ABSTRACT
The form of inference called “induction by simple enumeration” (which I shall call simply “induction”) has occupied, from Francis Bacon to Reichenbach, a very peculiar position in most accounts of scientific inference: it has been considered to be, like the hangman, necessary but unpleasant, and not to be talked of if the subject could possibly be avoided—except by those who, like Hume, refuse to be limited by the canons of good taste. For my part, I hold that the work of Keynes, considered in an earlier Chapter (Part V, Chapter VIII), suggests a change of emphasis, making induction no longer a premiss, but an application of mathematical probability to premisses arrived at independently of induction. Nevertheless, inductive evidence is essential to the justification of accepted generalizations, both those of science and those of daily life. I wish to make clear, in this Chapter, both how induction is useful and why it is not a premiss.
