ABSTRACT
This chapter examines cases of retroduction, and in what follows to try to give an account of its logical character. It shows that theoretical conclusions of far-reaching empirical importance are sometimes reached by methods which, according to past and present empiricist doctrines, are unempirical. The chapter examines a selection of particular episodes in the history of science to decide what the actual relation was in each case of observation to theory, and what part they played in the process of discovery. In the General Scholium at the end of the Principia Newton insists that theories must be deduced from the phenomena. In this philosophy particular propositions are inferred from the phenomena for whatever is not deduced from the phenomena is to be called a hypothesis. The phrase has been questioned by modern logicians who hold that from empirical facts neither general theories nor other particular facts can be derived by deduction.
