A corpus-based study of the prosody and information structure of English it-clefts and French c'est-clefts.

In English and French, clefts are a productive resource serving information structural purposes, which makes them a particularly useful device to fulfil specific communicative and rhetorical goals. This study investigates the use of cleft constructions in the two languages, first in their own right and then contrastively. Using data from comparable corpora of spontaneous dialogues, I show that clefts display a wider variation in their information structural behaviour than previously assumed. 

At the theoretical-methodological level, I develop a comprehensive framework for the study of information structural phenomena by providing i) an analytical model for the analysis of discourse-givenness/newness and ii) a fine grained methodology for the analysis of prosodically coded focus, which combines instrumental and auditory information. I give theoretical arguments for why information structural patterns have to be fundamentally distinguished from the syntactically coded specification relation between value and variable they are mapped onto. This produces an approach capable of studying the interplay between the different layers of information structure, which are distinct, yet interrelated, and the interaction between prosody and the grammatical structure of the cleft, which is not solely dedicated to (narrow) focus marking as has been argued.

At the theoretical-descriptive level, my qualitative and quantitative data-analyses establish that the two main dimensions of information structure, i.e. discourse-familiarity and prosodically coded focus, are operationalised differently in English and French clefts. In terms of the broad distinction discourse-given versus discourse-new, there are no major differences, although French clefts feature more anaphoric pronouns as value. However, focus assignment patterns contrast strongly, with it-clefts favouring narrow focus and c’est clefts broad focus. I also show that reduced clefts, which have received very little attention in the literature so far, are more than an informationally motivated variant with their own potential for achieving specific rhetorical effects.

The account presented in this thesis thus resets base-line thinking about clefts as specificational constructions enabling multiplicity and versatility of information structural realisations, whose overall profiles specialise differently in individual languages.