Composing Speech: Investigation and Application of Musical Expression Embedded in Spoken Language

Abstract

This dissertation explores the musical potential of emotive expression in cut-up speech sounds. Cut-up is a twentieth-century technique with roots in Dadaism in which one cuts “pre-existing material into radical juxtapositions” (BBC, 2015), made popular in literature by William Burroughs in the 1950s and 60s. Speech is used primarily to communicate information relating to the world around us, but it operates sonically. Therefore, it has inherent parameters that can be manipulated to inform how information is received. The ability to manipulate the inherent sonic parameters of speech is one way in which it can be emotively coded. Sung vocals with lyrical content in music differ from speech in that the roles of information communication and the manipulation of the sonic parameters are reversed. Where speech relies on the manipulation of sonic parameters to augment or diminish the information being conveyed, sung vocals that utilise lyrical content rely on the semantic content to augment or diminish the sonic characteristics of the voice. Sung vocals could therefore be thought of as sonic utterances that are semantically coded. These inherent parameters are shared by music as it also operates in the sonic realm. The researcher used electronic music production techniques to isolate the shared parameters between music and speech (pitch, rhythm, timbre, and dynamics), and composed expressive, accessible, and engaging musical works based on these parameters. Digital music technology has the capacity to explore the limitations of sonic expression, due to its capacity to manipulate recorded sound waves. Therefore, it equipped the researcher with the necessary tools to manipulate cut-up speech sounds with compositional intent. The objective of this research was to compose musical works that drew from popular music styles, with an aesthetic focus on rich, timbrally expressive vocal material created from recordings of speech, to understand the expressive capabilities of the chosen raw material (speech sounds). The methodological procedure was to record speech from various sources, edit (cut-up) the phrases to create brief clips that were divorced from semantic signification, present the edited clips to an audience, and analyse their responses. The researcher used the insights from this analytical process to inform the use of the same speech sounds in the compositional practice. The researcher presented 26 examples (brief composed cut-ups of speech sounds) to 45 participants in a survey group and eight South African music industry professionals in one-on-one interviews. The responses yielded qualitative data that was analysed using thematic coding, followed by statistical analysis using Spearman’s rank v correlation (1904). The results provided vague answers to the primary research questions, but ultimately supplied the researcher with various qualitative interpretations of how the speech sounds expressed meaning in a cut-up context. This informed the researcher’s creative practice in the musical application of cut-up speech. Although the interpretation of the qualitative data did not result in definitive answers to the research questions, the aim of this research to explore the musical application of emotive expression in speech was achieved. The understanding that a listener experiences music in an inter-subjective and inter-contextual manner, combined with the expressive nature of the raw materials, liberated the researcher to compose expressive music without the need to know each listener’s subjective experience of expression.