Comments on Arrow and on Lucas

Such a number of interesting and provocative things are put forward in Lucas and in Arrow (in this issue) that I would like to be able to dissect them in detail. For example, Lucas states that "the problem of controlling inflation has been 'successfully solved' in a scientific sense" (p. S405). I am tempted to explore the limits of the set of problems that might be said to be scientifically "solved" in the Lucas sense, but I fear that this would take up a good deal of space. Arrow deplores the use of homogeneity assumptions as a device for adding content to otherwise inconclusive economic theories, but presumably he does not believe that the Sonnenschien-Mantel-Debreu theorem is the ultimate conclusion regarding the useful content of the propositions of general equilibrium theory. Some restrictions of intermediate strength are presumably called for, and I would like to make the case that strong empirical generalizations about the nature of human actors-such as those describing the biological requirements for survival-are available and relevant in this connection. This too would take some space to develop in detail, though not as much as the first subject. Given the constraints, I must choose a focus. I will focus on what I take to be the main message that Lucas has put forward regarding the relationship between "adaptive behavior" and the rationality assumptions employed in standard economic theory. This message is recognizable as the intellectual descendant of similar discussions by Friedman and Machlup in the 1940s and 1950s, and the ideas involved remain extremely influential in the discipline today. I will present and comment on my own statement of this position, which I (following Blaug [1980]) will call the "Classic Defense" of the rationality-as-