ABSTRACT

Popper actually gives an oversimplified view of the theories of Bacon and Mill in suggesting that inductivism necessarily entails that repeated favourable instances will count in favour of a theory and that inductivism is as justificationary in spirit as astrology and the pseudosciences. He says (CR, p. 256):

Even so, if such an attempt were made, it would be, as Grünbaum (1976a, pp. 215-22) has shown, entirely in accord with the methodologies of Bacon and Mill. Bacon emphasized that negative instances were worth more evidentially than affirmative ones because theories were proved only when their rivals had been falsified. In other words, an observation provides true support for a theory only when it is at the same time a negative result for a rival. Mill (1887, p. 313) makes a similar point, explicitly contrasting the small value of a vast mass of repetitive experience with the true probative value of an observation which speaks for one hypothesis at the same time as it speaks against an alternative. There is no need to stress the similarity between these views and Popper speaking of evidential support arising only from falsifications of rival theories. Unfortunately for Bacon and Mill, however, positive support through eliminative induction will be given not only to the one hypothesis we are interested in but, as Popper points out (LSD, p. 419), equally to the countless other possible hypotheses also consistent with the evidence; but this is a difficulty which we shall see arises with Popper’s views as well.