> Is whether or not you play the score as written an issue? How do we even know this about composers that predate recording technology? Sometimes they pretty m

> Is whether or not you play the score as written an issue? How do we even know this about composers that predate recording technology? Sometimes they pretty much tell us directly, e.g. in treatises which discuss performance. Other times, we can look at early recorded media to figure out what the style was. But there's no question that classical used to have a hugely important tradition of improvisation, which was only gradually lost throughout the 19th-c., as technical ability in performance was regarded as more important than well-rounded musical aptitude. It survives today, e.g. among organists, who are still trained in both performance and composition. > There's a massive difference between pre-20th century classical music, which was very much the pop music of its day Well, not totally true. On the one hand, by the 19th century, pop music did exist already, e.g. Stephen Foster's songs; on the other hand, it was not yet radically different from classical music, being published in sheet music form, for a musically-knowledgeable public. It would be hard to tell apart a Stephen Foster song from the 19th c., from a classical "arietta" or "canzonet", of the sort that even e.g. Joseph Haydn composed, by any inherent criteria. Classical music forms did vary widely in "intellectual" ambition ("arietta's" and "canzonet's" being on the low end, together with dance suites, "divertimenti's" and "cassations"), but for a long time that ambition wasn't regarded as incompatible with some degree of reasonably-widespread appeal - for something on the "high" end, consider symphonies, instrumental concerti or opera overtures.