Focus on topic: information structure in the formal variety of Indonesian

Abstract

The correlation between form and function is the subject of a huge debate in linguistics. This thesis studies this correlation by investigating sentences (utterances) and their formal features, as in using language, people not only think about what to say – the informative content, but also how to order it – the structure. The study is a piece of research into the relationship between units of information conveyed in sentences and their morpho-syntactic devices (constructions, constituents and domains). It deals with important concepts in information structure such as topic, focus, topicalization and focalization. The data I use are the formal variety of Indonesian obtained from three political speeches on Pancasila. The data are categorized into canonical and non-canonical constructions, as the two have different information structure. In the former construction, a linguistic truism, that topic precedes focus, is argued to be relevant. It is also argued that conspiracy of syntax in the construction prevents new information from being placed in sentence initial position. On the other hand, in the latter, a different information structure is very likely to occur. Basing the analysis on Lambrecht (1994), on the one hand, I argue that in canonical constructions, topics strongly correlate to subjects. Only a few instances in the data show that subject is not topic. When this is so, they either refer to focus with the type ‘argument focus’ structure or refer to a part of a focused constituent with the type ‘sentence focus’ structure. On the other hand, I argue that in non-canonical sentences there are two constructions under scrutiny: word order restructure (preposing) and passive construction. I argue that their information structure is predictable to some extent. First, in preposing, NP object and PP is topicalized while VP and AdjP predicates are focalized. In addition, the preposed intransitive VP ada is used to show that subject and predicate are focused in a presentational sentence. Next, in yang sentences, the predicate may be topicalized or focalized, depending on the context where they occur. Second, in passive type 1, the subject is assigned either aboutness topic, continuing topic or new topic. If the VP predicate is preposed, it is part of the focused constituent. In passive type 2, the object is fully topicalized. I conclude that the correlation between, (i) subject and topic in canonically ordered constructions of formal Indonesian, (ii) different types of topic and subjects in passive type 1, and (iii) object and topicalization in passive type 2, are all predictable, rather than accidental or random. This, however, does not negate the possibility of individual variation by the language user. On the contrary, I draw no conclusion about the relationship between preposed predicate and focalization in the non-canonical construction. Thus, at this point, the conclusion is tentative and further research is required.