Found Form

I'm currently trying to prise myself away from my latest project and release it -- more on that soon, I hope -- and thinking about the next one. It got me onto the subject of how I handle large-scale forms and I thought it might be useful to try to set that down. Most of this blog is about the local details involving a note, scale or chord so this is a bit of a departure.

Form used to be a big problem for me. I could write or improvise a minute or so of music that I was happy with, but didn't seem to be able to expand that into something longer without losing my sense of purpose. For the last decade or so I've adopted a very specific solution to this without really verbalizing it, so here goes.

I always start with the length of the piece. Usually this is given to me in some way: for example, in The Chymical Wedding each instrumental part had an allotted lengths depending on its function, and the underscore for the text of course was determined by the time it took to perform it.

I think of this duration as my raw material, like a slab of marble.

I then need to "find a form" within that time. It seems as if that would be the hard part but in fact fixing the duration of the piece right at the start turned out to be crucially important.

Finding the form is a matter of distributing categories of sound (which are not yet really defined) in the available duration. This could be done with a graphic score, and I experimented with that, but I found a more convenient way. Since I do a lot of my work in a DAW, I use three or four tracks as my "score" and start putting things into them. The clips on these tracks could be MIDI or audio, and what they sound like doesn't matter at all (they're muted). What's important is how they look, since I'll use these as a visual cue for the form of the piece.

In the process of putting this together, I'm constantly developing my idea about what "story" the music is telling. I don't mean a narrative you could put into words so much as a pacing-out of the available time; things like "something will be revealed here", "this idea will continue and develop into something else over this duration" and so on.

I use the term "found form" in the title because these clips usually depend on whatever I have lying around. Indeed, quite often an audio clip will be stretched to fill the whole available duration, and its undulations then suggest formal elements that can be brought out. The audio itself isn't used but its shape is.

Once I have a "graphic score" like this, the next step is to interpret it in terms of instruments and musical elements. Already, this is a kind of composing in the literal sense of "putting things together". It's still rough work, like chopping at the marble block with a chainsaw, but that's OK; gradually I can transition to the finer tools. I've found that if I start with a tiny chisel, for example by making little phrases right away, I never make progress towards a whole piece. Those things need to be stored away and used when the form is clear.

What's surprising is that this process always feels productive, rather than staring at a blank screen or page waiting for "inspiration". It certainly doesn't commit me to anything -- forms can and do change right up until the final mix -- but it means that almost from the beginning I have a structure to rub up against and challenge. That friction is crucial to actually making music, in my experience, as opposed to talking or writing about it.

I was struck recently by hearing Samuel Andreyev's interview with Chris Dench in which he talks (about the 1:06:30 mark) about being a "top-down" composer who devises "force fields into which I sprinkle the notes in order to make them audible". I think that's very close to what I'm describing.

Anyway, I hope to release my new album soon, before the new college term gets properly underway, and indulge myself in the fantasy that is starting something new...