To break up some capricious and possibly ill-advised research,* I've been filling up ever-longer intermissions with the transcripts of Morton Feldman's late Middelburg lectures (published in two volumes by MusikTexte as Words on Music/Worte über Musik). Feldman's lecture style was famously engaging. He was a great substitute for a favorite anecdote-and-general-BS-delivering uncle, unschooled but educated, unpolished but in his own way erudite, and simultaneously vague and razor sharp.** Even in print you tend to hang on every word, especially, it seems, when you disagree most with him or when Feldman has mixed something up or even got it altogether wrong. The editor, Raoul Mörchen, does a great job of identifying obscurities and correcting errors, which are, sometimes, slips and, not infrequently, intentional slights on Feldman's part. (Hilariously, after a series of insults on Feldman's part about the recorder, the MusikTexte publishers themselves even step in with a footnote of their own to defend, as it were, the honor of the recorder as a serious musical instrument for new music.) Highly recommended.
* Under the motto "now that serialism is all over, or at most completely unfashionable, what can be learned, recovered, or renewed?", which has led to some strange and wonderful stuff, not only from the usual suspects on either side of the Atlantic but also to works and writings by people like Golisheff, Hauer, Eimert, and Heiß. There have been a number of musicological publications on this topic of late (see, for example, volumes by Grant, Whittall, Straus) but none of these have struck me as particularly useful for composers looking for productive possibilities in the serial residua. If we look for elements common to all of the music which might conceivably have flown under the flag of serialism (and the flag*** I'm waving is large enough to find serial elements in Music of Changes, In C, and I am sitting in a room...), we won't find much, but two ideas fundamental to serial practice — order/series and set/collection/gamut — are so basic to so much music, that there are certainly some historical cul de sacs with potential for new music.
**As a Californian, and hopelessly constrained by my dialect, I'm constantly impressed by Feldman's ability to do this, efficiently getting across rather pointed arguments with incomplete sentences without ever without interjecting the all-utility "you know" into every sentence.
*** My apologies for all the flag waving around here these days. Must. Find. Better. Metaphors.