McMullen argues in response to Kuhn that explanatory power ought to be considered among the virtues of a good explanation because explanatory power is a value which coincides with scientific realism. Somebody in class yesterday argued that perhaps there was a time in the history of science when treating explanatory power as a virtue (as copernicus, a realist, did in response to and over against Ptolemy), was useful. It was useful not because explanatory power is indicative of truth, but because theories which were preferred on the basis of having greater explanatory power made more and/or better predictions. Thus, the pragmatic anti-realist argument went, since we no longer need to value explanatory power today (i.e., we no longer need to be realists, since realism can be regarded as an accoutrement of mature science’s infancy) to do science successfully, we should give it up without a fuss.
I responded in private after class that I think it continues to be pragmatically useful to value explanatory power as a basic explanatory virtue and that, if I am right, then therein lies implicitly an inbuilt defeated for pragmatism. Scientific realism, on pragmatist grounds, ought to be preferred to pragmatism, because it is plausibly of greater use.
This analysis is probably shallow, but for what it’s worth, it seemed to me an interesting thought.