Metre and Melody in Dinka Speech and Song

In most European languages, lexical and grammatical distinctions are conveyed by sequences of consonants and vowels ('segmental phonemes'), which are readily represented in alphabetic writing. However, many aspects of linguistic sound systems - pitch, tone of voice, tempo - cannot be so readily written down; for example, the full stop at the end of an English sentence crudely represents a whole range of utterance-final effects, including lowered pitch, slowed tempo, and in some cases creaky voice quality. These 'prosodic' or 'suprasegmental' phenomena are easily overlooked, not only because they do not affect the meaning of individual words, but also because they are poorly reflected in writing. They are also the features most obviously overridden in song, which imposes its own metre and melody on the rhythmic and melodic patterns of speech. For these and other reasons, suprasegmentals, though integral to spoken language, are in some sense 'beyond text'.

Nevertheless, in many languages suprasegmental features function just like consonants and vowels to convey distinctions between words. The best-known cases are 'tone languages' like Chinese, where the difference between e.g. hua 'flower' and huà 'speech' is conveyed only by voice pitch. Such features pose a challenge for alphabetic writing, and therefore often also remain 'beyond text' despite their important role in meaning. They also pose a problem for song, because they create a conflict between the demands of the musical form and the linguistic meaning the song is intended to convey - not merely in mismatches between the musical mood and the overall meaning of the text, but in the very direct potential influence of pitch patterns and note durations on which words are expressed. Studying song and writing in a tone language thus provides interesting opportunities to refine a more general (i.e. less alphabetic, less Eurocentric, and less literary) notion of 'text'.

In this project we propose to study suprasegmental features in Dinka, a language of Southern Sudan. There are several reasons for choosing Dinka: it is a thriving language (some 2m speakers) with one of the richest known suprasegmental systems in the world; composing songs is a vibrant part of the Dinka cattle-based socio-cultural system and songs are used to communicate about all aspects of social, pastoral, and spiritual life; literacy in Dinka is limited and the orthography is still far from standardised, so that written Dinka is quite variable and thus potentially provides insight both into the suprasegmental system itself and into the socio-cultural contexts of writing and text creation. Moreover, the language is spoken in a variety of environments, from traditional pastoralist groups in the Dinka homeland, through urbanised groups in towns in the south, to significant communities of civil war refugees in Khartoum and abroad. The resulting variability in speech, writing and song is central to this proposal: our larger goal is to contribute to an understanding of how the non-textual suprasegmental properties of the Dinka language are mediated in different formats (speech, song, writing), and how this mediation is influenced by factors ranging from low-level properties of auditory and visual perception to the physical and cultural settings in which texts are created.