Sawu: a language of Eastern Indonesia

Abstract

This thesis is primarily a description of the Seba and Mesara dialects of Sawu (Chapters 1 to 7), but reference is made to other Sawu dialects. Chapters 8 and 9 place it in the wider context of eastern Indonesia. The Introduction provides a brief account of Sawu's language, speakers, islands and recent history. It also includes details of fieldwork, informants and data collected, together with a critical survey of the linguistic literature pertaining to Sawu. Chapter 1 is a phonology of Sawu which differs significantly from two earlier attempts by Radja Haba (1958) and Lee (ms). Chapter 2 delineates the distinctive characteristics of Sawu word classes. The Noun Phrase (Chapter 3) is characterised by little morphology, case prepositions and post-posed possessives and demonstratives. Common nouns are often preceded by a common article, and nouns in general can be unmarked for singular and plural. However, plural can be indicated by reduplication, and singular and plural by demonstratives. Counters are normally required for the specification of number, and quantifiers and relative clauses can precede or follow the head noun. An important section of this chapter is the detailed study of the semantic role(s) represented by each case preposition. Verbs (Chapter 4) are divided into two semantic classes: Action verbs and non-Action verbs. Like the Noun Phrase, there is very little morphology. It is restricted to verb agreement, a causative prefix, a reciprocal prefix and reduplication. Chapters 5 and 6 identify and define the large number of Sawu Clause Modifiers which include Excessive Adverbs and Particles. Sawu syntax (Chapter 7) begins by classifying verbal clauses according to the case-frames of their verbs. Non-verbal clauses are of two kinds: Interjections and Juxtaposed NPs. All clauses are, then, analysed according to their functions. We also look at negation, possession, comparison, co-ordination, complementation and deletion. Two final sections focus on the interaction of role and reference properties in the clause. The first looks particularly at word order and seeks to discover whether it is possible to predict which NP will be the leftmost. The second examines Keenan's (1976) Subject Properties and their distribution. We are able to conclude that in an intransitive clause the Absolutive Noun Phrase will be the subject and will nearly always be leftmost and that in a transitive clause there is no clearly identifiable subject and the leftmost Noun Phrase is usually Ergative or Absolutive. Ndao is usually regarded as a dialect of Sawu because of the large percentage of common lexicon. Chapter 8 examines this claim by comparing the grammars of Sawu and Ndao. It concludes that Ndao is now sufficiently different from Sawu to be regarded as a separate (but very closely related) language. Chapter 9 looks at a recent claim that Sawu and Ndao are languages with a Non-Austronesian substratum and a heavy overlay of Austronesian (mainly lexicon). I discuss the Sawu data in the light of this claim and put forward the view that there is a strong case for regarding it as Austronesian. Show more