Rationality, Meaning, and the Analysis of Delusion

Keywords: Capgras, Cotard, schizophrenia, radical interpretation, recognition, the first person

 

On what I will call a rationalist approach to delusion, delusion is a matter of top-down disturbance in some fundamental beliefs of the subject, which may consequently affect experiences and actions. On an empiricist approach, in contrast, delusion is a rational response to highly unusual experiences that the subject has, perhaps as a result of organic damage. Ellis and Young (1990) recently provided an empiricist analysis of the Capgras and Cotard delusions. I want to begin with some remarks on just why it is important to the empiricist approach that it should acknowledge the rationality of the subject's delusional responses to unusual stimuli. We will see that when the rationale for the rationality constraint is fully set out, it is questionable whether Ellis and Young's approach actually succeeds in giving its place to the rationality constraint. In conclusion, I will look briefly at the prospects for a rationalist approach and at what other approaches might be possible.

1. Rationality and Knowledge of Reference

It is often said that rationality on the part of the subject is a precondition of the ascription of propositional attitudes such as beliefs and desires. One simple reason for thinking that rationality is critical here is that unless you assume the other person is rational, it does not seem possible to say what the significance is of ascribing any particular propositional state to the subject. If you tell me that someone rational thinks that it is raining, then given that the person is rational and does not want to get wet, I know what kinds of behavior to expect. If, however, the person is not at all rational, then saying they have the belief has no implications at all for how they will behave.

This, though, is not the fundamental reason why rationality is important for ascriptions of belief and desire. This consideration only tells us that there must be some architecture to someone's propositional states, without telling us which architecture we have to assign. The real reason for thinking that rationality is important has to do with the relation between belief and meaning. Quine long ago ridiculed the notion of a "pre-logical people" (Quine 1960). Suppose someone claims to have discovered some savages so benighted that they have not yet managed logical inference. They understand the word "alpha," let us suppose, just as we understand the word "and," but they have not yet sorted out which inferences to make using it. From a complex sentence of the form, "A alpha B," for example, they do not infer A, and they do not infer B. Rather, in their benighted condition, they will wait to have as premises both "A alpha B" and A before they infer B. Moreover, they do not even understand from what premises they may infer "A alpha B." They mistakenly suppose that if they have managed to prove B from A as premise (perhaps together with some further assumptions), they can conclude from that to "A alpha B" (depending on those further assumptions alone). Quine's point was that in this kind of situation, we cannot sustain the charge of irrationality. We have to conclude that we have mistranslated "alpha," and that the foreigners are in fact making quite sensible use of a notational variant of "if . . . then . . . ." The meaning of "alpha" dictates what constitutes a rational way of using the term. It is not really possible for someone to have grasped the meaning of a term but to be using it in a quite irrational way. The finding of irrationality can always be traded for a finding of mistranslation. And we should always translate so as to find the subject rational in the use of a term by the lights of the subject's own understanding of the term.

This point does not apply only to logical constants. Suppose someone says to you, "The Statue of Liberty...