True Identity

True Identity

The classic identity theory was advertised as an empirical (“contingent”) theory not an a priori truth. That means it could turn out to be empirically false. I will argue that it is, but another identity theory could take its place. It was always sketchy about what states of the brain are identical to mental states and very few examples were given—in fact one, the identity of pain with C-fiber firing. It wasn’t claimed that pain is identical to some fiber (axon) firing, but specifically C-fibers. The former would be no good because all mental states would then be identical to the same physical state (neural firing), and hence identical to each other (pain would be identical to pleasure). No, it had to be C-fiber firing: that physical type of fiber must be the one with which pain is identical. So, it had better not be true that C-fiber firing is correlated with some other mental state too—pleasure also being correlated with C-fiber firing. Pleasure needs to be correlated with D-fiber firing. But what are these physical types of fiber? Are they of different lengths or diameters or degrees of myelination? There is no evidence of that, and it seems highly unlikely given what we know of the brain; in fact, it appears empirically false. What is necessary is that for each type of mental state there is a distinctive physical type of fiber correlated with it. But that is surely empirically false. The case isn’t like heat and light, say, in which the two natural kinds are correlated with distinctive physical correlates, viz. molecular motion and streams of photons, respectively. We don’t offer some woolly statement like “heat is some property of molecules” or “light is the behavior of tiny particles of some sort”; we need to be told specifically and distinctively what heat and light are to have an empirically viable identity theory. We likewise need to be told what it is about C-fibers that makes them uniquely what pain is–and similarly for every other mental state. Or else we are just saying that pain is some property or other of nerve fibers; and it certainly isn’t just their firing because all nerve fibers do that. The theory, as it stands, is hopelessly inadequate empirically. This identity theory is empirically false or incomplete or unexplanatory. We don’t have good empirical grounds for believing it.

But that doesn’t mean that no identity theory can be empirically true. And I for one think that some sort of identity theory must be true; the question is what kind of identity theory. It seems obvious that no current theory has what it takes, since we don’t have the kind of specific information about the brain that would satisfy reasonable conditions of empirical adequacy, i.e., a one-one mapping between mental types and brain types. What we know now is far too general, unspecific, lacking in detail. What precisely is it about the physical correlates of pain, in particular, that suits them to be pain and not some other sensation? Molecular motion won’t give you light and streams of photons won’t give you heat; where is the analogue for pain and pleasure? This is what is lacking, and we seem far from filling the gap. But I don’t doubt that something does, because the brain is clearly integral to the mind—there is nothing else in there. I would think some sort of conceptual revolution is going to be required in order to identify the right properties of the brain, but once this is achieved, we can imagine slotting the right term into an identity statement and getting an empirically adequate result. That is, the solution to the mind-body problem will take the form of an identity statement (assuming we don’t have to jettison the concept of identity!). The classic identity theorists were right that the mind is the brain (dualism is false), but they were wrong about the terms of the identity, i.e., the concepts and properties that need to be invoked. Whether this identity statement is discoverable by the human mind is another matter. There must be correlates in the brain that map systematically and beautifully onto states of mind, and these must be what states of mind are—given that dualism has been ruled out. But the candidates for identity currently available are far from empirically adequate. The statement “pain is C-fiber firing” is at the level of “heat is something to do with what objects are made of” or “light is some sort of impalpable stream”. Pain is indeed something to do with the properties of nerves in the brain, but that is far from being an empirically adequate identity theory of pain, true though it may be.[1]

[1] If the neural properties turn out to be anything close to what the panpsychists suggest, then the identity statement would be of the form “pain is identical to micro-psychological properties x, y, and z” or some such thing. It might also be necessary to find further properties of pain that can be paired with properties of the brain, as it might be phenomenological structural properties. The question is wide open and extremely confusing (as someone once said).