Kevin R. Busch

While Kant does not address the problem of induction often attributed to Hume, he does, by way of a transcendental deduction of an a priori principle of reflecting empirical judgment, address a distinct problem Hume raises indirectly. This problem is that induction cannot be justified so long as it presupposes some empirical concept applying to or some empirical principle true of more than one object in nature, a presupposition neither determined by nor founded on reason. I draw on Hume’s positive account of induction to motivate the following objection to Kant: in so far as induction can be justified, there is reason to doubt that it would be so in virtue of any a priori feature.

In his account of our belief in the Causal Maxim Hume argued, among other things, that it is not absolutely necessary for any event to be caused. Harold Noonan attempts an objection to Hume’s argument: in showing (i) the absolute possibility for any event to exist without its actual cause, Hume would not thereby show (ii) the absolute possibility for any event to exist uncaused. For this objection to succeed, Noonan needs two further assumptions: first, that Hume indeed could not move plausibly from (i) to (ii); second, that Hume needed to move from (i) to (ii) to show (ii). Both assumptions are false.