Abstract :
[en] Increasingly, scientists move across geographical borders and affiliate with multiple organizations; almost all collaborate with other researchers. However, despite the massive expansion and relative opening of science, disciplinary boundary-crossing remains challenging. Although interdisciplinarity is often favored rhetorically, bibliometric research finds relatively few cases of meaningful, sustained dialogue between disciplines. To understand why, this study explores such disciplinary boundary-crossing through borrowed concepts in the relevant case of entropy-disorder or uncertainty-at the intersection of economics and physics: "econophysics". Born and raised in physics, entropy has become an increasingly important concept within economics. To uncover its half-century career of crossing boundaries, we build entropy's semantic space from abstract embeddings, while topical clusters are identified via Natural Language Processing (NLP). Scientometric results show blurred boundaries between economics and physics in the semantic space, along with distinct boundaries between topical clusters. A journal-topic mismatch provides fresh empirical evidence of the long-hypothesized institutionalization of econophysics-in the shadow of physics. Simultaneously, the temporal analysis across a half-century uncovers the emergence of an economics camp: entropy has not been completely boxed in by physics. Contributing to the literature on the institutionalization of disciplines and on borrowed concepts, this case study advances the rare yet important semantic dimension of disciplinary boundary-crossing.
Scopus citations®
without self-citations
0