David Huron, Mei Yen Ch'ng, Kim Rasmussen, and Sarah Stockwell
Society for Music Perception and Cognition Conference, 1997.
Music theorists have long distinguished a variety of rhetorical devices used in Western compositions. One of the most pervasive rhetorical distinctions is between passages that are introductory, expository, developmental, transitional, or closing in nature. To what extent are these rhetorical categories identifiable by both trained or navie listeners?
Forty-three listeners were exposed to thirty-two, 20-sec. sound excerpts from recordings of string quartet movements by Haydn and Mozart. Participants were asked to classify the randomly presented excerpts as either expository, developmental, transitional, or closing. Excerpts were randomly selected from passages that were previously classified according to traditional sonata-allegro analysis. Excerpts were constrained so they did not start at the beginning of the analytic section or terminate at the end of the section.
Three groups of listeners participated. A "trained music" group consisted of university music students who were completing a course whose curriculum stressed the identification of music-rhetorical devices such as those found in sonata-allegro form. An "untrained music" group consisted of music students of comparable level who had only a rudimentary training in sonata-allegro form. A "naive" group consisted of non-musician university students who claimed to have little or no musical background or training and did not express particularly strong musical interests. All groups of listeners displayed significant abilities in classifying the excerpts according to rhetorical type.
Closing passages were most easily recognized; transitional passages were most poorly recognized -- and tended to be confused with developmental passages. With one exception, no differences were found between the three groups of participants. That is, musically trained and untrained listeners were found to be equally adept in their recognition abilities. This result implies that at least some aspects of musical rhetoric involve perceptual skills that are broadly shared by Western listeners.
The "trained music" participants showed a response bias against answering with the transitional category and scored at a chance level for this rhetorical type. Since transitions are comparatively brief and are considered minor rhetorical devices, this response bias is likely due to training.
In summary, listeners from all backgrounds were generally able to identify four rhetorical types from brief random musical excerpts. This suggests, for example, that there is something "expository sounding" in exposition passages -- independent of the musical context.