Orchestration, Perception, and Musical Time: A Composer's View

The construction of a comprehensive theory of orchestration adequate to deal with not only traditional instrumental music but also computer music has so far been hampered by two main factors: the enormous variety in the sounds available, and the exponential rate at which audio and computer technology has been developing. This paper outlines a theory of orchestration based on the relationship between sounds and their musical/temporal context that circumvents these problems. Since it is not tied to any particular system of sound synthesis, such a theory makes possible some useful generalizations that will not become obsolete with the next generation of synthesizers. Since all music involves placing sounds in organized temporal contexts, that is, in studying combinations of sounds, the musical situation sounds are embedded in is at least as significant as their acoustical makeup. The situation determines to a large extent how listeners will direct their attention. Of course, a given sound's musical use is often an outgrowth of its internal structure and character-certain sounds suggest specific contexts. For example, a loud, irregular burst of sharp percussive taps will generally not create a sense of repose. It is one purpose of this article to explore the relationship between sounds and their contexts. Before going further, I must make clear some of the limitations of this discussion. First, I will not speak in acoustical detail about sounds and their combination. An acoustical description of even an isolated sound does not adequately explain how that sound will be heard. Pierre Schaeffer points out