How does Kripke lay out the epistemic argument?

Question

The epistemic argument against descriptivism, as laid out by the Stanford Encyclopedia of Philosophy states:

Kripke's epistemic argument (1980, 78; 87) is closely related, but trades on epistemic, rather than metaphysical, modality. His argument is not that names are rigid in epistemic contexts. That would be a hard sell, as (10) is true on its epistemic reading (Kripke 1980, 103-4):

10. The Morning Star might have turned out not to have been the Evening Star

Instead, he argues that no definite description D has the same semantic value as the name ‘Aristotle' (say), because otherwise the sentence (10) would be analytic, and so knowable a priori.

11. Aristotle is D

But, having read the relevant pages in Naming and Necessity, I'm lost (just to be clear, I have read most of the book up to page 88 so far).

On page 78, we have:

If I use the name 'Hesperus' to refer to a certain planetary body when seen in a certain celestial position in the evening, it will not therefore be a necessary truth that Hesperus was ever seen in the evening. That depends on various contingent facts about people being there to see and things like that. So even if I should say to myself that I will use 'Hesperus' to name the heavenly body I see in the evening in yonder position of the sky, it will not be necessary that Hesperus was ever seen in the evening. But it may be a priori in that this is how I have determined the referent. If I have determined that Hesperus is the thing that I saw in the evening over there, then I will know, just from making that determination of the referent, that if there is any Hesperus at all it's the thing I saw in the evening.

I don't understand how any of this relates to the epistemic argument.

On page 87:

Thesis 5 says that the statement, 'If X exists, the X has most of the φ's,' is a priori if true for A... I think that my belief about Godel is in fact correct and that the 'Schmidt' story is just a fantasy. But the belief hardly constitutes a priori knowledge.

But how is refuting this thesis an attack on descriptivism? Descriptivism doesn't require that thesis, does it?

Furthermore, when does Kripke show that a statement being analytic means that it must be a priori? From what I've read of Naming and Necessity and Identity and Necessity, he seems to feel that you need some sort of argument or proof to show that the labels analytic, a priori, and/or necessary are interchangeable.

Am I misunderstanding here, or getting his arguments mixed up?

Answer 1

Descriptivism requires that proper names have their reference fixed in virtue of a description attached to them, which singles out a unique object in the world, and which is analytic: the meaning of a proper name can be expressed by a definite description.

Analytic means: in virtue of the meaning of words. It is implicit here that the meaning of words is known a priori by anyone who masters a language.

Kripke refutes the idea that we could have a priori knowledge of a definite description that would single out a referent for a proper name. We can use a proper name correctly (Gödel, Aristotle) while knowing almost nothing a priori about the referent. This is the argument on page 87, and also the second part of SEP entry you cite (except that "the sentence (10)" should be "the sentence (11)", there is probably a mistake).

Per my understanding, the passage on page 78 attempts to argue that necessity and analycity are different things, and that names are not rigid designators in an epistemic context (the first part of SEP entry). It doesn't seem to be directly related, but is probably used in the other argument (it is not necessary that Aristotle has such definite description).

Necessity and analycity were often interchangeable for descriptivists, but Kripke introduced a metaphysical necessity distinct from linguistic or logical necessity.