changing speeds
photo: soprano Bodies Electric
soprano Abby Powell with guitarists John Chang, Adam Negrin, Jason Sagebiel, and Martin Moretto
Working with arrays makes things go at a certain speed. The speed of sudoku.
I came up with a quick list of how to vary the speed. Please add to it and give me your thoughts.
A problem is misguided commitments to norms of futurity that not even Milton Babbitt subscribed to. I am now both doggedly "tonal" and doggedly faithful to all musical possibilities that emerged in the 20th C.
One Music!!-- like radio in France, but less like IRCAM.
1--Gut the array and make it Webernian for a spell. Take one of the tirichords in your array and go Webernian with it. (A Babbitt trick)
Changes rate of set-class change, but potentially preserves the rate of aggregation.
2--Multiply-partitioned arrays—
Another babbitt trick—my setting of Gillian Welch’s My Morphine demostrates this very clearly--
The 3\4 partition is slow and lugubrious; the 5\2; 1\2 partitions moves fast, like Schoenberg Op. 27 #4. See the array, below.
The aggregates tend to go by much faster in 5\2; 1\2 partitions because the 1\2 is an “accompaniment” to the two linear 5-note tunes. They make very palpable aggregates.
Next—
3--Something I do—express all the vertical intervals of an array over a pedal tone.
The result sounds like the opening of Brahms #1.
4-- Babbitt’s “weighted aggregates”. I have't tried this, but it's interesting in this connection, I think.
#3 and #4 could both subtly change the perceived speeds, no?
Next—
5--Look at all the *memorable* collections in your “passacaglia” and do other neighboring tones, creating other adjutant harmonies. This should be weighted to keep hold of the source collection that you’re embellishing.
Do this with two of the collections in succession and you begin to see local aggregation at a speed radically different from your first “passacaglia”.
To vary speed one must behave more like a tonal composer, one must relax one's commitment to certian TT orthodoxies.
My thinking tends to be too vertical, but what is vertical may be horizontal.
My Morphine's mulitply partitioned array--