Séminaire OLOFOS - Clash between the manifest and the scientific image of the world

The unity of evidential reasoning through the prism of credibility

Abstract : The claim that evidential reasoning obeys essentially the same norms and corresponds to roughly the same psychological reality in the various contexts where it occurs is popular nowadays. It is one central posit of Bayesianism, which is probably the dominant theory of scientific reasoning today. It is also made by David Schum in his ambitious Evidential Foundations of Probabilistic Reasoning (1994). In this book, Schum seeks to offer a general theory of evidential reasoning. The aim of my talk is to put this unity claim to the test and assess the prospects of such a general theory. To do so, I will focus on credibility. Indeed, while assessing the credibility of evidence seems to be an essential part of evidential reasoning in ordinary contexts (and for that matter in judicial contexts, which are Schum’s central concern), scientists do not seem to worry much about this problem. I will support that, although it is often overlooked and not easily visible, assessing credibility is an essential part of scientific reasoning.