ABSTRACT
Scientific theories, models, experiments, and the like are often subject to aesthetic assessment. Is there any good epistemic reason to prefer an elegant experiment to an inelegant one, a beautiful theory to an ugly one, a streamlined model to one that seems more like a Rube Goldberg machine? What are we focusing on when we make such assessments? I will argue that aesthetic factors are integral to good science. They are not mere instruments. Nor is their utility primarily practical. But there is no reason to think that they themselves are truth-conducive. Rather, they figure in what it takes to make a scientific construct—theory, model, experiment or whatever—acceptable. They do so, I suggest, because they are gatekeepers on acceptability: they play a regulative role.
