November & December 2014 Lunchtime Abstracts & Details

::: Confronting Philosophical Concepts with Historical Cases
Raphael Scholl, Visiting Fellow, University of Bern
Tuesday, November 4, 2014
12:05 pm, 817R Cathedral of Learning

Abstract:  The “dilemma of case studies” holds that the project of integrated history and philosophy of science cannot succeed. If we start with philosophical concepts and proceed “downward” to historical cases (the first horn of the dilemma), then we must always suspect that the cases were chosen so as to fit our philosophical preconceptions: the obvious dangers are selection bias and gerrymandering. If, however, we start with history of science and proceed "upward" to philosophy (the second horn), then we do not have any obvious warrant for generalizations: proper support for a philosophical concept cannot derive from its applicability to one, two or even several cases.
        The dilemma of case studies points to the need for an explicit methodology for integrated history and philosophy of science. In order to avoid the charge of selection bias, we need sound criteria for why particular case studies are chosen and how they relate to the philosophical concepts under scrutiny. And instead of worrying about “generalizing” from historical cases, we need to formulate robust procedures for dealing with either a match or a mismatch between concepts and cases.
        I will motivate and develop these ideas using concrete instances from works in HPS.

::: A Classical System with Entanglement
David Snoke, University of Pittsburgh, Dept. of Physics
Tuesday, November 11, 2014
12:05 pm, 817R Cathedral of Learning

Abstract:  I will present the details of a fully classical physical system which has a mathematical description with exactly the same equations as that of a quantum system with entanglement, including violation of an analogous Bell inequality. The difference from a quantum system comes only in the non-mathematical part of the description, in the interpretation of the wave function. While this does not strike at the roots of quantum foundations, it does tell us that some glib formulations about what is "uniquely quantum" need to be corrected.

::: Should the Debate Over Scientific Realism Go Local?
Leah Henderson, Carnegie Mellon University, Dept. of Philosophy
Tuesday, November 18, 2014
12:05 pm, 817R Cathedral of Learning

Abstract:  The No Miracles Argument (NMA) and the Pessimistic Induction (PI) are often regarded as pillars of the scientific realism debate. However recently there has been an influential attempt to reorient the debate away from these arguments and to ‘go local’. Localizers argue that the proper way to conduct the realism debate is to focus on simply looking carefully at the empirical evidence that supports each scientific theory on a case-by-case basis.
        The local challenge suggests that we need to get a clearer picture of the status of the NMA and PI. Both the NMA and the PI are instances of what might be called ‘reliability-based’ arguments. They look to the history of science to get a general indication of how reliable scientific method is and then use that reliability information to modulate our confidence in current well confirmed theories. In this talk, I will examine the question of how exactly these arguments work and how they are supposed to fit in with the evaluation of theories on the basis of ordinary scientific evidence.

::: Evolving to Generalize: Trading Speed for Precision
Cailin O’Connor, University of California, Irvine
Dept. of Logic and Philosophy of Science
Tuesday, December 2, 2014
12:05 pm, 817R Cathedral of Learning

Abstract:  Biologists and philosophers of biology have generally agreed that learning rules that do not lead to evolutionarily stable strategies (ESSes) will not be evolutionarily successful.  This claim, however, stands at odds with the fact that learning generalization---which cannot lead to ESSes when modeled in games---is ubiquitous.  I show that generalization, despite leading to suboptimal behavior, can allow actors to learn quickly.  I observe that previous analyses of the evolution of learning problematically ignored the short-term success of learning rules. If one drops this assumption it can be shown that learning generalization will evolve in many cases.

::: Bayesian Cognitive Science and the Nativism vs. Empiricism Controversy
Matteo Colombo, Tilburg University
Center for Logic and Philosophy of Science
Friday, December 5, 2015
12:05 pm, 817R Cathedral of Learning

Abstract:  Suppose that brains are Bayesian machines. What follows about the dichotomy “nativism vs. empiricism” in the cognitive sciences? To address this question, I firstly distinguish between different formulations of the idea that brains are Bayesian machines. Then, I clarify what commitments each formulation makes about the shape and origin of priors. Finally, I assess the relevance of such commitments, and more generally of the Bayesian approach, for the nativism vs. empiricism debate.