9:30 AM - 11:00 AM
Group discussion: Katz, J. (2015). Hip-hop rhymes reiterate phonological typology. Lingua, 160, 54-73.
11:00 AM - 12:30 PM
Language Variation and Change Group
Practice talks for NWAV 44:
Paulina Lyskawa (University of Maryland), Emilia Melara (University of Toronto), and Ruth Maddeaux (University of Toronto): "Heritage speakers abide by all the rules: Evidence of language-contact effects in Heritage Polish word-final devoicing."
Emilie LeBlanc (York University): "Vraiment vraiment intense: The use of intensifiers in Acadian French adolescent speech."
12:30 PM - 2:00 PM
Raj Singh (Carleton University): "On the preference for global over local accommodation"
It is commonly assumed that presuppositions typically project out of embedded positions (“global accommodation”). Under restricted conditions, this projection can be blocked, in which case we say that the presupposition has been cancelled (“local accommodation”). To the extent that the generalization is correct, it remains unexplained. However, we will point out ways in which the generalization has been misstated. A further complication is that theories typically invoke several mechanisms to make sense of the data. For example, the satisfaction theory involves at least a projection component, a strengthening component (to deal with the proviso problem), and a cancellation component (local accommodation). Similar concerns can be raised for other approaches.
The goal of this talk is to work toward a more descriptively adequate generalization of the relevant preference principle, and to suggest that the revised generalization hints at a principled explanation and a reduction in the inventory of presupposition-specific mechanisms. The key insight is that pragmatic presuppositions are determined using a set of alternatives derived by taking the projected presuppositions of alternatives to the utterance. Assuming this, default `global accommodation’ amounts to selecting the maximal subset of alternatives (the entire set), and `local’ accommodation amounts to selecting the minimal subset (the empty set). This eliminates the need for a cancellation mechanism, and converts the “global over local” question into one of why “max” is preferred to “min”.