What makes a good metaphor in science?

Abstract

Metaphors have mainly been discussed and studied as rhetorical devices in human communication and in literary studies. In philosophy and science, metaphors have been considered important as heuristic tools, for aesthetic embellishment, or in helping communicate complex ideas more simply; but as having no epistemic force. Philosophers and scientists are cautioned to avoid metaphors as explanatory resources. However, I present several examples where metaphors have played a significant role in the advancement of scientific knowledge and not as mere heuristic aids or decoration. This tension is worth addressing. I use recent work in the cognitive sciences to argue that some metaphors play a significant role organizing and coordinating scientific practices in relation to goals of inquiry, thereby promoting epistemic values (scientific explanations and more generally, understanding) in the process. Such metaphors therefore perform a key cognitive role. Good metaphors in the epistemic sense, then, are those that play such a cognitive role.