Abstract

Reductive Humeanism with regard to laws and chances, the view that law and chance claims are reducible to claims about Humean states of affairs, is a highly problematic doctrine. Even its most sophisticated contemporary version, espoused by David Lewis, leads to a litany of inadequately explained conflicts with our intuitions regarding laws, chances, and counterfactuals. The non-Humean alternative permits analyses that are adequate to the intuitive data. However, such analyses are insufficiently explanatory to be satisfactory. ;This dissertation motivates, elaborates, and defends a third type of view, a non-reductive Humean projectivism on which law and chance claims are analysed in terms of an expressivist semantics. The benefit of this strategy is that it provides detailed explanations of the intuitions that are problematic for reductive Humeanism without relying on murky non-Humean explanations. However, there are numerous prima facie worries. Many philosophers surmise that the non-factual character of such an analysis must render law claims subjective or mind-dependent, or indeed, must lead to the denial of all law claims and, plausibly, all claims with nomic commitments of any kind. Further, there are technical worries, as represented by the Frege-Geach problem. The dissertation is responsive to such concerns, including the following features: A novel motivation of Humean projectivism is provided with reference to the explanatory goals of science. Detailed explanations of the intuitions that confound reductive Humeanism are provided. Notably, the analysis of chance laws is not susceptible to the problem of undermining futures, and further, the Principal Principle is both vindicated and justified. The Frege-Geach problem is solved using semantical tools of the kind developed by Allan Gibbard and Simon Blackburn.