From rules to practice

The typographic "rule" that one should not (that is, never) separate two sentences with more than a single space (as opposed to the "rule" that one should (that is, always) separate two sentences with two spaces) has received some on-line attention of late. (For example here, with countless arguments for and against scattered through the 'sphere.) While the issue here is, ultimately, trivial, the exercise, Ithinks, is a useful one, a reminder of how often we confuse habitual practice with compulsion or necessity. Classical musicians, for example, get trained in the "common practice", a body of technique roughly identifiable with German/Austrian practice from the middle of the eighteenth century onward for some century and a half or so. While this is a remarkable slice of repertoire and the pedagogical tradition which has grown around it is remarkable in its own way, I think that that technique only really becomes its most exciting when one teases, no digs into, the edges of that repertoire, or — better yet — entertains the possible insights and advantages of traditions which are, with regard to the common line, heterodox. My favorite example of such an uncommon practice comes in works of Berlioz (a composer who, if I weren't avoiding the ten best/ ten most influential business would have a lock on a place in either list) who had absorbed the alternative tonal practices of the revolutionary era composers as well as aspects of text setting in French to derive a tonal practice that in terms of voice leading, harmonic rhythm, metric stress etc. continues to place the common practice in vivid relief. Resolve a 4/2 dominant to the 5/3 tonic? Sure. Use a mediant as a substitute for a dominant? Why not? Sustain passages with only the voicing of a chord changing? Of course! The compositional space between rules closely cut to a particular era's local tradition and all the possible resettings of the parameters in such rules is huge and — pace Schoenberg — that's where there's still plenty of good music to be made.