Categorial Inference and Convert Realism: Structuring Ontology Via Nomological Axiomatics

Following Sellars, this paper shows how ontological theorization becomes diachronically coordinated with and constrained by empirical science, and how the formal-quantitative kernel of scientific theories corresponds to qualitative-conceptual determinations at the structural level. To do this, we follow Johanna Seibt’s characterization of ontology as a theory of categorial inference, in which category construction and recategorization are seen relative to and modeled by the most general inferences and descriptive concepts of natural and scientific languages. This allows us to trace the preservation of inferential semantic structure across ontological theories in relation to model languages. We then provide some provisional indications to coordinate Seibt’s account with Jay Rosenberg’s “convergent realist” assessment of scientific theories in terms of Cauchy convergences and “counterpart modeling,” which defines epistemological conditions for articulating the preservation of formal structure in theories toward a limit or ideal point of enquiry.