Fact, Fiction, and Forecast

A short book that combines a 1946 lecture on counterfactual conditionals with three lectures given in London in 1953. The first part shows how hard it is to say when a counterfactual such as 'if that match had been struck, it would have lit' is true, and links the problem to the difference between laws and accidental generalisations. The lectures turn to induction. Goodman argues that Hume's old problem, why inductive inference is justified, is not the real difficulty; the new riddle is which inferences count as inductive at all. He defines 'grue' as applying to things examined before a certain time just in case they are green, and to other things just in case they are blue. Every emerald examined so far is both green and grue, so the same evidence seems to support the conflicting predictions that future emeralds will be green and that they will be blue. Goodman's answer is that projectible predicates are those entrenched by their past use in successful predictions.

The grue paradox is one of the most famous puzzles in philosophy, and this short book shows why induction is harder to understand than it looks.