## Apr 5, 2015

### Our Next Visiting Speaker April 10

Dr. Marshall Abrams, Associate Professor of Philosophy at the University of Alabama, Birmingham, will our next visiting speaker. He will be presenting a talk titled “Causal and non-causal probabilities in applied population genetics” on Friday, April 10 from 1:30-3:30 in McCain Hall, Room 185. His abstract is as follows:

I argue that scientific practice often depends on an implicit distinction between what I call causal probabilities and probabilities in other roles. Roughly, when we can manipulate frequencies by manipulating the characteristics of a chance setup that determines probabilities' numeric values, the probabilities can be considered causal probabilities. Philosophers often assume that only propensities and closely related kinds of probability play this kind of role. Even putting aside standard objections to propensities, it's doubtful that propensities can play all of the roles required by causal probability. Moreover, other causal probability interpretations have been proposed in recent years by Rosenthal, Strevens, and myself. In this talk I focus on elaborating and clarifying the concept of causal probability apart from any particular interpretation of probability, though. I use examples from applied population genetics to illustrate distinctions between causal probability and other, non-causal roles for probability. My discussion addresses roles played by probabilities in computer simulations and mathematical models as well as in the systems modeled. I also discuss evidential criteria for the existence of causal probability. Finally, I show how trying to make the concept of causal probability precise by generalizing Woodward's analysis of manipulation clarifies a difficulty in spelling out the sense in which frequencies can be affected by manipulating chance setups. This problem, though difficult, is rooted in assumptions central to scientific practice, and I suggest strategies for addressing it.

We hope to see you there!