A Hybrid Theory of Timbre

Since the advent of electronic sound production techniques, timbre has played an increasingly important role in contemporary music. Although there have been scattered attempts at systematizing the musical possibilities of timbre (e.g. Schaeffer, 1966; Erikson, 1975; Thoresen, 2004), there is as yet no consensus "music theory" of timbre. I take it that philosophy, psychology, and aesthetics may all contribute constructively to this endeavour. The goal of this project is to juxtapose evidence from these areas in order to motivate a two-stage theory of timbre perception.

Theories of timbre perception fall into roughly two categories. Helmholtz-style theories take timbre to be a function of the spectral composition of the sound wave; Gibson-style theories take timbre to be a function of the physical event type that produced the sound wave. While psychophysical studies of timbre tend toward the former theory (e.g. the work of McAdams), the latter theory has become popular in contemporary philosophy of sound (e.g. O'Callaghan, 2007). An adequate theory of timbre perception will predict both our categorization of sounds into timbres and our assessment of similarities between timbres. While the Gibson theory does a convincing job at predicting timbre categories across arbitrary changes in the sound wave (a violin still sounds like a violin when heard from a long distance outdoors, down a long hallway, through a wall, etc.), it fails miserably at predicting timbre similarities. A babbling brook sounds like a room full of people despite radical differences in the physical mechanism producing the sound; a guitar played through a wah pedal sounds like a baby crying. On this task Helmholtz does much better.

These considerations motivate a two-stage theory of timbre, a hybrid between the Gibson and Helmholtz approaches. Primitive categorization, i.e. timbre identity, supervenes on physical event types, yet similarities between these categories supervene on similarities between waveforms. I support this two-stage theory by demonstrating how it sheds light on several case study attempts at notating timbre, e.g. Stockhausen's translation of the score of Mikrophonie I from German into English, and recent attempts to use Thoresen's sound Typomorphology to notate electronic music for performance.