ABSTRACT
Theories are not verifiable, but they can be ‘corroborated’. The attempt has often been made to describe theories as being nei-
ther true nor false, but instead more or less probable. Inductive logic, more especially, has been developed as a logic which may ascribe not only the two values ‘true’ and ‘false’ to statements, but also degrees of probability; a type of logic which will here be called ‘probability logic’. According to those who believe in probability logic, induction should determine the degree of probability of a statement. And a principle of induction should either make it sure that the induced statement is ‘probably valid’ or else it should make it probable, in its turn-for the principle of induction might itself be only ‘probably valid’. Yet in my view, the whole problem of the probability of hypotheses is misconceived. Instead of discussing the ‘probability’ of a hypothesis we should try to assess what tests, what trials, it has withstood; that is, we should try to assess how far it has been able to prove its fitness to survive by standing up to tests. In brief, we should try to assess how far it has been ‘corroborated’.*1
The fact that theories are not verifiable has often been overlooked. People often say of a theory that it is verified when some of the predictions derived from it have been verified. They may perhaps admit that the verification is not completely impeccable from a logical point of view, or that a statement can never be finally established by establishing some of its consequences. But they are apt to look upon such objections as due to somewhat unnecessary scruples. It is quite true, they say, and even trivial, that we cannot know for certain whether the sun will rise tomorrow; but this uncertainty may be neglected: the fact that theories may not only be improved but that they can also be falsified by new experiments presents to the scientist a serious possibility which may at any moment become actual; but never yet has a theory had to be regarded as falsified owing to the sudden breakdown of a wellconfirmed law. It never happens that old experiments one day yield
new results. What happens is only that new experiments decide against an old theory. The old theory, even when it is superseded, often retains its validity as a kind of limiting case of the new theory; it still applies, at least with a high degree of approximation, in those cases in which it was successful before. In short, regularities which are directly testable by experiment do not change. Admittedly it is conceivable, or logically possible, that they might change; but this possibility is disregarded by empirical science and does not affect its methods. On the contrary, scientific method presupposes the immutability of natural processes, or the ‘principle of the uniformity of nature’.
