thoughtdraft: One difference between philosophy & history in the realm of argument refutation is a question of volume. Refuting arguments in logic & philosophy can be self-contained. Given aptitude, one can recognize a bad argument or a logical fallacy. Refuting historical arguments, on the other hand, requires a certain volume of knowledge (facts, cognizance of existing arguments, source data). Consequently there is a natural degree of trust involved in picking up a book of historical scholarship. One unversed in the history of mathematics should reasonably be able to pick up a history of mathematics & trust in its verisimilitude or integrity of scholarship, i.e. one takes what is stated on faith, in deference to a presumed authority. If this information is erroneous, the fallacies can go undetected for much longer than philosophical/logical fallacies, which are self-contained. How much can we trust historians to tell us truths?
I say this because I had been reading a magisterial work of synthesis on the history of ideas in the twentieth century by Peter Watson, which I found wonderful & compelling up until the point when I was assured on good authority that much of what he says about the history of mathematics (and a goodly amount more) is misleading, or simply wrong. A kind of disillusion set in. Even given the necessary leeway that one might afford popular history, there is still an important difference in degree between rigour and truthfulness, surely.