Annotation for Drake, Carolyn, and Palmer, Caroline

Accent Structures in Music Performance

Annotation (by Bill Tilghman):

This paper extends the authors' examination of three types of accent (See Drake, Dowling, and Palmer 1991) from the perceptual domain to the performance domain. Specifically, they conduct three experiments to investigate whether pianists systematically vary intensity (velocity with which the note is struck), interonset timing (the amount of time from the attack of one note to the attack of the next), and articulation (overlap between the attack of one note and the release of the previous note) in order to emphasize accent structures in the music. In the first experiment, pianists performed simple sequences containing only one accent structure. In the second, they performed the same short tunes used in Drake, Dowling, and Palmer (1991), in which accent types either coincide or are systematically displaced from one another. In the third, a single experienced pianist performed an exerpt from Beethoven's Sonata Op. 111, a more complex sequence containing both coinciding and conflicting accent structures.
Similar systematic performance variations were observed in all three experiments. Rhythmic grouping accents were the dominant factor in performance variations when the three types of accent conflicted, and the variations attributable to rhythmic grouping were consistent in all three experiments: the final events of rhythmic groups were consistently delayed, played louder, and separated (non-legato) in comparison to other events in the groups. Metric accents (first event in measure) and melodic accents (changes in contour: jumps and turns) were also articulated in performance, but the types of variation attributable to each were not consistent across all three experiments, and those related to melodic accents in particular were easily affected by the presence of other accent structures. The experimental results suggest that performers use expressive variation to highlight important aspects of musical structure, which may facilitate listeners' segmentation of the music into more analyzable units.
Keywords: