Einstein and the problem of confirmation by previously known evidence: A comment on Michel Janssen and Jürgen Renn's "Einstein and the Perihelion Motion of Mercury"

Abstract

In this paper, I comment on a recent paper by Michel Janssen and Jürgen Renn. In his published paper of November 18, 1915, Einstein presented a solution to the problem of the perihelion motion of Mercury and obtained the correct result of 43 seconds of arc per century. Before 1915, Einstein had established non-covariant field equations (the Entwurf field equations). But in 1915, he changed his mind and dropped these equations and was led to general covariance. In a manuscript written in 1913, Einstein and his best friend Michele Besso tried to solve the Entwurf field equations to attain the perihelion advance of Mercury. The end result arrived at by Einstein was 1821 seconds of arc per century. To make a long story short, the field of a static Sun produced an advance of the perihelion of Mercury of 18 seconds of arc per century. Einstein kept quiet about this result and continued to work on the Entwurf theory. Janssen and Renn ask: If we follow the prescriptions of Karl Popper, why did Einstein not accept that his Entwurf theory had been falsified? They further ask: Given the undeniable importance of Besso's earlier calculations, why did Einstein not invite Besso as a co-author of his November 1915 paper on the perihelion motion of Mercury? Besso did not even get an acknowledgment in Einstein's paper. After a brief review of Einstein's and Besso's early attempts at calculating the advance of the perihelion of Mercury, I discuss falsification and confirmation by previously known evidence and then propose an answer to the above two questions.