The Neoclassical Maximization Hypothesis: Comment

In a recent paper in this Review, Lawrence Boland argues it is futile to criticize the neoclassical maximization hypothesis. His argument is summarized as follows. 1) There are only two types of criticism of any behavioral once one has established its logical validity. One can argue against the possibility of the hypothesized behavior or one can argue against the empirical truth of the premise of the hypothesis (p. 1031). Boland hopes to show that neither of the two forms of criticism works against the maximization hypothesis. 2) One cannot prove the logical impossibility of the maximization hypothesis. Attempts to do so fail for one of two reasons: either one must assume that inductive proofs are necessary for true knowledge, or one must assume that true knowledge is necessary for maximization (see pp. 1032-33). Boland argues that neither assumption can be maintained. 3) Regarding empirical criticism, methodological problems of empirical refutations of economic theories are widely accepted (p. 1033). 4) Even were this not true, the logical form of the maximization (an statement) makes it neither verifiable nor refutable. [E]ven if one could prove that a consumer is not maximizing utility... this would not constitute a refutation of the neoclassical hypothesis (p. 1034), because one can always claim that the consumer was maximizing something, that is, something other than utility. 5) Finally, this does not imply that the is necessarily a tautology. The is best considered an untestable metaphysical statement which is an essential component of the neoclassical research program. Points 3 and 5 above attributed to Boland are acceptable as stated. They are related points: the empirical problems mentioned in point 3 are sufficient to establish the untestability of the maximization hypothesis, which supports the conclusion in point 5. The purpose of this comment is to criticize points 1, 2, and 4. Specifically, it is shown that though there are only two forms of logically compelling criticism, many other forms of criticism that are useful exist; and that given a broader interpretation of the term criticism, Boland's remarks about proofs of logical impossibility, though correct, miss the point of what critics of the maximization tried to do. An alternative to Boland's characterization of the maximization as an all-and-some statement, one more consistent with well-known developments in philosophy of science, is also presented. In the final section, it is shown that Boland's approach of necessity neglects some widely known and relevant developments in the methodological literature.