Equilibrium Selection in the Stochastic Iterated Prisoner's Dilemma

A recent spate of writings has argued that in a fundamental sense, science is reaching limits. The basic claims of this work is twofold. First, it is argued that for many areas of inquiry, notably physics, chemistry, and biology, science has provided an essentially accurate view of nature. The theories of evolution, relativity, and quantum mechanics are taken as primary examples of scientific success. Further advances on these successful theories about nature will be derivative of these existing broad theories. Second, it is argued that in those areas where scientific inquiry cannot claim great empirical successes, fundamental advances are impossible. Unverifiability may occur either because of the lack of distinct empirical predictions of the theories or because of the nonexistence of feasible experiments to allow discrimination. Cognitive and social sciences are often cited as examples of disciplines doomed to permanent failure. This paper argues that the claims about limits to science are essentially unpersuasive. First, I claim that there is no basis for concluding that a successful science is at the end of fundamental breakthroughs. Second, I argue that the claim that social science, as an example of a supposedly unsuccessful science, is incapable of producing successful models, is unsupported. Here, I focus on economics given my professional training, but the arguments apply more broadly. In short, it appears more plausible that there are fundamental barriers to our ability to identify bounds on science than that we have reached such barriers, so that it is the epistemology of science rather than science per se which may face fundamental limits.