Abstract The generalisability of results from large-scale studies of science has often relied on the extent to which the studies can adequately control for differences across fields. However, disciplinary differences are not well understood. In this paper, we examine the extent to which disciplinary differences can be detected from inconsistencies in the use of a single word: serendipity. Drawing on Wittgenstein and an ordinary language analysis, we examined how and where authors use the term serendipity in astronomy, psychology, and psychopharmacology. Our results indicate normative differences not only in researchers' attitudes as to what constitutes “sound science” but also substantive differences in their empirical, theoretical, and methodological emphases. This contrasts with policy discussions that treat science largely as a singular entity, in which we might expect notions like serendipity to carry the same meaning across all fields. For science policy, our work suggests that the sciences are different, and the differences matter.
Scholte et al. (Mon,) studied this question.