Functionalism, Psychology, and the Philosophy of Mind

摘要

A great deal has been claimed for functionalism; in my view, considerably more than it can or should try to deliver. To overestimate its scope renders it vulnerable to a number of objections which cannot easily or plausibly be evaded,1 so that it is easy to conclude by undere stimating its power and potential in its true and proper domain. In this paper I sketch my own version of the scope and limits of functionalism, arguing both for the promise of its scope and the extent of its limitations. First, though, some ground-clearing: a few preliminary remarks which I hope will be recognized as obvious and trivial, (i) To describe a theory as 'functionalist' is elliptical shorthand; 'being a function' is as such an incomplete predicate. Functions are functions of things: cutting is the function of a knife, herding sheep that of a sheepdog, and seeing ( vide Aristotle) is the job of the optical apparatus. Functions are thus functions of structures, and often of complex structures at that. Further, some structures physically cannot perform certain tasks meringues can't carve joints. It often therefore proves unwise in psychology to concentrate exclusively upon the functional design of the systems examined, because the applicability of many predicates will often depend, as Gunderson puts it,2 not so much on how the robot is programmed but rather upon how the program is roboted. To keep this important, if obvious, point in mind I shall throughout the rest of the paper talk not of 'functionalism' but of 'S-F theory', where 'S-F' stands for 'structural-functional'. This point has implications to which we shall return.