Abstract

I argue that one standard account of the problem of induction, that which is commonly attributed to Hume, is misguided, in that it places unwarranted emphasis on the denial of necessary connexions between events. Once this emphasis is removed, then a way open to a solution, or dissolution, of the problem is revealed. ;I attempt to show first, some recent scholarship notwithstanding, that Hume can fairly be read as propounding scepticism about induction, and that this scepticism in large measure depends upon the claim that physical events are all contingent. But then I argue that Hume has failed to support this contingency thesis. Once it is admitted that there may be necessary connexions between events, then the problem of induction as standardly characterised appears less threatening. ;While this particular sceptical thesis may be removed, however, its place is taken by a more general scepticism, under which it is complained that unless the conclusion of an argument is entailed by its premisses, we have no reason to accept that conclusion. But, I argue, although we cannot rebut this scepticism, neither can the sceptic provide any reasons for us to take him seriously. A toleration of the inefficacy yet the resilience of sceptical considerations is, I suggest, Hume's major philosophical insight.