Tonal adaptation across musical modality: A comparison of Sambla vocal music and speech surrogates

This talk presents a first systematic comparison of tonal textsetting in vocal music and instrumental surrogate speech for a single language, Seenku, spoken by the Sambla people of Burkina Faso. The Seenku tone system boasts four level tones and multiple contour tones. This tonal density is exploited to create a musical surrogate language on the balafon, a kind of resonator xylophone; by encoding the tones and rhythm of the language on the notes of the pentatonic scale, musicians can communicate with each other and the audience without ever “uttering” a word. Mapping of tones to notes depends upon the mode of the song, with the highest tone level (super-high) typically anchored to the tonal center of the mode. Tone is also implicated in the melodies of vocal music, with Sambla songs obeying cross-linguistic principles of tone-tune association: maximizing parallel movement between tones and melody and minimizing direct opposition, where, e.g. tone goes up but the melody goes down.

We find a divide in how tone is adapted to surrogate speech vs. sung music: the balafon surrogate language encodes lexical and morphological tone, but not the output of postlexical tone rules such as tonal absorption or downstep, whereas vocal music appears to take as its input surface level of tone. I suggest that these differences follow from the communicational constraints and function of each system. Specifically, surrogate speech uses no segmental information, and so content-ful tone (lexical and grammatical) must be easily recoverable, leading to strict mapping between tones and notes and an avoidance of phonological processes that would obscure meaning. Sung lyrics, on the other hand, use all regular linguistic material (tones, consonants, vowels), allowing tone to be realized more freely. Further, sung melodies are inherently musical and aesthetically-driven, whereas surrogate speech draws on musical elements but is primarily a system of communication.

Finally, I show how the two systems are brought together in balafon renditions of sung music, yielding “surrogate song” where the balafon encodes sung tonal melodies, thus incorporating looser tonal encodings and the application of postlexical tone into instrumental “language”.