Hypothetical Universe
Hypothetical Universe: a functionalist critique of Lerdahl-Jackendoff
Presented at the Society for Music Perception & Cognition (SMPC) conference, Las Vegas, Nevada, June 2003.
Abstract
This paper presents a critique of some assumptions from generative linguistics as applied to music, and proposes alternative and complementary approaches derived from functional linguistics, providing a less hypothetical approach that may prove more consistent with actual lived experience. Lerdahl & Jackendoff’s 1983 book “A Generative Theory of Tonal Music” (GTTM) has stood for twenty years as required reading for work at the border of music and language. However, the reliance in their theory on musical scores, rather than the sound source implicitly assumes a direct and consistent correlation between a score and a listener’s perception. While this is consistent with work in generative linguistics, this assumption is not warranted, as it ignores the stage of production that necessarily intervenes. In short, no accommodation is made for the difference between reading music in silence and hearing a performance.
Evidence from analysis of sound examples presents systematic discrepancies between scores and performances of those works. If a cognitive theory is to describe music as it is experienced, it is essential to accommodate these discrepancies. While experiments can and have exploited computer-generated sound in precise relationship to the score, this approach fails to acknowledge that surface variation in the sound is an inherent part of the normal human experience of music.
While the methodological choice in GTTM of seeking to describe end-state analyses rather than on-line temporal processing is noted, such a choice limits the scope and relevance of the theory. In relying heavily on written documents rather than sound objects, this theory becomes more a tool for understanding analytical and interpretive intuitions than a cognitive theory of sound perception. The current paper shows how acoustical analyses of various performances create problems for the hypothesized “experienced listener,” rendering a theoretical framework based on readings from a score suspect. The musical examples are drawn directly from those in GTTM.
While the strength and value of experimental approaches is acknowledged, analysis from real-life examples can supplement and enrich experimentalism, and provide a needed counterbalance to a purely experimental tack.
Presentation files:
Hypothetical Universe text of talk (pdf)
Hypothetical Universe data (xls)
Hypothetical Universe PowerPoint slideshow (ppt)
Sound Files:
