Last week I organized a workshop with the lengthy title 'From cognitive science and psychology to an empirically-informed philosophy of logic' (more like an abstract, really). I have written on the general approach before here at New APPS, and it then generated quite some interest. Since I had to write a report for the sponsors anyway, I thought I might as well post the report here too, in case anyone is interested. So here it is!
On December 7th-9th 2010, the workshop ‘From cognitive science and psychology to an empirically-informed philosophy of logic’ took place in Amsterdam. The goal was to bring together logicians, philosophers, psychologists and cognitive scientists to discuss the interface between cognitive science and psychology, on the one hand, and the philosophy of logic on the other hand. More specifically, the goal was to investigate the extent to which (if at all), and in what ways, experimental results from these fields may contribute to the formulation of an empirically-informed philosophy of logic, one that takes into account how human agents, logicians and non-logicians alike, in fact reason.
Johan van Benthem opened the workshop presenting his thoughts on the very idea of a philosophy of logic where ‘facts matter’ (to gloss the title of one of his papers). He elaborated on how philosophy is still thought to be a purely a priori enterprise in many circles, but argued that there is hope for change! The workshop then continued with Keith Stenning on the emergence of classical logic in human reasoning. He contrasted a purely social-constructivist account of classical logic as a societal discovery/invention to a naturalistic account according to which even young children have some grasp of the principles underlying classical logic. He proposed a synthesis of these two views, illustrated with evidence from empirical studies of students' logic learning. Shira Elqayam was up next, speaking on the purported normative status of formal systems for psychological research more specifically and for the concept of human rationality more generally.
The afternoon started with a double feature of Michiel van Lambalgen and Alexandra Varga, both speaking on causal inference and reasoning in infants. Michiel introduced the event calculus, a general framework to reason about causation, and argued that it can be used to model some of the experimental data on infants. Alexandra gave continuation to the analysis, looking specifically into the psychological findings of selective reenactment in infants. The first day ended with a thought-provoking presentation by Fred Sommers, one of the founders of the ‘natural logic’ school, on what he described as a ‘logibraic’ approach to basic inferential patterns such as the inference from ‘Not all dogs are friendly’ to ‘Some dogs aren’t friendly’.
The second day began with Helen de Cruz on animal logic; she presented a comprehensive overview of the research on deductive reasoning in non-human animals, and discussed implications of these findings for the debate on animal rationality. After Helen we had Maria José Frápolli and Stavros Assimakopoulos on the concept of logical constanthood. On the basis of conceptual and linguistic considerations, they argued for the bold view that the conjunction is not a logical constant. The morning ended with Catarina Dutilh Novaes on formal languages in the practices of logicians. She argued that, more than just extending the mind (in the sense of Clark and Chalmers), formal languages in fact ‘alter’ the mind in that they may counter some of our reasoning biases.
The afternoon of the second day was a ‘Californian’ afternoon, with three talks by people coming from UCSD. The first talk was by Ben Sheredos and Tyler Marghetis, and they took Frege as their starting point to evaluate the prospects for a psychologistic approach to logic. They argued that many of Frege’s claims can be countered by recent empirical findings in cognitive science, but that some of Frege’s suggestions, in particular concerning the role of notation, are in fact corroborated by recent findings. We then had Adam Streed on expressivism as an attractive middle course between Platonism and psychologism in logic. He argued that Brandom’s version of expressivism is not amenable to empirical data, but that expressivism as such can be formulated in an empirically-informed manner. Finally, Rafael Nuñez closed the day with a talk on the prospects for an embodied grounding of logic. He drew on data from cross-cultural speech-gesture studies, among others, to argue that some of the essential ingredients of what we call logic today are already present in everyday informal conceptual systems.
The last session kicked off with Jeff Pelletier, presenting a surprising and illuminating combination of different topics he has worked on: psychologism in logic, the semantics of generic terms and default reasoning. He convincingly argued that they are all different facets of the same story, and presented data from a pilot empirical study on how people interpret generic terms. Mark Zelcer was next, on a cognitive neuroscience approach to the sorites paradox. He argued that the sorites paradox appears paradoxical to us because of our brain's architecture and arises from the conflicting interactions of neurologically distinct systems. The very final talk was by David Over, on what he described as a new paradigm psychology of conditionals. He presented the probabilistic, Bayesian paradigm, which is greatly supported by experimental results showing that people judge the probability of the indicative conditional, P(if A then B), to be the conditional probability, P(B|A).
All in all, the consensus was that it was a very successful event, and that an empirically-informed approach to the philosophy of logic is worth being further explored. To be continued…
Recent Comments