled into workaday middle age. It comes to all of us sooner or later.
Frayn, however, doesn’t approve of all that picking of nits. ‘Reading a philosophical journal, I’m sometimes reminded of Dawkins’s Law of the Conservation of Difficulty, which states that obscurantism in an academic subject expands to fill the vacuum of its intrinsic simplicity.’ So be it. (I’m not sure that Dawkins is the ideal guide here; whatever his virtues, a feeling for the hardness of hard questions pretty clearly isn’t among them. Wasn’t it Dawkins who announced that ‘selfish gene’ theory had discovered the meaning of life?) In any case, the range of issues Frayn takes on is staggering. Here are some of the subtitles of some of his chapters: ‘the nature of laws’; ‘an almanac of causal lineage’; ‘the structure of space and time’; ‘the world as numbers’; ‘intention and purpose’; ‘the act of deciding’; ‘thinking of thoughts, speaking of things’; ‘the truth functions of fiction, the fiction-functions of truth’; ‘sense and syntax’; ‘how words have meanings’; ‘the gift of analogy’. And these topics are not treated narrowly: one gets a whole spectrum, from Frayn on quantum mechanics to Frayn on the psychology of perception, to Frayn on the ontology of numbers, to Frayn on Chomskian linguistics, to Frayn on personal identity, to Frayn on the phenomenology of dreaming, with many, many intermediate stops. Could anybody conceivably have views worth hearing on all those topics? Could it be that the gods no longer punish hubris?
I don’t mean to suggest that The Human Touch lacks a provenance or a unifying theme. Quite the contrary: it belongs to a philosophical tradition that reaches back at least as far as Kant and which persists in the neo-pragmatism that is as close as anything gets to being the current philosophical consensus. Frayn’s version goes something like this: back in the good old days, it seemed perfectly clear that everything turned around us, both astronomically and otherwise. The moon, the sun, the planets and the stars were put in place to be the backdrop to a moral drama of which God was the audience and we were the heroes and villains. What we did or failed to do mattered to the whole scheme of things; it was what the whole scheme of things was for.
But then came Copernicus and the scientific world view, and it turned out that the universe is quite a lot bigger than we had supposed, and that we aren’t at the centre of it after all. Maybe there isn’t any centre; maybe it goes on and on in every direction for ever and ever. And God only knows what’s become of God. The moral seems to be that we had badly overestimated our role, which turns out to be pretty peripheral. Here we are, lost in the stars; and what does it matter to them what happens to us? Even if we did something worth reporting, we’d be long gone before the stars got the news.
There are those of a Woody Allen temperament who find this revised worldview not bearable. Angst and dread overcome them; they languish in existential loneliness. Or at least they tell me that they do. I guess I sympathise, though it’s not a mood I find it easy to get into. If the spiral nebula Andromeda isn’t very interested in me, so be it. I reciprocate: I have at most a theoretical interest in it. And I don’t much care if it doesn’t much care whether, stuck in traffic, I will miss the first act of the opera I was planning to attend. I care. I care a lot. Why isn’t that enough? What business is it of the spiral nebula Andromeda? Does the spiral nebula Andromeda have any idea what opera tickets cost these days?
There is, however, a philosophical tradition that seeks to reinvert the Copernican inversion, and to this tradition Frayn’s book belongs. True, the scientific story seems, at first blush, rather an affront to our sensibilities. But the second blush is less dire. After all, the scientific story is just a story; indeed, it’s a story that we made up for, as pragmatists like Frayn like to say, the practical purposes that we have in hand. The universe
is big, it’s small … because you and I and some of our friends say it is. If we weren’t here in the audience … the whole show would have gone for nothing … It would not be odd or awe-inspiring – or even banal. It would have no characteristics at all. And if it had no characteristics, then in what sense would it be anything? In what sense would it exist? … So we are perhaps not after all such nobodies. We are not for nothing. The middle of things is not an entirely inappropriate place for us to be.
Well, that’s a relief; I feel ever so much better now.
The basic idea is to undermine the authority of science (and, indeed, the authority of common sense) by launching a general attack on the notions of truth and knowledge. What a Copernican astronomy taketh away, a relativist epistemology giveth back. As the blurb on the front flap of Frayn’s book puts it:
What would [the] universe be like if we were not here to say something about it? Would there still be numbers, if there were no one to count them? Or scientific laws, if there were no words or numbers in which to express them? Would the universe even be vast, without the very fact of our tininess and insignificance to give it scale?
Why, yes, it would. (Frayn often treats rhetorical questions as though they were philosophical arguments; it’s a bad tactic.) The universe would still be just the size it is even if there weren’t astronomers to measure it. And water would still be H2O even if there weren’t chemists to analyse it. And water would still run downhill, and there would still be hills for it to run down, even if none of us were here to take note of its doing so. You can’t pin the natural order on me, Frayn; I’m not guilty. I didn’t make the universe; I wasn’t even there at the time.
How on earth can anyone seriously suppose otherwise? That’s a long story, and it comes in a lot of versions. There is, however, a gaggle of fallacies that generally get committed when a philosopher tells it, and Frayn’s book is no exception. I’ll mention two or three mistakes that I take to be central.
First of all, Frayn is something of an old-fashioned positivist; sometimes he is so quite explicitly. ‘I must surely be a little positivistic here once again. I must limit myself to what is observable – if only by myself.’ Why must he, I wonder? Positivists thought they could impose a priori epistemic constraints that scientific theories must acknowledge. The idea was that what can’t be observed can’t be, tout court. The limits of our experience set bounds on the ways that the world could be (or, in some versions, on the ways that the world could intelligibly be said to be; epistemology and semantics often take in one another’s wash in the positivistic sort of philosophy). But positivism wasn’t a raving success, and it’s now widely thought to have been completely wrong-headed. It’s one question what’s the case, and another how we know what’s the case. Very often, we’re able to answer the first sort of question even though we can’t answer the second. In particular, we’re rarely in a position to say just what it is about our experience (or about anything else) that warrants our claim to know that a proposition is true. That being so, it’s not an argument against the proposition being true that we don’t know how we know that it is. The sun will rise tomorrow morning; I know that perfectly well. But figuring out how I could know it is, as Hume pointed out, a bit of a puzzle.
Frayn’s discussion of perception offers a clear case of his general tendency to confuse epistemology with metaphysics. I have ten toes, I can see that from here (I like to type with my shoes and socks off). Now, there’s a lot we don’t know about seeing. For example, seeing is sometimes unveridical; one sometimes takes oneself to see what, in fact, one doesn’t see; and, since nobody has a general account of when this happens, it’s not clear how I know that this isn’t one such occasion. Also, as Frayn rightly points out, seeing is shot through with interpretation. There’s quite a long inferential route between the patterns of light that fall on the retina and the perceptual beliefs that their falling there eventually engender; between, that’s to say, my having ten toes and my knowing that I do. That’s all very interesting (it really is very interesting). But it’s not – it doesn’t begin to be – a reason for doubting that the perceptual beliefs we have are very generally true.
To be sure, interpretation can be a source of error or bias. But Frayn keeps forgetting that it also can perfectly well be conducive to truth. Sometimes our interpretations get things right; that’s why interpreting is worth the bother. So, for example, the most familiar instances of perceptual inference (the perceptual ‘constancies’) regularly bias perception in the direction of truth; the interpretations are better guides to what is actually going on than the data that they interpret. Double the distance to the target and you halve the size of its retinal image. But there are inferential mechanisms in perception that correct this geometric effect so that things generally don’t seem to get smaller as they move away (they just look further off). That’s just as well since, as a matter of fact, the size of things doesn’t vary with their distance; it’s only the retinal images that shrink. In such cases (and they’re the rule rather than the exception) it’s precisely because perception is saturated with inference that it is epistemologically reliable.
There are lots of cases where we know more about how the world works than we do about how we know how it works. That’s no paradox. Understanding the structure of galaxies is one thing, understanding how we understand the structure of galaxies is quite another. There isn’t the slightest reason why the first should wait on the second and, in point of historical fact, it didn’t. This bears a lot of emphasis; it turns up in philosophy practically everywhere you look.
For example, nobody is really a solipsist about other minds; everybody knows that everybody else has beliefs and desires out of which they act, just as one does oneself. But it’s a serious question how one knows that other people have minds; and it’s not a question that psychologists are able to answer to anybody’s satisfaction. Is it an inference from, as one says, one’s own case? Or is it a kind of belief that one is simply born with? Or did one learn it at granny’s knee? (If so, where did granny learn it?) The fact is, we don’t know. But that we don’t doesn’t matter if the question is, say, whether people drink because they are thirsty. In particular, your being thirsty doesn’t depend on whether I have a story that says that you are. Of course it doesn’t: your being thirsty is about you, not about me. It wasn’t me who ate the salted peanuts; you did. People say: ‘Is there anything to drink? I’m thirsty.’ Why would they say that they were if they weren’t?
So one thing that’s wrong with Frayn’s arguments is his partiality for inferences from epistemic premises to metaphysical conclusions. Frayn isn’t alone in this of course; quite a lot of 20th-century anglophone philosophy made it a matter of principle to make this mistake. It turns out, at the limit of this sort of philosophical nuttiness, that it takes two to see a tree; you can’t see one unless there’s somebody around (actually or counterfactually) to interpret you as seeing one. But surely this is back to front? Surely it’s the seeing that warrants the interpretation, not the other way around? Stories are made to conform to the world, not vice versa.
Another thing Frayn gets wrong (and here too he doesn’t lack for company) is his persistence in what I’ll call ‘all or nothing’ arguments. So, for example:
What, for example, could have more gratifyingly distinct spatial frontiers than a car? … But now follow it through time, from its beginnings in vague discussions between designers and sales directors [to when it] undergoes compression, meltdown, absorption into the fabric of other cars, into tin cans and bicycles. When did it start being a car … ? Somewhere this side of the preliminary discussions, certainly. When did it cease? Somewhere before its transmutation into cans of baked beans.
The implication is that, since there’s no fact of the matter about when a thing starts to be a car (or ceases to be one), there is likewise no fact of the matter about whether a thing is a car; it may be a car according to your story but not according to mine and, in principle, there’s nothing to choose between the stories. So, it’s all or nothing: if there’s no matter of fact at the margins, there’s none in the middle either.
I look out of the window … I tell you that the sun is setting … But, even here, in this simple factual report of what is before my eyes … there is also a performative element … I am deciding that the sun is setting … even though we have no agreement on what precise relationship between sun and horizon constitutes the sun’s setting … All narration and description … is indissolubly subjective because it involves selection.
I’m not saying the bridge is open because it is; it’s open because I say it is.
And finally, with a flourish: ‘The story is the paradigm. Factual statements are specialised derivatives of fictitious ones.’
Piffle. Much of what we know is organised around clear cases, so what’s indeterminate about the marginal Xs can be a plain matter of fact about the paradigms. How many legs can Bossie have before she becomes not a funny kind of cow but a funny kind of centipede? How small can Bossie be before she’s too small to be a cow? (As small as a bread box? As small as an atom?) How big can she be before she’s too big to be a cow? (As big as Australia? As big as the universe?) And what if she walks upright and serves tea to friends? And what if she speaks Latin? What if she turns on and off on Tuesdays? Search me. Or search a biologist; if he doesn’t know, nobody does. But that Bossie as she stands, in full sunlight with four legs and flies, is a cow: that isn’t up to me, or to you, or to anybody else. Bossie is a cow without caveats, a cow sans phrase, a cow tout court. Nor is her being such merely the asymptote that the indeterminate cases converge towards; that gets things backwards: cows grade off from Bossie, not the other way around. That Bossie is a cow is story-independent.
Frayn is the kind of philosopher who can’t quite believe that what he believes is mostly true; that, by and large, things are much as we all suppose them to be, and that we suppose them to be that way mostly because that’s the way they are. And yet, on the face of it, that’s surely the view that has much the most to recommend it. As a matter of fact, there’s no competition; it’s the only story that anybody has a glimmer of how to tell. It’s one thing to remark that there could be other stories; it’s something quite else actually to tell one that is remotely plausible. No doubt, there’s plenty to worry about at the fringes of what we believe; quantum entanglement really is hard to swallow, and I, for one, can’t get my head around black holes. But Bossie? And the car in the garage? What’s the likelihood that we’ve got it all wrong about them? How could we have? What on earth would conceivably explain Bossie being in my story if not Bossie being in the world?
I will tell you a philosophical joke. Once upon a time, a visiting scholar presented a lecture on the topic: ‘How many philosophical positions are there in principle?’ ‘In principle,’ he began, ‘there are exactly 12 philosophical positions.’ A voice called from the audience: ‘Thirteen.’ ‘There are,’ the lecturer repeated, ‘exactly 12 possible philosophical positions; not one less and not one more.’ ‘Thirteen,’ the voice from the audience called again. ‘Very well, then,’ said the lecturer, now perceptibly irked, ‘I shall proceed to enumerate the 12 possible philosophical positions. The first is sometimes called “naive realism”. It is the view according to which things are, by and large, very much the way that they seem to be.’ ‘Oh,’ said the voice from the audience. ‘Fourteen!’