Karl Popper (1902–94):
The critical rationalist never tries to establish a theory about the world; he does not believe in 'foundations'. Yet he may believe — as I do — that if we produce many competing ideas, and criticize them severely, we may, if we are lucky, get nearer to the truth. This method is the method of conjectures and refutations: it is the method of taking many risks, by producing many (competing) hypotheses; of making many mistakes; and of trying to correct or eliminate some of these mistakes by a critical discussion of the competing hypotheses. I believe that this is the method of the natural sciences, including cosmology, and I think that it can be applied also to philosophical problems. ... Nothing in our intellectual life is exempt from this process of criticism and error elimination.
— "Beyond the Search for Invariants," in Karl Popper, The World of Parmenides: Essays on the Presocratic enlightenment, ed. Arne F. Petersen, (London and New York: Routledge, 1998), pp. 146–222; this passage from pp. 152–53.