Abstract

It has become commonplace in the theory of concepts to distinguish between questions about the structure and questions about the ontology of concepts. Structural questions concern the way concepts are composed of, or otherwise related to, other concepts (or non-conceptual constituents), while ontological questions concern the metaphysical nature of concepts: how concepts exist (if they exist); what kind of entities they are. A tacit assumption in discussions about the structure and ontology of concepts seems to be that structural and ontological questions can be treated more or less independently of each other. Correspondingly, it might be assumed that one can rather arbitrarily combine structural analyses of concepts with any hypothesis about their ontological make-up. In this paper, we would like to question this assumption with regard to a particular recent development in the structural analysis of concepts, the so-called frame theory. As a matter of fact, most frame theorists adopt a subjectivist theory of concepts. Yet, subjectivism attracts philosophical criticisms, which originate mainly in the writings of the later Ludwig Wittgenstein. While Wittgenstein appreciated the cognitive dimension of concepts, he came to reject the subjectivist’s causal analysis of mental explanation and their ensuing reification of concepts as causally operative mental particulars. Instead, Wittgenstein suggested that concepts are socially imparted norms of classification and inference – i. e., abstractions from collective word-usages and other practices associated with categorizing and higher-order thinking, which may serve as instructions for individual agents on how to discriminate and interact with things of different kinds. Applying the Wittgensteinian critique of subjectivism to the frame analysis, the we not only find that the frame approach quite nicely dovetails with what the pragmatist has to say about concepts and categorization, insofar as frames seem particularly well-suited to depict the defeasible and context-dependent default character of our usual criteria for semantical understanding. Indeed, pragmatism, with its emphasis on the holistic, indeterminate and normative nature of our conceptual activities, seems able to explain why concepts are frames. This shows that there exist at least some interconnections between structural and ontological analyses, which must not be ignored, especially in interdisciplinary investigations on concepts. Even if frame theory is not strictly incompatible with a subjectivist ontology, it does license an abductive inference to a pragmatist ontology of concepts, as frames precisely formalize the insights into our actual conceptual practice that led Wittgenstein to develop his pragmatist view.