ABSTRACT. In this talk, I will present the core ideas and structure of my current book project, Unfinished Business. The main motivation for the project is this: Theories of justification in epistemology have focused on explaining what makes attitudes justified that we adopt as conclusions of our reasoning. But little attention has been paid to attitudes we adopt while our reasoning is still in progress, i.e., before we have finished making up our minds about how our evidence bears on the questions we want to answer.

I defend two novel claims: The first is that those transitional attitudes, which are formed during ongoing deliberations, play interestingly different roles in our reasoning and decision making than attitudes that serve as conclusions of our reasoning. Secondly, the fact that these attitudes play different roles supports the idea that they should be evaluated according to their own standard of rationality, which I call pro tem rationality.

With those new resources in hand, we can answer a variety of philosophical questions in novel ways, such as “How should we model logical learning?”, “How should we respond to higher-order evidence?”, and “Should we believe our philosophical views?”

ABSTRACT. Recently, Dardashti et al. (2019) proposed a Bayesian model for hypothesis justification by analogical inference. We show that the confirmation impact the evidence from the source system has on the hypothesis about the target system can go down when increasing the degree of similarity between the two systems in their model. We then develop an alternative model and show that the direction of the variation of the degree of confirmation always coincides with the direction of the degree of similarity between the two systems in this model.

ABSTRACT. Justification is a central aspect of dealing with one’s own and others’ epistemic claims. There have been arguments aimed at showing that claiming certainty about a given (empirical) proposition is never rationally justifiable. But that would be tantamount to rendering the expression certainty useless. The goal of this paper is to make plausible the thought that it is rational to claim cer- tainty after a finite inquiry that takes the form of deduction. Deduction, as conceived here, consists on the one hand of a question-answer-game that intro- duces premises and on the other hand of inferences. If only such questions are asked that can be answered in the given context, then the deduction using these answers in its inferences is a rational justification to claim certainty about its conclusion.

ABSTRACT. Computational modelling and simulations are often compared with experiments. It has been argued that these methods should be distinguished from experiments, that they are theoretical or that they require new epistemology. Many of these position are based on the intuition that as these methods rely on computation which can be reconstructed as series of deductive steps, the results they produce are conclusions of deductive arguments. Through a model-theoretical reconstruction of in silico experiments, I will demonstrate that the deductivist framework is not entirely adequate to capture their epistemology. Consequently, deduction is not an adequate criterion demarcating simulations from experiments.

ABSTRACT. In On Certainty, Ludwig Wittgenstein provides not only a brilliant solution to a previously intractable philosophical problem, but also the elements of a new way of approaching this and longstanding, apparently unresolvable, problems. In these notes, he reconceives the problem of radical skepticism as a kind of philosophical “disease.” His approach to the problem, which I emphasize is similar to the treatment of disease, has two main goals: (1) bring about an awareness in the philosopher this kind of extreme skepticism is not a serious methodological approach,

and, with this awareness,

(2) an attempt to replace this radical skepticism with a practical, Common Sense framework.

Implicit in Wittgenstein’s approach are strategies found in a contemporary approach to psychotherapy known as Cognitive Behavioral Therapy (CBT). As this paper will show, these strategies, along with philosophical methods, seek to diagnose and treat irrational thoughts and beliefs that often emerge in the discipline.

ABSTRACT. This paper presents an argument for the epistemic rationality of the representativeness heuristic. In this heuristic, agents rule out doxastic possibilities according to the cognitive goals they have and the tasks they are required to perform. The argument for its rationality involves appealing to two different senses of rationality: one that needs to take into account the cognitive limitations of agents and another that evaluate the results of cognitive processes regardless of agents limitations. This result allows to understand why epistemically rational capacities are still in play in cases where agents can be criticized for making mistakes.

ABSTRACT. Learning that a peer disagrees with us gives us a reason to believe that we are mistaken – states the so-called `Defeat' assumption at the core of the peer disagreement debate. In particular, learning of a peer disagreement gives us a reason to believe that we have an attitude which is either defective truth-wise – call it `alethic defeat' – or rationality-wise – call it `rationality defeat'. In an outright model of belief the mistaken attitude at the core of alethic and rationality defeat is respectively a false and an irrational belief. B ut this cannot be the case in a credal framework: after all, such a thing as a true credence does not even exist. In this talk, I investigate what the credal version of alethic defeat looks like. Firstly, I show how the question of alethic credal defeat falls within the scope of the disagreement debate. Then, after claiming that the fairly obvious appeal to inaccuracy comes with philosophically heavy commitments, I turn to what credences are taken to be for a principled answer.

ABSTRACT. Bayesians pursue a unified theory of epistemic and pragmatic rationality. A split has emerged between Standard Bayesians and Imprecise Bayesians. The latter argue that using sets of probability functions to represent beliefs is a more powerful formalism for modelling epistemological concepts like justification and ignorance.

We investigate the pragmatic side of this debate via simulations to compare their short-run performances in a classic decision problem. Our results reveal the Ignorance Dilemma: the features of Imprecise Bayesianism which make it such an epistemologically powerful representational framework for modelling epistemic states also cause it to underperform in many decision problems. We explain the trade-off and discuss some implications for Bayesian epistemology.

ABSTRACT. This paper presupposes that beliefs determined from credences by a rational belief binarization method are well-justified. To compare different belief binarization methods, we propose a new rationality norm that explicates the so-called conjunction requirement for justification. Roughly, this says that if two propositions are justifiably believed by two independent credences, then believing their conjunction is also justified. After formalizing this norm concerning belief binarization, we prove that the expected utility maximization procedure inspired by Hempel (1960) and the Camera-shutter rule (Lin and Kelly, 2012) satisfy the new norm, while the stability theory of belief (Leitgeb, 2017) does not.