Abstract

Certain fundamental philosophical disputes, in contrast to disputes in the empirical sciences, are characterized by the persistence of disagreement. This has led some to endorse conventionalism, the view that the 'facts of the matter' partly depend on our conventions and that disagreements persist because both sides to the dispute employ different conventions. What does it mean to say that the facts of the matter partly depend on conventions? My thesis is concerned with this question. It has four parts. ;Part I examines how some matters of fact may depend on convention. I argue that while versions of conventionalism which can be construed in terms of one of the familiar dependence-relations are intuitively plausible, most interesting versions of conventionalism cannot be so construed. To maintain the claim that some range of facts depends on convention, conventionalists need to explain how the features they take to be conventionally determined systematically covary with conventions. ;Part II provides the formal tools to model conventionalist dependence-relations, tools that respect the methodological assumptions of conventionalists and reflect the logic of conventionalist discourse. The framework developed is also useful for perspicuously formulating other philosophical accounts that take some aspect of reality to depend on human practices, such as neo-Kantian, projectivist and response-dependence accounts. ;Part III investigates how to make philosophical sense of the dependence-relations invoked by conventionalists. I critically examine several conventionalist accounts in the literature, and, employing the tools developed in part II, I propose various explications of how a range of facts may depend on convention. ;Part IV classifies conventionalist accounts according to what kind of dependence-relation they invoke and critically discusses the interest and plausibility of ontological conventionalism