Colloquium organized in partnership with the SND laboratory of the Université Paris-Sorbonne on June 8 at the Maison de la recherche and on June 9 at the Collège de France.
With the support of :
Organizers: Alexandre Declos (Collège de France) - Julien Tricard (Sorbonne Université)
The problem of induction, as formulated by Hume in his Treatise on Human Nature, lies at the crossroads of metaphysics and epistemology. On the epistemological level, it consists in finding a justification for inductive inferences: how can nomological and general hypotheses, which relate to unobserved facts, be legitimized on the basis of a finite number of experimental facts? From a metaphysical point of view, the problem is to demonstrate a principle capable of underpinning all our inductions. If we could show that the future will resemble the past, we might find in this truth a sufficient basis to legitimize inductive inferences. But how can we demonstrate this metaphysical principle, if not by induction, and thus by presupposing the validity of inductive inferences?
This is the problem bequeathed to us by tradition since Hume. The aim of this colloquium is to examine this legacy, and to question the possibility and means of providing a metaphysical foundation for inductive inference today. How do the epistemological and metaphysical problems of induction fit together? Is there a need for a metaphysical principle of induction, and if so, how should it be formulated?