Aarhus University Seal / Aarhus Universitets segl

How to Make sense of history

The status of predictions in the appraisal of theories

Oplysninger om arrangementet


Mandag 28. november 2011,  kl. 14:15 - 16:00


Department of Science Studies, Building 1110, room 214

Abstract: Many of use share the intuition that a theory that successfully predicts a certain piece of evidence should receive more confirmation than a theory that merely accommodates it. Despite the strength of this intuition, this intuition is at odds with some very important historical examples. The philosopher Worrall and the historian Brush have argued independently that some of the prima facie most impressive successful predictions in the history of science in fact do not support our intuition. These cases concern Fresnel’s successful prediction of the ‘white spot’, Einstein’s successful prediction of light bending around strong gravitational fields, and Mendeleev’s successful prediction of new chemical elements. Some of the recent debates have focused on the latter prediction. Whereas some see temporal predictivism—the view the abovementioned intuition has come to be associated with—undermined by the historical facts of the appraisal of Mendeleev’s periodic table, others believe that this is not so. In this talk I shall revisit and assess the arguments that have been made on either side. I shall side with the temporal predictivism camp, but for different reasons than the ones usually mentioned (which I think are the wrong reasons). However this is not where we should stop. In order to reconcile Mendeleev’s case with other studies on predictions (mentioned above), I want to outline a view that I call local-symptomatic predictivism. According to this view, predictions are of import to the scientific community only insofar as they serve as indicators of another theoretical property being genuine, which might otherwise be not assessable. I shall argue that the property in question, at least in the case of Mendeleev, pertained to the coherence of Mendeleev’s table. The predictivism that I defend is local because not in all contexts are predictions in their role as indicators of a particular theoretical property required in order for the value of a theory to be assessed.

Speaker: Samuel Schindler holds a B.Sc. in Cognitive Science (Osnabrück) and an M.A. in History and Philosophy of Science from the University of Leeds. At the University of Leeds he also earned his Ph.D. in Philosophy in 2008. After a short research visit at the Department of Philosophy at the University of Stockholm, and a temporary lectureship at the University of Birmingham (UK), he was a research fellow at the University of Konstanz from 2009-11, and visited the Center for the Philosophy of Science at the University of Pittsburgh in 2010. He won one of the prestigious Emmy-Noether research grants from the German Research Foundation (DFG), but instead accepted the appointment as Associate Professor from the Center for Science Studies at the University of Aarhus.  

Samuel Schindler’s research interests include classical methodological questions such as theory-appraisal and theory-choice under the consideration of historical case studies, the philosophy of experiment, and the nature of scientific explanation. His writings have been published in the The British Journal for the Philosophy of Science, Synthese, and Studies in History and Philosophy of Science.