ABSTRACT
In the previous chapter I have stressed that induction is never a method. Popper, however, makes an important additional point. So far as the previous chapter is concerned, I have simply said that there is no rational machinery for passing from observational premisses to an inductive generalisation but that a hypothesis* is attained by some mental jump. Now this way of putting the matter would suggest that the observations recorded in the premisses are made first and that only subsequently does there arise the question of how a conclusion is obtained. But Popper points out that this is not the actual procedure of science, because the hypothesis or at any rate some hypothesis is entertained in mind before the observations are made. The time factor is important, for it finally disposes of the possibility of induction as a method. If the reader will consult any example of 'inductive' discovery, he will see at once that at all points of an investigation some hypothesis precedes experiment and observation, and that experiments are planned in the light of a hypothesis in order to test it. Here Darwin's famous reflexion comes to mind:
How odd it is that anyone should not see that all observation must be for or against some view if it is to be of any service!1
