Changing conceptions of pitch structure and timbre: A modest proposal

Two areas of musical exploration of interest to composers for which the computer is unusually well‐suited are microtonal systems and “new” timbres. Notions about how to achieve these musical extensions are doubly theory‐dependent in that they depend on both our theory of perception and our theory of the stimulus. Most extant attempts to do microtonal music have used small‐integer ratios as the fundamental description of the musical elements involved and the basic entities to which human perceivers are presumed sensitive. For timbre, the universe of possibilities has traditionally been characterized in terms of spectral variables (possibly time‐varying), and more recently in terms of projections on axes in a “subjective” multidimensional space. In the present paper, alternatives to prevailing conceptions of both pitch and timbre are considered. A symmetry‐oriented group‐theoretic approach to pitch structure that eschews ratios will be described; the resulting view of pitch systems leads to some novel ideas about how to do microtonal music on a computer. For timbre, an approach that focuses on the dynamics of the sound‐producing activity rather than on the resulting spectra will be described. This leads naturally to an interesting alternative to currently popular methods for generating timbres and “timbre transformations” on a computer.