[PLing] Talk by Eva Csipak

[Viola Schmitt]

Dear colleagues,

everyone is cordinally invited to a talk by Eva Csipak (University of 
Konstanz) with the title `Conditionalizing referentially-used definite 
descriptions'. You will find the abstract below.

When: 16.45, Friday, Nov 24
Where: Übungsraum 3, Germanistik (main building)

All the best,

Viola Schmitt


Title: Conditionalizing referentially-used definite descriptions


Abstract: Utterances containing definite descriptions can make true 
claims, even
if the descriptions themselves fail to denote ("referential use";
Donnellan 1966 and much subsequent literature). Whether the meaning of
definite descriptions is systematically ambiguous or a purely
pragmatic phenomenon remains a matter of debate.
I present empirical evidence that supports assuming referentially used
definite descriptions have a covert epistemic modal: if-clauses can
originate inside these DPs.
(1) Her spouse, if they ever got married, just walked in.
Crucially (part of) the matrix clause is taken to be true, i.e.,
someone walked in, even if the if-clause is false and "her spouse"
fails to denote. We observe that the if-clause does not receive a
standard interpretation as the antecedent of a hypothetical
conditional, and it only occurs when the if-clause is parenthetical or
postposed.
I argue that referential uses of "the a" denote a unique individual x
and presuppose that in certain epistemically accessible worlds, x is
a. The if-clause p restricts these epistemically accessible worlds to
those where p holds.
I show that this correctly predicts that if-clauses can only attach to
referentially used definite descriptions, not attributively-used ones,
only epistemic readings of the modal are possible, and counterfactual
uses are impossible.