Katedra logiky a metodológie vied pozýva na dvojicu podujatí s prof. Jonom Williamsonom (https://jonwilliamson.uk/) z University of Manchester.

Workshop on Idealizations

Termín a miesto konania: 4. 2., 13:00 – 16:30, G140

Jon Williamson: The role of idealisation in the epistemology of causality

Abstract: In this talk, I shall describe three strategies for pursuing the epistemology of causality. One strategy is to make strong theoretical idealisations and offer methods that guarantee success in cases where those idealisations hold. For example, Donald Rubin's potential outcomes framework and Judea Pearl's graphical causal modelling framework have both led to very fruitful streams of theoretical results. A second strategy is to make strong metaphysical assumptions and idealisations to motivate theories of causal enquiry. For example, probabilistic theories of causality can be used to motivate certain statistical approaches to causal inference, while agency or interventionist theories of causality steer one towards specific kinds of causal models, and dispositional or mechanistic theories are sometimes used to motivate methods such as process tracing. A third strategy is to try to avoid strong assumptions and idealisations in the hope of developing accounts of causal enquiry that are more general and less controversial, but nevertheless informative. The evidence-based evaluation movement can be thought of as situated in this camp, as can Evidential Pluralism, a recent theory of causal enquiry. I shall argue that substantial progress can be made by pursuing the third strategy.

Lukáš Bielik: Idealisations in Logic and Epistemology

Abstract: Idealisations and idealised assumptions are part of scientific models. When evaluating competing scientific models in relation to a target system, the standard consideration is to consider the question of how well the model provides accurate predictions or how well it can explain the phenomena of the target system (Weisberg 2013; Frigg and Hartmann 2024). These models contain so-called descriptive idealisations. On the other hand, normative disciplines such as logic, decision theory, or (formal) epistemology come with normative models of how we should reason, make decisions, or change our beliefs respectively. Part of normative models are so-called normative idealisations (Colyvan 2013). The aim of my paper is to show in what sense normative idealisations differ from descriptive idealisations, what is the relationship between a target system and a normative model, and how one can think about the preference of one normative model over another, competing model. Using examples of both normative and descriptive idealisations from both logic and formal epistemology, I show that the preference of one normative model over another is a matter of instrumental justification.

Pozvaná prednáška: An Objective Bayesian Approach to Inductive Logic

Termín a miesto konania: 5. 2., 13:30 – 15:30, G239

Jon Williamson: An objective Bayesian approach to inductive logic

Abstract: The objective Bayesian approach to inductive logic appeals to Edwin Jaynes' principle of maximum entropy, which holds that one should use the probability distribution with maximum entropy, from all those that fit the evidence, to draw inferences, because that is the distribution that is maximally non-committal with respect to propositions that are underdetermined by the evidence. This talk provides an introduction to the logic and its motivation, explaining how it overcomes some problems with Carnap's approach to inductive logic and with the subjective Bayesian approach. It also describes a range of recent results that shed light on features of the logic, its robustness and its decidability, as well as methods for performing inference in the logic.