This paper argues that every inductive inference is covertly deductive and that validity might involve different kinds of necessity (logical, nomological, metaphysical). The main distinction is between inductivism, the view that some inferences are genuinely inductive, and deductivism, the notion that all inferences are covertly deductive. Section 2 presents the problems of inductivism, highlighting its conceptual deficiencies and advancing deductivism as an alternative. In section 3, I argue that the epistemic commitments of inductivism and deductivism are fallibilism and infallibilism, respectively. (...) Section 4 attempts to clarify the metaphysical assumptions of both positions and suggest a modal interpretation of probability that illuminates some puzzles regarding chancy counterfactuals. Section 5 concludes. (shrink)