I heard a fantastic performance of one of my pieces the other night. Comments were generally positive, as (at least out of courtesy) they always are, but one listener voiced his opinion that the piece could have been longer, confirming a suspicion I’d had since I finished writing it almost a year ago.
That’s a tough criticism to answer, especially since almost everybody else I asked felt that the piece’s length wasn’t a problem. Is the piece really too long? How would I know whether it is or not? If it really is, what can I do to make sure it doesn’t happen again? Is it better to err on the side of “too short” or “too long” (I think I already know the answer to this one)? There was another piece on the program for the same instruments by a well-regarded young composer that was almost exactly the same duration—nobody said his was too short. What gives?
I think the bottom line is that there are psychological “computation cycles” of varying length that we must undertake to make sense of a piece which are independent of the piece’s internal proportions. Furthermore, based on last week’s experience, these cycles seem to fluctuate widely from listener to listener. In other words, a three-minute piece might work well as is, but stretching it out (with a phase vocoder, for instance, or by rewriting it around the same shape but with expanded surface details) to three times its normal length may not produce a successful nine-minute piece. And even if you think it does, the person sitting next to you might disagree. Is the solution to write music based on one’s own rate of dramatic metabolism, even if few others share it, or to analyze the literature, develop a hypothetical “normal” metabolism, and take that as one’s model?
In short, is it my job to write music that moves at your pace or mine? This depends, of course, on how accurately I can determine what “your pace” is–that I’d get it right is hardly a foregone conclusion. Brahms did something akin to this focus-grouping when he played piano reductions of his pieces for friends. Despite this precedent, however, I’d feel somewhat phony distributing questionnaires and counting standard deviations; for the time being, I’ll devote my efforts to zeroing in on my musical metabolism and writing music that better articulates it, chi-squares be damned.