Announcing Canon #84 – “Chalk.”
Writing this piece has been a week-long journey. I was about to stop working on it before the week was up, and I was going to say that it’s the sort of creation where the finish line isn’t clear. With most of my canons I reach a point where I’m confident that every note is in the right place and nothing should be changed. But with some pieces, like Canon 84, I could change a whole bunch of notes and these edits might not make the piece better or worse, just different (this apparent malleability is what led me to the name “Chalk”). With such pieces one has to declare completion without the feeling of arrival. One has to accept that the so-called “final” version of the piece is just one of many possibilities with merit.
I was thinking this way about Canon 84 until I tried to go in and make some changes. Boy, was it difficult! The canon is dissonant enough that you might think it would be possible to change a few sharps to flats without the difference being apparent, but this piece doesn’t work that way. While I kept the name “Chalk,” I discovered that the piece does not lend itself to easy modification.
One of my main constraints in writing the piece was that the relationship between the lines should be “prickly.” Dissonances are emphasized, particularly the harshest ones, minor seconds and major sevenths. Another constraint was that each phrase should avoid establishing a tonal center. I think of the piece as an experiment where atonality is pursued, but not combined as it often is with other destabilizing elements like erratic rhythms, disjoint melodic contours, and an avoidance of phrasal repetition; instead, the atonal phrases try to be smooth and easy to follow, and they all adhere to the same rhythmic pattern: they’re all variations of the same idea.
To avoid establishing a tonal center I use some basic heuristics like trying to include pitch classes evenly, avoiding note repetitions that create emphasis, avoiding scalar patterns, and being very careful about perfect fourths and fifths as melodic intervals, always trying to cancel out their tonal implications. It’s easier to do this if you allow your melody to be disjoint and chaotic; it’s harder when you seek continuity.
There were a few reasons why I wanted to change this piece after my first attempt to wrap it up a few days ago. I kept noticing little violations of my heuristics for atonality, places where a note is stated one too many times in a phrase, or where a scalar pattern seems to imply a temporary tonal center. But almost every edit I tried to make in search of a “purer” atonality seemed to cause other problems, like introducing a vertical consonance between the lines and interrupting the “prickly” sonority.
And then I spent some time listening to one line of the canon alone, and I wondered if the phrases were too repetitive. I usually consider the sound of each line alone as an important part of the creative product. Sometimes there’s a trade-off between the beauty of the single line and the beauty of the two parts together: you can make choices that hurt the single line while enhancing the two-part whole. It would seem that the two-part whole should be given precedence because it’s what the listener is actually going to hear, but I take it as a matter of pride that the canon’s leader could be played alone, without the follower, and it would still be interesting. In the case of Canon 84, I struggled to edit the phrases so that when played outside the canon, they would sound more like distinct utterances and less like variations on the same thought. And I think I achieved this. I made a beautiful single-part piece with plenty of variety. But I scrapped the effort because the added variety made the canon harder to follow when heard in two parts. I realized that the repetitiveness of the phrases is actually an asset here, serving as a welcome anchor.
So this piece which I thought I could poke and prod turned out to be highly resistant to any kind of reshaping. I caught and corrected a few errors in the strict chromatic imitation this morning, and then decided not only that I would stop working on the piece, but that piece the didn’t want me to work on it any more anyway! The finish line was clearer than I thought.
That’s how I stopped working on the piece, but how did I get started on it? I was looking through some of my earlier canon scores and I remembered that I used to do something in my outlining process which, for whatever reason, I haven’t done recently. Most of my canons begin as outlines where each part is an uninterrupted sequence of whole notes. But I used to sometimes leave gaps or whole rests in these sequences, writing the phrase structure directly into the first version of the outline, instead of waiting till later to create phrase boundaries by adding rests where there had been notes: erasing parts of the outline in the same way you might give shape to a sculpture by removing parts of the original stone block. Leaving gaps in the outline makes the writing process easier because each gap frees you from a bunch of logistical problems. You don’t have to generate new material that seamlessly connects with the preceding material while playing well with the other part.
So I set out to make an outline with gaps, like I used to, and started wondering what I could “do” with those gaps. I came upon the idea of changing the interval of imitation after each gap, so instead of having a “canon at the octave” or a “canon at the fifth” I would would have a canon where each phrase in the follower imitates the leader at a different interval. (I’ve used a shifting interval in previous pieces like Gallium and Palladium, but circumstances were different in those pieces, long story.) I wondered how the effect of such a shifting interval would differ between a tonal canon and an atonal one. I decided to try things out in an atonal context. I gradually developed the core material of the piece, a set of six phrases where the follower (soprano) imitates the leader (bass) a major sixth above, then a minor second above, then a minor sixth above, then a tritone above, then a major seventh above, and finally a minor second above. Having done this, I hoped to extend the piece by inverting the parts, but it didn’t work because I had used a lot of fourths; these turned into fifths which didn’t have the proper “prickliness.” I decided instead to have the piece repeat without an inversion. But I found I could introduce some variety into the repeat through a few transpositions. In the repeat, the follower’s first phrase is transposed so the interval of imitation is an octave, and the third phrase is transposed so the interval is a major seventh. Both transpositions came about through experiment and were selected because they preserved the requisite prickliness. The leader’s final phrase is also transposed an octave higher to increase the tension at the conclusion.
What does the piece mean to me? I find it kind of thrilling to hear an interaction between the voices that is calculated, on the one hand, but chaotic on the other. It’s a loud, sharp entanglement that’s neither an argument nor an embrace. I like how the phrases are “free” from the magnetism of a tonal center but at the same time they have a feeling of direction and drive. The soundscape and phrasal repetition reminds me of earlier canons Flint and Zebra Marble but those are much more tightly patterned than this one.