If there is a pun here, it is an illuminating one
We had the second meeting of our Language Classics group yesterday after a brief hiatus. We read John McDowell’s ‘On the Sense and Reference of a Proper Name’. We found that there was a lot to say about this one, and being McDowell it wasn’t always easy to follow on a first reading. I think it’s one of my favourite papers in philosophy of language though. We talked about a couple of things in particular.
Some of what McDowell says about the link between his theory of meaning and theory of mind caused us trouble. He wants to deny ‘psychologism’, which he takes here to be the temptation to think of a theory of meaning as the mechanism in the mind that explains linguistic behaviour. It’s clear that he wants to avoid this picture, and he takes himself to be following the Wittgenstein of the Philosophical Investigations when he does so. What I in particular found odd is what seems to be a consequence drawn from this about those beliefs that involve bearerless names. McDowell writes about this situation:
“No belief is expressed by [the subject's] words: they purport to express a belief which could be described in the transparent style, but since no appropriate belief could be thus described, there is no such belief as the belief which they purport to express.”
Clearly there is a link between having a belief, having a belief ascribed to one by somebody else, and there being an object to feature in their ascription of your belief. I have to say I find this suggestive, but puzzling.
The source of this view is the claim that language is to be treated as behaviour to be made sense of by the attribution of propositional attitudes and of clauses describing a theory of meaning. This is where the pun comes from, a theory of sense is part of a theory of making sense of human beings. There’s something right about this but it’s hard to be clear about the best formulation.
We also had a side discussion about how McDowell’s view would try and handle puzzle cases like Kripke’s Pierre. Of course this came after McDowell’s paper which was published in 1977. I think the consensus was that he would not try and solve the puzzle but would merely point out that, as the puzzle arises, his descriptive theory of sense would require that the names cannot be substituted, and possibly that the sentences in different languages cannot be translated, without giving a misleading picture about Pierre. This would be a case where his theory describes but does not explain.
Thanks to everyone involved for an interesting discussion. I think we might read Donald Davidson’s ‘A Nice Derangement of Epitaphs’ next time.
RSS - Posts
Mark Sainsbury has a paper on how a McDowellian might handle the other Kripke cases. Suppose that your solution to Frege puzzles is to introduce two distinct homophonic axioms into your theory of meaning, so we have:
For all x, ‘Hesperus’ denotes x iff x is Hesperus, and
For all x, ‘Phosphorus’ denotes x iff x is Phosphorus.
(These are classically equivalent to McDowell’s axioms. They’re weaker in the free-logic Sainsbury favors to handle the problems about empty proper names you mentioned.)
It looks like there’s pressure to add two distinct axioms for ‘Paderewski’, since we don’t want to interpret Peter’s utterance of ‘Paderewski is not Paderewski’ as a manifest contradiction any more than the ancients’ utterances of ‘Hesperus is not Phosphorus’ – that wouldn’t render Peter’s utterances intelligible in the desired fashion. But two distinct axioms i) seem undesirable when we’ve only got one name on our hands, since it’s in effect positing massive ambiguity in proper names and ii) require a departure from homophony. So, problem for McDowell-style minimalist accounts of sense.
The paper is in Philosophical Books, published in 2004 as part of a symposium on Sainsbury’s ‘Departing from Frege’.
Thanks for that, I’ll look at that paper when I get the chance.
I wonder just how strong the worry is for McDowell. You might say that there’s only an ambiguity if a single name denotes more than one object. But that doesn’t happen either with ‘Hesperus’, ‘Phosphorus’ or ‘Pederewski’ as far as I can see. It would also seem odd to ban homophonic names from having different referents, for example I know of a philosopher called ‘Graeme Forbes’ and a Ph.D. student called ‘Graeme Forbes’. If we accept McDowell’s claim that what he’s doing is describing a confused and confusing linguistic practice then it’s not a problem for his theory that he has to allow for that situation, and the situation that some people call Venus by one name in the morning and another in the evening.
The Pederewski case is harder because it looks like we have to say that people have beliefs about him qua pianist and qua statesman. Which looks like a rich theory. It’s that which would really scupper McDowell.