Dory and Viola

Here's a new piece!



Laying in bed one night recently, I had the thought to build something out of loops using the first four or five steps of a C Dorian scale (Bb and Eb, if you don't know) on each viola string.  Each string would be its own voice (so, four voices, with four loops each on the C, G, and D string and five on the A) and each loop would be of differing lengths and could include only one pitch, but each pitch could be played more than once and in more than one manner (legato, staccato, etc.).  Ableton Live 10's Session View, based as it is on clips, can extend this Enoan technique by allowing randomization of clip order, velocity, and other parameters.

I haven't played my viola since before the lockdown, mostly because I just don't have the energy these days to practice.  But this piece really wanted the viola voice, so I relied on Ableton's Orchestral Strings sample pack, which I've used many times before.  Its voila samples in have some significant shortcomings, which were especially apparent in my last piece, but the requirements of this one were such that it wasn't as much of a problem.

While I appreciate the concept, I'm mostly not gaga about algorithmic music per se, as it often feels flat to me and doesn't seem to go anywhere.  However, my intention here was to use the algorithm as a foundation and then to perform effects on it in a way that would create some sense of development and direction.  Compared to most of my other music, this approach much more like that made on a modular synthesizer in which a patch is set up, usually with one or more generative sequences, and the performer manipulates the mix of those sequences and various effects to produce a musical piece.  A key challenge in that modality is to design effects and how they relate to each other in such a way as to be musical and performable.  After much head-scratching and flow-charting, I settled on a routing structure that would allow me to manipulate sections of the work live with a feedback delay, a harmonizer, and a buffer-based sequencer and then feed the whole thing through my favorite reverb.

The harmonizer is a relatively old (I'm guessing Max 6?) Max for Live device called M4L.dl.13.Harmonizer (I couldn't find documentation for it beyond a reference in an undated tutorial index).  I like it because it has a feedback loop, so the transposition you pick stacks up:  if you choose a two-step transposition, you get whole-tone harmony; with three steps, you get diminished chords; with five steps, you get quartal harmony, etc.  Additionally, it incorporates per-channel delays, giving a kind of pseudo-arpeggio effect.  I manipulated the direction and distance of the transpositions to give the algorithm a sense of mood and movement. 

The sequencer I used randomizes pitches and limits them to whatever notes (or microtones) you want.  In this case, I continued using the dorian scale; i.e., any sound in the buffer would be repitched to any note in a dorian scale in the first octave up or down from that sound.  This has the effect of creating a new dorian "key" centered on whatever note goes into the buffer; in other words, if the note being played is D, then the sequencer plays random notes from a D dorian scale (no sharps or flats).  When you have multiple notes going into the sequencer, this can get a bit chaotic, but it somehow feels to me still tonal, if strongly chromatic, at least compared to allowing the sequencer to play in twelve tones. 

I also experimented with controlling or randomizing different parameters of the loops themselves.  For example, I had hoped to map velocity to attack, with low-velocity notes having long (1500ms or more) attacks; my experiments with this were ultimately fruitless, though, and I abandoned the goal. 

I enjoyed being able to incorporate performative components into this.  I did several takes of it, all of which came to about the same length, giving me some confidence in the naturalness of the flow.  In the absence of my viola (or at least the energy to invest in it), I expect to be looking for more ways to make my electronic music more "live."