The ever-estimable Lisa Hirsch has two items (here and here) in response to a Teachout article on musical complexity. For music criticism, use of the term complexity is something of a blunt instrument. One has to be very specific about the materials or means or relationships which are being described as complex and even when identified as complex then there are no clear or generally accepted ways of measuring relative complexity, whether within a single work or among several works.
Most critically, in the thick of a real musical event, it is not always clear — and this, in the best cases, is a matter of compositional design — whether composer, performing musician, or listener can catch or is supposed to be able to catch, some or all details or a more general gestalt, or even to move attention between levels of detail, or if the experience of whole and parts is designed to be ambiguous. Most works of music which have the capacity to stay fresh on repeated listenings — renewable musics, one might say — offer a reserve of additional details and relationships. To the extent that musical complexity is quantitative, filling a work with detail and internal or external connections, the central issue for composition is projecting and framing that content so that it can be heard. From this point-of-view, the most self-consciously minimalist composer is as concerned with managing complexity as the overtly complexist or maximalist; the critical difference, I believe, lies in how the relationship between musical materials and audition is drawn.
By my own measure, a work like Alvin Lucier's I am sitting in a room is a more effective compositional projection of an extremely complex set of internal, acoustical elements and external, and to some extent, extra-musical connections than the elaborate pitch constructions of many a high serial work which are lost to a surface which fails to project the construction in anything but the most impressionistic manner.
Central Javanese Gamelan music, Karawitan, is my favorite example of this: it is possible to hear a classical gamelan work coarsely, attending, for example, only to a single trunk melody carried by the loudest instruments, but experienced players and listeners learn, over time, to attend to more and more instrumental and vocal lines, learning to hear not just simultaneities but the ways in which individual lines relate to one another but also to the whole piece as well as to the repertoire and style in general.