Abstract

I offer an analysis of the concept of scientific innovation. When research is innovated, highly novel and useful elements of investigation begin to spread through a scientific community, resulting from a process which is neither due to blind chance nor to necessity, but to a minimal use of rationality. This, however, leads to tension between two claims: (1) scientific innovation can be explained rationally; (2) no existing account of rationality explains scientific innovation. There are good reasons to maintain (1) and (2), but it is difficult for both claims to be accepted simultaneously by a rational subject. In particular, I argue that neither standard nor bounded theories of rationality can deliver a satisfactory explanation of scientific innovations.