A reworked version of this article can be found here.
Playing on one of my Kalimbas, I realized what might be the reason that there seems to be a preference for somehow rough and smoky timbres in African music. Many instruments have special devices that create a buzzing or otherwise “noisy” sound, totally against the European ideals of a clean and clear sound. The Kalimba I was playing (that is a lamellophone) has two wholes on the back side that enable me to create a “wawa” effect. A little chain on top of the lamellae makes a rattling noise. Another of my lamellophones has a crown cap fixed to it that produces a buzzing sound.
Some African Xylophones have gourds hanging underneath with holes covered with a membrane made from spider silk. Some harps have devices made from lizard scales fixed next to the string so the vibrating string will touch them, producing some kind of buzz or rattle.
On one hand, these devices make the sound louder. But I think there is more to it.
Most languages of the Niger-Congo language family are tonal. This means that there are words or grammatical forms (or both) distinguished only by intonation.
Consider you would speak such a language. If you hear a conversation dampened through a wall, you might only be able to distinguish the changes of the tone pitch. For a single word, there might be hundreds or thousands of words with a given intonation pattern, but if you hear a longer utterance, this ambiguity will go away. You might be able to understand the sentence from the pitch pattern alone. Our brains have a remarkable ability to reconstruct language in a noisy environment or when part of the speech is damped away.
Africans have used this effect in long distance communication in drumming signals. The drummer, normally using something like a two-tone slit gong, will drum the intonation pattern of a message. The hearer, with some training, can learn to understand it even if the information conveyed by consonants and vowels is lost. This type of communication was used, for example, in the ancient Congo Empire to transmit messages to and from the capital, enabling its rulers to get and send information with an unprecedented speed. Yoruba drummers use a similar effect with their talking drums to recite praise poems or other texts.
In a tonal language, a piece of music played on an instrument like a xylophone, a lamellophone or a string instrument might evoke language associations in the listener by imitating tonal patterns of language.
This is where, I think, the buzzing sounds come in. If the sounds are noisy, containing a broader range of frequencies, the listener’s perceptive system can more easily perceive it as language. Such instruments will “talk” or rather “whisper” better, providing more acoustic material from which the listeners brain may form the impression of consonants or other language sounds. This is, at this time, just a hypothesis, but it could be tested in the lab by playing more or less “buzzy” melodies to speakers of such languages and ask them to associate language with it.
Interesting aesthetic effects are possible based on the connection of music to tonal languages. The text of a song will normally constrain the melodies possible with it, so a composer will chose a fitting melody. The ethno-musicologist Gerhard Kubik in one of his articles describes an interesting example from Uganda, where a musician (Evaristo Muyinda) plays the “Enanga” harp. The song is about a historical battle (The Battle of Nsinsi), so the main topic is war. The starting line is translated “The battle of Nsinsi killed men…”. Now the composer chose a melody fitting this text. Then he composed a second voice that musically fits to the first one according to the musical principles of this culture. The interesting effect is now that the second voice that was chosen for purely musical purposes now evokes bits of text that are totally unconnected to the text of the base line. These text lines are, in the performance of the song, interspersed into the song. They translate: “She is beautiful, Nanjobe is beautiful” and “the civet cat got cubs”. So the music evokes unrelated text passages from totally different thematic areas like love or nature that give a strong poetic contrast with the text about the war, but is connected through the music. This obviously results in aesthetic effects that can only be fully appreciated by a native speaker of a tonal language, but which we can understand at least on a theoretical basis.