Compressed 96kbps MP3.
This web-page documents a composition completed during the author's study of a Masters in Music, Composition (Studio) at Goldsmiths College, University of London. The composition was written for a week-long take away paper in the final examination, the requirement being that the composition make use of ten audio recordings supplied by the tutors (John Drever, Michael Young, Ian Stonehouse) with no accompanying information.
Spectrogram of complete composition (0-22khz grayscale); time-domain
waveform overlaid (yellow):
Compressed 96kbps MP3.
This piece is about taking things away.
Given ten very distinct source recordings to work with (and no indications regarding their origin or context), I selected segments to take away from each, keeping the remainder for further transformation and arrangement. Some of the selections specifically focus on residual aspects within the original material; for example, from the instrumental sound sources, I frequently retain only the ends and beginnings of gestures (including the spaces between).
As the remaining material for use was still plentiful and diverse, I considered means of transformation that might reduce their quantity and qualities, and begin to scuplt my own coherence on the sounds. I continued with a selective segmentation of the chosen sounds, from around one second in duration down to smaller and smaller segments of selections.
By sequencing these micro-selections end-on-end, and finely controlling changes in the selection point and length, I was able to create timbral streams with imposed gestural structures in real-time. Smooth variations in the duration allowed me to generate ‘bouncing' and amplitude modulation, whilst smooth or rapid variations in the onset point allowed me to control the content of the stream, thus shaping spectral timbre, overall volume, smoothness/roughness of the grain cuts, etc. As the durations shorten and the selections approach a pulse, the sounds approach to amplitude modulation, and then finally reduce to simple pulse-waves, obliterating the semantic reference of the sounds. After generating many various gestural streams and arranging them in parallel, I once again used selection and segmentation to take away elements, reducing the quantity and diversity of the material.
While this is the principal means of transformation and arrangement used, I have also ‘taken away' content in the frequency domain (through spectral noise gating and spectral delay) and allowed existing material to segment other material (using side-chain gating). The selective removal continued to the macro-scale; I began removing sections from the piece arbitrarily, sometimes leaving the empty spaces in their absence.
What I found quite interesting is that taking things away and aggregating the residue not only imposes a coherence to diverse sources, it also enables a more liberal, gestural approach with the material. The approach permits a freedom of movement between concrete and abstract sound forms, a fascinating region to explore. I have therefore consciously avoided creating convincing physical environment cues (such as natural reverberation, timbre masking, spatial techniques) yet attempted to retain a mimetic physicality in the timbre streams themselves. The referential origins of the sounds may be heard as I do not consciously attempt to conceal them, but I hope I have allowed them to function in playfully musical forms without recourse to such literal references.
Each part of the piece approaches this grey area in a different way, and there is a gradual evolution of these techniques through the piece. I begin the piece with the segmentation plainly obvious. These are clearly real sounds that are broken, but an arrangement is evident, opening the space for subsequent development of these ideas. In the second part (1:17), more voices sounds are introduced, with more fluid segmentation. The wind instrument for example is still audible as such, yet is gestural in a way a wind instrument cannot be. The third part (2:44) introduces frequency domain segmentation, contrasting a more immersive yet artificial sonic environment. The final part (5:42) reaches extreme segmentation such that the sounds no longer retain original timbral content; they have become completely abstract, as microsonic pulses. They are given an unrealistic physicality however reverberating in a space of almost recognisable streetcar horns and choral voices.
Subjective descriptions of supplied anonymous source recordings: