The workshop focuses on philosophical and methodological problems connected to causal inference. The topics covered by participants range from econometric modeling to medical research. The list of participants includes Sebastian Køhlert (How Empirical is Empirical Modelling Really? On Probabilism in Econometrics), Jan-Willem Romeijn (Shrinking and extremizing: two studies in meta-analysis), and Mariusz Maziarz (How to make inferences from inconsistent empirical results?). All the participants are welcome but please register by sending an email.

How do deliberating groups work? And can we design deliberative processes that can guarantee well-informed decisions? In this talk I will introduce, in a light way, a number of features of deliberative processes that I consider central, show their relevance for research in logic, economics and linguistics, and highlight some challenges for the development of a science of deliberative processes.

Previous studies of the neurocomputation of quantifiers have shown that complexity impacts which brain areas are activated in processing. However, there are critical issues with the previous experiments. Firstly, the formal properties of the quantifiers that are grouped together have not been properly controlled. Secondly, the poor temporal resolution of fMRI prevents studying the different stages of quantifier computation, specifically which cognitive resources are recruited at different stages. Taking the quantifier classes in Szymanik (2016) into account, this study was interested in whether the memory component required to compute proportional quantifiers ('most', 'fewest of') was reflected in online neuronal activity, as recorded by EEG. In particular, we expect that event related potentials associated with memory should be modulated differently by proportional quantifiers, compared to cardinal and Aristotelian ones. In this talk, I will present behavioral and neural data, and hope to discuss the theoretical significance of these.

The Causal Inference Lab will meet on Monday to discuss the following paper:

Nadya Vasilyeva, Thomas Blanchard & Tania Lombrozo (2018), Stable Causal Relationships Are Better Causal Relationships doi.org/10.1111/cogs.12605

Everyone with an interest in causal reasoning is very welcome to come along!

We turn our attention to two of the best papers at ISMIR 2019. The selected papers use two popular deep learning models, convolutional neural networks and transformers, to tackle loosely related tasks: predicting modality and chord transcription. The goal is to get beyond the technicalities of these deep learning models, and also discuss their assumptions and broader implications.

I explore the implications of the Tense Phrase deletion operation known as sluicing (Ross 1969) for the semantic and pragmatic literature on the Free Choice effect (Kamp, 1973; von Wright, 1969). I argue that the time-honored ‘I don’t know which’-riders on Free Choice sentences, traditionally taken to show that the effect is pragmatic, are sensitive to scope. Careful attention to such riders suggests that these sluices do not show cancellation on Free Choice antecedents in which disjunction scopes narrower than the modal.

Speakers: Mark Jago (Nottingham), Peter Hawke (Amsterdam / St Andrews), Aybüke Özgün (Amsterdam / St Andrews), Janneke van Lith (Utrecht), Johannes Korbmacher (Utrecht), Maria Aloni (Amsterdam).

The literature on question-embedding predicates has generally focused on the restrictions of certain predicates and the complements they can take as argument. However, the discussion becomes even more convoluted when we take the analysis cross-linguistically. My presentation explores a construction that exists in Spanish, but not in English, which allows a speaker to unambiguously report a question that this construction demands a more detailed analysis of question-embedding predicates. I will present such an analysis as well as introduce a test in order to highlight a correlation between a property shared by all verbs that share this construction that take this construction, which is not shared by predicates that do not. It's going to be very verby!

Abstract: A paradefinite logic is a logic that is both paraconsistent and paracomplete. In this talk we introduce a framework for models of paradefinite set theories based of Thierry Libert's work in paraconsistent set theory. We then present a model of paradefinite set theory which can be seen as the result of enriching the classical von Neumann universe of sets with various non-classical sets. We will also discuss the axiomatization of the theory of this model.