Saul Kripke, Naming and Necessity



Saul Kripke, Naming and Necessity

Metaphysics and Epistemology

I. Rigid Designators

What does it mean for something to be a RD?

Only proper names, Kripke says, are RD. So, if “Nixon” is a proper name, then “Nixon” refers to Nixon in all PWs, even those in which Nixon isn’t called Nixon, and even those in which Nixon doesn’t have any of the properties we associated with Nixon.

On the other hand, a non-rigid designator is a description, like “the 38th president of the US”. This is non-rigid because it can actually pick out vastly different objects in different possible worlds. We know this because we agree that “Nixon could have not been the 38th president of the US” is true.

II. Three theories of Naming

Theory A: Frege and Russell: descriptions fix the sense of a proper name

What this theory means is that a proper name is synonymous with some description. So, whatever description properly (in the actual world) picks out some referent gives the meaning of the proper name.

Aristotle = “teacher of Alexander” and “pupil of Plato”

Then,

1. Two people could know one and not the other of Aristotle. So when someone says

“Aristotle died in 456BC” this sentence will have different meanings for these two people.

2. We could replace “Aristotle” in “Aristotle is the teacher of Plato” with “teacher of Plato” to get “the teacher of Plato is the teacher of Plato” which is a tautology.

But this is clearly not a tautology given that someone else could have been the teacher of Plato.

The Undisclosed Problem:

Frege and Russell are saying that a rigid designator is synonymous with a non-rigid designator. It seems as if F and R are assuming that non-rigid designators are actually rigid. But they are not. So we need descriptions to be non-rigid or we have this problem.

Theory B: Donnellan: Names are Mere Tags

Let’s say that name D designates x. Then the meaning of D is the property of designating x. D even rigidly designates x, so solving the problem of theory B.

Consider AL and CG.

CG designates Cary Grant.

AL designates Cary Grant.

So the two designators mean the same thing, and they are both rigid.

1. AL is a famous actor means the same as CG is a famous actor, since CG and AL are synonyms.

Suppose Ann knows CG from his days as a thug named AL. She also knows CG as the star of MGF.

But Ann does not know that AL became CG.

Now Ann understands “AL is a famous actor” and “CG is a famous actor”. If CG and AL are really synonyms, then she should agree that all propositions true of CG are true of AL.

But she does not. She rejects “AL is a famous actor”

So then they cannot be synonymous.

Underlying Problem: Any two rigid designators that refer to x will always have the same meaning.

Theory C: Kripke: Names are Rigid Designators

Apparently, what we need are two rigid designators having different meanings (unlike in B) but which have their meanings not because they are synonymous with some descriptive content (theory A).

A description is used to fix the referent of a name. each proper name has a description “the F” which gives the descriptive content of N and which picks out the actual F as the object N designates.

CG is associated with “star of HGF” so that CG rigidly designates whoever is actually the star of HGF.

1. CG is not synonymous with “star of HGF”. This must be so, because “CG” is rigid and “star of HGF” is not. In one PW, “star of HGF” designates Reagan, in another it designates Grant.

So this fixes the problem with Theory A (Frege and Russell)

2. CG is not synonymous with AL. Here we need to rigid designators pointing to the same x to not have the same meaning. The meaning of a rigid designator is the descriptive content that allows us to fix the reference of the proper name, in this case, the descriptors in both cases are different. So the meanings of the two proper names are different.

So this fixes the problem with Theory B (Donnalen)

III. The Necessary A Posteriori and the Contingent A Priori

The theory of rigid designators has some consequences for epistemology and metaphysics, and especially for the claim that what is necessary is always a priori and that what is contingent is always a posteriori.

Assume CG = AL.

1. What is the modal status?

2. What is the epistemic status?

Take #1.

Is the identity true?

a. It is clearly true at the actual world.

b. But what about other worlds?

It is true that the non-rigid designators used to fix the reference of AL could be true of some other person in that world. That’s okay, because CG is not synonymous with the method of ascertaining the referent.

So it is true in all possible worlds, and so that’s its metaphysical status.

Take #2.

The sentence itself is a necessary truth. So, according to the traditional view, it must be a priori.

Can Ann come to know that CG = AL without going out and collecting sense-experiences?

So what we have here is an example of a necessary a posteriori sentence.

The Contingent A Priori

“Stick S at time t0” = “meter”

1. What’s the modal status of this claim?

2. What’s the epistemic status of the claim?

Take #1

Clearly the metaphysical status of the proposition is contingent. It could very well be the case that this stick was longer or shorter than one meter. So it might be one meter in this world, but it need not be in all worlds.

Take #2

What’s the epistemic status of the claim? Well, we introduced the rigid designator “meter” by means of a certain descriptor – “Stick S at time t0”

As such, the proposition is clearly known a priori. It is merely a matter of reflection to see that the stick is one meter. Clearly we don’t learn this. We introduce it, so it can be known without investigation.

So, in a way, Kripke is saying that we are using the descriptor – which is identical with the object in the actual world – as a kind of contingent “ladder” upon which to reach the actual referent. So, “the stick S” is the ladder that I use to reach the object, “meter”. Once named, “meter” is what it is. It is metaphysically necessary. But the “Stick S” is not so fixed, as it can differ in PWs.

Does this mean that all descriptors and their names are contingent a priori?

“Aristotle is the teacher of Plato”?

Is “the teacher of Plato” contingently identical with Aristotle? Yes.

IV. Hesperus = Phosphorus

1. The case of contingent identity. How can an identity be contingent? Does that even make sense? How can A=A only be contingently true?

Note here that Kripke’s point here is extremely intuitive. Some identities do, in fact, seem to be contingent. But if so, then it looks as if his theory of naming is right.

2. How can Kripke’s point here about contingent identity be used to argue for the identity theory of mind? Namely, that “pain” = “C-fibers” is not a necessary truth?

Some say that this is a correlation between two different things. But Kripke is saying here that this is not so; pain and C-fibers are one and the same thing in the actual world, but that the identity is not necessary. So pain can be something other than what is actually is in the actual world.

3. Could it have been the case that Hesperus is not Phosporus?

Clearly someone could use the names differently. But this isn’t what we mean.

Isn’t it the case that there’s a PW in which H isn’t P in advance of our discovering that they are the same?

Kripke’s point here is that if we mean “it might be true that H=P and it might be true that it is false” then we must be clear about the sense of “might” we are talking about. To say that it might be one way or the other doesn’t mean that however it turns out, it’s not necessary. It is. The “might” refers to our knowledge.

Metaphysical necessity: the way things must be

Epistemic necessity: how things must look to us

Things can clearly look true or false but that doesn’t mean that they can be true or false.

The Goldbach’s theory case. However this turns out, it will be necessary. But I don’t know how it will turn out. It looks like it could go either way. But it really can’t.

Mention that the difference pops up in showing how Descartes’ proof for the existence of God fails because it equivocates between the two notions of possibility.

V. Fodor on Kripke

Stage One: Conceptual Analysis

Philosophy used to be about things. But then, after WWII, it became a discipline that concerned itself with concepts of those things.

Some advantages:

1. We don’t need to find the Good, just have a grasp of its concept to investigate it.

2. The findings of empirical analysis are revisable, but conceptual analysis isn’t.

3. Conceptual analysis, however, can be awfully stupid.

Stage Two: Quine

1. No conceptual truths. If so, then no conceptual analysis.

2. Thus, analytic philosophy is methodologically unemployed.

Stage Three: Kripke

1. We need a notion of necessity that doesn’t relate to concepts

2. Drop the notion that necessary truth is linguistic or conceptual, it’s metaphysical.

3. We use possible world semantics now, but not presupposing concepts.

4. Analytic philosophy used to talk about tracing conceptual relationships, now they talk about relationships between possible worlds.

5. Think about Putnam. If “water” = H20, then it is true in all PWs. But don’t we want to say that water could be lots of things? So what gives us the modal intuition that water has an actual essence?

6. New move: modal intuitions are not fallible.

7. But how do we learn these things? Metaphysical necessities can be learned by examining philosophically relevant intuitions.

8. These are not about concepts, but about what is possible (from the account of naming)

9. So analytic philosophy was doing the right thing (analysis) but for the wrong reasons.

10. But what are modal intuitions made of?

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download