Online research in philosophy

A chronic difficulty for functionalism is the problem of instantiations of a functionalist theory of mind which seem to lack some or all of the mental states--especially qualitative--we want to attribute to minds the theory describes. Here I discuss one such counterexample, Block’s system S, consisting of the population of China organized to simulate a single mind as described by some true, adequate, psychofunctionalist theory. I then defend a version of functionalism against this example, in part by an adaptation of Dennett’s notion of “stances”. A true, adequate theory, as Block understands it, would be appropriate to Dennett’s “design” or (at best) “intentional” stance; but a genuinely true and adequate theory should instead coincide with a “personal” stance. Hence, if system S does instantiate such a theory, we must impute to it mental states, even qualitative, whether or not it “really" has them. Hence Block’s counterexample lacks force.

In this paper I distinguish three alternatives to the functionalist account of qualitative states such as pain. The physicalist-functionalist holds that (1) there could be subjects functionally equivalent to us whose mental states differed in their qualitative character from ours, (2) there could be subjects functionally equivalent to us whose mental states lacked qualitative character altogether and (3) there could not be subjects like us in all objective respects whose qualitative states differed from ours. The physicalist-functionalist holds (1) and (3) but denies (2). The transcendentalist holds (1) and (2) and denies (3). I argue that both versions of physicalist-functionalism inherit the problem of property dualism which originally helped to motivate functionalist theories of mind. I also argue that neither version of physicalist-functionalism can distinguish in a principled way between those neurophysiological properties of a subject which are relevant to the qualitative character of that subject's mental states and those which are not. I conclude that the only alternative to a functionalist account of qualitative states is a transcendentalist account and that this alternative is not likely to appeal to the critics of functionalism.

Ned Block’s Chinese Nation Argument is offered as a counterexample to Turing-machine functionalism. According to that argument, one billion Chinese could be organized to instantiate Turing-machine descriptions of mental states. Since we wouldn’t want to impute qualia to such an organized population, functionalism cannot account for the qualitative character of mental states like pain. Paul Churchland and Patricia Churchland have challenged that argument by trying to show that an adequate representation of the complexity of mind requires at least 10 30,000,000 homunculi. As such a large collection of Chinese is beyond comprehension, the intuitive force of Block’s example would be undercut. I argue that Churchland and Church land erroneously assume that every possible input-state combination in the human Turing-machine table must be assigned a homunculus. I attempt to restore the intuitive force of Block’s thought experiment by pointing to a way to simulate the human mind that does not require any such staggering number of homunculi.

The functionalist view of the nature of the mind is now widely accepted. Like behaviorism and physicalism, functionalism seeks to answer the question "What are mental states?" I shall be concerned with identity thesis formulations of functionalism. They say, for example, that pain is a functional state, just as identity thesis formulations of physicalism say that pain is a physical state.