We are happy to announce a talk by Nadine Bade (University of Potsdam) in the Semantics Colloquium.
The talk will take place on campus in IG 4.301.
Title: Word learning with visual animations as a window into the Triggering Problem — introducing a new paradigm
Date: May 5
Time: 4 pm – 6 pm ct
In this talk, I will discuss a new methodology using a word learning task with visual animations to approach the triggering problem for presuppositions, which is the long standing theoretical issue of predicting which entailments of natural language expressions end up presupposed. Recent literature discusses the need of an algorithm do determine what parts of meanings become presuppositions, that is an explicit rule that predicts a trivalent output (T,F, #) of a bivalent input (T, F) (Abrusán, 2011; Schlenker, 2021). Earlier arguments involving iconic presupposition triggers suggest that presuppositions are generated productively from iconic expressions, which could not encode the presupposition conventionally (Schlenker, 2019; Tieu et al., 2019; Schlenker, 2021). We conducted two pilot experiments using a word learning task with a change of state verb to argue for the success of the methodology in displaying productive lexical triggering in action. Our data show that, in line with most theories on triggering mechanisms with change of state verbs (Abusch, 2010; Abrusán, 2011; Schlenker, 2021), the initial state is treated as a presupposition. This is evidenced by the behavior under negation (and quantifiers), but also by the fact that training against the predictions of triggering algorithms is less effective than training for it. This last finding is a novel finding, revealing that training with visual communication can effectively be used to answer theoretical questions regarding the triggering problem.