We found a match
Your institution may have rights to this item. Sign in to continue.
- Title
RATIONAL CHOICE: A SURVEY OF CONTRIBUTIONS FROM ECONOMICS AND PHILOSOPHY.
- Authors
Sugden, Robert
- Abstract
I have focused on three features of the received theory of rational choice. First, I have examined the philosophical foundations of expected utility theory, concentrating on the version that commands most support among modern theorists-that of Savage. I have argued that Savage's axioms are much stronger than can be justified merely by an appeal to an instrumental conception of rationality. There is no adequate justification for the requirement that preferences are complete. Nor is there for the transitivity and sure-thing axioms, once it is recognised how Savage's axioms impose restrictions on what can count as a consequence. Second, I have looked at the assumption of `common knowledge of rationality', which underlies game theory. I have argued that this assumption is not sufficient to lead rational players to Nash equilibria. More controversially, I have argued that there can be games for which this assumption is incoherent. The heart of the problem is that Savage's theory of rational choice effectively requires each individual to be able to assign subjective probabilities to all events; what it is rational to do then depends on those probabilities. In game theory, the possible acts of opponents are treated as if they were Savage events, so that what it is rational for A to do depends on the probabilities he assigns to B's actions. But at the same time, A is assumed to know that B is rational in the same sense, so that the probabilities A assigns to B's strategies depend on what it is rational for A to do. A related kind of circularity occurs in some sequential games, such as Centipede. Here the problem is that A cannot work out what it is rational for him to do unless he can predict the consequences of playing each of his possible first moves. He cannot do this without first working out whether, on observing a particular move by A, B will infer that A is rational or irrational; but this depends on what it is rational for A to do. These circularities make...
- Subjects
RATIONAL choice theory; ECONOMICS; UTILITY theory; GAME theory; ECONOMIC demand; SOCIAL choice
- Publication
Economic Journal, 1991, Vol 101, Issue 407, p751
- ISSN
0013-0133
- Publication type
Article
- DOI
10.2307/2233854