A philosopher wishes to include in a comprehensive view certain ingredients; this may be impossible; or, it may be possible, but beyond that philosopher’s capacity.
It is important, therefore, to distinguish between a philosopher’s desiderata for a theory, and his theory’s answering or not answering these desiderata. In different stages of his career, a philosopher may consider different desiderata and assume they do not conflict; but this assumption may be false.
In various places I have ventured to apply these general and rather obvious considerations to Popper’s theory of science. Let me air one simple further example before going into the major example of the present essay. In a recent comment on criticisms by Bartley, Popper exempts himself from replying to these criticisms, on the ground that Bartley, though he had been a prized student of Popper’s, chose to overlook quite a few aspects of Popper’s philosophy ([‘Remarks’], 88). For example, his own problem-orientation since rather early in his philosophical career (96). No doubt, Popper has taken problem-orientation to be a desideratum for the theory of science fairly early in his philosophical career, and his quotation from an old text of his fully supports this claim. Popper, however, considers this as evidence that his own theory of science does answer this desideratum. We may say to this, in the absence of any theory of problems and of problem-solving, it is difficult to know how to judge this claim. We may say, alternatively, that every given task implies the problem, how best can one perform the given task? The task of finding a hypothesis, then, or of choosing it, of accepting or rejecting it, of refuting or confirming or verifying it, are all problems which render each and every known theory of science problem-oriented, if not practically all philosophy. Granting that not every task is a problem, I do not find Popper’s early philosophy of science - his theory of scientific character as refutability and his theory of learning from experience by refutation - to be particularly problem-oriented.
In his lecture courses Popper himself has presented a more problem-oriented philosophy of science. Scientists explain puzzling phenomena and try to render their explanations less ad hoc or less arbitrary than these look at first. There is much in common between Popper’s two theories of science, the oral and the written. For example, reducing arbitraries is done by corroborating, and attempts to corroborate are the same as attempts to falsify (the difference between corroborations and refutations is not in preparation, in the question addressed to nature, but in outcome, in nature’s answer). But Popper claims that these two theories are identical. I have elsewhere ([‘Nature']) explained (at least) to my own satisfaction that this is not so; I shall have occasion later to mention a few counter-examples to this, as well as to show how one may present a more problem-oriented theory of science, which may indeed be viewed as a variant of Popper’s.In the meantime, I wish to return to Popper’s desiderata and to the theories which possibly satisfy them. Let us begin with the question, how well his various desiderata harmonize. In particular, I shall examine his desideratum of refutability ([Log/c]) and his desideratum of search for truth ([Conjectures] [‘Aim']); they seem to agree with each other with ease; but the claim that they do is a hypothesis which may merit critical examination.
The ancient dichotomy between nature and convention, where nature stands for truth and convention for fiction, applies both to the philosophy of science and to social and political philosophy. A third view replaces fiction by bona fide error, and postulates degrees of approximation of convention to nature. This view is very close to one which has been proposed by Sir Karl Popper both in the philosophy of science and in social philosophy. He also declared institutions to be like scientific hypotheses and science to be a social phenomenon. This parallel between his two gradualist theories - of science and of society - may really be a version of modified conventionalism, though Popper himself favors a somewhat narrower version - of modified naturalism or modified essentialism.
I.