From John Stuart Mill onward, the problem has been approached by attempting to analyse: ‘How is it that physics is so successful?’, and applying the answers to backward sciences [sic], such as psychology and the social sciences, and to a lesser extent the biological sciences. Mill took over views from Bacon and others that the methods of science were fundamentally inductive (inductivism). Mill’s law of causality is a generalisation from multifarious observations (see note below).
With the method of deduction one starts with original ideas of unclear origin, i.e. general hypotheses, and then tries to prove these hypotheses. The hypothesis is provisional. When estab-lished by some kind of proof, it becomes a theory. However, you never get beyond the stage of a hypothesis. The last word may be said on some scientific problem, but, if it is said, we cannot know it; hence, the whole distinction between theory and hypothesis breaks down, i.e. all theories are hypotheses and never more.
But the reverse is not the case. All hypotheses are not theoretical. Hypotheses are of two kinds – (a) general or universal, as in science (these could be called theories), (b) special or individual, e.g. a medical diagnosis.
To sum up: The aim of science is not certainty. It is a human effort and in consequence shares human imperfection.