Kripke Identity and Necessity

One of Kripke's main questions is "Can there be contingent identity statements? (Contingent just means, not necessary, and so the question is the same as "Are all true identity statements necessary?)

Is that a question about language or a metaphysical question? There is certainly a trivial sense in which it's a question about language—as phrased, it is about statements. However, the real question is whether there could be things that are distinct but could have been identical or a thing that could have been several.

Our answer to the linguistic question is going to be deeply tied to answers to metaphysical questions about how objects might differ from the way in which they actually are.

If that is what Kripke is interested in (and he was, in fact, originally interested in the issues in the article because he was interested in whether mind and brain could be the same), why does he talk about identity statement instead? Why, that is, does Kripke employ (as Quine has suggested) what Quine calls "semantic ascent"? One part of the reason is to avoid getting tangled in Plato's beard, but that is only part of why Quine suggests the strategy: it is a way of avoiding getting tangle in all sorts of issues about how to correctly express what you mean. The issues are subtle and concern aspects of things we aren't accustomed to talking about, and so we don't have a good vocabulary for discussing them. Words, on the other hand, are relatively straightforward things, right there on the page, and much easier to talk about.

Quine's Descriptivism

In "Ontological Relativity," Quine argues that the references of our words are profoundly indeterminate by arguing that everything we say, all of our descriptions has multiple interpretations on which exactly the same sentences come out true. On some of them, "rabbit" refers to rabbits, while on others "rabbit" refers to undetached rabbit parts. What is preserved in all the alternatives? The language, the descriptions used. Quine argues that we don't know what rabbits are by arguing that all of our descriptions of rabbits fail to say unambiguously what rabbits are.

Many philosophers, and others, had thought that we knew what rabbits are in a quite different way: we have some rabbits around, as "samples." Quine seems to disallow that as a way of learning about rabbits, and many philosophers, cowed by Quine, consented. Kripke in today's reading is arguing against (Quine's) descriptivism, that is, against the view that getting all of the assents and dissents to language correct is enough. He, nearly single-handedly, opposed descriptivism, and many, perhaps most philosophers have now signed on.

Kripke focuses the whole debate on names, and only later generalizes. How do we know to what a name refers? According to Quine, we see the name used in association with lots of descriptions, and that gives us the name. Kripke thinks that names "tag" (Ruth Marcus) objects: that a name can be just tied to an object not in virtue of any description.

So, the debate is, to a first approximation, whether names are abbreviated descriptions (like "Pegasus" abbreviates "Pegasizes") or tags.

Quine has argued that we should always take names to be abbreviated descriptions, largely to avoid Plato's beard. Kripke says that that is not adequate when we discuss counterfactuals. (A "counterfactual" is a statement about what things would be like in some situation that actually hasn't or won't happen.) Kripke thinks that when we say, for example, "Shakespeare might not have written Hamlet," it is clear that we intend to be saying something about Shakespeare, the bard himself, not just about someone who happens to Shakespearize. Quine's view loses something important, namely, the ability to contemplate counterfactual about particular individuals, not about mere descriptions. One way to make that clear might be to consider the claim that "Nothing that is actually true of Shakespeare (no description of him) had to be true of him, and so Shakespeare might have been so completely different that no description would necessarily pick him out." If that were true, Kripke would clearly win, since Quine would have no hope of being able to express that. Kripke doesn't, in fact, believe that, and so he attacks in slightly less direct way. He claims that names often function as rigid designators that pick out the object they name not in virtue of any description.

If a name isn't stuck onto an object in virtue of a description, what "glues" it there? Kripke talks about "baptisms." When you name an object, you need to indicate what object you are naming somehow (pointing, describing, ostending, dunking, …). But, once the object has been named, the name sticks to it because it is the same object. How can I use a name for an object I've never seen? Because I acquire the name from someone who has. Each use of a name names the object it does because there is a causal chain between the use and the object.

What can Quine say against that? It won't work for abstract objects, and, since all objects are to some degree like abstract ones, it doesn't work at all.

Kripke mentions "natural kinds" as abstractions that Quine can't handle.

-- ShaughanLavine - 30 Jan 2008 - 01 Feb 2008