Abstract
Carnap’s thought not only played a pivotal role for the development of formal semantics and modern philosophy of science, but also engendered the profound methodological reorientation that distinguishes analytical from traditional ontology. Historically and systematically, Carnap’s formal approach to category theory is the primary source of influence on the three research programs that have given analytical ontology its distinctive profile: the design of constructional systems, the investigation of the expressive power of first order theories, and the meta-linguistic reduction of abstract terms. Each of these research programs takes off from Carnap’s deflation of traditional category theory as set out in the Aufbau and modified in Syntax. None of these programs, however, follow Carnap in his claim that category theory, even in the attenuated sense of a derivational systematization of cognitive contents, may ultimately be separated from metaphysics or remain metaphysically “neutral.” Carnap’s “neutralism thesis” is taken to imply the stronger “elimination thesis,” his explicit anti-metaphysical pronouncements that stand and fall with semantic verificationism. However, I will argue, first, that the neutralism thesis is historically and systematically independent of the elimination thesis. Second, unlike the elimination thesis, the neutrality claim and Carnap’s attempt to cash it out in the Aufbau is important for the methodology of contemporary ontology, at least where it follows constructivist predilections. Third, Carnap’s neutrality claim in the Aufbau is supported by an implicit conception of reality as invariance-structure. Yet, just as the structuralization of referential domains may be able to warrant neutrality, it threatens the explanatory aspirations of constitution theory. This tension between neutrality and explanatory power sheds new light on the notion of “foundedness,” Carnap’s curious restriction on the interpretation of the system’s basic relation that has struck some as outright incoherent.