Nature of Probability and Rational Credibility

Abstract

As we have seen, in retort to the other argument of the Humean skeptic, that we have to know for certain that empirical statements must be true in order to validate them, Lewis holds that it is the validity of empirical knowledge as probable judgemnts only which requires to be assured. Since an empirical generalization is subject to tests and revision in the light of further empirical findings, we should hold it not with certainty but with mere probability. Now the question is whether this probability warrants our belief in empirical generalizations or laws. By rejecting our empirical knowledge as invalid because they are not necessarily true, the skeptic may be suggesting, as many other philosophers usually do, that some principle like the principle of the uniformity of nature is requisite for the purpose of validating empirical knowledge. Lewis’s answer to this is that no such principle is requisite for that purpose. He argues that no rational proof can establish that empirical knowledge must correspond to objective facts or that predictions must be infallible.