Network 1 (2017)
For digital synth and interactive network
Take 1
Take 2
Take 3
Take 4
Take 5
All takes are unedited
Duration: from 20 to 30 minutes
Performances:
- 23-Jun-2017 - Spektrum, Berlin
- 27-Apr-2017 - BEAST FEaST, Birmingham
I play a DIY digital synthesiser developed in Max. Three modules of a complex recursive FM synthesis are controlled with an XY pad, which interpolates between preset nodes, all already involving some sort of autonomous behaviour. I look in the interpolations for the areas that produce the most unexpected results, the ones that still sound new to my ears even after playing the synth for a while. Its output is sent to an interactive system, also developed in Max, which analyses some of the features of what I’m playing and creates new layers of synthesis loosely related to it, mostly sine waves, and white noise filtered through banks of resonant filters, all the frequencies somehow related to FFT frames of the synth’s output.
The synth is designed to respond promptly to my gestures, although my improviser’s instinct suggests the term attentive tinkering rather than improvisation in this case, while the interactive network of autonomous modules generates slower and more tangential responses. These responses also define how long a section will last without letting me know in advance, forcing me to stay challenged and listen even in the noisy miasma. Some of the limits are hard-coded, as the interactive system evolves through different phases only in response to specific gestures.
I’m writing this quite a few years after the composition of this work. As I go ahead and try not to reject everything I’ve done, I’m also learning, a posteriori, the clearer influences behind some of my music. The sound world and the approach come from David Tudor, the League of Automatic Music Composers, and all the live noisy tinkerers whose appetite for determining what works best before the performance was very scant. Even the algorithmic melody in just intonation that I ended up including in the materials to tinker with live (the part that would probably work best as an “excerpt”) is hidden somewhere in the middle of the performance, as another unexpected element to wait for. I was thinking: many of the great musicians working in the context of emergence avoided melodies, chords and familiar elements because they were too close to our expectations about form and taste, so to stay true to that spirit, after three or four decades, I’d try to imagine something they wouldn’t have predicted, even though Jim Horton and David Behrman worked extensively with just intonation. I should avoid a defining element: if the work can be easily branded as algorithmic just-tuned patterns, whatever-ism or any of those top-down classifications that seem so important to talk or write about music if we have only 30 seconds to form an opinion, I’m not doing my job. The weird melodic parts will come to light only with some patience. And they will probably end unexpectedly, because they often aren’t in line with my expectations either.
Tudor said “it is important to me that the audience senses the presence of a live musician. It makes all the difference in the world.” I dutifully share this feeling: even if my activity as improviser is severely limited by the controller, I’m still here, looking for ways to find a balance between my tinkering and the autonomous modules of the network, listening for the areas of the interpolation where I find my expectations unconfirmed and keeping my ear open to change the plan if the interactive net gives me something hard to deal with.
I ended up including the last five takes of the work, unedited. As is customary with emergence, multiple iterations of the same work are the only way to see whether the system's behaviour is producing new results each time. Maybe the number five was an unconscious homage to Neural Synthesis, of which I have five versions in my music library: four on Neural Synthesis Nos. 6–9 (Lovely Music) and one on Tudor Plays Cage & Tudor (Atonal).