On the Evolution of Induction in Russell's Philosophy

Russell's philosophy can be divided into two distinct periods. In the first period, under the influence of his predecessors, he endeavored to preserve foundationalism and in the second period, he tries to minimize foundationalism. In his early works, he accepts induction because of its inherent intuition and widespread use in life and considers inductive inferences as probable, not certain. But in the second period, by analyzing empirical phenomena, he likes what a pure empiricist would do, infer the postulates that are the bases of empirical sciences. These postulates have a supra-logical nature, and interestingly, induction is not among these postulates. In fact, according to Russell, the science and the ordinary understanding that are resulted by non-demonstrative inference require these postulates. In this paper, we try to understand Russell's notions on induction, therefore, we investigate it in detail in the mentioned two distinct periods. For the first period, Russell's notions and one of the examples of probabilistic proofs presented for it will be presented. And for the second period, his reasons for changing his approach and his new perspective, as well as the supra-logical postulates governing the non-demonstrative, along with some of their properties, are presented. And finally, though briefly, some of the criticisms are discussed.