Pragmatic and semantic aspects of focus in Hungarian

Edgar Onea

Universität Stuttgart

edgar.onea[kukac]ing.uni-stuttgart.de

Hungarian was widely supposed to reserve a distinct syntactic projection for focus, which was to be associated with some exhaustive or identificational operator compositionally yielding a focus interpretation. Recently É.Kiss (2006) and Wedgewood (2005, 2006) suggested that a focus projection and its interpretation as a focus operator may not be necessary. Instead, a predicative projection, usually occupied by verbal prefixes, is postulated, which is supposed to trigger, e.g. by means of conversational implicatures, appropriate focus interpretation.

There are models of how an exhaustive reading of narrow focus relates to the PredP hypthesis in Hungarian, e.g. assuming structured meaning and conversational implicatures, however if one assumes that focus and focus interpretation mechanisms are language universals one would also desire a theory that integrates the semantic import of PredP into a more general focus interpretation model.

In my talk, I shall accept the PredP hypothesis and reject any compositional syntactic relevance of focus. On this basis I shall attempt to outline some ideas for answering two important questions:

1)Why a focussed constituent moves to PredP in Hungarian, if focus is not a syntactic feature?

2)How should focus interpretation be modelled, such that a general model of (free) focus interpretation could be simply applied to Hungarian data? Moreover, such a model should be able to account for both exhaustive and non-exhaustive focus interpretation, and both for pre-verbal and post-verbal focus.

My basic idea is that a great deal of underspecification is present at the interface between sentence meaning and discourse representation, thus not only classical presuppositions but most of the sentence meaning can be resolved in the same way as presuppositions. In other words, with some exceptions, it is linguistically underspecified, what is presupposed and what is non-pressuposed in a sentence. In this approach intonation generally can ease the integration of the information associated by sentences in the discourse representation by disambiguating presupposition and assertion to a certain extent. While topic intonation generally insures a presuppositional interpretation, focus includes a predicative feature, which blocks a presuppositional reading. Yet, there is no rule stating that the presence of focus would structures the meaning of the sentence (e.g. sensu Jacobs (1983)). As an answer to the first question, it will be stipulated, that the predicative feature triggered by prosodic highlighting triggers the movement of narrowly focussed elements to PredP in Hungarian. Focus interpretation itself is, however, a pragmatic implication issue, which can be modelled by a matter of exhaustivity of the focussed constituent in the given discourse representation structure. However, this reading can only arise, if the focussed element is the sole non-presupposed predication in the sentence, since the focus interpretation rule, not interacting with syntax, cannot distinguish between parts of a proposition. According to this approach, thus, exhaustive focus interpretation only occurs if the focussed element can be conceived as the only predication in a sentence, every other element insuring solely the correct integration into a discourse representation structure.