A performance last fall of Le marteau sans maître left me once again astonished about an unresolved tension in the piece, a tension between structural concerns — the interlocking cycles of two, three and four movements and the systematic orchestration and scoring patterns in particular — that were made explicit, expressed at the immediate surface of the music, and the note-to-note continuity, the precise sense of which was not aurally recoverable beyond the most general and impressionistic considerations — contour, registration, and to some extent pitch complementarity. Although Le marteau is a piece which I have listened to, intently, for more than 30 years, and has some qualities — not least its notorious coolness — which I actually find attractive, the absence of a compelling connection between local continuity and global order makes for an increasingly frustrating listening experience. Moroever, there are some scoring pattern changes and choices of instrumental register in the piece that are almost shocking in their clumsiness.
The problem here might be described as being too schematic. A schematic diagram uses topological abstraction to present the relationship between the functioning parts of a process, system, or organism in a clear fashion by ignoring actual size, scale and precise orientation and by eliding detail. Of course, such simplification is often extremely useful, but when the process, system, or organism is something as precious and sensuous as a piece of treasured music, then the loss of precision may very well come at the expense of those details, dimensions or parameters of the music which we value most.
[There is a widespread misunderstanding that Schenkerian analys suffers from the same such loss. This misunderstanding is probably due to some weak pedagogy, in which the act of analysis is seen — and, tellingly, not heard — as the production of ever-more reduced graphs, a confusion of product with process. Schenkerian analysis can, in fact, be an excellent strategy for the recovery of detail and dimension.]
In the Left End of Newmusicland — where one might likely locate the radicals, experimentalists, & minimalistas as well as some of the complexers — the relationship between a music and its construction is a persistent theme. * But one without a persistent resolution. In some music, it's an advantage if the way in which the music was put together is clear, while in other music, the advantage goes with hiding any traces of the compositional work. Piano Phase and I am sitting in a room are two examples of music in which such clarity is essential, while Music of Changes is an example which finds advantage in opacity. And still, for all the immediate aural evidence of their construction, in neither Piano Phase nor I am sitting in a room is there a lack of mystery in terms of either sonic quality or musical continuity, while in Music of Changes — the construction of which is all-but-impossible to recover — there is a prevailing sense of precisely the purposefulness and restraint one would expect in a tightly constructed work. I think that the reason these three pieces work for me (and Le marteau doesn't) is that the clear and opaque elements are complementing and reinforcing one another in optimal ways. The immediacy with which the constructive, procedural elements are comprehensible in the Reich or the Lucier facilitates hearing the music made possible by and above and beyond the process, while the anticipated ergodicity of Cage's surface soon reveals itself to be much more discrete and eventful than expected (it is music that really does change!)
A bit of constructive technique that I have find useful of late involves composing with pre-determined scoring patterns. I have sometimes combined these with rhythmic structures akin to Mr Cage's famous square root method and have found Gray codes ( and, specifically, Beckett-Gray codes) to be particularly useful scoring patterns. Now, this probably sounds like a lot of abstraction — and even, as I later discovered, with the Gray codes, some fairly sophisticated maths — but, my goal has never been to let this become so abstract as to not relate audibly to the immediate musical concerns of a musical continuity.
* Morton Feldman is here unavoidable: "It appears to me that the subject of music, from Machaut to Boulez, has always been its construction. Melodies of 12-tone rows just don't happen. They must be constructed....To demonstrate any formal idea in music, whether structure or stricture, is a matter of construction, in which the methodology is the controlling metaphor of the composition...Only by 'unfixing' the elements traditionally used to construct a piece of music could the sounds exist in themselves--not as symbols, or memories which were memories of other music to begin with."