One of the themes of this blog, if such a thing could be said to exist, is the endeavour to see music in its wider setting (society, culture), through exploring links with other disciplines (language, science). In that regard, I’m always grateful to receive invitations to talks in Edinburgh University’s Institute for Music in Human and Social Development (IMHSD).
On Tuesday 2nd November, I attended a talk by Professor Bob Ladd entitled Suprasegmantel phonemic distinctions in Dinka speech and song. The Dinka people form the largest ethnic grouping of Southern Sudan. Allow me to quote Professor Ladd’s own summary of Dinka song tradition:
Making and singing songs is an integral part of Dinka culture. Songs are used to chronicle all aspects of individual and communal experience: to tell stories, to insult rivals or enemies, to praise family or cattle, and so on. Songs are typically sung solo or in unison, accompanied (if at all) by clapping or simple drumming. Rhythm is generally a simple regular pulse, and song segments or phrases may be of different lengths with no overarching metrical structure. Scale is uniformly pentatonic.
For those who, like me, are interested in languages but are a little vague about the vocabulary of the science of linguistics, permit me to attempt to unpack the title of the talk – Suprasegmental phonemic distinctions in Dinka speech and song:
- Segment – the individual sounds which make up speech
- Phoneme – the smallest segment is known as a phoneme e.g. the word bad has one only syllable, but three phonemes: b – a – d
- Suprasegmental – a phenomenon can be described as suprasegmental when it takes place over two or more segments e.g. prosody, tone, stress.
Professor Ladd described to us his work as part of a wider project – Metre and melody in Dinka speech and song . Specifically, he and his colleagues are exploring how a language which relies on musical phenomena (pitch, duration, timbre) for meaning is set to music. Do the two languages intuitively come together? Is there a clash of pitch and duration imperatives? If so, which one yields and when?
Three musical components of Dinka prosody (a Nilotic language) were featured:
- Tone – there are four tone phonemes – high, low, rising, falling
- Quantity – there are three lengths of vowel – short, medium & long
- Voice Quality – there are two voice qualities – modal (normal voice) and breathy (somewhere along the journey from whispering to normal speaking)
The combination of these sound options, when mixed with seven possible vowel sounds, allows for 168 possibilities, most of which occur in regular usage. At first glance, it would be impossible to believe that such a spectrum could be reduced in any way without meaning being compromised.
One further feature essential to understanding the rhythmic aspect of setting of words to music is that most stems are monosyllabic – consonant-vowel-consonant or consonant-glide-vowel-consonant.
Here is some example of such singing:
[kml_flashembed movie="http://www.youtube.com/v/lz6aPMsdY5I?rel=0" width="425" height="350" wmode="transparent" /]Despite the many musical features of this language, it would seem that linguistic constraints are over-ridden by musical ones, without any obvious loss of understanding. Professor Ladd’s own parallel with this was that we can easily understand people when they whisper, despite the loss of pitch and timbre involved.
I found myself wondering whether – given the monosyllabic nature of the language, and the prevalence of the pentatonic scale – there was a tendency to align important words e.g. verbs with structural notes of the scale (do-mi-so) and less important words e.g. prepositions with the less important ones (re-la). It seems that this hasn’t (yet) been explored.
I found this a thoroughly engaging talk, not least because it made me realise how much we take for granted in the field of word setting. Possibly, this is because our culture is one which leaves word setting to experts. I look forward to discovering more about the project.