| Ruth Millikan and others advocate theories which attempt to naturalize wide mental content (e.g. truth conditions of beliefs) in terms of (teleological) functions, where the latter are in turn based in part on facts concerning past natural selection. While I support basing content on functions which are constituted by facts about the past, I argue that it is a mistake to base content on selection. This dissertation works out an alternative concept of function which is a more appropriate basis for a theory of mental content. In particular, I define a concept of function in terms of (a) comparisons between certain past probabilities, and (b) comparisons between fitnesses of certain ancestors, without regard to facts about selection. This account clearly requires naturalistic concepts of probability and fitness. Furthermore, fitness must not depend on selection. The propensity interpretation of fitness and its elaborations, which define fitness in terms of probability, in particular propensity, provide a framework for a naturalistic concept of fitness which does not depend on selection. But (along with others) I argue that propensity is not a suitable basis for fitness. A large portion of the dissertation is devoted to a sketch of a fully naturalistic concept of higher-level objective probability, a concept which can provide a foundation for probability-based theories of fitness and other higher-level uses of probability. This concept of probability is defined in terms of the structures of mechanisms and in terms of very general facts about certain sets of actual events. This sort of probability has causal connections to observed relative frequency without being equivalent to relative frequency, and is consistent with lower level probabilities with arbitrary values. Although the focus of the dissertation is on objective probability, on the foundations of teleological theories of content, and on the foundations of a theory of teleology suitable for supporting such theories, my discussion has implications for more general issues concerning scientific realism and the foundations of evolutionary biology. |
