Ori Barel’s Algorithmic Music for Player Piano

from the Amazon.com Web page for the album begin discussed

This Friday Albany Records will release the second album of music by composer Ori Barel. The CD consists of a single composition entitled Centrifugal Force for Player Piano, organized into four parts, all slightly more than ten minutes in duration. As usual, Amazon.com is processing pre-orders for this new recording. Barel’s interests as a composer include algorithmic compositions, working primarily with the algorithms behind cellular automata and Markov chains. These sources may be taken as polar opposites, since Markov processes are stochastic, while the behavior of cellular automata are rigidly deterministic.

While these mathematical techniques are radically different, it is worth recalling an observation that Virgil Thomson made for The New York Review of Books when he wrote an article about John Cage. By the time that article was written, Cage had produced a large corpus of works, all of which were based on scores constructed as a result of tossing three coins and interpreting the results as they would apply to the I Ching, an approach to divination involving the construction of a six-line hexagram. Each line either had a gap in the middle or was a single solid stroke; and the coin-tossing process would select the type for each line and designate which of them would change from one type to its opposite. Hence the title of one of Cage’s longest compositions created this way, “Music of Changes.”

Thomson chose to compare Cage with his “polar opposite” among composers of the time. György Ligeti had written an impressive analysis of a Boulez composition for two pianos in which every property of every note in the score had been strictly determined by a set of rigid constraints. Thomson explored this comparison because he argued that, from a listener’s point of view, you could not distinguish the Cage piece from the one by Boulez. Thomson’s point was that extreme complexity could result from either a highly deterministic algorithm or an entirely random process; but, regardless of the source technique, the result would still be referentially opaque.

From a philosophical point of view, Cage embraced that opacity. His philosophy was that a composer simply creates a listening experience. What the mind of the listener makes of that experience is not relevant to the acts of the composer. One might say that listening amounts to an extremely personal approach to listening triggered by the most impersonal approach to creation. I have not yet established an equally clear perspective on Boulez’ intentions. However, in her footnotes for The Selected Letters of John Cage, editor Laura Kuhn observes that the initially close friendship between Boulez and Cage began to fall apart when Cage committed himself to chance techniques.

By way of an appendix to this lengthy attempt to establish context, I should observe that I had been experimenting with both deterministic and stochastic approaches to composing music since the summer following my freshman year (1964), when I was providing coding support to graduate students in the Moore School of Electrical Engineering at the University of Pennsylvania and could use my spare time to create computer programs of my own design. I derived a lot of fun from those and subsequent activities but never really hit on anything that I found particularly satisfying. I suppose the reason is that, by the time I had progressed to doctoral research, I was more interested in how software could assist the efforts of a real composer rather than rise to the level of an ersatz one.

Listening to the music on Barel’s new album, I have to say that I could not distinguish the determinism of cellular automata from the statistical behavior of Markov processes. I can also state that I did not particularly care about that distinction. In other words, as an attentive listener, I find myself following in Thomson’s footsteps and have no compunctions about that frame of mind. From that point of view, I have to say that 40 minutes of Barel’s stuff was more than I could take. Even the ten-minute duration of an individual part began to strain my attentiveness. While there are any number of ways in which original composition may emerge from different mathematical models of complexity, the music itself ultimately resides in the listening, which is the point Thomson was trying to make. I have to wonder how Barel approaches his own results as a listener.