Abstract

It seems natural to suppose that Artur Schnabel’s occasionally inaccurate performance of Beethoven’s Hammerklavier would have been even better had it been accurate throughout. In the present paper I defend this supposition against a sceptical argument which purports to show that we have no good reason to believe it. The sceptical argument, which draws on some plausible-seeming thoughts about aesthetic properties, concludes that, because we cannot know whether this or that (as-yet-unachieved) musical result is so much as possible, we have no good reason to suppose that the accurate version of Schnabel’s performance should be preferred to Schnabel’s own. In attempting to defuse this sceptical argument, I open up, and begin to explore, some new and potentially interesting questions about musical creativity and about the species of rationality that it requires and exhibits