Cybernetic Improv
A Duet in C In Sen Pou


Photographs by Timothy Nohe

"Cybernetic Improv" was an outdoor performance consisting of computer generated music, improvised electric guitar, and spoken word.

A program I'd written, "Pentatonic Hands," randomly generated music from a Japanese "In Sen Pou" scale, while I accompanied on guitar and discussed both random number generation and the "In Sen Pou" scale.

Hear the performance

Read the script

My software, which I named "Pentatonic Hands," is an alleatoric musical composing engine, embellished with animation. A Shockwave embed is to the left, for you to play with. If you don't have the Shockwave plugin, you can download it here.

One of six notes on a chosen pentatonic scale (including the octave) is randomly generated in one of two sound channels, producing a chance-based, harmonic interval.

The sounds were generated by mixing two tracks of a saw-toothed wave, flanged from either direction. This was then mixed with a duplicate track that had been pitch-shifted one octave down.

The graphics are scanned images of my left hand in a variety of positions, oftentimes wiggling under the scanner.

The code assigns a random number from 1 to 6 to two variables, and triggers one of six notes in both sound channels based on a series of "IF...THEN" statements.

In order to appropriately accompany "Pentatonic Hands" as it composed, I restricted my guitar playing to the same In Sen Pou scale: C, D flat, F, G, and B flat.

My improvised playing was purely melodic, and had a floating tonality that matched the ungrounded counterpoint of "Pentatonic Hands."

My interpretation of In Sen Pou maintained a definitive key center; despite the lack of substantial cadences, a Japanese minor feel in C was established.

The spoken word component of the performance consisted of statements about computer random number generation interspersed arbitrarily with statements about the Japanese in sen pou scale, these being the two most fundamental theoretical tools at my disposal.

The juxtaposition of statements was intended to lend a sense of random architecture to this spoken performance component.