Theoretical Aspects of Bantu Tone (review)

The book contains eight papers of varying lengths (14–100 pp.) from varying theoretical perspectives. All treat the distribution of Bantu tone which, although complex, is startlingly uniform in its complexity, as laid out by these papers, at least on the eastern and southern side of the continent where the languages treated here are spoken. One paper, ‘AUX in Bantu morphology and phonology’ by ScottMyers(231–64), is, however, less focused on Bantu tone. Myers is more interested in presenting phonological (tonological), morphosyntactic, and even psychological reasons for breaking the Bantu verb complex down into the separate constituents, Aux and Verb Stem.

Of the phonology papers, three adopt optimality theory, another uses lexical phonology, and another an eclectic approach. Two less formally oriented articles focus on diachronic issues to trace the evolution of tonal patterns. All of the papers employ autosegmental mechanisms at the ‘language-of-observation’ stage, where the relevant facts are laid out for the reader. An issue which most face is that of domains and morphological conditioning, when and where rules apply, and how domains can be represented. Moving from beginning to end (the authors are arranged in alphabetical order), readers will find some repetition, both on the theoretical side, e.g. explaining align, as well as on the descriptive, explaining tone doubling and Meeussen’s Rule. Nonetheless, the painstaking and detailed analysis is impressive. Moreover, all of the authors present original data. Given the obstacles to data collection and analysis faced by Africanists, one can’t but be overwhelmed by the accomplishment.

Only some of the papers make what could be characterized as truly theoretical contributions. ‘Optimal domains theory and Bantu tonology: A case study from Isixhosa and Shingazidja’ by Farida Cassimjee and Charles W. Kisseberth (33–132) makes substantive proposals as to the universal constraints operating on Bantu tonal systems and argues for a supplement to the formal apparatus of optimality theory. ‘Tonal domains and depressor consonants in Ikalanga’ by Larry M. Hyman and Joyce T. Mathangwane (195–229) looks at the interacting phenomena of high tone spreading, a feature common to the Bantu languages treated in the volume, and depressor consonants, a feature found uniquely in southern Bantu. Depressor consonants form a class of obstruents that lower fundamental frequency, thus, ‘depress’ tones. The interaction of the two illustrates the need for several layers of rule application, as is provided by lexical phonology and prosodic domain theory. Denis Creissels’s article, ‘Expansion and retraction of high tone domains in Setswana’ (133–94), continues the detailed consideration of Tswana, turning up a typologically unusual case of tone spreading facts. Tswana has both local and nonlocal tone spreading with the former limited to one, two, or three tone-bearing units. To handle the somewhat complicated tonological facts, he proposes ‘empty syllables’ in lieu of what have often been treated with floating tones.

Two papers are diachronically oriented: ‘Tone reduction vs. metrical attraction in the evolution of Eastern Bantu tone systems’ by Gérard Philippson (315–29) and ‘Semantic/pragmatic considerations on the tonology of the Kongo noun phrase: A diachronic hypothesis’ by Jean A. Blanchon (1–32). Due to its brevity, the former does not provide enough details to prove convincing. Blanchon treats the tonal marking of definiteness in the Kongo languages and shows how it has led to case-marking, beginning with the well-known association of definiteness and subjecthood.