Evidential Support First, Probability Second
When
Please join us this Friday, 11 April, for our next Philosophy Colloquium. As per usual: Maloney Seminar Room, Social Sciences 224, 3-5p. Our guest this week is Shyam Nair, who is an associate professor of philosophy at Arizona State University. Shyam works in ethics, epistemology, and philosophical logic.
Title: Evidential Support First, Probability Second
Abstract: Is E evidence for H? Is E better evidence for H than E' is for H'? These questions have been studied by approaches that make use of probabilities. They (i) provide an account of what Pr is intended to represent and (ii) on this understanding of Pr, claim that E is evidence for H iff Pr(H|E) > Pr(E). Though there are many differences among the approaches, they are all reductive accounts of evidential support: What it is for E to be evidence for H is understood in terms of whatever the theory says Pr represents where Pr itself is not representation of evidential support.
This talk is about how to accept that E is evidence for H iff Pr(H|E) > Pr(H) while rejecting reductivism. The core ideas are these: First, one can directly characterize the notion of evidential support qualitatively and develop a quantitative representation of this qualitative structure. Second, though this representation is not probabilistic, it allows us to define another numerical representation that is probabilistic and together these entail that E is evidence for H iff Pr(H|E) > Pr(H). Third, there are interesting applications for this way of thinking that merit further consideration. The talk primarily focuses on the first and third of these points.