890 LANGUAGE, VOLUME 73, NUMBER 4 (1997) the common characteristics of the languages described . In all languages, there is a basic aspectual opposition perfective vs. imperfective. In addition, there are three moods—realis, virtual, and necessity. In some languages, the aspectual opposition is basic; in others, the modal opposition. In all languages, tone interacts with these categories. Negation is in general sentence-final (or, in the case of discontinuous morphemes , one part of the negation morpheme is in sentence-final position), which is quite remarkable typologically. Paulette Roulon-Doko discusses the verbal system in Gbaya (25-80). She provides us with a wealth of data, not only in surface form but also in underlying form. This is especially helpful since the tonal alternations are quite involved. She includes discussions on serial verb constructions, negation, and nominalizations. Banda-Linda (81-112) is treated by France Cloarec-Heiss. She emphasizes the tonal interactions in her article. This is justified by the fact that there are three verb classes, distinguished by different tonal patterns, and that each mood has a distinct tonal pattern conditioned by these verb classes. However , this emphasis does lead to a scant semantic discussion. At times, I was unsure of the semantic distinctions between certain categories. However, a glossed folktale is included as an example. Pascal Boyeldieu discusses Yakoma (113-39). His emphasis is on the semantic side of the issue, although the tonal interactions are not neglected. There is an extensive discussion of auxiliaries, which denote various aspectual notions such as iterative, progressive, etc. An appendix gives some short texts to illustrate various modal distinctions. In Sango (141-64), treated by Marcel Diki-Kidiri , tone does not play a very important role m the grammaticalized verbal system. Rather, the different aspectual and modal distinctions are expressed by distinct morphemes. The most interesting part of the description concerns so-called neo-auxiliaries, which give us a glimpse into the history and future of the language. This section is recommended for anyone interested in grammaticalization. The final paper, again by B (165-97), is a description of Zande, the only language among the five described in which tense is part of the grammaticalized system. Tense combines with aspect and definiteness vs. indefiniteness to create a complex system. In addition , there is an extensive section on negation. This book is useful for a wide variety of scholars, from phonologists to typologists. If one wants to get a quick impression of the verbal system of the Ubangian languages, this book is ideal. For a more thorough study, one needs to refer to the individual grammars listed in the references. [Ferdinand de Haan, University of New Mexico.] A grammatical sketch of Shona, including historical notes. By Siegmund Brauner. Köln: Koppe, 1995. Pp. 66. In a grammatical sketch, the linguist expects to quickly obtain an overall feel for the way a language operates and at the same time to discover what is unusual or typologically unexpected. I am sorry to say that this book aids neither of these endeavors. In addition, the book is badly translated, presumably from German, and there is little editorial oversight (even the title should be changed, especially considering the bad translation). For example, on page 37 the reader is told, "The hiatus between two vowels is deleted by the semiconsonant IyI.' Misspellings abound, e.g. 'deminutive' for 'diminutive' (p 28), 'proclithic' for 'proclitic' (p 25), and 'tee' for 'tea' (p 61). And sometimes there is both: 'The suffigation is the modern productive way to derívate causative verbs' (56). The author's understanding of even traditional Bantu grammar (as laid out by Doke) is questionable. For example, the Bantu tense-aspect prefixes are called 'infixes' at one point (p 46ff.), while later on verb extensions (suffixes) are also considered infixes (p 55ff.). The author designates as 'ideophones' words glossed 'question marker' and 'good evening' (18). It is clear that he has not carefully read even a work he references (George Fortune, Ideophones in Shona. 1962, London: Oxford University Press.) when he says that ideophones are 'imitations of sounds' (p 62), for Fortune himselfexplains that they are much more than that. The book may be useful as a reference...