3 min read

Volitional Composition

Volitional Composition

I am currently in a period of flailing around, trying to figure out what the next big thing I'm going to sink my energies into is going to be. This often happens between projects, and I have noticed that during these times a pattern tends to emerge. It has happened again almost immediately after getting Komoy Live System off the ground. It looks like this:

  1. Start a fresh, carefully organised framework for some kind of new generative music system that's going to solve all my problems forever.
  2. Get quickly bored of making generic mini-systems like '4/4 kick drum', 'euclidean melody', 'arpeggiator' etc. and, as a proof of concept, concentrate on more bespoke, one-time-use mini-systems to demonstrate that the system overall has potential.
  3. Start bending the supposedly generic parts of the main system around the bespoke stuff to make it sound cooler.
  4. Realise that the system itself has become so specialised that the only thing it can make is itself, over and over again.

System as song. I normally end up leaning into this, because having created a song is always exciting and, honestly, I don't really know how to avoid it anyway. I think I also quite like the idea that song-writing actually has to be like this. That to make something useful, it is appropriate and necessary to reinvent quite a lot of the wheel each and every time. To not just write a song, but to design and build a structure for the song to manifest inside of.

For Komoy Live System, this is a practical concern, involving the building of a material (well, digital) structure. It is a set of quite specific conditions that is able to spit out many different flavours of song, but the songs it creates will (I think) always be recognisable as coming from that system.

More broadly, I think this concept works, er, metaphysically too. In the sense that, putting aside all the practical parts of composing, in order for a song to exist, for it to have intent, you first have to build the world/conditions from which the song can emerge. I contend that each and every song has one of these — its own world — whether the composer was aware that they were building it or not.

Or, to put it in a cooler sounding way: a song is a soundtrack to itself.

I didn't really mean to get into all this in this particular devlog, but whatever. This approach to composing I'm trying to describe kind of fits with a concept I've been thinking about a lot lately, that I've decided to call volitional composition. It hinges upon this dialectical idea of the act of composing being to simultaneously create a world and also charting an intentional path through it in the shape of a song.

I have decided that this idea is useful in a not-figured-out-how-this-is-at-all-useful-yet-but-feels-like-it-should-be kind of way. So I'm letting it percolate in the background for now. More in due course.

Back to my earlier pattern recognition though. This time, with KLS, I haven't even gotten as far as making a song. I've just not got the patience to keep going through the motions of getting it up and running when I can already sense it is quietly changing its purpose into becoming some kind of replicr, 2019-era 65 song-maker rather than a multipurpose generative music system.

So I am immediately pivoting. As I optimistically set out in the previous post, the original plan was to have some amount of generic generative patches all synced to some kind of central clock patch. And then many more custom patches that do the more unique things.

Instead, for now, I am going to concentrate on just having one central mixer/sync patch that has some kind of A/B crossfader thing and an optional clock that other patches can choose to hook into if they want. And then all the other patches can be self-contained generative-systems-as-songs, free to be as complex or simple as I want them to be.

Why am I doing this at all? Dunno. But I have learned that during these between-project periods, this is a question to avoid asking as much as possible. Instead the best thing to do is just switch off any kind of critical analysis (as if....) and trust in whatever intuition is leading me to want to experiment with weird Max patches yet again. Worst case it'll probably end up with me making some basic techno bangers.