:: Article

Without mirrors

3:AM: The idea of practical deliberation being ‘open’ or ‘settled’ is a key to how you go about answering the issue of the asymmetries of causality. Can you explain this and say whether as with your views about times arrow and truth your metaphysics link functional explanation to why we think about the world as we do (even though science tells us that there are no asymmetries in their theories). Is that right?

HP: Suppose you are deliberating about whether to do something – whether to stop reading your email and go out for lunch, say. There are some things you know, like what time it is, whether you are hungry, and so on, and other things that you may not know but think of as ‘available for knowing’, before you make your choice. All that counts as settled, in the terminology you mention to here. But there are some things you don’t know before you make up your mind, starting, obviously, with whether you’re going to go out to lunch or not. (If you thought you knew that, you couldn’t think of yourself as still deciding whether it would happen.) Things in the latter category are open, in this terminology.

Just as with our immediate choices (like whether to go for lunch), anything we take ourselves to be able to influence by our choices (like whether we are hungry in mid-afternoon) must also be in this open category, and for the same reason. To think you can control it is to think that your present choice ‘trumps’ any existing evidence you might have, as Jenann Ismael puts it. Ramsey was the first to appreciate this point, too, I think. He puts it like this: ‘Any possible present volition of ours is (for us) irrelevant to any past event. To another (or to ourselves in the future) it can serve as a sign of the past, but to us now what we do affects only the probability of the future.’

As I noted earlier, Ramsey proposes that this is the basis for the distinction between cause and effect. He doesn’t go into the question why this distinction seems to line up past-to-future, but as I said above, I think that that is plausibly explained by a contingent fact about us, namely that we also deliberate about later actions. As I noted, creatures elsewhere might do it differently, and in that case they would divide up the world differently into what was settled and what open, from their point of view.

Another interesting possibility, which Ramsey misses, is that not everything in the past need count as settled, from our point of view. It isn’t incoherent to think of some of it as open and under our control (as an indirect consequence of some future action), so long as we don’t think we could find out about it before we decided whether to perform the action in question. This possibility was first noticed by Michael Dummett in the 1960s. It amounts to admitting a bit of retrocausality, even in a world in which the prevailing forward direction of causality reflects our contingent temporal perspective, in the way I have described. I think it is going to be big news in quantum mechanics, incidentally.

3:AM: Last year Naturalism Without Mirrors came out. And in it we get a further look at how you do (or don’t do, depending on how we look at this) metaphysics. You don’t so much ask ‘What are we looking at?’ but rather ‘What are we talking about?’ What is the difference between the two questions in terms of what answers we might give and what we might suppose metaphysics is really about? You have a brilliant riff on this concerning the transparency of language where you say its not just that ‘… as an Australian apostle might have put it, that language shows us the world “as through sunnies, darkly”… ‘ but that ‘the glassy metaphor itself is entirely empty.’ So are you a kind of pragmatist? The metaphysics suggests that perhaps Peirce might be your kind of pragmatist, but the title of your book and this language orietated approach to the subject is kind of solid with Rorty, Wittgenstein and Brandom, and that’s where many people seem to put you. Are you doing metaphysics or are you deflating metaphysics so far that it doesn’t count as metaphysics anymore?

HP: The alternative question isn’t ‘What are we talking about?’ – that leaves us just where we were, trying to think about the things. I want to recommend that we think instead about the language, by asking ‘Why are we talking this way – what role do these concepts play in the lives of creatures like us?’ That’s why I say that I’m recommending that we do anthropology, in place of metaphysics.

Yes, certainly I’m a pragmatist. In my view, the most helpful way to characterise pragmatism is to say that it approaches a range philosophical issues in the way I just mentioned, by asking about the practical role that philosophically interesting concepts (e.g., that of causation) play in our lives – by looking for explanations, and genealogies, in broadly naturalistic terms (i.e., by starting with the assumption that we are natural creatures in a natural environment).

I don’t know much about Peirce, so I’m not sure whether I’m his sort of pragmatist, but I’m happy to be seen as kind of solid with Rorty, Wittgenstein and Brandom, as you put it. (We’re, you know, like that.)

3:AM: The ‘without mirrors’ is your way of signaling a major theme in the book, which is your rejection of the view that language mirrors nature, that we somehow represent nature and that our thoughts cut nature at the joints. We’ve seen this in your approach to time and causation. Can you say something about why you think a ‘semantic ladder’ from linguistic concerns down to ontology is not feasible?

HP: The context for my use of this metaphor is the issue whether if you start off asking questions about language, you can nevertheless end up asking the usual kind of metaphysical questions about (non-linguistic) things. On the face of it, the answer is ‘yes’. If you start with something you take to be a referring term, say “X”, and then ask the question ‘What does “X” refer to?’, it looks as though a question about language – about the term “X” – is directing your attention to the world.

But if you are a deflationist about semantic notions like truth and reference then, as Quine pointed out, this is a kind of illusion. In this case the question ‘What does “X” refer to?’ is just another way of saying ‘What is X?’ – which isn’t a question about language at all. So deflationary semantic notions can’t provide a ladder to take you from genuinely linguistic issues to metaphysical issues. That was my point.

Simon Blackburn makes a similar point in terms of what he calls ‘Ramsey’s ladder’, that takes us from P to ‘P is true’ (and back again, of course). He points out that Ramsey’s redundancy theory of truth – an early version of what we now call deflationism – implies that this ladder is ‘horizontal’, in the sense that it doesn’t take us anywhere new. Again, as Quine says, we have the same subject matter in view at both ends. (Blackburn himself complains about philosophers who take advantage of the fact the ladder is horizontal in order to climb it, and then announce a better view from the top.)

3:AM: You make a distinction between two species of expressivism – local and global types. Simon Blackburn’s quasi-realism is the acme of the local expressivist position but you think even this fails and that its global expressivism that we need. Can you say what this distinction seperates and why only the global species is useful?

HP: Local expressivists rely on the idea I mentioned in my answer above, that there is a bifurcation in language between descriptive and non-descriptive uses of declarative utterances. They take expressivism to be appropriate on the non-descriptive side of the line, and some sort of representationalist story on the other side of the line. (Blackburn’s quasi-realism is distinctive in paying a lot of attention to the question why some declaratives that are not descriptive – ‘Cruelty is wrong’, say – neverthless behave as if they were.)

That combination of expressivism on one side of the line, representationalism on the other, is off-limits for me, because I reject the idea that there is such a line – I reject the bifurcation thesis, as I put it above. In principle, this would be compatible with being either a global expressivist or global representationalist, depending on who wins when the wall comes down. But in my book, both figuratively and literally, it’s the expressivists who win. The sorts of questions that local expressivists ask about what they take to be non-descriptive claims – questions about use and function, cast in something other than the semantic terms that their representationalist opponents employ – turn out to be good questions to ask everywhere. Hence global expressivism, as I call it.

3:AM: Your view is both deflationary and expressivist, and for philosophers like Crispin Wright, John McDowell this is simply not done. How come you think you can make it work? Why are they mistaken?

HP: One way in which local expressivists cashed out what it was supposed to mean to be non-descriptive was in terms of the idea that non-descriptive utterances ‘don’t have truth conditions’, or are not ‘truth-apt’. But writers such as McDowell and Wright pointed out that that if we are deflationists about truth, having truth conditions and being truth-apt is cheap – anything that meets the basic syntactical criteria gets to qualify, more or less. This was thought to be a big problem for expressivists (who did tend to want to be deflationists themselves, on the whole).

The solution is to note that expressivists typically made two proposals about their target vocabulary – moral claims, say – one negative and one positive. The negative thesis was that moral claims lack truth conditions, or are not truth-apt. The positive thesis was whatever non-representationalist story expressivists proposed instead about what the target vocabulary is ‘for’ – e.g., that its function is to express affective attitudes. Deflationism undermines the negative thesis, but it doesn’t touch the positive thesis. On the contrary, it implies that we need such non-representationalist stories everywhere, because the representationalist’s semantic vocabulary – truth, references, and the like – has been ruled too ‘thin’ to do any theoretical work in telling us about the function of various bits of language.

So in a sense the old objection got things exactly backwards, in my view. Deflationism is indeed a problem for local expressivists such as Blackburn, but the problem is not with their expressivism, but with their residual representationalism about what they regard as the genuinely descriptive vocabularies. Deflationism supports global expressivism, not global representationalism.

3:AM: If an anthropologist was to discuss the issues you do, conceptions of time, causation and so on, would they be looking at anything different from you, only with more field work? So firstly, is it philosophy you’re doing here or is it converging with anthropology? And secondly, if what you think we should be doing is thinking about what we’re saying then is there a role for xphi who look and see if what philosophers say the folk are saying is actually what they are saying?

HP: It’s converging with anthropology, certainly, or perhaps some mix of biology, psychology and anthropology. Does that mean that it’s not philosophy? Not in my book, but if someone wants to be precious about the use of the term ‘philosophy’, that’s not going to bother me. I certainly think that experimental work is relevant. For example, I think that the recent fascinating work on development of causal concepts, both in human infants and in primates, is helpful in bringing into focus questions about the functions of these concepts – what they are for, as it were. (We learn about this by learning about what difference it makes if creatures don’t have them.)

I’m less clear about whether my approach has something to learn from some distinctively philosophical kind of empirical work. Since I’m not wedded to the idea that there’s any sort of interesting metaphysical fact of the matter about what the folk are locking on to, in using a particular term, I’m a little bit skeptical about whether there’s something for xphi to investigate, that’s not already investigated somewhere else. (It would be compatible with this that some of the work called xphi is the same kind of thing done elsewhere – my kind of anthropology, for example – but with a sharper eye for the philosophically interesting questions.)

3:AM: Is there no room for, say, a Frank Jackson analytic metaphysics in your view? And what would you say to people who say that you’ve changed the subject: they might have been bad at it, but they really wanted to ask about fundamental reality and why time moves in one direction and all that, rather than be told that all we can say is that’s how things seem to us because, well, that’s the way our minds are. In a sense , people do want to ask what God sees. Is this just by-gone metaphysical talk that won’t die out, like the monarch? Oh – and is it harmful?

HP: Well, I do want to change the subject, but not in most cases by just walking away from the old one. I try to say what I think is wrong with it first. For example, I’ve argued that at least some versions of the Canberra plan are problematic because they rely on semantic or representational presuppositions that turn out not to be amenable to analysis by the program’s own lights. Again, I have offered McTaggart-style objections to A-theoretic conceptions of time, and I could give more examples of this kind.

More generally, I try to identify common ground with my opponents – common explananda, for example – so as to be able to argue that a pragmatist approach provides better explanations, in the cases in question. (Accounting for our sense of the direction of causation is a case in point here.) So I haven’t just changed the subject. I’ve tried to explain why it seems to me that it should be changed.

Most people won’t change their minds, or their philosophical spots, of course, and you get to the point where there’s nothing to say. Does that matter? Probably not, in my view. We need some stubbornness and intellectual inertia to sustain our research programs, and it’s only philosophy, after all. The scope for any serious harm is rather limited!

3:AM: Finally, are there five books (other than your own which we’ll all be dashing out to read straight after this) you could recommend to the pragmatist deflationary metaphysicians here at 3:AM to take us further into your world?

HP: You could start with Rorty’s classic, Philosophy and the Mirror of Nature . After that, well, definitely some Brandom – perhaps start with his recent collection Perspectives on Pragmatism, and work back from there. And some Blackburn – here I recommend starting in the middle, with Essays in Quasirealism , a collection of his major early papers. And some Horwich, for classic deflationism – his collection From a Deflationary Point of View, for example. Finally, Amie Thomasson’s Ordinary Objects is an excellent recent example of someone taking the deflationary approach seriously in metaphysics.

If you pressed me for a sixth I’d have to break the rules and mention Expressivism, Pragmatism and Representationalism, which is my Descartes Lectures from 2008, with commentary essays by Blackburn, Brandom, Horwich and Williams, and a long new postscript and replies by me.

Richard Marshall is still biding his time.

Pages: 1 2

First published in 3:AM Magazine: Friday, September 14th, 2012.