Alignment was written in the spring of 2004, for the empty-hand gestural instruments commotion and contingence. Both of these instruments consist of gloves with sensors attached. Commotion uses two bend sensors, a two-axis accelerometer, and a head-mounted camera to detect mouth-shape, and is mapped to a complex 64-sample-array wavetable playback synthesizer. Contingence uses five fingertip-mounted FSR pressure sensors and two bend sensors, and controls pitch, amplitude, and timbre of an FM synthesizer.
Writing for a “new” digital instrument, one without a history and tradition of performance- practice or a composed repertoire, it is easy to become overwhelmed by the potential variability of sound, and forget that in standard practice, much is never notated. While I could probably compose for and notate the control of every musical parameter available in the instruments, this would be foolhardy: it is unlikely that the performers would be able to process so many streams of information simultaneously, and the result would probably not be very musical. Instead, I tried to focus on the a few important parameters and assign others to the interpretation of the performers. Thus, for “contingence” pitch and rhythm are notated precisely, and timbre more generally, and for “commotion,” playback speed and loop length might be indicated, but delay level and individual sample choice be left to the performer.
The decision to map the “contingence” index-finger pitch control to a pentatonic scale was mostly practical – in order to allow discrete pitch selection with predictable results we found 12 semitones/octave too fine a resolution for the finger/sensor. The decision to use the hirayoshi pentatonic in particular was purely compositional – it includes a great variety of available intervals which most notably contain two minor seconds. This, I believe, allowed the composition to present the illusion that the resolution of pitch selection available to the instrument was smaller than it actually was.
The notation of the part for “Contingence” is mostly conventional, with some graphic elements. The addition of modulation to the sound is usually indicated with written words (i.e.: pure sound, full modulation), much as timbral change is indicated with more standard instruments (i.e.: pesante, sul tasto). Standard notation practices are used for pitch, amplitude, and rhythm. Graphic notation is used at rehearsal marks D and H in the second section, and I found (perhaps not surprisingly for an instrument of this type) that the performer’s interpretation of the drawn gestures was much more physical than in the precisely notated sections.
For the part of “Commotion,” graphic notation was actually necessary. The first section adds fairly standard graphic “shapes” to what is essentially extended conventional notation. Rhythms are notated precisely in order to present strong performance goals but with the understanding that playback is approximate, and elements of standard notation practice are used whenever practical (i.e.: staves, fermatas, rests). The notation of the second section was much more challenging and is very different. While trying to develop notation standards for “Commotion,” I realized that the two most important controllable parameters of the instrument, rather than being pitch and rhythm (the timing of sound- events over “score-time”), were in fact the complexity of the sound (number/choice of samples played) and the periodicity of the sound or playback speed (in “instrument-time” rather than “score-time”). Devoting score space to BOTH of these variables simultaneously meant either modifying standard notation in ways that I considered to be counter-intuitive for the performer and difficult to read (using color changes to communicate changes to linear parameters, for example) or using a more dramatically different – but surprisingly simple – approach. Conventional notation can be considered to be a graph of pitch (y) against score-time (x), and on the two-dimensional score surface other information must be presented symbolically or as text. The second-section Alignment graph uses loop-length/complexity (y) plotted against playback speed (x), and the passage of score-time, which is not on an axis, is instead represented by the relative thickness of a line to be followed by the performer. This notation approach, rather than extending standard practice, uses the same tools that were used to develop standard notation in the first place, but assigns the axes to more instrument-relevant parameters.
The detailed compositional choices in Alignment were driven largely by the instrument design and the needs of the performers, the most obvious example of this being the choice of pitches for “contingence.” Pitch rows were chosen with careful attention to the finger position relationships between adjacent notes. Position changes are almost always step-wise, with larger jumps arranged so that the second note uses one of the two reference positions (1 – completely open hand, or 5 – completely closed hand).
The piece itself is divided into two sections, each using a different compositional approach. The first section attempts to attain maximum compositional control of the resultant sound by changing focus rather than asking unreasonable precision from the performers. This section treats sample playback in “commotion” more as a timbral variable, with individual samples acting mostly as input material for the delay line for the shaping of large-scale sound shapes, rather than having individual aesthetic function. The movement is short, but slow in pace, consisting of only four large sound-shapes with focus on slow timbral shifts in the sound. The first event is the simplest in shape and most ordered, with each subsequent event more and more complex, until the fourth, in which the performers are encouraged to create a chaotic texture that collapses into the more complexly organized second section. The second section is faster, and attempts direction of musical variables with much finer detail. Despite and because of this, the section also involves significant amounts of performer improvisation. The section is divided in turn into eight shorter subsections or “phrases” and a coda, allowing the performers to “align” the 2 parts correctly even though one part is written on a graph and cannot be read linearly.