Igor Douven

It is known that evidential support, on the Bayesian definition of this notion, is intransitive. According to some, however, the Bayesian definition is too weak to be materially adequate. This paper investigates whether evidential support is transitive on some plausible probabilistic strengthening of that definition. It is shown that the answer is negative. In fact, it will appear that even under conditions under which the Bayesian notion of evidential support is transitive, the most plausible candidate strengthenings are not.

It has seemed natural to model phenomena related to vagueness in terms of graded membership. However, so far no satisfactory answer has been given to the question of what graded membership is nor has any attempt been made to describe in detail a procedure for determining degrees of membership. We seek to remedy these lacunae by building on recent work on typicality and graded membership in cognitive science and combining some of the results obtained there with a version of the conceptual spaces framework.

A common objection against deflationist theories of truth is that they cannot do justice to the correspondence intuition, i.e. the intuition that there is an explanatory relationship between, for instance, the truth of ‘Snow is white’ and snow's being white. We scrutinize two attempts to meet this objection and argue that both fail. We then propose a new response to the objection which, first, sheds doubt on the correctness of the correspondence intuition and, second, seeks to explain how we may nonetheless have come to have that intuition.

According to van Fraassen, constructive empiricism yields a better account of science than does scientific realism. One particularly important advantage van Fraassen claims his position to have over scientific realism is that the former can make sense of science without invoking (what he calls) pre-Kantian metaphysics. In the present paper the consistency of van Fraassen's position is put in doubt. Specifically, it will be argued that van Fraassen faces the paradox that he cannot do with nor without the pre-Kantian metaphysics he abhors.

External world skeptics are typically opposed to admitting as evidence anything that goes beyond the purely phenomenal, and equally typically, they disown the use of rules of inference that might enable one to move from premises about the phenomenal alone to a conclusion about the external world. This seems to bar any a posteriori resolution of the skepticism debate. This paper argues that the situation is not quite so hopeless, and that an a posteriori resolution of the debate becomes possible once it is recognized that the skeptic holds overly defensive and ill-motivated positions vis-à-vis both evidence and inference, and that more reasonable ones are available. In stating these more reasonable positions, as well as in showing how they make possible an a posteriori resolution of the skepticism debate, the paper draws on the machinery of Bayesian epistemology.

The two-envelope paradox is a paradox in game theory. This paper offers a solution to it in three steps. The first step is to recognize that the paradox is dependent on how one represents the game-theoretic situation that gives rise to it. In the second step, we note that the representation used in the paradox violates a familiar symmetry requirement. And in the third, we derive the solution from the symmetry requirement plus the given that we are dealing with a zero-sum game. © 2012 Elsevier B.V., All rights reserved.

According to Stalnaker’s Hypothesis, the probability of an indicative conditional, $\Pr(\varphi \rightarrow \psi),$ equals the probability of the consequent conditional on its antecedent, $\Pr(\psi | \varphi)$. While the hypothesis is generally taken to have been conclusively refuted by Lewis’ and others’ triviality arguments, its descriptive adequacy has been confirmed in many experimental studies. In this paper, we consider some possible ways of resolving the apparent tension between the analytical and the empirical results relating to Stalnaker’s Hypothesis and we argue that none offer a satisfactory resolution.

According to the deontological view on justification, being justified in believing some proposition is a matter of having done one's epistemic duty with respect to that proposition. The present paper argues that, given a proper articulation of the deontological view, it is defensible that knowledge is justified true belief, pace virtually all epistemologists since Gettier. One important claim to be argued for is that once it is appreciated that it depends on contextual factors whether a person has done her epistemic duty with respect to a given proposition, many so-called Gettier cases, which are supposed to be cases of justified true belief that are not cases of knowledge, will be seen to be not really cases of justified belief after all. A second important claim is that the remaining alleged Gettier cases can be qualified as cases of knowledge. This requires that we countenance a notion of epistemic luck, but the requisite kind of luck is of a quite benign nature.

In recent years, various computational models have been developed for studying the dynamics of belief formation in a population of epistemically interacting agents that try to determine the numerical value of a given parameter. Whereas in those models, agents’ belief states consist of single numerical beliefs, the present paper describes a model that equips agents with richer belief states containing many beliefs that, moreover, are logically interconnected. Correspondingly, the truth the agents are after is a theory (a set of sentences of a given language) rather than a numerical value. The agents epistemically interact with each other and also receive evidence in varying degrees of informativeness about the truth. We use computer simulations to study how fast and accurately such populations as wholes are able to approach the truth under differing combinations of settings of the key parameters of the model, such as the degree of informativeness of the evidence and the weight the agents give to the evidence.

To save antirealism from Fitch's Paradox, Tennant has proposed to restrict the scope of the antirealist principle that all truths are knowable to truths that can be consistently assumed to be known. Although the proposal solves the paradox, it has been accused of doing so in an ad hoc manner. This paper argues that, first, for all Tennant has shown, the accusation is just; second, a restriction of the antirealist principle apparently weaker than Tennat's yields a non-ad hoc solution to Fitch's Paradox; and third, the alternative is only apparently weaker than, and even provably equivalent to, Tennant's. It is thereby shown that the latter is not ad hoc after all.

According to Putnam the reference of natural kind terms is fixed by the world, at least partly; whether two things belong to the same kind depends on whether they obey the same objective laws. We show that Putnam's criterion of substance identity only "works" if we read "objective laws" as "OBJECTIVE LAWS". Moreover, at least some of the laws of some of the special sciences have to be included. But what we consider to be good special sciences and what not depends upon our values. Hence, "objective laws" cannot be read as "OBJECTIVE LAWS". It follows that the reference of natural kind terms cannot be fixed by the world, not even partly. The final conclusion applies to a variety of realisms.

In a famous critique, Goodman dismissed similarity as a slippery and both philosophically and scientifically useless notion. We revisit his critique in the light of important recent work on similarity in psychology and cognitive science. Specifically, we use Tversky’s influential set-theoretic account of similarity as well as Gärdenfors’s more recent resuscitation of the geometrical account to show that, while Goodman’s critique contained valuable insights, it does not warrant a dismissal of similarity.

Inductive logics purport to specify, for any given hypothesis and any given evidence statement, whether and, if so, to what extent the evidence statement should bear on our confidence that the hypothesis is true. If we agree that there can only be one true answer to questions of this sort, then the project of inductive logic faces a serious difficulty, namely that the many different systems that have been proposed in the literature rarely reach an unanimous verdict. In this paper I investigate the possibility of settling empirically the question which of all the extant inductive logics is the correct one, or, if the correct inductive logic should still await formulation, which of the extant inductive logics is the least incorrect. My main point will be that the fact that empirical investigation of inductive logics necessarily requires the use of some inductive logic at the metalevel does not render the whole project futile. I also discuss and eventually dismiss some well-known Bayesian arguments to the effect that the debate about the correct inductive logic can be settled on a priori grounds and has in fact been settled in favour of Bayesianism.