Martin Archer: A game of poker with Morton Feldman (1995)
for cello, synthesizer and electronics

The following notes were written by the composer to accompany the recording of this work on the Discus label (DISCUS 6CD) performed by Angie Rosenfeld, cello, and Martin Archer, synthesizer and electronics.

This composition was realized using a Yamaha MDF2 datafiler as a compositional tool. This spirit-being is generally used for midi data storage rather than as a musical instrument, but I have found mine to be a powerful familiar over the last years. I began by recording a couple of dozen fragments, (mostly just a few seconds), from the synthesizer to the datafiler. Next I used the file append routine to make multiple copies of each fragment, and then to shuffle individual files and then blocks of files back and forth according to a roughly predetermined plan. I did this until I had generated about thirteen minutes of music from my original fragments, a process which I likened to dealing a deck of cards, and which prompted the name of the composition. I then deleted various bits which I didn't like, and replaced them with some new material to arrive at the finished main piano part. Next, I superimposed a slow arpeggio at 70 bpm over the entire piece using a second sequencer. I was delighted to find that the arpeggio was virtually undetectable in the context of the fuller music around it, but that nevertheless it brought a subtle and variable element to each repeat of the original material. I completed the synthesizer part by adding the various drones which are introduced some way into the piece. Up until this point I had imagined that this was going to be a work for solo synthesizer. However, when I ran the sequences back some time later, I felt the music was too austere and decided to add a cello part; I had recently completed another piece with Angie, and decided to create a substantial part for the work in hand. The cello part is conventionally notated, with the exception of some improvised phrases in the middle, but is synchronized only approximately to the synthesizer part. It is divided into ten sections each with a defined starting point in relation to the piano, but reliant thereafter on the skill and judgment of the player to interpret repeats, rests and dynamics.

I find the process described above, which blends clear intention, improvising skill, chance correlation, pure trial and error, and a creative approach to new music technology, to be very attractive in its blurring of the reasons behind who did what, when, how and why. I have no reservation about opening up this naive and haphazard methodology to general scrutiny. No room to describe how much I love Feldman's music.