Grammar and information structure. A study with reference to Russian

Abstract

This thesis is dedicated to the study of information structure (IS) dividing information into given/new, salient/backgrounded etc. There is an information structure represented in the discourse mental model and an information structure encoded in the grammar, which indirectly reflects it (as the tense system of the language indirectly reflects our mental representation of time), and the interface between the two. Our conceptual system is known to be very rich, and only a fraction of the information contained in it can be expressed by the grammar. The main question of the dissertation is what information from the discourse IS system passes through the narrow channel at the interface and gets encoded in the grammatical IS system and by what means, and what information is lost? This question requires extensive research because there is little agreement both about the means of encoding in the grammatical IS system and about the meanings encoded. To answer it, IS-related word order variation and prosodic phenomena are analyzed (primarily in Russian, but also in several other languages). The following claims are the most important for this thesis. First, it argues that IS notions in the grammar are relational (such as ‘more or less accessible’: e.g. A is more accessible than B) rather than categorical (such as ‘given’ or ‘new’: e.g. A is given, B is new). An IS model capable of encoding and interpreting such notions is developed. It is based on two IS scales (accessibility and salience). The means of encoding are syntactic configurations (defined in terms of the order of merger: e.g. A is merged above B, and interpreted according to the interface rule). They result from EM and independently motivated movement or are derived by means of IS-related ‘free IM’ (as it was introduced in Chomsky (2005)). Crucially, relational IS notions cannot be encoded by means of IS features, which constitutes an important argument for configurational and against feature-based IS models. Second, prosodic IS phenomena are deduced from syntactic configurations in the proposed model. This sets it apart from other configurational IS theories (including the major ones developed by Reinhart, Neeleman and Szendröi), which rely on some form of prosodic encoding. Third, a novel model of the EPP in the Tense domain is proposed for Russian (considering EPP-driven movement is essential for this work because IS-related reorderings are ‘superimposed’ on it). This model makes an interesting addition to the EPP typology (showing that Russian is not anomalous in this respect and patterns with English). More globally, it removes one of the most significant arguments for dissociating the EPP and agreement. References:

1. Chomsky, N. (2005). On phases. Ms., MIT, Cambridge, MA.

2. Neeleman, A., & Reinhart, T. (1998). Scrambling and the PF interface. In M. Butt & W. Geuder (Eds.), The projection of arguments: Lexical and compositional factors (pp. 309-353). Stanford, CA: CSLI Publications.

3. Reinhart, T. (2006). Interface strategies: Reference-set computation. Cambridge, MA: MIT Press.

4. Szendröi, K. (2001). Focus and the syntax-phonology interface. Doctoral dissertation, University College London.