The American share
NewMusicBox editor Frank J. Oteri has been quarterbacking a useful discussion (in which I participated): How much American music should American performers and ensembles program each season? In a poll of readers (largely interested parties – NewMusicBox is geared to U.S.-based composers and those interested in contemporary art-music), a near-majority favored 26 to 50 percent:
http://www.newmusicbox.org/chatter/chatter.nmbx?id=6283
In an exchange of e-mails last season with Steven Smith, the newly named music director of the Richmond Symphony – also, presumably, an interested party, as he is an active composer – I posed a related question:
Q: In this orchestra's current schedule of 12 classical subscription programs – with, let's say, 40 pieces performed – how many works that are unfamiliar or "challenging" to listeners would you consider to be appropriate?
A: Given that there are always many factors involved in programming, such as number and length of works on a single program, choice of soloists, repetition of "warhorses," inclusion of chorus, budget considerations, "festival" programming (that is, focusing on a particular theme or composer), and artistic development of the orchestra, I think that the number of "new" works could be anywhere between 7 [and] 16.
Practically speaking, budget considerations rank almost as high as audience tolerance of the unfamiliar, which most American art-music is to most symphony concertgoers, in making programming decisions. The 18th- and 19th-century warhorses are in the public domain; most modern and all contemporary scores are still under copyright. In times of financial stress, it's tempting, or even necessary, to skip compositions for which you must pay to play.