Marshall Abrams (University of Alabama) will be giving his talk on Friday, October 23, 2015 in BUCH A202 at 3pm.
Probability in mechanistic explanation
The widespread use of mechanism concepts in scientific research has been the focus of much work in recent philosophy of science. It’s become clear that the widely-cited account of mechanistic explanation offered by Machamer, Darden, and Craver, and framed in terms of organized activities of entities within mechanisms, should be amended to allow activities to occur probabilistically. However, there’s been little discussion of the nature of the probabilities involved. Interpretations of probability (e.g. Bayesian, frequentist, propensity, mechanistic probability) specify ways that conditions in the world count as probabilities. Which interpretations of probability are consistent with the roles of mechanisms in science? I argue that relevant interpretations of probability must define what I call “causal probabilities”, in which manipulation of properties that realize probabilities would also manipulate frequencies of outcomes. In addition, I argue that some activities in mechanisms may involve only “imprecise probabilities” defined by sets of probability distributions. The role of activities in mechanisms allows us to simplify and broaden arguments by Terrence Fine and his collaborators that the concept of imprecise probability found in some extended Bayesian theories can also extend objective probability concepts. My presentation won’t take for granted familiarity with mathematical probability theory, philosophy of probability, or philosophical discussions of mechanisms.
This event free and open to the public, faculty, staff and students