Composing for Choreography

This week, choreographer and dancer Lauren Sherlock joined me in the studio to experiment with expanded movement techniques in the system. Lauren and I have worked together in the past on improvisatory dance/music projects and have received funding to develop another that embraces movement interaction to produce an integrated choreomusical work in which choreography doubles as a musical expression. In the development of the system, working with Lauren is extremely beneficial as her choreographic skill-set brings a fresh perspective to the work; her movement language is much more expanded than my own, her ability to memorize movement sequences allows for more structured experimentation, and dividing system use from back-end operation between the two of us makes the example recording process much smoother. 

A key point of analysis each time we meet is the difference in experience between Lauren using my examples and her own. The general trend is that when using examples recorded by myself (when she is absent) the experience is much more exploratory, as Lauren is unaware of the characteristics of the patch. When Lauren uses her own recorded examples, she often feels more confident in performing the system due to having predictable places and poses she can return to. In this session, having a place and pose that generates silence proved to be very beneficial to her experience. 

In recording Lauren’s examples however, we are yet to land on a truly effective method. The priorities in this process are to balance predictability with exploration; to provide reliable points and spaces that lend themselves to reactive improvisation. Using Tai-Chi moves as a basis, we attempted a handful of methods in this session to varying degrees of success. Generally, we found that recording more examples risked Wekinator equating one example with another when they were similar in pose, which made for a jumpy outputs of the models. Less examples with more space between them however resulted in less reactivity when moving throughout the space. These findings must only be taken as indicative however, as perhaps the former could be remedied by less drastic sonic outputs recorded while the latter could be solved by more drastic. The main takeaway is that there are many elements to be considered in the recording process. The question for me currently is whether more planning of pose/sound relationships will benefit the process.

This session also underlined the importance of documenting patches, both in Wekinator and on the synthesiser due to Lauren beginning to evaluate positive interactions in the system which we would like to preserve. This led me to realise that interactions that I might have viewed as arbitrary are, in fact, substantial to a choreographer making efforts to devise movement sequences coherent with sonic results. This became increasingly apparent as potential structures for our piece began to reveal themselves through the addition and subtraction of various layers: the harmonic bass, the resonant chord, the harp, birds and plucks etc. 

This has relevance to the broader project by revealing that coherent compositional results with the system can be achieved by exploring sonic interactions through improvisation layer by layer, but composing the way these layers change in relation to one another throughout the space. Conceptually, this is a bit like writing for an orchestra in which every player will react to every movement of the conductor, so the art is in defining how the movements are to be interpreted differently by each section. While the harmonic bass and resonant chord heard in the demonstration video were often audible, I found interchanging live harp input, plucks played with a keyboard and a beat-repeat effected field recording of birds as the melody layer to be very effective as leading musical devices. 

In this demonstration, Lauren is exploring examples recorded by herself based on Tai-Chi movement. The early part of the video in which she is centred displays composed interactions, while the latter half depicts an improvisatory approach to exploring the interactive space. 

Modulations (existing as one model in Wekinator and one parameter controlled in the synthesiser) in this session consisted of:

Audible in video:

Harmonic Series Wavetable Oscillator Timbre

Harmonic Series Wavetable Oscillator Colour

Oscillator Filter LFO Freq + Amp

Spectral Resonator Resonance (blend between field recordings and resonant chord)

Granulation Mix

Granulation Density

Granulation Size

Not featured in the video:

Sample Beat Repeat Repeats

Sample Beat Repeat Pitch

Harp Granulator Size

Harp Granulator Feedback

Harp Granulator Reverb