This essay examines and criticizes a set of Kantian objections to Parfit's attempt in Reasons and Persons to connect his theory of personal identity to practical rationality and moral philosophy. Several of Parfit's critics have tried to sever the link he forges between his metaphysical and practical conclusions by invoking the Kantian thought that even if we accept his metaphysical theory of personal identity, we still have good practical grounds for rejecting that theory when deliberating about what to do. The argument between Parfit and his opponents illuminates broader questions about the relationship between our metaphysical beliefs and our practical reasons.

Robert Brandom argues for a “pragmatic phenomenalist account” of knowledge. On this account, we should understand our notion of justification in accordance with a Sellarsian social practice model, and there is nothing more to the phenomenon of knowledge than the proprieties of takings-as-knowing. I agree with these two claims. But Brandom's proposal is so sketchy that it is unclear how it can deal with a number of much-discussed problems in contemporary epistemology. The main purpose of this article is to develop and defend a pragmatic phenomenalist account of knowledge by resolving those problems. I argue, in particular, that this account can accommodate both the lesson of the Gettier problem and the lesson of reliabilism simultaneously.

In this article we aim to see how far one can get in defending the identity thesis without challenging the inference from conceivability to possibility. Our defence consists of a dilemma for the modal argument. Either “pain” is rigid or it is not. If it is not rigid, then a key premise of the modal argument can be rejected. If it is rigid, the most plausible semantic account treats “pain” as a natural-kind term that refers to its causal or historical origin, namely, C-fibre stimulation. It follows that any phenomenon that is not C-fibre stimulation is not pain, even if it is qualitatively similar to pain. This means there could be phenomena that feel like pain but are not pain since they are not C-fibre stimulation. These possible phenomena can be used to explain away the apparent conceivability of pain without C-fibre stimulation. On either horn of the dilemma, the identity theorist has ample resources to respond to Kripke's argument, even without wandering into the contentious territory of conceivability and possibility.

We begin by discussing some logical constraints on the psychological approach to personal identity. We consider a problem for the psychological approach that arises in fission cases. The problem engenders the need for a non-branching clause in a psychological account of the co-personality relation. We look at some difficulties in formulating such a clause. We end by rejecting a recently proposed formulation of non-branching. Our criticism of the formulation raises some interesting questions about the individuation of person stages.

This article shows how the “problem of moral luck” and Sartre's concept of “bad faith” are mutually illuminating, since both have to do with confusions about how much we control, or are controlled by, our situations. I agree with three recent proposals that the problem of moral luck results from certain epistemic malfunctions. However, I argue that the problem cannot be dissolved by overcoming these malfunctions because they are rooted in bad faith. Against the currently dominant interpretation, I argue that bad faith is an inescapable condition of human being because, while we are both free and factical, we can only exist as one or the other at once (much as a “duck-rabbit” drawing can only be seen as duck or rabbit), thus always keeping an aspect of our nature hidden (though in plain sight) from ourselves.

In this article I argue that Hume's empiricism allows for time to exist as a real distinct impression of succession, not, as many claim, merely as a nominal abstract idea. In the first part of this article I show how for Hume it is succession and not duration that constitutes time, and, further, that only duration is fictional. In the second part, I show that according to the way Hume describes the functions of the memory and imagination, it is possible to explain how we are able to perceive a distinct impression of succession.

Many of the results of Timothy Williamson's Knowledge and Its Limits depend upon his argument that many, if not all, of our mental states fail to be luminous in the sense that if we are in them, then we are in a position to know that we are in them. The purpose of this article is to show that his argument is unsound. I conclude by distinguishing between partial and total luminosity, and by arguing that even if mental states are not totally luminous, they are at least partially so.

This article offers a critical analysis of John Broome's conception of practical reasoning. I first introduce his main claims and then point out some of the difficulties raised by the notion of “double expression” and by some aspects of the cognitivism which he explicitly endorses. I then emphasize two consequences of these criticisms: one concerning the link he sees between belief and intention, and the other concerning the idea that our practical reasonings are inextricably linked to our theoretical reasonings. Finally, I argue that the problem Broome seems to be facing has its source in the difficulty of distinguishing clearly between reasoning (a relation between mental states) and the description of its content (a relation between propositions).