|dc.description.abstract||The aim of this study is to present an analysis of the morphology, syntax, semantics and phonology of the applied, causative and passive verb extensions in Tonga.The syntactic and semantic analysis is undertaken in the context of Government-Binding Theory whereas the phonological analysis employs Underspecification theory and Feature Geometry.The need for such a study arose out of our realisation that there was no available study on any aspect of Tonga that dealt with the four levels of linguistic analysis namely phonology, morphology, syntax and semantics in the context of a linguistic theory. This is because the available studies on Tonga have concentrated on phonology at the expence of other aspects of linguistic levels of analysis and that the few grammars that purport to discuss all aspects of Tonga do this from merely a taxonomic point of view.It was hoped that by imdertaking such a study we would contribute to the general theory of grammar as the approaches the study employs were mainly developed by observing non-Bantu languages. Since Tonga is one of the regional official languages that are taught and examined at primary and secondary school levels in Zambia a study such as ours could be used by both teachers and writers of Tonga teaching materials although this would require that the work is recast and revised to make it more accessible to people in these categories.Even if the researcher is a native speaker of the language investigated she collected the bulk of the data from various kinds of texts on Tonga such as novels, readers and grammars. Also she validated her data using other speakers of the language.A major contributibn of our study to contemporary linguistic theory is our argument that morphology is best dealt with in the lexicon and our illustration on how the verb extensions imder study in Tonga and Bantu in general can be generated in the lexicon.By and large our work has shown that although the theoretical models we employ can be applied to the analysis of verb extensions imder study in Tonga, certain aspects of these models would need to be revised without shaking the foundations of the theories.Chapter one is the introduction and outlines the statement of the problem, rationale, objectives, methodology, theoretical framework of the study, literature review and discusses some aspects of Tonga verbal morphology.vii Chapter two presents and discusses various arguments on the generation of verb extensions. In the discussion it is assumed and argued that derivation and compounding are entirely dealt with in the lexicon, while inflection is handled both by syntax and the lexicon.Chapter three gives an account of the thematic roles of applied, causative and passive verb arguments and the argument structure of applied, causative and passive sentences. It is shown that, whereas the argument associated with the applied extension can have several thematic roles, the argument associated with the causative extension and the argument associated with the passive verb has always out thematic role. Similarly, it is revealed that, whereas the argument associated with the applied extension can occupy different positions in a sentence depending on its theta-role, the argimient associated with the causative extension and the arguipent associated with the passive extension is always the external argument, occupying, that is, the sentence-initial position.
Chapter four outlines the phonological processes that affect the applied, causative and passive verbs. It is noticed that these are in two categories. There are those that apply to the verb extensions and those that apply to some other elements in the extended verb due to the presence of the extension.
The examples given in this study, are numbered separately for each chapter. Notes are given at the end of each chapter. We have adopted two categories of nimibering for this study. The appendices are numbered in Roman numerals while the main work is numbered in Arabic numerals.||en_US