On the Tension between Belief Formation and Argument Evaluation

Abstract

There exists certain circularity between the formation of some epistemic atti- tudes of an agent and the way she assesses her available arguments. For the sake of simplicity, I will restrict my attention to the case of beliefs in what follows. The mentioned circularity arises when one tries to embrace two principles that, taken separately, seems to be intuitively acceptable: 1. When an agent assesses her available arguments, she should take into account her beliefs with respect to the premises. In this sense, arguments with believed premises should be taken to be stronger by the agent than arguments whose premises are not believed. 2. The beliefs of an agent should be partially determined by the evaluation she performs of her available arguments. To be more precise, if an agent is considering her doxastic attitude towards a proposition P, then she should rst assess her available arguments about P and form her belief consequently (for instance, by believing P if her strongest argument is in favour of P) The tension between both principles becomes clearer when one compares works in formal argumentation (endorsing the rst one) to works in epistemic logic (endorsing the second one). During my talk, I will propose a formal model that allows us to (consistently) endorse some version of both principles. After that, I will explain that the epistemological view that underlies this model corresponds to some sort of foundationalism. In this view of founda- tionalism, we can distinguish between basic beliefs and argument-based beliefs. Within the second class, another distinction can be drawn between strong beliefs (based on deductive arguments) and weak beliefs (based on defeasible arguments). To conclude, I will point out what kind of epistemic actions can change each type of beliefs.