Thomas Augustin - Reflections on the Foundations of Probability and Statistics: Essays in Honor of Teddy Seidenfeld

This series deals with practical and social philosophy and also foundational issues in philosophy of science in general that rely on methods broadly based on rational choice. The emphasis in the Series A is on well-argued, thoroughly analytical and philosophical rather than advanced mathematical treatments that use methods from decision theory, game theory and social choice theory. Particular attention is paid to work in practical philosophy broadly conceived, the theory of rationality, issues in collective intentionality, and philosophy of science, especially interdisciplinary approaches to social sciences and economics.

This Springer imprint is published by the registered company Springer Nature Switzerland AG

The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

About 5 years ago, we noticed that Teddy Seidenfeld, although as active and cheerful as always, was approaching his 70th birthday(!). We then decided to put together a Festschrift to celebrate Teddys many contributions to philosophy, statistics, probability, game theory, and related fields. This volume covers a subset of those themes; going through all topics that have received Teddys attention would require a series of books. We hope the reader will continue this celebration by reading the many papers written by Teddy himself with his many collaborators.

Teddy was born in New York City in May 1948. He took a major in mathematics and philosophy at the University of Rochester and then a PhD in philosophy at Columbia University. He started his academic career as an assistant professor at the University of Pittsburgh, moved to Washington University in St Louis, where he was an associate professor, then to Carnegie Mellon University, where he is now the H. A. Simon Professor of Philosophy and Statistics. In his distinguished career at Carnegie Mellon, Teddy has worked as head of the Department of Philosophy, as graduate director of the logic and computation program in philosophy, as co-director of the machine learning graduate program, and as chairman of the Faculty Senate.

Rather than write a dry review of Teddys career, we decided to interview Teddy to offer his own thoughts on past events. We are most grateful to Teddy for providing witty and detailed answers to our many questions in fact, we had to limit ourselves to a small fraction of our questions to keep the interview within a reasonable size. The reader will find the interview in the first chapter of this book, ranging from Teddys experiences as a baseball fan and chess player to his formative years as an undergraduate and graduate student. The reader will also find insightful answers on Teddys collaboration with Henry Kyburg and Isaac Levi, his advisers at Rochester and Columbia, respectively, on Teddys long-standing collaboration with his colleagues Jay Kadane and Mark Schervish at Carnegie Mellon, as well as commentaries on Teddys perspective on research.

Teddy has written extensively on the foundations of probability and statistics, individual and group decision theory, and on the interface between philosophy and statistics. His contributions range from deep mathematical results on finitely additive or infinitesimal probability to general consequences of coherence and consensus in statistical reasoning and decision-making. His perspective is always based on solid mathematical arguments. Indeed, Teddy seems to have a particular interest in examining the limits of proposed theories, for instance, by visiting cases where measure-theoretic probability theory leads to unexpected and sometimes unpleasant conclusions. Thus Teddy has produced, with his collaborators, a collection of gems on finitely additive probability, for instance, looking at the consequences of finite (not -)additivity to conglomerability, sequential decision theory, and convergence theorems. Teddys analysis of decision theory, and of sequential decision theory in particular, has led to a refined understanding of connections between acts and beliefs, normal and extensive forms, and unbounded utilities. And Teddy has often returned to questions related to convergence and consensus, carefully mapping cases that go beyond the usual convergence-to-certainty results at the core of probability theory. Consensus, of course, does not arise only through convergence; Teddy has also examined the shared preferences of decisions makers and how those preferences may, and may not, coalesce into common preferences.