Hume's problem of induction

In January 1739, a 27-year-old Scot named David Hume published the first two books of A Treatise of Human Nature in London, anonymously. In Book 1 he asks how we know about cause and effect. We never see one event compel another. We see only that events of one kind regularly follow events of another kind. We then expect the same to happen again. But on what basis? We cannot prove it by reason, because it is not contradictory to imagine the future being different. We cannot justify it by experience either, since arguing that the future resembled the past before, so will again, uses the very assumption we are trying to prove. This is the problem of induction.¹³

In his later Enquiry (1748), Hume used the example that the sun will not rise tomorrow is no less intelligible a claim than that it will. He concludes that our belief in causes comes not from reason but from custom or habit, a natural instinct that we cannot do without. He worked in the tradition of Locke's empiricism and aimed to bring Newton's experimental method to the study of the mind, and he used the doubt of ancient skeptics, though he held that no one can live as a thorough skeptic. The Treatise, he wrote, fell dead-born from the press. He recast it more attractively in the Enquiries, and the book gradually became one of the most important in philosophy.¹²

Hume also argued that the self is only a bundle of perceptions and that you cannot derive what ought to be from what is. Immanuel Kant said that Hume interrupted his dogmatic slumber and spent his Critique of Pure Reason answering him. Karl Popper responded by saying science works by testing and refuting, not induction, and Nelson Goodman's grue puzzle sharpened the problem.¹³