Presented at the University of California, Los Angeles, Department of Systematic and Ethnomusicology, 1998 Feb. 27.

In the training of Western musicians, the canon of harmony and voice-leading rules has been considered one of the essential foundations of the art-music craft. Of course not all composers accede to the norms of traditional harmony and voice-leading -- nor should they. Rules ought to be followed only where the composer agrees that the goal or goals implied by the rules form wothy musical objectives.

What are the goals of voice-leading? And how do the traditional rules contribute to achieving these goals? Over the past 15 years, significant advances have been made in answering these questions. In this talk I will review this research, and show how traditional voice-leading rules (such as the rule forbidding hidden or exposed octaves) can be formally derived from experimentally-determined properties of the human auditory system. The research not only informs us about the perceptual and cognitive nature of the voice-leading canon, but also informs us about different types of voice-leading -- such as polyphonic vs homophonic vs close-harmony writing (e.g. barbershop), as well as about non-Western textures such as heterophony.

Changes in part-writing practice over the centuries imply that Western music history can be interpreted from a cognitive and perceptual point-of-view. The results similarly imply that cognitive and perceptual interpretations may be proposed to account for differences in musical practices between cultures.