The role of categories of knowledge, or disciplines, in science has not previously been explored in scientonomy. While disciplinary communities devoted to the production of knowledge are a modern phenomenon, the practice of dividing knowledge into categories is a universal feature of science. Although at any moment of time, many questions and theories can be part of a given discipline, not all of these are essential to the discipline. We show that two components are essential to a discipline: the discipline’s core questions as well as the discipline’s delineating theory, a second-order theory that identifies these questions as essential to the discipline. If the questions of one discipline are a proper subset of the questions of another discipline, the former discipline is a subdiscipline of the latter. Since a discipline is a complex entity consisting of questions and a theory, epistemic agents can take epistemic stances towards disciplines. A discipline is said to be accepted if its core questions and its delineating theory are all accepted. To illustrate the applicability of these new concepts, the transition from physical to biological anthropology is discussed.