Puzzling out science

Abstract

W hat do the natural sciences know, and how do they know it? I approach our topic with a little diffidence, since I am nei ther a scientist nor a historian of science, not even a specialist in the phi losophy of science, but a plain epistemologist. Indeed, recalling Sheldon Glashow's reaction to an invitation to discuss what prophets of the end of science present as Grave Epistemological Issues: forgive me i f I thrash about a bit--it's not easy to beat a dead oxymoron, I feel more than a little diffidence! I, too, shall thrash about a bit; in my case, however, not out o f pure exasperation, but in hopes o f separating the chaff, the absurdities o f radical critics who profess to believe that we can no longer imagine that science is an objective endeavor, bu t must acknowledge it to be a subjective and relativistic project, operat ing out o f social attitudes and ideologies , 4 f r o m the ser ious ques t ions to which an h o n e s t epistemologist might be able to contr ibute something. 5 Once u p o n a t ime (the phrase is a warning that what follows will be cartoon history) it was taken for granted that science enjoys a peculiar epistemic authori ty because o f its uniquely objective and rational m e t h o d of inquiry. Successive efforts to articulate what that uniquely rational m e t h o d might be gave rise to u m p t e e n c o m p e t i n g vers ions o f what I shall call the Old Deferentialist position: 6 science progresses inductively by accumulat ing true theories conf i rmed by empirical evidence, by observed facts; or deductively, by testing conjectures against basic statements and, as falsified conjectures are replaced by corrobora ted ones, improving the verisimilitude o f its theories; or instrumentally, by developing theories which, though not themselves capable o f t ruth or falsity, are efficient ins t ruments o f predict ion; or, etc., etc. Such obstacles as Quine 's thesis o f the unde rde te rmina t ion of theories even by all possible observational evidence, ~ Russell Hanson 's and others ' o f the theory-dependence of observation, s Goodman ' s new riddle o f induction, 9 though acknowledged as tough, were assumed to be superable, or avoidable.