This book "supplies a vocabulary of English words and idiomatic phrases 'arranged … according to the ideas which they express'. The thesaurus, continually expanded and updated, has always remained in print, but this reissued first edition shows the impressive breadth of Roget's own knowledge and interests."
AUTHOR: Idiata, Daniel Franck TITLE: What Bantu Child Speech Data Tells Us About the Controversial Semantics of Bantu Noun Class Systems SERIES: LINCOM Studies in African Linguistics 67 PUBLISHER: Lincom GmbH YEAR: 2005
Mantoa Rose Smouse, Linguistics Program, University of Florida, Gainesville
The purpose of this book is to provide a fresh look at the semantics of noun classes in Bantu languages. The author draws from an analysis of data on acquisitional strategies employed by children and tries to link this with the historically acclaimed diachronic and synchronic perspectives. Drawing from a variety of Bantu languages, the author concludes that there is a correlation between the acquisition and evolution of the noun class system in that children do not pay attention to any semantic information in their organization of nouns. This book is concise, readable and useful as both an introduction to the study of noun classes in Bantu and a resource for potential research questions.
The book comprises of a preface, an introduction and five chapters. The preface by Gérard Philippson provides an overview of the state of research concerning noun classes in Bantu. In particular, Philippson highlights the issue of how semantic categorization continues to dominate discussions despite empirical evidence from recent studies on loanwords (Batibo, 1992) and child language (Demuth 2000) which suggest that there is a need to explore other avenues.
Chapter 1 provides a brief definition of noun class in Bantu. The remainder of the chapter is dedicated to a historical analysis of noun classes. This chapter also addresses the question of the relation between ''noun class'' and ''noun classifier'' and provides tables showing differences between the two.
Chapter 2 describes the current understandings of the semantics of Bantu noun class systems within diachronic and synchronic approaches. Although the author discusses both approaches, the diachronic works receive more attention and elaborate discussion including diagrams of semantic classifications. The synchronic discussion is limited to a paragraph. However, the author provides some interesting insights and elaborates on this issue in chapter 3.
In chapter 3, the author provides empirical evidence in support of the synchronic proposal. Based on a analysis of 5 North Western Bantu languages (although in the introduction the author says eight), the author concludes that the semantic nature of noun classes is eroding and that the phonological and morphological criteria is employed with some remnants of semantic criteria. Some demographic details and data based on each of the languages, namely Mmala, Ikota, Pove, Isangu and Inzebi are provided.
Chapter 4 deals with studies on child language acquisition in Bantu languages. It opens with a brief discussion of the role of cognitive and linguistic determinants in child language acquisition. The author analyses five acquisitional studies on Bantu noun classes (Siswati, Sesotho, Tswana, Zulu and Isangu).
Chapter 5, provides an ''experimental psychological examination on the conceptual categories of young Masangu children'' (104). The Isangu data is based on fieldwork carried out by the author. The chapter provides an outline of Isangu noun class and agreement system and shows how the acquisition of nouns by Masangu children reveals a ''principle of organization''. Further, the author demonstrates theoretical links between noun classes and ontological categories. In addition, the author addresses three questions often raised about Bantu noun classes: ''(1) the existence of prototypical class representations, (2) the innate influence of basic ontological categories on acquisition and (3) the relationship between the acquisition of these systems'' (134).
The conclusion pulls together recent formulations of the semantics of noun classes and argues that the data presented supports the view that children do not pay attention to any semantic information in the organization of noun classes. However, the author points out that , with regards to agreement markers, the only semantic feature in children’s data is the distinction between [+animate] and [-animate] which is evident in the acquisition of agreement markers. The book concludes with further questions concerning Contini-Morava’s (1997) proposal that the belonging of a noun to a category is founded on several criteria. I comment on this issue below.
The author should be commented on breaking away from the traditional approaches to the study of grammar in Bantu. Indeed, as Gérard Philippson (preface) puts it, the author presents food for thought to those interested in the study of Bantu Noun classes. There are some issues however, that the author may have overlooked.
At the beginning of chapter three, the author indicates that the synchronic semantic system is eroding. He illustrates this by providing data from five North Western Bantu languages. A comparison of the characterization of these noun class systems indicates that none of the semantic criterion holds. Having compared the noun class prefix system, the structure of the prefixes, the noun class genders and the semantics of these languages, the author concludes: ''noun prefixes do not have (anymore?) any correspondence at the semantic level...noun classes cannot (more?) be useful as the bases of categorizing nouns at the semantic level'' (48). Further, the author raises the question of whether there are common patterns of semantic evolution of the classes inside different languages. Although he does not attend to this question directly, the author notes that the system is very far from the Proto-Bantu system, and that the criterion taken into account are phonological and morphological with certain semantic survival.
In addition to the diachronic and synchronic proposals, Mc Laughlin (1997) provides a third proposal which adds to Contini-Morava. Based on a study of noun classification in Wolof, a member of the North Atlantic sub-group, Mc Laughlin proposes that a noun class system that appear incoherent from a semantic point of view suggests that noun classification should not be viewed as a result of a synchronic or diachronic process or a combination of both. Rather, noun classification, ''because it is not only an artifact of the human mind, but also an artifact of human language'' (1), ''incorporates a variety of different and intersecting parameters, namely semantic, morphological, and even phonological and sociolinguistic parameters'' (24). Given this third proposal, one can begin to attend to the question raised by the author of the patterns of semantic evolutions. From his data, all of the factors Mc Laughlin points out (with the exception of sociolinguistic factors) are evident. However, because the author subscribes to the synchronic means of classification, the third proposal is not explored.
Concerning the data on child language acquisition in chapter 4, and the evidence of the influence of linguistic factors, I find the treatment of the various languages (Zulu, Tswana, Pedi, Siswati and Isangu) very uneven. Some languages are treated in more detail. In particular, given that the author bases his conclusions about the influence of linguistic factors on acquisition on Zulu data, it would have helped to see some of that data. Furthermore, I do not think the author clearly indicates the link between the child acquisition data and the semantics of Bantu noun classes in the same way he does with the Isangu data. Nevertheless, he gives an innovative way to look into the process of noun classification.
OTHER CONCERNS The book is well organized and there is a nice flow from one chapter to the other, however there are some issues that I think can be attended to when/if the book is revised: (1) The font is very large and therefore makes the spacing appear smaller than usual (2) The Foreword ( pages 7-10) is the same as the Introduction (15-18) with the exception of a few words like 'but' and 'however' (3) Typographical errors (i) Page 86 & 87 section number 4.1 appears twice (ii) Page 88, Siswati data, imi-khwa is glossed as 'knives' in (a) and (b), but as 'white men' in (c) (iii) Page 88, paragraph, In the process...are realized by children by in... (iv) Page 89, section 4.1.3. Tswna (v) Page 90, section 4.1.5, ... As for as we know (vi) Page 129, ''the awaited answer is bolded'' (vii) Page 131, Table 5, Percentages of the right answers according to the age rackets (viii) Page 159, NP Nouminal Prefix (ix) Back cover ''The present book, which comes back on that (exiting?) subject...
Contini-Morava, E. 1997. ‘Noun classification in Swahili: a cognitive-semantic analysis using a computer database in African Linguistics at the Crossroads, edited by R.K. Herbert.
Guthrie, M. 1948. The Classification of Bantu Languages. London.
Mc Laughlin, Fiona. 1997. ‘Noun classification in Wolof: When affixes are not renewed’ Studies in African Linguistics Volume 26, Number 1.
ABOUT THE REVIEWER:
ABOUT THE REVIEWER
Mantoa Rose Smouse is currently writing her PhD Dissertation at the University of Florida. Her primary area of interest is syntax, in particular control and the theory of movement in Sesotho, a Bantu language spoken in South Africa and Lesotho. Smouse also has interest in Morphology and Second Language Acquisition. In the past few years, she worked on areas that highlight the interface between Morphology and Syntax. Her ultimate goal is to bring these three areas of research together by studying the significance of the findings from the syntax-morphology interface on second language learning.