Local and metrical tone shift in Nguni
In this paper I argue that the Nguni languages have both a metrical tone shift rule and a non-metrical (local) tone shift rule which precedes and feeds assignment of metrical prominence (accent). As Goldsmith, Peterson, and Drogo [1986] and Peterson [1989b] have argued, a metrical tone shift rule be...
Guardado en:
| Autor principal: | |
|---|---|
| Formato: | article |
| Lenguaje: | EN FR |
| Publicado: |
LibraryPress@UF
1990
|
| Materias: | |
| Acceso en línea: | https://doaj.org/article/e1584e75c91b43fd91aa083f1dbf925e |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| Sumario: | In this paper I argue that the Nguni languages have both a metrical tone shift rule and a non-metrical (local) tone shift rule which precedes and feeds assignment of metrical prominence (accent). As Goldsmith, Peterson, and Drogo [1986] and Peterson [1989b] have argued, a metrical tone shift rule best accounts for the fact that the rightmost high tone in most words surfaces on the antepenult. Not all high tones shift to the antepenult, however; instead, they shift one syllable to the right. Earlier metrical analyses accounted for (some) of these cases by proposing rules of accent shift. The present analysis accounts for all these cases by a local tone shift rule ordered before accent assignment. This approach not only is more general and more natural than the accent shift analyses, but it also straightforwardly accounts for a number of tonal phenomena which were inadequately treated in earlier analyses. |
|---|