ABSTRACT
Idealizations, the need for conventions in empirical inquiry, the gap between theory and evidence are all familiar features of scientific investigations into both the natural and the human realm. Commonsense idioms coexist comfortably with scientific ones when scientific theories are applied to phenomena individuated by concepts in ordinary language. Sententialists would retort that, just as a computer can instantiate a variety of languages which differ in their level of abstractness, so a human being can truly be regarded as, among other things, an instantiation of a theory which postulates operations on sentences. If cognitive psychology retains enough of the intentional structure of folk psychology to permit the core notion of action directed by an agent in virtue of that agent's beliefs, then the deepest threat of scientific psychology to moral responsibility will have been repelled.
