Abstract

Abstract. My starting point is Hume’s sceptical doubt as to the necessity of causal connection. This doubt consists in challenging the metaphysician to found rationally the necessity of the causal relation—that is to say, to prove that it is necessary that the same effects should always follow from the same causes. According to Hume, this challenge cannot be met, because the only instances that could possibly rationally found causal necessity are experience and logical necessity. Now, I can certainly affirm on the basis of experience that causality has never failed in the past; but that does not allow me to prove that the same will apply in the future. As for logic, it prescribes only noncontradiction; and there is nothing contradictory in conceiving that objects might behave differently tomorrow than they did today, under exactly the same conditions. Hume’s conclusion is that our adhesion to causality proceeds from a vital habit, an accustomedness that persuades us that the causal sequences we have observed in the past will be repeated identically in the future. This conclusion is sceptical in the sense that it disqualifies the metaphysical claim that one can discover through a priori reasoning the necessities that govern nature. However, it is possible to draw a very different conclusion from the same premises. For, rather than conceding that reason cannot prove causal necessity a priori, why not affirm that reason reveals that all causal relations are contingent, that it opens up the possibility that, in the future, completely different effects could follow from exactly the same causes? In that case, reason would have discovered, beneath the apparent fixity of empirical constants, a chaos indifferent to all causality. This is the hypothesis to be examined, and we shall try to show that there exist precise reasons to take it seriously.