CRISSP is happy to announce a new installment in the CRISSP Seminar series:
Lecturer: Luis Vicente (University of Potsdam)
Title: Free vs. bound variables and the meaning of gaps
Date & time: Monday 22 February 2016, 15h30
Location: CRISSP/KU Leuven Brussels Campus, room 4402
Potts (2002) argues that the gaps of parenthetical “as”-clauses (and, by extension, of at least some other gap-containing parentheticals) are resolved purely grammatically. In (1), the gap demonstrably denotes a variable over propositions: given that the host of adjunction (the root clause “Ames was a spy”) is a proposition, the meaning of the “as”-clause can be computed by defining “as” as a function that allows the host to act as the binder of the gap.
(1) Ames was a spy, [as the FBI suspected ___]
In general, Potts predicts a type-identity effect between the host and the gap. In this talk, I discuss a class of “as”-clauses where this effect doesn’t hold —e.g., (2), where the gap is a variable over propositions and the host is an individual. We will see that computing the meaning of these gaps requires assuming two independent variables: first, a variable bound by the host of adjunction (as in the standard Pottsian cases) and second, a free contextual variable resolved by anaphora to a discourse antecedent. Properly, this is the same configuration we find in ellipsis sites that contain a movement trace. I will discuss how this analysis can be extend to the cases originally discussed in Potts’ work.
(2) Ames and, [as the FBI suspected ___], Hanssen were spies.