Functionalism and type-type identity theories

Token-token identity theorists do not and need not deny that it may frequently be the same (kind of) brain state which on different occasions fills the functional rôle definitive of a given mental state. That is not at issue. What is at issue is whether functionally-oriented identity theorists should make two claims or three claims. The two claims they customarily make are, first, that each instance of a mental state is an instance of a brain state, and, secondly, that being in a mental state is having in one a state filling the relevant functional rôle. But to be in a mental state is to have that state in one. To be in pain is to have pain, to desire water is to have desire for water, and so on; just as to be poisoned is to have poison in you. (It is to have what is poison for you at the time, of course; and likewise for pain, desire and so on.) Our paper has been about a third sort of claim - relating particularly not to being in a mental state, nor to instances of that state, but to the mental state itself. We have argued that functionally-oriented identity theorists can and should make, in addition to the first two claims, the third type-type identity claim that mental states are brain states. Consequently a token brain state is a token of pain in a derivative sense. What makes it a token of pain is that it is a token of the type of brain state which realizes the pain-rôle for the organism at the time.