Ramsey's solution to the first problem
Now that we have seen how to solve the problem of defining one mental state without circularly assuming that we have already defined some others, let's see if we can see how to do this for belief.
If we were to try to do this for belief, we should need many more letters than “X” and “Y.” We call these letters “variables,” and they function in a way I shall explain in the chapter on language. But the procedure would be exactly the same. We would first write down all the claims about beliefs and desires and evidence and action that we think have to be satisfied by a creature that has a mind. This body of ideas is what is sometimes called our “folk psychology”: it's the shared consensus of our culture about how minds work, the “theory” we learn as we grow up. If we join all the claims of folk psychology together with “and's” we will have one very long sentence, and that will be our functionalist theory of the mind. Call that sentence MT (for “mental theory”). From MT, we would then take out all the mental terms referring to beliefs and desires and replace them with “variables.” The result of this we can call MT*. Finally, for each variable we should write “There exists a... “ in front of MT*, and we would have a new sentence, which didn't have any mental terms in it. That sentence is called the Ramsey-sentence of the theory MT, because the British philosopher Frank Ramsey invented this procedure. The Ramsey-sentence of MT says, in effect, that something that has a mind has a large number of internal states—one for each variable—that interact with input and with each other in certain specific ways, to produce behavior. (I called the final version of the simple-minded theory of pain “R,” because it's the Ramsey-sentence of the simple-minded theory of pain.)In 1.4 I said that many philosophers who have thought about the other-minds question have wanted to be able to define mental states in such a way that it was always possible, at least in principle, that somebody else should know what is going on in your mind.
Notice that this functionalist theory, set up in the way Ramsey suggested, seems to make this possible. For Ramsey's method allowed us to define pain in terms of its causes and effects, its functional role, in such a way that if we have evidence that someone's internal states would make them react in certain public ways—brow wrinkling and the emission of loud noises—in response to certain public events— pinpricks—we have evidence that they are in pain. It allowed us to do this without requiring that we know anything about the other internal states—in this case, worry—except that they too would have certain causes and effects, which could, in the end, be seen to show up in what people do. For the Ramsey-sentence of MT is true of someone if and only if he or she has a system of internal states that produces the right pattern of responses in output—in this case, brow wrinkling and loud noises—to input—in this case, pinpricks.In the more complex case of beliefs, as we saw, we can proceed in a similar way. But here, just because the case is more complex and there are so many more internal states, it may be very hard, in practice, to discover that the right complex pattern of dispositions to respond to input exists. So, while allowing us to take mental states seriously, functionalism also allows us to believe that they might be very difficult—indeed, practically impossible—for anyone, except perhaps the person who has them, to find out about. (I'll say something about how a functionalist might explain our knowledge of our own states later, in section 1.11.) It is in this sense that functionalism is a halfway house between Descartes and behaviorism. For Descartes, as we saw, left open the possibility that someone could have mental states that no one else could know existed even in principle. Functionalism denies this. Any evidence of the existence of the right (extremely complex) pattern of dispositions will be evidence of your mental states. For behaviorism, every mental state is nothing more than a disposition to respond to input. Functionalism denies this also. What someone with a certain belief will do when stimulated depends, the functionalist claims, on other internal states as well.
1.9