A Theory of Plausibility
Final Report Abstract
The traditional problem of information scarcity calls for methods of defeasible reasoning (default reasoning, ‘non-monotonic’ reasoning). The more recent problem of information overload, which has become particularly pressing in our ‘post-truth era’, calls for methods for the rational revision of beliefs and opinions. The theory of plausibility proposed in the project offers the resources that are needed to cope with both problems. The monograph resulting from this project develops a comprehensive philosophical theory of plausibility that unifies a model for beliefs and a model for expectations within a single mode of representation. It covers, among other things, (i) the representation of beliefs and belief states; (ii) the philosophy underlying defeasible reasoning and belief revision, interpreting plausibility as doxastic preference; (iii) rational “doxastic choices” in the face of imperfect discrimination (indistinguishability, negligibility), incomparability (incommensurability, multi-dimensionality) and context effects; (iv) reflections on the relation between qualitative representations of cognitive states and quantitative representations as well as their dynamics. The approach taken is first and foremost a qualitative approach; it covers categorical belief and categorical expectation as well as comparative strengths of belief and expectation. However, quantitative representations (probabilities, rankings) are discussed that help us understand the status of important principles of the qualitative side. The second part of the project concerned conditionals that are linked to the topic of the first part by (various variants of) the so-called Ramsey test. Recently, a number of authors have expressed dissatisfaction with the dominant analyses of natural-language conditionals as “suppositional conditionals”. The latter are true or accepted if the consequent is true/accepted on the supposition of the antecedent. But this can happen although the antecedent is completely irrelevant (or even somewhat adverse) to the consequent. In natural-language conditionals, however, the antecedent is typically meant to support the consequent, or make it more plausible. The logical form of such conditionals will thus be more complex than the suppositional theory would have it. The principal investigator of the project developed an account of “differencemaking conditionals” and “dependence conditionals” as two ways of encoding this idea, using modal and probabilistic modellings, identifying the valid logical principles and comparing this account to its strongest competitor. Keywords: Plausibility, defeasible reasoning, belief change, rational choice theory, conditionals, evidential support
Publications
-
The logic of conditionals. In E. N. Zalta (ed.), Stanford encyclopedia of philosophy. Stanford University: Metaphysics Research Lab
Egré, P. & Rott, H.
-
Evidential Support and Contraposition. Erkenntnis, 89(6), 2253-2271.
Rott, Hans
