Statements on Isolation, Memory, and Acceptance

Abstract

The dissertation will be comprised of a fixed-media electronic composition approximately 20 minutes in length in two movements. Spoken word recordings of poetry and prose, drawn from several literary works, comprise the entire sound source for the work. In the first movement, fragments of text are drawn primarily from House of Leaves by Mark Danielewski. In the second movement, text is drawn entirely from the Emily Dickinson poem informally known as After great pain, a formal feeling comes, catalogued as number 372. To me, the texts serve to represent contrasting emotional states: the fragments in the first movement convey fear, anxiety, tension, and claustrophobia; the Dickinson text conveys resignation, yielding, and desolation. While these texts differ greatly in their content, they all carry certain sonic properties which enables them to undergo similar processes of manipulation: all are consistently dramatic and yet poetic, and all use imaginative and abstract imagery (“the covering memory permanently hitched to everything preceding it” and “the nerves sit ceremonious like tombs”) which enables the composition to rely on the timbre and aesthetic of the language rather than on the specific content. No original sound synthesis is present; the sonic material is instead drawn from the processing of these recorded voices. The first movement, Stories misremembered, is characterized by the technique known as granular synthesis. Samples of the vocal sounds are increasingly reduced in length until they reach a grain size of 50ms or less, at which point hundreds to thousands of these samples are layered over one another to form a timbre that resembles pink noise from a spectral perspective. This composite noise is then very narrowly filtered to a series of frequencies to a degree that the waveform of each band is nearly sinusoidal, with rapid and minute variations in amplitude stemming from the speech patterns of the two vocalists. This process is underscored by a number of other vocal samples which have been manipulated primarily via time stretching to form low frequency pedals that serves the purpose of mirroring narrative aspects of the source texts. The second movement, Stories forgotten, uses the sonic character of the end of the first movement as its starting point. It begins with a series of narrowly filtered clouds of granulated sound underscored by a recitation of the Dickinson poem by six vocalists. This is comparatively static and should be clearly perceptible to the audience - albeit still manipulated, primarily by means of time stretching and splicing of phonemes. This movement was conceived as a clear background and foreground texture; the foreground consists of occasional statements of clearly audible text fragments while the background texture consists of a succession of sustained frequencies. Whereas the filtered frequencies of the first movement are drawn exclusively from an even-tempered tuning system, the frequencies of this second movement are drawn from the first 1000 prime numbers, with the rationale being that ratios of prime numbers are more likely to be irrational and thus heard as less consonant. These frequencies are always presented in pairs, with the frequency ratios of those pairs being irrational ratios which slowly approach – but never reach – the consonant ratios of 4:3, 3:2, and 2:1. The end result is a gradual but inconclusive approach towards consonance, and the end effect on an audience is an almost imperceptible tension which pervades the movement. The two movements together form a narrative and a commentary on the musical nature of the human voice as it is used to articulate these texts. An additional narrative aspect, conveyed through the ordering of the text fragments, in the first movement conveys a narrator pursued through an enclosed space by a beast of their own psyche; the second movement conveys the narrator accepting their role as both victim and perpetrator and resigning themselves to an impending profound loss. An analysis of these two movements using spectrographs reveals that the initially complex timbre of the voice is gradually destroyed until its waveforms almost approach something sinusoidal, before the process reverses itself and the complexity of the voice is once again revealed. The piece is being written primarily using SuperCollider, a programming language developed in the 1990s for the procedural manipulation of sound, and also involves the use of Audacity, Logic Pro, Pro Tools, and Reason.