The Syntactic Expression of Information Structure and the Architecture of Grammar
The overall goal of A1 for the third phase of the SFB 632 lies in the formulation of a model for the interaction of narrow syntax and information structure. Much current work on this topic assumes the base hypothesis of the so-called cartographic approach (Rizzi 1997 et seq), where information-structural notions like topic and focus are directly represented in narrow syntax as formal features heading functional projections. The cartographic approach predicts, by design, that the effects of information structure on syntax are categorical (i.e., factor F invariably triggers structure S). However, available evidence (Keller and Alexopoulou 2001 and others) indicates otherwise –i.e., information structure effects on syntax are optional/gradient (factor F triggers structure S only in a certain percentage of cases). In order to account for the observed data distribution, we follow a model (cf. Fanselow and Lenertová 2010) where narrow-syntactic operations are triggered exclusively by morphosyntactic features and therefore cannot be influenced by information-structural factors. We refer to this as the Strong Modularity Hypothesis (SMH).
This approach raises the question of how to deal with the influences of information structure on syntactic structure (e.g., focus and topic movement, among others). Part of the project will consist in the development of an Optimality Theoretic model of grammar where the output of core syntax (a set of candidate structures) is inputted to an evaluation component. This component contains, among others, a number of constraints linking narrow-syntactic configurations and information-structural factors, which allow us to determine the degree of markedness of specific structures given specific contexts. Consequently, the influence of information structure on narrow syntax falls out from the fact that these constraints will rank some candidate structures higher (i.e., less pragmatically marked) than others. Additionally, we will explore OT models incorporating weighted constraints in order to account for the gradience and the optionality inherent to the syntactic expression of information structure.
In order to support the theoretical work just defined, we are planning a number of empirical studies (both acceptability rating and production tasks) aimed at precisely determining the strength of the different pragmatic factors involved. Note that part of the goal of A1 is to ascertain the adequacy of the SMH in languages with prosodic and morphological systems different from those investigated during the second phase of the SFB 632. As a consequence, we plan to carry out these experiments on a number of different languages.