Spectra as Chords
James Dashow · Computer Music Journal · 1980
This article describes a general computational principle which is applied to a variety of spectragenerating situations, thereby providing a means of deriving classes or families of chordal structures. The techniques described here are the results of a continuing effort to find coherent means of organizing the audio source materials selected for a given work or body of works. In much contemporary music, there seems to me to be a lack of a direct relationship between the composer's technical concepts and the musical effect that the chosen methods actually produce. This is particularly true of music generated by the a priori application of arbitrary theoretical constructs to the organization of audio events. Theorists have invoked everything from statistics to stochastic processes; from serialism to biorhythms; from deterministic to random procedures; often in the name of scientific and/or mathematical certitude. Thus it is arguable-and indeed, many composers would so argue-that methods of composition have only an indirect bearing on the hearing of (Lerdahl and Jackendoff 1979, p. 167). It is more than probable that one could substitute any other contemporary method for serial in the preceding observation. I do not by any means wish to imply that successful compositions are not possible with such methods. The point is that the method often has little or nothing to do with the success or even the musical content of the composition. The composer's skill, taste, and sensitivity usually come into play, shaping the theory-generated materials to correspond to musical goals. If possible, the composer will impose upon these materials (or find) such relationships that suggest that the music expresses some sort of overall structure. This structure may be hierarchical or not, but in any case it is not just an arbitrary succession of random events (I am obviously excluding from this discussion those pieces that are random, or designed to be played differently every time, or rely heavily on visualtheatrical elements). But for each piece, one has the problem not only of understanding the immediate context, but also discovering its givens (i.e., its axiomatic values )2 out of which the musical context is created. Each composer tends to have a unique set of axioms. The so-called intuitive composer usually gets right down to the task by letting these (not always well defined) axioms generate the materials according to that composer's musical needs. The theoretical composer imposes his or her axioms onto the materials derived from an a priori application of whatever theory he or she has decided to use. The axioms, as well as the composition, must be made to work. The composer is no longer working within a well-defined or well-known system whose axioms are generally understood on multiple levels. Neither are the technical procedures within the system directly related to the perceived sense of the music.