Guest Speaker: Dimitrios Skordos

We are happy to announce Dimitrios Skordos' talk (University of Calgary, https://profiles.ucalgary.ca/dimitrios-skordos) on November 7th, 3:10-4:30pm at SS560A; the title and abstract of his talk are presented below. There will also be a small reception after his talk at the department lounge. 

For questions, please contact Prof. Myrto Grigoroglou (Guest Speakers Committee) at m.grigoroglou@utoronto.ca

Reasoning with disjunction under the universal quantifier
What do we envision as the state of affairs when someone utters “Every boy has an apple or an orange”? Assuming three boys, how many of them would need to have apples and how many would need to have oranges for this to be a good description? This depends on whether one derives an exhaustivity inference from this utterance and if so what kind of exhaustivity inference they derive. Recent work (Crnič et al., 2015), looking at how adults reason with utterances like these, concludes that hearers can either be logical (derive no distributivity inference), or derive weak distributive inferences, such that at least one boy has an apple and at least one boy has an orange. Importantly, strong distributive inferences, such that no boy has both an apple and an orange and not every boy has the same fruit, are argued to be unavailable. Work on children (Pagliarini et al., 2018) suggests that they can sometimes derive distributive inferences, but it is unclear if those are weak or strong. I present new empirical data from preschoolers and adults tested under the same paradigm that provide evidence that children derive weak distributive inferences. Adults, on the other hand, prefer strong distributive inferences. I will discuss these results in the context of the literature and speculate about why children appear to be different from their adult counterparts. I will also discuss implications for language development.