Theories as Conventions
I pulled this quote from a blog of a friend of mine. He is a graduate student in economics and interested in the philosophy of science. I have excerpted parts of my responses (numbered 1 and 2) to his question and have left them incomplete in order to initiate what I hope to be a heated debate…
"Philosophers of science *should* match their program theory with past historical instance. It would be important to show that, for instance, validation has worked in the past. That is, philosophers of science should practice their own discipline. That said, it is insufficient to criticize a philosophy of science theory merely on the grounds that it does not align with history. However, it must be shown that any philosophy of science theory in the past has worked, even though it may not explain the entire history of progression in science. A Popperian, for instance, would attempt to falsify other theories in the philosophy of science on the basis of poor empirical support or disconfirmation.”
(1) I would ask that you explain why, if the theory is strictly normative, the theory of science needs to answer to historical validation. I understand the need for validation if the theory is sold as a descriptive/normative account of science, but for a normative account, I would argue that it would be something akin to a category mistake to require empirical justification for some theory that is not making any empirical claims.
(2) For Popper, his theory of falsificationism is supposed to be initially adopted as a convention, to be justified solely according to its consequences (problem-solving magnitude, though he does not use these words). Popper conceptualizes scientific endeavors as essentially problem-solving endeavors (philosophy in general is also based on problem-solving). As such, hypotheses are proposed solutions to these problems and are only useful insofar as they are good at solving the problem that they have been generated to solve. The problem for a theory of science is, for Popper, a means of successfully demarcating scientific claims from non-scientific claims. He suggests that Falsificationism is well equipped to demarcate science and to uphold various other characteristics that must be upheld in order for some enterprise to be called scientific (various empirical strictures, a conception of progress, etc). So his theory of science is conventional, in that, it does not initially come with a full-fledged rational justification-package, but is to be provisionally accepted as a means of solving the problem at hand. If it solves a problem (the problem of demarcation), then we keep it, if it does not, we provide another solution and examine how well it solves the problem. So a methodological theory is not to be critiqued by another theory and its particular methodological processes, but by how well that theory solves the particular problem that it was proposed to solve. For example, Popper didn’t critique the Positivist’s verification criteria of meaning according to whether it is subject to falsification, but because it did not adequately solve that which it was suppose to solve; i.e, the problem of demarcation.