As I prepare for a semi-improvised performance at the NIME conference this week, I find myself thinking a great deal about improvisation and indeterminacy (more on the distinction between the two in a bit). Specifically, why is indeterminacy still looked upon with such suspicion in the new music world, 100 years after John Cage’s birth? Oh sure, certain aleatoricisms have been tamed and have found their way into the standard notational vocabulary—familiar gambits like box notation and feathered beaming. At this point almost no one objects to their use, and they’ve entered into common performance practice. But paradoxically, this renders them almost wholly determinate in sound, encouraging rote and mechanical use. Casually deployed, they can become conspicuous signposts that announce “this is a new music piece” and not much else.
Thankfully, thoughtful composers continue to develop new notational gambits, but this presents its own challenges, and puts an extra burden on performers to absorb this new information. Composers can try to take on as much of that burden as possible by making the notation as clear and vivid as they can; in this respect, I admire David Smooke’s approach. In a recent post on his toy piano concerto, Smooke describes a subtle and flexible notational system capable of loosening some musical parameters (e.g. rhythm, time) in order to make other parameters (e.g. texture, ensemble coordination) easier to control. Of course, some aspects of this notation are not exactly new (the unmeasured preludes of Couperin and Pandolfi come to mind).
So far, so good. But when indeterminacy gets bigger and scarier, people’s attitudes start to change. We can distinguish between indeterminate notation meant to evoke a specific sound, and indeterminate notation that is meant to prompt or provoke the performer in some way. The former is almost universally sanctioned; the latter is still controversial. I wish it wasn’t, because it’s a powerful locus of creativity. Too often I’ve encountered the attitude that, by leaving too much up to the performer, the composer has abdicated his or her professional duties. I hope that this is mostly due to misunderstanding, but I worry that it’s an impossible ideological divide. In this provocative model of indeterminacy, a little willful notational obscurity is even desired, because it compels the performer to engage with the piece deeply. That is, if they’re not put off right away—it’s a tactic that requires great trust between performer and composer.
I wonder if the resistance to indeterminacy is somehow a part of the long hangover from the musical culture of deliriously extreme specificity which dominated the last century. Perhaps ironically, even this musical movement culminates in a kind of indeterminacy of ability in the music of Brian Ferneyhough. Here, the notation is about as specific as you can get, but the near-impossibility of it shifts the nexus of indeterminacy from the details of the notation to the capabilities of the performer.
Some of my music also ends up occupying an awkward middle ground between indeterminacy and specificity, and I’d like to defend this awkwardness if I can. In Mobile I for violin and electronics, the pitch content of the electronics is unspecified, but descended from the spectral content of the violin, ensuring that it remains musically connected to the violinist’s performance. Like Smooke’s concerto, the end result has a particular texture that is unique to the piece. I wonder, though, if it’s too vague to satisfy those fixated on specificity, and too predictable to appease those who prioritize exploration.
This also raises questions about the distinction between indeterminacy and improvisation. Often this distinction seems more semantic to me than anything, and this is especially true when it comes to music with live electronics. If we view the computer as a willed agent, then surely it is improvising. On the other hand, if you break the program down into its rule-based or stochastic components, then it is merely another layer of notation, this time in the form of code. Part of this is due to technological limitations; often the tools are not as reliable or predictable as we would like.
But working with these limitations can also lead to novel, inventive solutions. This crosses my mind many times while working with Mimi (Multi-modal Interaction for Musical Improvisation, a software system for live human-machine improvisation designed by Alexandre François). As I practice with Mimi, the fear of a less-than-perfect improvisation is often present, and so there is a strong temptation to make the performance as predictable as possible. However, Mimi won’t let me. Every time I sit down with the system, it does things I don’t expect and can’t predict. In the end, I think this is why I like indeterminacy—it compels me to do the things I am afraid of.