Fwd: s éance du jeudi 6 février 2014 14h salle 533, 5è =?ISO-8859-1?Q?me_=E9t._?=Olympe de Gouges

[carmen sorin]

Excuser les envois multiples

carmen.sorin at linguist.univ-paris-diderot.fr

le 6 février à 14h, Jeremy Kuhn (NYU) nous parlera de

*Massy yet chunky: All quantifies over covers*

The quantifier *all* leads a dual life (Dowty 1987): in some cases ("All
the students touched the ceiling"), it necessarily distributes the
predicate down to individuals; in other cases ("All the students
gathered"), the predicate seems to hold of the group as a whole. This
paradox is perhaps most famously evident in the distinction between the
predicates *gather* and *be* *numerous*: both are intuitively "collective
predicates," but only the former is grammatical with *all*.

In fact, however, the entailment patterns are not homogenous even among
*gather*-type predicates: depending on contextual plausibility and lexical
semantics, predicates may be distributed down to every single pair of
individuals ("All the students disagreed"), or merely down to geometrically
salient pairs ("All the students held hands").

Based on this observation, I propose an analysis in which all quantifies
over contextually salient covers of the NP. Of relevance, there is a close
relationship between this proposal and Champollion 2010's definition of
Stratified Reference: rather than requiring exhaustive subdivision, both
merely assert the existence of *some* way of dividing the event into
sufficiently small parts. It turns out that this extension of Champollion
2010 successfully provides an entailment-based diagnostic to distinguish
*gather*- and *numerous*-type predicates: only the former have Stratified
Reference over participants.




carmen.sorin at linguist.univ-paris-diderot.fr
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/parislinguists/attachments/20140203/df894d6a/attachment.htm>