University of Minnesota
School of Physics & Astronomy

Physics and Astronomy Calendar

Friday, March 25th 2016
Speaker: Bruce Glymour, Department of Philosophy, Kansas State University
Subject: Evolutionary Biology and Inertia in Theory Change: A Preliminary Indictment of Explanatory Commitments
Refreshments served at 3:15 p.m.

Kuhn famously argued that scientific paradigms are immensely resilient to empirical evidence against their core theories. I offer a tentative and contentious diagnosis of one such case in evolutionary biology. Post-Synthesis evolutionary theory has been characterized by three nominally distinct theories of natural selection—classical population genetics and its extensions, quantitative genetics, and the halfway house occupied by models employing variants of the Price equation. Notwithstanding their important differences, all share the idea that selection is to be understood in terms of differences among types in one or another measure (generally called fitness) defined as some function or partial function of a probability density over reproductive success. Models implementing that idea immediately confront some intractable problems that limit their explanatory and predictive power. There are alternative conceptions of selection which do not face exactly those problems, and the mathematical tools requisite to them were available either before or roughly contemporaneously with the Synthesis itself. While more orthodox models generally employ the analysis of variance or co-variance in both discovery and explanatory contexts, the alternative models rely on regression and path analysis, and in so doing generate importantly different kinds of explanation and are vulnerable to a different suite of errors. In this paper I delineate (some of) the problems plaguing traditional models, and explore the idea that their continued dominance in both evolutionary population biology and philosophy biology is owed in large measure to a prior commitment to the explanatory importance of one kind of non-causal, statistical explanation.

The weekly calendar is also available via subscription to the physics-announce mailing list, and by RSS feed.