Nous aurons le plaisir d'entendre un exposé d'Orri Stefansson (Collège d'Études Mondiales), intitulé : "How Should We Evaluate Chances?".
Résumé : It is standardly assumed that chances are not carriers of value, but are only instrumentally valuable in virtue of the outcomes with which they are associated. Learning that you have won a lottery, and that your chances of winning were, say, 0.000001, is on this view neither more nor less desirable than just learning that you have won the lottery. I call this view—implicitly assumed, for instance, in expected utility theory, risk analysis and orthodox economics---Chance Neutrality. The aim of this paper is to question the claim that Chance Neutrality is rationally required. I will do so by first showing that given Chance Neutrality, the Principal Principle---according to which a person should align her credence with her beliefs about objective chance---entails Linearity---the idea that the value of a lottery is equal to the sum of the values of the lottery’s prizes discounted by their probabilities. I then argue that the Principal Principle is a requirement of (both practical and theoretical) rationality but that Linearity is not. Hence, Chance Neutrality is not rationally required. I conclude with some remarks on implications for the Ellsberg paradox.