Demonstratives, referent identification and topicality in Wambon and some other Papuan languages

Abstract In Papuan languages like Wambon and Urim demonstrative forms are used both in contexts of referent identification, e.g. as demonstrative operators in noun phrases, and in topicality contexts, e.g. as topic markers with adverbial clauses and phrases, recapitulative clauses, new topic NPs and given topic NPs. Using notions from the Functional Grammar framework (Dik, 1989), I present a non-unified account of the demonstrative forms: helping the addressee to identify referents by giving deictic hints like ‘close to speaker’ and orienting the addressee about the topical cohesion of the discourse are two separate functional domains in language. This ‘two-domain’ hypothesis, which views the demonstrative forms as having two synchronically unrelated functions, explains the fact that in Wambon and Urim the demonstratives show important differences in form and behaviour depending on whether they are used for referent identification or for expressing topicality distinctions. The ‘two-domain’ hypothesis explains such formal differences but cannot explain the formal similarities between topic markers and demonstrative operators in Papuan languages like Wambon and Urim. To explain these formal similarities I suggest a diachronic development: in several Papuan languages topic markers developed from demonstrative operators. In the relatively well-documented Awyu-family of Papuan languages this process can be traced: in Wambon, the resumptive demonstrative pronoun- eve integrated in the preceding NP as a topic marker in stative clauses with a very transparant dichotomous topic-comment structure. In Korowai, also of the Awyu-family, the clitic -efè, function as a demonstrative operator and functions solely as a topic marker