In a strangely real way for me, I’ve always been writing this piece. There is a slow grind of trying things, and hoping that this time, they’ll fit. I know I’m kidding when I say that I’m always writing the same piece, only maybe there was always a part of me that wanted more change faster. To metamorphosize definitively. To kill off my past self, in an exquisitely brutal way. To break free. That word, freedom, always has that same inevitable allure. Mesmerizing like little else. Palpably hypnotizing. So I started talking, running my mouth like I do, about what was holding me back. Categories of sound, whole strata of expression that where, or rather, I had made forbidden to approach. What was lacking, past just going for it, was the ability to translate between modalities. I measure my own freedom in capacities. What tasks am I able to complete? The change might not have been measured in leaps and bounds by anyone else’s yard stick but for me, this was a big deal. The scope of my craftsmanship got expanded working on this piece. So what I’m talking about here in practical terms is extracting polyphonic pitch information from sound and translating it into useful information. I was then able to further sculpt the translated data in order to make other, transformed reflections of originals. These new entities were completely set apart from material that I had been able to summon previously. At least that’s the way I felt about it: Really inspired artistically by having broken new ground technically. There it is again, that internal emotional state, unprovable as ever, relatable only through story.

This piece came off of the heels of working on Mapping The Valleys of The Uncanny for what I felt was a really long time. The format of the residency was that we had a week at Ställberg in the beginning of the summer and then there was the concert at the end of the summer. In the beginning of the summer I had the idea that I wanted to return to the idea of anchoring the piece in the place. This was just as much a trope for me going into as it reads to you now. I really did want to leave my comfort zone [chuckles]. What I ended up with in practice was running a bunch of tuba glissandos through a big rattly tin box. There was the idea of intentional contamination of “field recordings” (any recording done outside of the studio is a very wide casting of the net). I think the term reAmping is more plausible. Also the more conventional strategy of reRecording sounds in a room, as a reverb became very integral to the work. The utility of the impulse responses that Mats and David recorded was enormous. Also the particular flavour of running sound through the room in realtime with all of it’s unique nonlinearities became very apparent in comparison. Recording various material through the box and Maskinhallen at double the samplerate and then playing them back at half the speed also proved rewarding. When I got back home I expanded on my experiences. I did a lot of reAmping of materials I had created in Ställberg through different transducers. I was able to make various metal objects vibrate creating complex nonlinear connections between the low end and the high end of the spectrum. The less noble the metal the better it seems, so far at least. Still exploring that space.

In terms of form, I guess it’s rondo. That is really a very fancy form if you know me, or my way of relating to form (I’ve been ignoring it, actually). For the longest time it seems, I’ve been carrying around the idea that my music should have no form, because finite realisations where always a matter of staying practical. I felt that all of my compositions where actually states, infinite in terms of time, the way that places are. You could enter into them and stay there, listening, as long as you cared to. Thinking of the boundlessness of time itself has always inspired me most of all. I guess you could say that a lot has changed. Getting back to the subject at hand; The form of the piece is ABACADAEAFAGAHA basically. If I’m being very particular then all of the A’s are more like A1, A2, A3 etc. because as we progress through them, they get more and more inclined to present a more dualistic, or layered tonality. This saltiness is more noticeable towards the end.

Also another little tidbit is that F is split in two parts in the 8 channel version. Halfway through the two ensembles in the inner most quad configuration switch places. I felt it was an interesting kind of movement. I could hear it, no problem. I guess you could say it is the only actual spatialisation going on in the piece. That change occurs over the span of one minute. All of the things that appeared to be moving are actually due to how the synthesis was split up in four stereo files. Have a look at that code here. I enjoy how the line between what is an overtone in a spectrum and what is a new pitch in a linear sequence is blurred. The choice of which speaker every new pitch arrives in adds a sense of being veiled by the sound around you. It is a kind of phenomenological result of the synthesis. The synth definition was originally written by Marcus Pal, then later adapted in collaboration with friends.

In the synth definition the envelope segments are expressed in percentages.
e4 through to e1 tell you how far into the note value that the envelope should have reached it’s maximum level after the attack time, after which the release time begins immediately. So an e1 value of 0.01 will yield a 1% duration for the attack and a 99% duration for the release. c4 through to c1 denote the curvature of the envelope segments. hr is the harmonicity ratio of the operator. mi means modulation index, which is the modulation amount by which that oscillator will modulate the next. The last oscillator (e1) doesn’t have a modulation index value because it isn’t modulating anything else. I like to think that because we are listening to the different operators on their own, as well as hearing their effect on each other, spread out around us in the room, and that these operator frequencies are harmonic ratios of the fundamental, something special happens. At least to my ears I am reminded of how the natural overtone series can be used in additive synthesis. The sensation is further heightened for me by SuperCollider’s uncanny knack for delivering such clean synthesis, owing to negligible round off errors in the calculation of waveforms at the lowest level. This becomes especially important for me where modulation indexes are concerned. Without this level of detail, FM can otherwise become a very round about way for me to make white noise. Anyway, this .scd is now out in the wild. You go ahead and have a mess a round with it and let me know how it goes.

The bottom half of the file is where the patterning of events happens. The durations between the events are pretty straightforward, this is slowly evolving music. The key to how I can get away with what I’m doing in the keys that control which pitches get chosen is that aforementioned slowness. I think I picked it up from Morton Feldman. Slowing things down makes strange pitch sequences a lot more palatable. It let’s me do things like sequence scales. I think I was attracted to that idea initially because I was in awe of the height of abstraction there. If we compare it to sequencing enharmonic transposition for example, then sequencing scales looks to me like it would happen at a much higher altitude. It has that titillating quality I am such a tremendous sucker for. For a long time I felt like I was cheating doing this but I got over that after I did Map. The reason was that I had some really uplifting interactions talking to folks about how I had organized the pitches after that first show. I think at this point I’ve started thinking of it as something akin to a trademark move. But \scale is not the only key here. Most times I’ll start with just one scale and sequence \degree and \octave and try to get something going. When I have something that works started, I’ll then want more variation. I’m all about variation. That’s what leads me to find things like different scales having different amount of degrees in them interesting and then proceed to sequence \degree grinning. This kind of thing is tricky though. I have to go very slow. I change one tiny thing at a time, and just focus on that. Then after I’m satisfied, I move on to the next thing. Also using \Pseed to find a good spot in the river to step down into helps a lot after I reach the point where I have more than a few keys effecting what pitch comes out.

This piece was originally composed for 8 channels and had it’s premiere on the second of August 2019 at Ställbergs Gruva as part of their three day festival “Att återta det levande”. I was able to spend an inordinate amount of unusually intense time writing this piece, thanks to the generous support of Musikverket, who supported Ställberg’s residency program Timbre. The residency culminated in our contribution to the festival in the form of a concert. The entire concert was six hours in total. It started at six in the evening and ended at midnight. The concert was held in Maskinhallen, which has a breathtakingly beautiful room tone. It is my favourite resonant space. Myself, Maria W. Horn, Mats Erlandsson, Lisa Stenberg and David Granström each wrote about an hour’s worth of music for the concert. I am especially grateful to Anton Andersson who did the lights, Eric Sjögren who was our director and Veine Bartos who was our carpenter, without whom neither the residency nor the concert would have been possible. I had the time of my life and I felt part of something much larger than myself. That really is the way I’ve always wanted to feel. I am so happy and grateful to have been a part of this project that I feel like my heart might cave in on itself.