Chaotic Systems as Simple (But Complex) Compositional Algorithms
Rick Bidlack · Computer Music Journal · 1992
Chaos is the generic term used to describe the output, under certain conditions, of nonlinear dynamical systems. It is generally believed that what we call chaos is a behavioral characteristic found in a vast number of real-world phenomena. Chaos, in other words, describes important aspects of the way things in the real world move and change with time. As such, chaos is of potential interest to composers who work with computers, because it offers a means of endowing computer-generated music with certain natural qualities not attainable by other means. Several composers and researchers in the computer music community have recently described their work using chaos and its mathematical cousin, fractal geometry. Applications of chaotic dissipative maps to sound synthesis have been explored by Truax (1990) and Waschka and Kurepa (1989). The use of chaotic systems as generators of higher level events, such as note sequences, has been described by Pressing (1988), Di Scipio (1990), and Gogins (1991). This article is concerned exclusively with the use of chaotic systems as generators of note events. In such a context, chaos is employed as an algorithm for making choices having to do with note parameters such as pitch, dynamic level, rhythm, and instrumentation. Rather than viewing the output of chaotic systems as music in its own right, however, it is probably best to consider such output as raw material of a certain inherent and potentially useful musicality. Clearly there will be as many ways to apply chaos to musical decision-making as there are composers interested in doing so. In light of this, the discussion below emphasizes a breadth of behaviors that may be elicited from nonlinear dynamical systems, while making use of fairly neutral mappings of chaotic sequences into a generic musical space. One hopes that these examples will be sufficiently compelling in their own right to inspire other composers to develop their own more extensive and idiosyncratic manipulations of the output of the systems.