CARNEGIE-MELLON UNIV PITTSBURGH PA PITTSBURGH United States
Pagination or Media Count:
Humes problem of induction has important philosophical consequences for many scientific endeavors. Here I implement it in cosmology, a task that, to my knowledge, has not been undertaken in the literature. I argue that the problem is particularly acute in the multiverse setting, and that the problem may still lurk beneath the surface for John Nortons material theory of induction, despite Nortons claims to the contrary. In Section 2, I motivate the discussion by introducing the Cosmological Principle, which cosmologists use routinely for induction. Section 3 contains a careful reconstruction of Humes problem. This reconstruction enables us to transform the temporal version of the problem as Hume originally presented it into the spatial version of the problem as it appears in cosmology. The conversion occurs in Section 4. In Section 5, I argue that the spatial version of the problem reveals that multiverse cosmological models are vulnerable to the problem in a novel way. In Section 6, I implement the disentangled spatial and temporal flavors of induction to demonstrate that the material theory of induction faces a dilemma either it is rendered inert, or it is susceptible to Humes problem. Finally, in Section 6, I address the critic who maintains that Humes problem is old news in philosophy of science and that all sciences must, of necessity, take on some risk of induction.