Synthesis Thinking


The most striking difference between composing for acoustic instruments and virtual synthetic tones has to do with expectation. When I sit down to play piano, I have an excellent idea how it will sound ahead of time and how the instrument—whether it’s a real piano or a virtual one—will respond to my touch. The layout of the keyboard is also important: the piano’s territory is familiar and I have a repertoire of go-to moves, chord shapes, and hand positions upon which I can build music. With the piano, I know what to expect in terms of touch, sound, and layout. But when I sit down to play synthesized tones (triggered by a keyboard or a pad controller), my expectations are dashed because I have no idea what to expect before I begin exploring. Whether I’m listening to presets, triggering samples, or tweaking patches of my own, the central problem is that there’s no end to where the sound stops changing. I can take a sound and roll off some of its frequencies, truncate its attack or decay, filter or modulate or swap out its waveforms, and so on (and on and on), until the sound has become something else. With a few twists of a virtual knob or fader, a piano sound could be an oozing bass or a screeching bell. How did I get there? I sometimes think that in electronic music there’s no there there, because any sound can become any other.

The fact that synthesized tones have no endpoints challenges my sense that sounds should have a stable and clear identity. Should they? Where did I get this idea anyway? Maybe from playing percussion—from accumulating a body knowledge of how these instruments respond to my touch. From the instruments’ potentials forever lying just beyond reach of my skills. From having watched my teachers play the drums and mallets better that I could/can—more musically, more consistently, with more control—and then chasing after that sound and technique. When I’m exploring synthesized tones, I’m going after a sound, but I’m not sure what or where the sound is that I’m looking for. Maybe I’ll recognize it when I hear it, but when I’m chasing a sound I don’t yet know I’m not in control of the process. Instead, I’m open to the possibility that I could hear something that re-orients my notion of what is musical. If there’s a technique I’m chasing after, it might be a more refined sense for how different sounds can fit together. 

Despite the ambiguities of the process, working with synthesized tones offers lessons. 

First, sounds are fluid. Just because a sound sounds one way doesn’t prevent you from changing it into something else, so it’s useful to be perpetually open to changing your sound. 

Second, suspend your expectations. Just because you expect to hear one thing doesn’t mean you won’t find something else that is more delightful. (If you do, go with that.) 

Third, sounds don’t have to have a stable and clear identify. Embrace uncertainty. If you can’t describe why you like a sound, that’s okay—sound perception often works on subliminal levels. As the artist Robert Irwin says: “Intuition is about sensing facts before they materialize.” 

Fourth, let the process guide you. Hearing an unexpected sound and then responding to it (e.g. with further tweaking) is a robust methodology for moving forward on a project. Nassim Taleb describes this methodology with a phrase I love: We shall be guided by what is lively. 

Finally, the lessons of working with synthesized tones are an analogy for thinking effectively. Applying the concepts of fluidity, suspending expectations, embracing uncertainty, and following a process is a way to synthesize yourself anew as you try to figure something out.          

On Concepts Out Of Context


I find myself reading (fiction and non-fiction) whilst on the lookout for interesting standalone phrases. This may say more about what I notice that what the authors wish to highlight, but the phrases are nevertheless concepts to think about. Here’s a few I’ve been pondering: 

reverse our usual method  


obstacles to dynamism 

minimal gestures 

bottom-up cooking 

a scene which generated its own weather. 

Production Workflows: Working With Call And Response

Attachment-1 copy.jpeg

A technique which has served me well in composing, when I’m layering one part upon another, is to think in terms of call and response. In most situations, a new part takes shape as I listen to what I already have and try to figure out how to complement that. Keeping in mind the possibility that I won’t ever need the part I’m figuring out (I often mute or delete vast swaths of parts I make), I nevertheless sit there and try to come up with something that fits.

I’m learning to key in on little things in the music that I think of as “calls” that I can respond to. Calls are almost always rhythmic in nature—events that reoccur in regular ways. For example, maybe a bass part plays around the downbeat of every two bars (or once every 8 beats). If I’m layering another melodic part, instead of playing over the bass part, I can respond to it on the downbeat of every second bar (and every 8 beats thereafter). This creates a slow-moving call and response between the bass and the melody part. Recording a second part in this responding way to a first part requires a bit of restraint. My tendency is to want to play each part continuously, so I have to hold back, waiting for the part’s turn to respond to the bass part’s call.

Things get compositionally more interesting as you layer more and more parts, because your musical puzzle is becoming increasingly complex and thus you need to listen more closely to where the available spaces are and where the potential calls in need of responses are hiding. For example, let’s say that my second melodic part is a set of chords and my next (third) part is some kind of bell sound. (I like bells.) As I try to figure out where to place the bell, I now relate it to both the bass part on the downbeats and the responding chords every second measure. I can use the bell to cut the distance between the bass calls and chord responses by half, or I can go the other way and play the bell just once every two or four cycles of the bass.

Layering parts using this call and response method is one way to get the music listening to itself at the various layers of its rhythmic action. No matter how I choose to relate a new part to what is already recorded, the goal is to make the groove as lively and tensile as possible. In essence, I don’t just want a piece to merely have a beat; I want every part of the piece to be it’s own kind of beat. Adding parts in call and response thickens the texture of the piece and animates the music into a more unified and pulsating whole.