To compose music is to dictate some type of organization to a series of sound events. This is true even of John Cage’s aleatoric compositions; he doesn’t dictate the sounds themselves, but he dictates their organization quite carefully according to some abstract schema. I find aleatoric music stupid and boring, however. I feel the listener ought to be able to perceive the organization that the composer has imposed on the sound events.
How do composers of avant-garde or experimental electronic music organize their pieces? As far as I can determine, they don’t. No formal organization lies behind the surface phenomena.
In researching this question, I’ve been reading a wonderful book by Thom Holmes called Electronic & Experimental Music. He documents the early years of experimentation and discovery in loving detail. Primitive means were employed toward equally primitive ends. Folks who enjoy modern music technology may be excited, humbled, or horrified to learn that in the 1950s, a studio in Italy — at that time one of the leading institutions in Europe for the exploration of electronic music — had a lowpass filter with six selectable cutoff frequencies. Technologically, we’ve come a long way in a very short time!
Holmes discusses the ideological divide between the French and German schools of electronic music during that period. The Germans had come from the serialist camp, and preferred tones that were generated electronically. The French were pioneering what later came to be called electro-acoustic music or musique concrete, in which “found sounds” were subjected to manipulation and recombination using tape recorders.
The importance of serialism (another compositional school that leaves me cold) to the question I’m exploring is, as Holmes points out, that serialism got rid of the idea that notes were related to one another through the organizing principle of chords. In serialism, notes relate only to one another in a sort of egalitarian way.
This weakens the music in two ways. First, there is no hierarchy of pitch classes, and thus no concept of dissonance and resolution. Second, the egalitarian relationships of the notes (the tone row) are much more difficult for the human brain to process and perceive in the act of listening.
Meanwhile, over in France, electronic pieces were being composed by assembling a bunch of sounds first and only afterward working out how the piece would be structured. In other words, when you’re assembling the sounds you have no idea what you’re doing. The relationships among the sound events are, again, egalitarian rather than hierarchical.
I could do the same thing today with a modular synth and computer recording. I could record a bunch of sounds from the modular as digital audio and then afterwards try to figure out how they might be assembled into a piece of music. The process of assembly would be infinitely easier than splicing bits of tape, which is how they did it in the 1950s. But when I try doing that, I’m stopped dead in my tracks. I have far too many options, and no sensible way of choosing one option rather than another. It’s a jungle of choices, and not only is there no path through the jungle, I don’t even have a machete to hack my way through the mess.
The root of the problem is that each recorded audio clip is its own entity, its own private world. Like a note in a tone row or a found sound recorded on tape, it has no relationship to anything else. I can give one audio clip a relationship to another by putting one before or after the other, or by layering one on top of the other, but such relationships (simultaneity and succession) are not, in the end, very interesting. They’re arbitrary. Why put Clip A first and Clip B second, rather than the other way around? There’s no reason to prefer one order to the other.
This is quite different from the situation in tonal harmony. In tonal harmony, the progression G7-C has quite a different meaning from the progression C-G7. One is a movement toward stability, the other a movement away from stability toward an as-yet-unknown future. And listeners who have been exposed to common-practice music understand that quite intuitively.
But I don’t think the solution is to retreat into tonal music. The strengths of a modular synthesizer lie in its highly variable shaping of timbre. If you want to compose tonal music, there are better tools.
Possibly glitch music suggests a way forward. The foundation of the organization might be rhythmic rather than based on pitch classes. If a sound-oriented, non-tonal piece is in a strict meter at a steady tempo, the possibility of dissonance and resolution — in this case, rhythmic dissonance — reappears. When a given sound wanders off the beat, either through syncopation and hemiola or through gradual drifting, and then locks back onto the beat, a perceptible hierarchy of events is created. This type of hierarchical relationship, in which the rhythmic grid functions rather the way a tonal center functions in tonal music, can possibly be used to create larger forms.
I’ve been unconsciously avoiding this approach because it seemed too solidly rooted in dance music. I find four-on-the-floor kick drum about as boring and useless as serialism. But once you throw the kick drum overboard, things might get interesting. It’s an idea worth exploring, anyway.