Seminar on History and Philosophy of Science
In statistical physics and philosophy of physics, it has been standard to follow Ludwig Boltzmann and impose a Past Hypothesis on the boundary of the physical space-time. According to the Past Hypothesis, the "initial" state of the universe is in a very orderly (low-entropy) state. In this talk, I would like to explore an alternative hypothesis, motivated by the (in)famous Principle of Indifference. I will argue that both theories, on pain of empirical inadequacy, require certain objective self-locating prior distributions. It follows from reasonable premises that the two theories are in fact empirically adequate to the same extent. This is a surprising result, for it leads to deep and puzzling consequences for the epistemic justification for our beliefs about the past (including the prosaic ones that we and our surroundings were "younger"' in the past). We will then think about what this might mean for the general issues in philosophy of science about theory choice and pragmatic considerations.