On the tendency of scientific knowledge to be correct

“The best scientific opinion of the present time has a better chance of truth, or of approximate truth, than any differing hypothesis suggested by a layman. The case is analogous to that of firing at a target. If you are a bad shot you are not likely to hit the bull’s-eye, but you are nevertheless more likely to hit the bull’s-eye than to hit any other equal area. So the scientist’s hypothesis, though not likely to be quite right, is more likely to be right than any variant suggested by an unscientific person.”

“Perfect rationality consists not in believing what is true but in attaching to every proposition a degree of belief corresponding to its degree of credibility. In regard to empirical propositions, the degree of credibility changes when fresh evidence accrues. In mathematics, the rational man who is not a mathematician will believe what he is told. He will therefore change his beliefs when mathematicians discover errors in the work of their predecessors. The mathematician himself may be completely rational in spite of making a mistake, if the mistake is one which at the time is very difficult to detect. Whether we ought to aim at rationality is an ethical question.”

— Bertrand Russel in Human Knowledge, 1948