Jan-Willem Romeijn\, University of Groningen
\nResiliency in psychiatric science
What is a good classification scheme for mental disorders? This question is of great importance for clinicians and psychiatric researchers\, but also for philosophers of science. The problem of finding the right classification scheme arguably resembles a problem well-known in philosophical circles\, to wit\, the problem of the reference class. In my talk I propose a solution to this problem that goes back to Brian Skyrms' seminal 1977 paper "Resiliency\, Propensities\, and Causal Necessity". In it he develops the idea that chances are robust degrees of belief. I will elaborate on this idea and argue that it suggests a particular anti-reductionist approach to the problem of psychiatric classification\, in which causal inference and statistical model selection play a key role.