102 Chapter 5

are not. So perhaps it would be best to begin by trying to soften up the conviction that true beliefs are obviously valuable. It is my suspicion that part of the explanation for the fact that people take true beliefs to be of obvious value lies in a cluster of philosophical metaphors that have long ago become part of the fabric of commons en se wisdom. According to these metaphors, people's beliefs constitute their picture of the world, their internal mirror of reality. Or, to alter the image slightly, our beliefs are maps by which we steer." If beliefs are like pictures or mirrors or maps, then true beliefs are like pictures that resemble their subject, or like undistorted mirrors, or like accurate maps. And surely we'd prefer to have veridical pictures of the world and undistorted mirrors. Surely we're better off steering our way around the world with accurate maps. If having true beliefs is like having a good picture of the world or an accurate map of reality, it seems simply perverse to suggest that having true beliefs is not a situation to be valued.

All of this begins to unravel, however, when we recognize that talk of beliefs as pictures or maps is just a metaphor, and one which cannot be easily unpacked. If one is a materialist, then beliefs must be located somewhere in the material universe-presumably in the brain. But there are no pictures or mirrors or maps in the brain, not literally. I believe that Oswald shot John Kennedy and that Interstate 5 runs from Solana Beach to La [olla, but there is nothing in my brain that .looks like John Kennedy being shot, nor is there anything that has the shape of the southern California freeway system. Even if one rejects materialism, the metaphor of beliefs as pictures or maps of extramental reality is all but impossible to unpack, as both Berkeley and Kant knew all too well. Consider:

I believe that there are exactly four prime numbers between ten and twenty.

I believe that if Robert Kennedy had not been assassinated, he would have been elected President of the United States.

I believe that considerations drawn from the theory of evolution cannot support the view that normal people are generally rational.

I believe that we do not now have a good theory about the nature of consciousness.

For each of these beliefs, and for countless others, it seems to make no sense at all to suppose there might be a map or a picture of what is believed. The contents of these beliefs are just not the sorts of things that can be pictured. The moral I would draw here is that the metaphor

Do We Really Care Whether Our Beliefs Are True?

103

that beliefs are pictures or maps or mirrors is both obscure and profoundly misleading. The conviction that true beliefs are valuable cannot be sustained by the analogy between true beliefs and accurate maps, or between true beliefs and portraits that resemble their subjects, since when pushed those analogies quickly come unglued.

5.2 Belief and Truth

All right, then, beliefs are not much like pictures, and true beliefs are not much like accurate maps. But before considering whether having true beliefs is valuable, we will have to say something about what beliefs are and what it is for them to be true. That is a project that poses some serious difficulties, however, since both of these issues have been the focus of a great deal of philosophical dispute, and radically different views abound. My strategy for dealing with this rather untidy situation will be to start, in this section and the two to follow, by sketching what I take to be one of the more plausible positions about what beliefs are and what it is for them to be true. In 5.5, I will argue that if the account of true belief that I've presented is on the right track, then few of us will find having true beliefs to be intrinsically valuable. I'll also explore some of the obstacles to be overcome by those who would argue that true beliefs are instrumentally valuable. Of course, even if you're convinced by everything in 5.5, I will not have established that it's not valuable to have true beliefs. Rather, what I will have shown is that this conclusion follows if you accept the account of beliefs and what it is for them to be true developed in 5.2-5.4. Thus, in 5.6 I'll try to generalize the argument by showing that the essential features of the account of true belief developed in 5.2-5.4 will be shared by any plausible alternative account.

Now, what about belief? The theory I'll be using as a backdrop for my arguments assumes that beliefs are real psychological states, not explanatory fictions like the lines in a parallelogram of forces. Thus the theory rejects the sort of instrumentalism about belief that Daniel Dennett has made popular of la te. 3 The theory embraces the so-called token-identity hypothesis, which claims that each instance (or token) of a belief is identical with some neurophysiological state or other, though it does not endorse the type-identity hypothesis, which holds that the same belief type in different individuals is always identical with the same neurophysiological state type. Subtleties aside, what these assumptions amount to is the claim that belief-state tokens are brain-state tokens.

Unlike most brain tares, how ver, and unlike most everything else in the universe, beliefs have semantic properties. They are true or false.