ABSTRACT
In addition to the fact that the two rhythms are quite similar to each other with respect to the exact locations of their attacks, they are in fact identical to each other if they are represented by their rhythmic contours. e rhythmic contour of a rhythm is obtained by coding the change in the durations of two adjacent inter-onset intervals using 0, +1, and −1 to stand for equal, greater, and smaller, respectively. e durational patterns of the fume-fume and son timelines are, respectively, [2-2-3-2-3] and [3-3-4-2-4]. erefore, both rhythms have the same rhythmic contour: [0, +1, −1, +1, −1]. Rhythmic contours are relevant from the perceptual point of view because humans have an easier time perceiving qualitative relations such as “less than” or “greater than” or “equal to” than quantitative relations such as the second interval is four-thirds the duration of the rst interval. It has also been found that o en the reduced information contained in the contour is su cient to eectively describe certain types of music.‡ On the other hand, two rhythms with the same contour may also sound quite dierent, as is the case for the 16-pulse and 11-pulse rhythms with inter-onset intervals [4-3-2-3-4] and [3-2-1-2-3], respectively.§ erefore, used in isolation or in a context where the intervals can vary widely, the rhythmic contour suers from
severe drawbacks as a representation from which to extract meaningful rhythmic similarity features.*
John Cherno has suggested that for all practical purposes there is not much dierence between the binary and ternary versions of the “standard” African bell pattern (fumefume) when perceived relative to their underlying duple metric beats, [4-4-4-4] and [3-33-3], and that these regular beats play a perceptually important role. However, while there is little doubt that these duple metric beats in uence perception, it is not at all clear that this in uence propels the listeners’ judgments of the two versions toward greater similarity. It may be argued to the contrary, that the duple underlying beats esh out rather than camou age their dierences. Figure 10.3 shows the binary and ternary versions of the veonset standard pattern superimposed on their duple metric structures, to more accurately examine their perceptual role. For either rhythm, imagine playing the metric beats (each highlighted with a ring) with the le hand on a bass drum, and the rhythm with the right hand on a woodblock. Let us denote with the letters R, L, and U the events consisting of striking the instruments with the right hand, le hand, and both hands in unison, respectively. While it is true that the sequence of onsets that describes the union of the metric beats and rhythm onsets yields the same alternating pattern for both the clave son and the fume-fume, namely [U-R-L-R-L-R-U], and although both rhythms start and end on the rst and last beats of the cycle, these properties by themselves are not su cient to engender greater perceived similarity. On the contrary, feeling the duple meter makes the listener more keenly aware of the dierences in the placements of the third and fourth onsets of the rhythms, which in the clave son falls squarely in the middle of the interbeat intervals, whereas in the standard pattern fall closer to the beats, creating greater syncopation. In the fume-fume pattern, the third onset is twice as close to the second beat than to the third beat, and the fourth onset is twice as close to the third beat than to the fourth beat.
