Musical Composition as Experiment (and vice versa)
by Christopher Dobrian
Conversations on the Artistic Process
University of California, Irvine
January 25, 2012

Normally in undertaking a study of the artistic process, one would research and draw upon the ideas and practices of artists past and present, and try to discover some underlying principles of the artistic endeavor. In this fifteen minute talk, however, it would be futile to try to summarize what others have said in the past about the artistic process. There are at least as many different artistic processes as there are artists. The point of this series of panel discussions is to hear about various individuals' personal artistic processes, in hopes that combined they will give us some insights. So I'll simply talk about my own way of working in hopes that some of my personal experience and beliefs will be of interest. I'll begin by making a few simple statements about my beliefs and my working process as a composer, and I'll then show a couple of illustrative examples. Of course, one of the necessities of music is that it requires time to hear it. So even if I just play a couple of two-minute musical examples with a couple of minutes of explanation for each, that's already more than a third of my speaking time.

At a panel discussion such as this I always feel an obligation, an imperative, to say something controversial that will stir up debate. The problem is that to be controversial you have to either make a naively overreaching claim of universality that others disagree with, or you have to try to impose your belief on others against their will. But I'm not interested in doing either of those things, and given that I'm supposed to be talking about my own artistic process rather than making universal proclamations about it, who's going to find anything to debate about what I say about my own thoughts? Nevertheless, I'll try to say one or two things that are at least mildly provocative.

For starters, for me, I don't think there's anything that unique or special about the process of making art that differentiates it significantly from any other process of making something. I find that artists spend a lot of time feeling different, thinking that nobody understands them and their art, and that their work is uniquely important in society, as if they alone have creativity and special insight. "Imagine a world without art." Well of course that would be unpleasant, but so would a world without vegetables, a world without crockery, a world without shelter. An artist's work is no more or less exalted or important than the work of a farmer, a potter, or a carpenter.

Presumably art need not only be functional like a building. It can exist simply for its aesthetic value, its beauty, or so people say. "Art for art's sake." From that expression I take two meanings. One is that art need not have an obvious "function" to be of value, and the other is that art need not be "about" anything outside of itself. These are popular notions among artists, so perhaps in the interest of being just a little controversial I'll say here that I'm not that interested in art for art's sake. I'm certainly not going to come out "against beauty". I'm just saying that, for me, beauty alone is not enough. I understand the pleasure of simply soaking in a bath of lush romantic string harmonies or new age "niceness", but for me that's not very interesting for very long. I want to listen to music that challenges me emotionally and intellectually, and that's what I hope to produce in my own music. That by no means rules out the importance of beauty. That's where technique and training comes in. One learns the skills and techniques -- be it of woodworking, representational drawing, or musical counterpoint, harmony, and orchestration -- that enable one to craft things elegantly and attractively. And it is those same skills and techniques, usually developed through hard work and practice much more than by any mystical inborn talent or inspiration, that enable one to communicate one's ideas effectively. Maybe it's because I have no actual talent that I believe this. I think hard work and highly developed technique -- combined with the development of creative and intriguing ideas -- are a more believable explanation of my artistic process than the role of "talent" (whatever that is) or "inspiration".

I really do believe that art deals in ideas, more than in objects. For me, the beauty of the object is not what I revere, but rather the combination of thoughts and techniques that led to the object's existence. Arts does communicate ideas and express emotions to us, often in ways that would be impossible by other means. But I don't think that personal expression of emotion is what makes art special, either. In fact, let me see if I can say something else controversial. I think it would be presumptuous of me as an "artist" to think that the thing I made is of value just because I made it and it expresses what I was feeling. I think that I have an imperative to dig deeper than that, to explore and examine the ideas I want to express, to try to say something that is in some way, however small, new and different and an actual addition to the existing body of knowledge and the existing body of music.

So for me composing consists of manipulating ideas, and endeavoring to express them in a beautiful or elegant way. But so does the writing of nonfiction prose, the solving of a mathematical proof, or the development of a computer program. In all those cases, one takes an idea and one develops it, manipulating it into a statement of some kind that in some way provides new information, sheds new light on the topic, and presents it in a way that communicates to others. Indeed, programmers and mathematicians and writers speak often of beauty and elegance in their work. When they use those terms, they're usually referring to a combination of ingenuity and economy. E=mc2 is important for its physical implications in the world outside of mathematics, but it is "beautiful" and "elegant" as an equation because it says so much, and something so profoundly insightful, with such remarkable economy of means.

And therein lies, for me, the challenge of composing. For me it's very much the same challenge as writing a clear, concise, and hopefully insightful speech. It's not much different from writing a computer program that solves a problem and performs a task in a particularly ingenious and elegant way. Composing is very much like writing nonfiction or programming. One strives to develop an idea thoughtfully and make a statement that is ingenious in its construction and economical in its means. Although the idea I'm trying to express may be complicated and perhaps even antithetical to common notions of musical beauty, my challenge is to use my composerly technique to shape it into an elegant musical statement, often in such a way that its complexity and difficulty is almost completely masked.

That tension between complexity and difficulty on the one hand, and elegance, beauty, and ease on the other hand is crucial to the composing process for me. It gives me a resistant force against which to struggle as I try to make attractive and interesting music from a (possibly unmusical) initial concept.

In my own composing, I tend to be interested in exploring (some would say "inspired by") ideas from outside of, or tangential to, musical discourse. The fun and challenge comes in trying to demonstrate the relevance of those ideas to music, by devising for myself a musical experiment. I don't mean an experiment in the sense of trying to make a scientific proof of a hypothesis in an ironclad universal way, but rather to demonstrate or "prove" to myself, to my own satisfaction, something about musical cognition (the ways I understand music) or compositional technique (the ways I compose music) by composing a piece of music that demonstrates my "findings" as I explore the idea. Often, because my exploration is more rational and technical than intuitive, it's useful to employ computer programming as a means to generate music that allows me to measure empirically the sonic effect of a particular compositional technique or idea.

I'll demonstrate that experimental approach in two very short musical examples. They are taken from older pieces I made in the 1990s, but they're useful because they demonstrate pretty clearly the single, fairly straightforward, idea being explored, unlike some of my more recent pieces in which the experimental concept itself and methodology can get pretty abstruse.

Twenty years ago I was doing a lot of reading about information theory and the mathematics of probability, and I was also reading about chaos theory and the mathematical description of order and disorder, negentropy and entropy. I was also reading Emotion and Meaning in Music by Leonard Meyer, in which he posits some ideas about music cognition that were related to information theory, specifically the notion that as events occur in a piece of music the listener develops mental models of what is likely or unlikely to occur next. I wanted to test those ideas in a fairly rigorous experimental way. So I wrote a computer program that would compose and perform music on a computer-controllable piano, in which the choice of notes was made using statistical probabilities, relative likelihoods of certain things occurring or not occurring. My goal was to see if I could create a new sort of musical "order" in which we perceive statistical tendencies in the note choices more than traditional melodies or chord progressions. The challenge was to make it also be musically interesting, not by traditional cultural standards but more by our ability to appreciate the types of changes that occur globally in the music. I won't explain the exact methodology any further at this time, but I'll ask you to listen to this short excerpt from the beginning of the piece, paying attention to how things change (or stay constant) in terms of the pitches chosen, the high-low pitch range in which they tend to occur, the relative sparseness or density of note occurrence, the regularity or irregularity of the rhythm, and the tonality or atonality (the relative coherency) of the pitch choices. [Example 1.]

Sometimes one is called upon to write music in which the function of the music is more important than the conceptual basis behind it. When I'm writing a score for a play by Robert Cohen or an occasional piece of chamber music for a jazz concert by Kei Akagi and friends, my fellow artists and the audience aren't particularly concerned with any highfalutin theoretical conceptual underpinnings of the composition. They just want it to sound interesting and appropriate to the situation. But I still want it to be interesting work for me to compose it, and I want it still somehow to explore a new concept, in hopes of making the music somewhat novel within the given constraints of its intended function. By imposing an abstract "experiment" on my compositional process, usually one that is in some way contrary to my natural intuitive inclinations, I create for myself that "resistant force" against which to struggle, which ideally pushes me to compose music that is unlike what I have done before, while still being aesthetically attractive, striving for those notions of "beauty" and "elegance" I spoke of, and while still being appropriate to the expressive and artistic desires of my collaborators.

In the mid 1990s I wrote a composition for piano and computer, a sort of mini concerto in which the live pianist is accompanied by a quasi-intelligent synthesized orchestra. For that piece I initially settled on the conceptual theme of "reflection", which implies the notion of symmetry with respect to a line or a plane, which in turn implies the notion of musical shapes in the form of symmetrical structures in a multi-dimensional geometric space, and which also evokes the inherent symmetry of the human body. So the challenge that I set for myself was to compose music that consisted only of symmetrical pitch structures, and in which the computer's music was always some sort of direct reflection of the live pianist's music. Furthermore, and most importantly, I resolved that every single thing the pianist did with one hand must be precisely reflected by the exact same physical configuration in the other hand. That's admittedly a somewhat ludicrous premise, but it provided me with a strong constraint, a strong resistant force, that limited what I could write.

In this example [Figure 1], it's easy to see graphically the symmetrical up-down nature of the musical arpeggio figures. And anyone who reads music and knows a little music theory can also probably see (with some study) that each phrase is precisely symmetrical in terms of its intervalic pitch structure around some axis of pitch symmetry. What's perhaps not so obvious, except to pianists, is that whatever the pianist does with one hand is reflected exactly, in terms of the hand's shape on the keyboard, by the other hand. An additional challenge, of course, was to get the computer to recognize the structure and expressive characteristics of the player's performance and reflect that directly in real time. In this example, the player plays a passage once, and then the second time through the computer accompanies that with an inverted version of what the pianist is playing. While that's going on, the computer is also calculating the player's tempo, so that at the end of the passage the computer can take over the pianist's music and play it on a synthesizer while the pianist becomes the follower and plays something new -- still always perfectly symmetrical in terms of hand position -- along with the computer. The piece is called There's Just One Thing You Need To Know. [Example 2.]

The common thread that I see in my own work is to give myself a question, often about some extramusical idea that I find relevant to musical cognition or the compositional process, devise an experiment, often with the aid of a computer, to demonstrate to myself something about that question and its role in music, and use that question and that experiment as the conceptual basis for every decision I make in that particular composition. This lends a certain unity to the musical materials that I generally find desirable, and it pushes me to compose music that is -- I hope -- novel and intriguing, while still being attractive even to someone who is not consciously aware of the guiding concept.