Topic, Focus and Quantification

Abstract

In this working paper I explore the possibility of fruitfully combining some aspects of the contemporary Prague School perspectives on topic and focus (Haj i�ov4 and Sgall et al) , and other contemporary work such as Rooth's, Krifka's, and Kratzer 's on focus -sensitive constructions , with the kind of analysis of quant ificational structures found in the work of He1m and Kamp.In particular I am interested in seeing how far one can push the thesis that topic (or probably rather "focus framen) corresponds to restrictive clause (or domain restriction) and focus to nuc lear scope in tripartite structures.In surveying a range of focus -sens itive constructions , we ob serve that most of them are quantificational in some sense , and require something like a tripartite structure for the ir interpretation .I pursue the suggestion that the quantificational role of focus and focus -frames is a natural extension of their discourse role : in a discourse context , the set of alternatives provided by the focus -frame locates the (nnew• ) conversational content with respect to common ground or background ; with focus -sensitive operators , the focus -frame 's set of alternatives contributes to the specification of the domain to be quant ified over (or of some analogous argument of other essentially binary operators .)I will discus s some apparent problems and possibly conflicting generalizations .I will not claim to settle the issue of the extent to which the correlation between focus -frame/focus and restrictive clause/nuc lear scope is gramm aticized in languages like English , and in fact mo st of the hard questions relating to how and where focus relates to the grammar are left open by the informal observations discussed here .