In section 1, I develop epistemic communism, my view of the function of epistemically evaluative terms such as ‘rational’. The function is to support the coordination of our belief-forming rules, which in turn supports the reliable acquisition of beliefs through testimony. This view is motivated by the existence of valid inferences that we hesitate to call rational. I defend the view against the worry that it fails to account for a function of evaluations within first-personal deliberation. In the rest of the paper, I then argue, on the basis of epistemic communism, for a view about rationality itself. I set up the argument in section 2 by saying what a theory of rational deduction is supposed to do. I claim that such a theory would provide a necessary, sufficient, and explanatorily unifying condition for being a rational rule for inferring deductive consequences. I argue in section 3 that, given epistemic communism and the conventionality that it entails, there is no such theory. Nothing explains why certain rules for deductive reasoning are rational.

Metaphysical theories are often counter-intuitive. But they also often are strongly supported and motivated by intuitions. One way or another, the link between intuitions and metaphysics is a strong and important one, and there is hardly any metaphysical discussion where intuitions do not play a crucial role. In this article, I will be interested in a particular kind of such intuitions, namely those that come, at least partly, from experience. There seems to be a route from experience to metaphysics, and this is the core of my interest here. In order to better understand such ‘arguments from experience’ and the kind of relationship there is between this type of intuitions and metaphysical theories, I shall examine four particular cases where a kind of experience-based intuition seems to motivate or support a metaphysical theory. At the end of the day, I shall argue that this route is a treacherous one, and that in all of the four cases I shall concentrate on, phenomenological considerations are in fact orthogonal to the allegedly ‘corresponding’ metaphysical claims. An anti-realist view of metaphysics will emerge.

The paper argues that dualism can explain mental causation and solve the exclusion problem. If dualism is combined with the assumption that the psychophysical laws have a special status, it follows that some physical events counterfactually depend on, and are therefore caused by, mental events. Proponents of this account of mental causation can solve the exclusion problem in either of two ways: they can deny that it follows that the physical effect of a mental event is overdetermined by its mental and physical causes, or they can accept that the physical effect is overdetermined but claim that this is unproblematic because the case is sufficiently dissimilar to prototypical cases of overdetermination.

Nihilism is the thesis that no composite objects exist. Some ontologists have advocated abandoning nihilism in favor of deep nihilism, the thesis that composites do not exist_O, where to exist_O is to be in the domain of the most fundamental quantifier. By shifting from an existential to an existential_O thesis, the deep nihilist seems to secure all the benefits of a composite-free ontology without running afoul of ordinary belief in the existence of composites. I argue that, while there are well-known reasons for accepting nihilism, there appears to be no reason at all to accept deep nihilism. In particular, deep nihilism draws no support either from the usual arguments for nihilism or from considerations of parsimony.

We argue that giving up on the closure of rational belief under conjunction comes with a substantial price. Either rational belief is closed under conjunction, or else the epistemology of belief has a serious diachronic deficit over and above the synchronic failures of conjunctive closure. The argument for this, which can be viewed as a sequel to the preface paradox, is called the ‘review paradox'; it is presented in four distinct, but closely related versions.

I argue that free will and determinism are compatible, even when we take free will to require the ability to do otherwise and even when we interpret that ability modally, as the possibility of doing otherwise, and not just conditionally or dispositionally. My argument draws on a distinction between physical and agential possibility. Although in a deterministic world only one future sequence of events is physically possible for each state of the world, the more coarsely defined state of an agent and his or her environment can be consistent with more than one such sequence, and thus different actions can be “agentially possible”. The agential perspective is supported by our best theories of human behaviour, and so we should take it at face value when we refer to what an agent can and cannot do. On the picture I defend, free will is not a physical phenomenon, but a higher-level one on a par with other higher-level phenomena such as agency and intentionality.

Utilitarianism is often rejected on the grounds that it fails to respect the separateness of persons, instead treating people as mere “receptacles of value”. I develop several different versions of this objection, and argue that, despite their prima facie plausibility, they are all mistaken. Although there are crude forms of utilitarianism that run afoul of these objections, I advance a new form of the view—‘token-pluralistic utilitarianism’—that does not.

According to anti-atomism, we represent color properties (e.g., red) in virtue of representing color relations (e.g., redder than). I motivate anti-atomism with a puzzle involving a series of pairwise indistinguishable chips. I then develop two versions of anti-atomism.

Discussion on whether Hume's treatment of induction is descriptive or normative has usually centred on Hume's negative argument, somewhat neglecting the positive argument. In this paper, I will buck this trend, focusing on the positive argument. First, I argue that Hume's positive and negative arguments should be read as addressing the same issues (whether normative or descriptive). I then argue that Hume's positive argument in the Enquiry is normative in nature; drawing on his discussion of scepticism in Section 12 of the Enquiry, I explain a framework by which he provides what I call consequent justification for our inductive practices in his positive argument. Based on this, I argue that his negative argument in the Enquiry should similarly be read as normative in nature.

Are there disjunctive properties? This question is important for at least two reasons. First, disjunctive properties are invoked in defense of certain philosophical theories, especially in the philosophy of mind. Second, the question raises the prior issue of what counts as a genuine property, a central concern in the metaphysics of properties. I argue here, on the basis of general considerations in the metaphysics of properties, that there are no disjunctive properties. Specifically, I argue that genuine properties must guarantee similarity-in-a-respect among their instances, and must inhere in their bearers (two notions to be clarified). Disjunctive properties would fail both requirements. I compare the case of disjunctive properties with “mere Cambridge properties,” determinable properties, and functional properties, and show how my conception of properties remains compatible with determinables and functional properties while ruling out disjunctive and mere Cambridge properties.

This paper gives an account of proxy agency in the context of collective action. It takes the case of a group announcing something by way of a spokesperson as an illustration. In proxy agency, it seems that one person or subgroup's doing something counts as orconstitutes or is recognized as (tantamount to) another person or group's doing something. Proxy agency is pervasive in institutional action. It has been taken to be a straightforward counterexample to an appealing deflationary view of collective action as a matter of all members of a group making a contribution to bringing about some event. I show that this is a mistake. I give a deflationary account of constitutive rules in terms of essentially collective action types. I then give an account of one form of constitutive agency in terms of constitutive rules. I next give an account of status functions—of which being a spokesperson is one—that also draws on the concept of a constitutive rule. I then show how these materials help us to see how proxy agency is an expression of the agency of all members of the group credited with doing something when the proxy acts.

Shared activity is often simply willed into existence by individuals. This poses a problem. Philosophical reflection suggests that shared activity involves a distinctive, interlocking structure of intentions. But it is not obvious how one can form the intention necessary for shared activity without settling what fellow participants will do and thereby compromising their agency and autonomy. One response to this problem suggests that an individual can have the requisite intention if she makes the appropriate predictions about fellow participants. I argue that implementing this predictive strategy risks derailing practical reasoning and preventing one from forming the intention. My alternative proposal for reconciling shared activity with autonomy appeals to the idea of acting directly on another's intention. In particular, I appeal to the entitlement one sometimes has to another's practical judgment, and the corresponding authority the other sometimes has to settle what one is to do.

This paper explores the psychological analogues of a cluster of arguments that have played an important role in motivating a now widespread, reference-based approach in philosophy of language. What I will call the psychological analogues of Kripke-style arguments provide a substantial motivation for a reference-based approach to concepts. Insofar as such an approach is rarely given serious consideration, the availability of these arguments suggests the need for a rethinking of some foundational assumptions in philosophy of mind and other branches of the cognitive sciences.

In this paper, I begin by defending permissivism: the claim that, sometimes, there is more than one way to rationally respond to a given body of evidence. Then I argue that, if we accept permissivism, certain worries that arise as a result of learning that our beliefs were caused by the communities we grew up in, the schools we went to, or other irrelevant influences dissipate. The basic strategy is as follows: First, I try to pinpoint what makes irrelevant influences worrying and I come up with two candidate principles. I then argue that one principle should be rejected because it is inconsistent with permissivism. The principle we should accept implies that it is sometimes rational to maintain our beliefs, even upon learning that they were caused by irrelevant influences.

1. Introduction

2. Reward-Guided Decision Making

3. Content in the Model

4. How to Deflate a Metarepresentational Reading

 Proust and Carruthers on metacognitive feelings

5. A Deflationary Treatment of RPEs?

 5.1 Dispensing with prediction errors

 5.2 What is use of the RPE focused on?

 5.3 Alternative explanations—worldly correlates

 5.4 Contrast cases

6. Conclusion

Appendix: Temporal Difference Learning Algorithms

It is often alleged that, unlike typical axioms of mathematics, the Continuum Hypothesis (CH) is indeterminate. This position is normally defended on the ground that the CH is undecidable in a way that typical axioms are not. Call this kind of undecidability “absolute undecidability”. In this paper, I seek to understand what absolute undecidability could be such that one might hope to establish that (a) CH is absolutely undecidable, (b) typical axioms are not absolutely undecidable, and (c) if a mathematical hypothesis is absolutely undecidable, then it is indeterminate. I shall argue that on no understanding of absolute undecidability could one hope to establish all of (a)–(c). However, I will identify one understanding of absolute undecidability on which one might hope to establish both (a) and (c) to the exclusion of (b). This suggests that a new style of mathematical antirealism deserves attention—one that does not depend on familiar epistemological or ontological concerns. The key idea behind this view is that typical mathematical hypotheses are indeterminate because they are relevantly similar to CH.

The purpose of this paper is to defend a conception of language that does not rely on linguistic meanings, and use it to address the Sorites and Liar paradoxes.

It is widely agreed that sentences containing a non-denoting description embedded in the scope of a propositional attitude verb have true de dicto interpretations, and Russell’s (1905) analysis of definite descriptions is often praised for its simple analysis of such cases, cf. e.g. Neale (1990). However, several people, incl. Elbourne (2005, 2009), Heim (1991), and Kripke (2005), have contested this by arguing that Russell’s analysis yields incorrect predictions in non-doxastic attitude contexts. Heim and Elbourne have subsequently argued that once certain facts about presupposition projection are fully appreciated, the Frege/Strawson analysis of definite descriptions has an explanatory advantage. In this paper, I argue that both Russell’s analysis and the Frege/Strawson analysis face a serious problem when it comes to the interaction of attitude verbs and definite descriptions. I argue that the problem observed by Elbourne, Heim, and Kripke is much more general than standardly assumed and that a solution requires a revision of the semantics of definite and indefinite descriptions. I outline the conditions that are required to solve the problem and present an analysis couched in dynamic semantics which can provide a solution. I conclude by discussing some further issues related to propositional attitude verbs that complicate a fully general solution to the problem.