Understanding Focus: Tune, Placement, and Coherence

We present a semantic analysis of focal stress and prosodic tunes in English dialogue. We use data involving two distinct tunes to argue for modelling focus and tune jointly. This allows us to combine some independently motivated principles to tackle some previously confounding problems. We model focus as triggering a presupposition (Geurts & van der Sandt 2004a), which in turn is interpreted via its rhetorical connections to the discourse context and the proffered content of the utterance that introduced it; and we assign tunes perlocutionary effects that also interact with coherence (Schlöder & Lascarides 2015). We regiment this in Segmented Discourse Representation Theory (SDRT, (Asher & Lascarides 2003)), and use the formalism to predict intuitively compelling anomalies and implicatures of various tunes in various contexts.