Special Seminar sponsored by the Suppes Center for the History and Philosophy of Science Residual Phenomena as Evidence

A six part lecture series by Professor George E. Smith, Emeritus, Tufts University and Professor Teru Miyake visiting at the Radcliffe Institute, Harvard University

General Abstract: Residual phenomena involve systematic discrepancies between theoretical calculations and observation. Two classic examples from the 19th century are the 43 arc seconds per century discrepancy between the rate of precession of the perihelion of Mercury calculated according to Newtonian theory and the observed rate; and the systematic departures of each individual real gas from the ideal gas law – revealed one-by-one by Regnault during the 1840s and two decades later singled out by Clausius and Maxwell as a potential source of evidence on how molecules of different gases differ from one another. In these and many similar cases the discrepancy led to prolonged research trying to identify the cause of the discrepancy. Contrast this with the individual departures from a least-squares curve-fit obtained in linear regression analysis, which have no prima facie claim to resulting from some specific cause rather than from just the choice between linear and other options for the baseline curve. The question we shall be exploring in the seminar concerns the conditions under which residual phenomena do and do not have claim to being physically meaningful and thus for identifying the cause of the residual phenomenon in question.

The seminar will meet on Tuesdays and Thursdays at 5:15pm, May 24th - June 14th, 2018.