Epistemic Akrasia and the Nature of Mental Fragments

João C. Miranda

LanCog, University of Lisbon

24 February 2023, 16:00 (Lisbon Time – WET)

Faculdade de Letras de Lisboa

Sala Mattos Romão (Departamento de Filosofia)

Abstract: An agent S is epistemically akratic if and only if (i) S believes that p and (ii) S believes that they shouldn’t believe that p. The main debate about epistemic akrasia concerns its characterization and explanation: what is needed is an account of what mechanisms make cases of epistemic akrasia possible. Greco (2014) and Kearl (2020) have defended fragmentalist accounts of epistemic akrasia: there is not one unique belief-formation system, and the possibility of epistemic akrasia rests on the possibility of conflict between the different systems. I’ll argue that their versions fail, for their responses to pressing worries about the meta-epistemological theory that underlies them – epistemic expressivism – are unsatisfactory. I will then rescue fragmentalism by proposing a version that focuses, not on the linguistic/non-linguistic nature of the systems, but on the constraints under which beliefs are formed. Doing so will allow me to appeal to well established literature in psychology about a distinction between explicit reasoning and heuristic-based reasoning (Gigerenzer, Todd and the ABC Research Group, 1999; Kanheman, 2011). I’ll argue that, not only is there better empirical support for my version of fragmentalism, but it also avoids the problems that plagued Greco and Kearl’s account.