Simha Arom and Susanne Fürniss (1992) assert that Aka polyphony is fundamentally pentatonic, yet does not operate under any sense of absolute pitch or fixed interval size. Their groundbreaking study of this phenomenon documents various instances in which the realization of a particular scale degree can vary by up to a full semitone. In order to investigate the governing principles behind this variability in pitch, Arom and Fürniss presented a group of Aka with ten variously tuned versions of a song. What they found— much to their surprise—was that the Aka accepted every single one of them as authentic. The authors therefore surmised: “the key relevant to the structure of [an Aka] song is not to be found in the scale...[but] according to the melodic contour belonging to that song” (168).

Arom and Fürniss then presented the Aka with versions of the song that went beyond variation in tuning and actually distorted the contour of the melodic line by inverting some or all of its pitches. The Aka immediately identified the alterations, thus confirming the authors’ hypothesis. This in turn led Arom and Fürniss to propose the use of pitch contour graphs in place of western staff notation in transcribing and representing Aka melody.
In her later analytical study of Aka vocal polyphony, Fürniss reiterates this idea, declaring outright: “a graphic representation as proposed in Arom and Fürniss (1992) may be closer to the vernacular conception than transcription in staff notation” (2006, 169). Nevertheless, melodic contour does not play any discernible role in her paradigmatic organization of variants for each vocal part in the song, despite being an extremely salient byproduct of the principle of commutation she describes whereby a given note may be replaced by another located a scalar neighbor, fifth, or octave away.

This paper provides an alternative paradigmatic organization of this material using Robert Morris’s (1993) Contour-Reduction Algorithm (CRA) as the primary criterion for comparison. Based on the Gestalt principle of boundary salience, the CRA deduces both a basic shape and a variable number of intermediary levels for a contour by marking peaks and valleys as structurally significant, and removing “passing tones” and repetitions in successive stages until no further reduction is possible.

This alternative paradigmatic ordering of variants suggests a new organizing principle based on two prevalent syntactic constraints: (1) new contour variants are introduced in strictly descending paradigmatic order; and (2) leaps along the paradigmatic axis only skip over variants that have already occurred. These constraints ensure that the pitch contour transformations in the sequence are introduced in a more gradual and carefully prepared manner than the quasi-random unfolding suggested by Fürniss’s orderings.

Nevertheless, it is important to note that these syntactic constraints do not represent explicit musical rules the Aka self-consciously adopt as they are learning and performing their songs. Such a thing would be anathema to the Aka, who, as Fürniss notes, rarely, if ever explicitly acknowledge even the basic parts and patterns of the polyphony. “Indeed, they are immanent concepts that are never taught to the musicians as such. Many singers don’t know them and learn about the parts only when there are too many errors in the performance” (176). As Fürniss rightly points out later on, however, “As in language, musical rules are neither necessarily explicit nor formally taught, but the fact that people don’t speak about them doesn’t mean that they don’t exist” (202).

Moreover, the overtly linear thematic developmental processes that these syntactic constraints entail—not to mention the quasi-Schenkerian, Gestalt-based analytical methodology from which they emerged—undoubtedly raises the specter of colonialist “armchair analysis”; that is, the blind imposition of western preconceptions onto a musical culture in which they have no rightful place. On the other hand, both Kofi Agawu (2003) and Martin Scherzinger (2001) have called attention to the implicit dangers of othering African musics and cultures by exaggerating their differences and minimizing their commonalities with the west, thus unwittingly undermining any genuine attempt to meet them on their own terms. Scherzinger drives the point home with a rather striking rhetorical analogy, asking “Is the concept of human rights an export imposing one sector’s views onto others?... Does the West have a monopoly on the construction of humanitarian principles?”.
Digging a little deeper into Aka musical practice, one finds a particularly vivid illustration of this dynamic at work. Again, though rarely used, it seems the Aka do have a term for the specific melodic variation technique employed in these variants: kété bányé. This literally translates as “take a small path alongside of the large way.” The metaphor resonates surprisingly well with the underlying principle of the syntactic constraints, suggesting that perhaps our western teleological inclinations might not be so unfamiliar to the Aka after all. To immediately dismiss the contour-based syntactic constraints on grounds of a de facto western bias would be just as misguided as blindly accepting them out of sheer cultural ignorance.

This paper thus proceeds to further examine improvisational process and melodic syntax in Aka musical practice through the lens of the CRA, focusing in particular on important issues surrounding considerations of cultural context, analytical methodology, and interdisciplinary research.