This Pain and the Concept, PAIN

To try to clear up some suggestions (mostly for myself) I made last time, let's consider some identity statements, coupled with counterfactual statements:

(1) This table is made of wood.
(1*) This table might have been made of ice.

(2) This table is a packing case.
(2*) This table might have been a coffin.

(3) Cicero is Tully.
(3*) Cicero might not have been Tully.

(4) Heat is the motion of molecules.
(4*) Heat might not have been the motion of molecules

(5) Pain is a brain process.
(5*) Pain might not have been a brain process.

What distinctions might we draw here, and do they matter? The first thing that sticks out is that the 1-3 examples include either indexical reference or reference to particulars; 4-5 include general terms of reference. Also, the verification conditions for the 3 are ordinary, by which I mean they can be verified by the experience of the linguistically competent man on the street, given access to the particular referents. The latter 3 require sophisticated equipment, and maybe a lot of education to verify. I won't explore whether or not this matters here.

1-2 make reference to the composition of particulars, 3 includes only two names denoting a single object, 4-6 include reference to the composition properties. I'll try to unpack this below a little, because I think these distinctions are significant; they may affect which among these, if any, should be understood as necessary identities, such that it’s counterpart (*) is (metaphysically/logically?) impossible. I suggest that the case of the particular identities is importantly different from that of the general ones.

In 1-2 there is pull to the intuition that, once verified, the identities could not have been otherwise. When we go to the (*) counterpart sentences, it is not at all clear that were the table made of ice, or were the table a coffin put to another use, that we’d be talking about the same table in either (*) case. Maybe it is necessary that this table is made of wood, etc. ; there seems to be a straightforward sense in which indexicals rigidly designate.

With regard to 3, similar considerations apply. We may individuate reference for ‘Cicero’ and ‘Tully’ by different contingent facts functioning as descriptions about one and the same individual, but that we associate contingent facts with different names doesn’t imply that identity of the referents is contingent. Regardless of what descriptions apply, talk of Cicero and talk of Tully is talk of the same man. The (*) counterparts look impossible here, too; maybe proper names are good candidates for rigid designators, too.

The situation is less clear when we get to 4-5. We should look at them in turn. In 4, as was indicated, we refer to a property (not an object) and endorse its identity with a physical phenomenon. On the one side we have a property of a kind of subjective, apparently private, experience, the sensation of ‘heat’. On the other, the motion of molecules taken to be the same thing. As Kripke notes, we apparently use the contingent fact that ‘heat’ causes the sensation of ‘heat’, to identify ‘heat’. Crucially, there is supposed to be an analogy between contingent reference conditions for properties like ‘heat’ and contingent reference conditions for names like ‘Cicero’. We pick out Cicero by the contingent fact that he was a statesmen; so says Kripke, we pick out ‘heat’ by the contingent fact that it produces a particular kind of experience/sensation.

It is not clear at all to me that there is not some question begging going on here. If not, Kripke’s language is extremely confusing; he can’t mean to say we pick out ‘heat’ in the world by the contingent fact that it causes such and such a sensation. He must mean that we pick out the motion of molecules in the world by the fact that it produces such and such a sensation. But if that is right, and the sensation is only contingently related to the motion of molecules, then the identity is contingent. Kripke hasn’t shown that HEAT and MOTION OF MOLECULES are conceptually connected in such a way as to make the identity necessary. He would’ve had to show that there is something to the concept HEAT over and above the sensation. So, the (*) sentence does not express an impossible state of affairs in this case (4*), due to an important, if only implicit, disanalogy with cases of particular-object reference: we can imagine Cicero not being a statesmen, but we cannot imagine him not being Tully; but, unless HEAT is more than just a sensation correlated with a lot of molecular agitation, we can imagine that it is not identical with molecular motion.

What about the case of pain? In 5, we refer to a (less controversially, at least on the face of it) apparently private kind of state of consciousness, and endorse its identity with a neurological process. Kripke thinks this case is importantly different from (4): that is, we do not pick out states of pain by the contingent fact that we sense pain in a particular way. But here Kripke ignores the fact that we are no longer expressing an identity statement about a particular (and it may be that this owes to his target, strict identity, but his arguments against materialistic identity as a program appear to be defeasible); His crucial move seems to be this:

[it] 'might be true of the brain state [that we pick it out by the contingent fact that it affects us in a particular way], but it cannot be true of the pain. The experience itself has to be this experience, and I cannot say that it is a contingent property of the pain I now have that it is a pain.’

Fair enough, but that’s not what (5) says, and the material identity theorist doesn't need to be so strict. Even Place, in particular, might have been able to argue that Kripke conflates (5) with (7):

(7) This pain is a brain process.
(7*) This pain might not have been pain.

Now, consider:

(8) This heat is the motion of molecules.
(8*) This heat might not have been the motion of molecules.

Switching from general claims to specific claims does change the climate. But how so?

Presumably, PAIN should be read here as I indicated HEAT might be read above: there maybe nothing more to the experience of pain than the sensation or experience. Grant Kripke that this pain or that pain is necessarily pain; (7*) is impossible. But this does not imply that (5*), ‘Pain might not have been a brain process,’ is impossible. In other words, it may be necessary to claim that that 'this pain is this brain process,' but it need not be necessary that 'pain is a brain process': contra Kripke, ‘Pain is a brain process’ might really be a contingent identity statement, even though cases of particular pain might express necessary identity relations between particular experiences and whatever underwrites them.

Since we can imagine PAIN in creatures very unlike us (like Lewis’ Martian) we don’t need to identify PAIN as a kind with 'pain' as referring to this or that sensation/experience. I think the same goes for HEAT. This or that sensation of heat rigidly designates the experience as the reference, but it doesn’t rigidly designate HEAT as the motion of molecules. Since the identity is plausibly a contingent one, the materialist is not required to answer Kripke’s challenge that ‘he has to show that [pains without brain states] are impossible.’