Epicurus and Epicureans were famously antagonistic toward Platonic metaphysics and the dialectical style and technique pioneered in the Academy. However, there are Platonic methodological and doctrinal themes in Epicurus's epistemology, theology, and politics.

This dialogue is on the nature of thought and thinking. The five disputants are Socrates, an imaginary neuroscientist from California (whose opinions reflect those of contemporary cognitive neuroscientists), an Oxford don from the 1950s (who employs the linguistic analytic techniques of his times), a Scottish post-doctoral student, and John Locke (who speaks for himself). The discussion takes place in Elysium in the late afternoon. They examine the idea that thinking is an activity of the mind or the brain, whether the medium of thought consists of words or ideas, whether thoughtful speech is speech accompanied by thought, whether thinking, i.e. reasoning and inferring, is a process, and what is meant by the claim that ‘thinking is the last interpretation’. The dialogue ends when the protagonists go to dinner, but will be resumed after the meal.

I argue against Greco's account of the value of knowledge, according to which knowledge is distinctively valuable vis-à-vis that which falls short of knowledge in virtue of its status as an achievement and achievements being finally valuable. Instead, I make the case that virtuous belief is also an achievement. I argue that the nature of knowledge is such that knowledge is finally valuable in a way that virtuous belief is not, precisely because knowledge is not simply a success from ability. The value of knowledge lies in the positive responsiveness of the world to an agent's epistemic virtuousness.

Defended by scholars such as John McDowell and Julia Annas, the naturalism of second nature (NSN) claims that the virtues are part of a rational second nature instilled through moral education. While NSN emphasizes that rationality, fully developed, results in autonomy from nature, it is considered a sort of naturalism because the development of rational second nature unfolds through entirely natural processes. Critics object that NSN does not utilize human nature as a standard of evaluation, which is a problem for a view that claims to be a sort of naturalism. This paper attempts to meet this charge by introducing a novel way to understand the normative significance of human nature. It argues that NSN supports an analysis of human nature as the raw material of the virtues and, as such, the basis of a kind of aesthetic evaluation. Specifically, human nature is the basis of humaneness, a kind of beauty that belongs to what elevates or refines human nature. Thus, according to the fortified naturalism of second nature the ethical significance of human nature is explained by recognizing how the virtues exemplify a kind of beauty that only humans can have.

The guilt left by immoral actions is why moral duties are more pressing and serious than other reasons like prudential considerations. Religions talk of sin and karma; the secular still speak of spots or stains. I argue that a moral staining view of guilt is in fact the best model. It accounts for guilt's reflexive character and for anxious, scrupulous worries about whether one has transgressed. To understand moral staining, I borrow Christine Korsgaard's view that we construct our identities as agents through our actions. The contribution of immoral actions to self-constitution explains why moral obligations have priority and importance.

A long line of writers on Evans – Andy Hamilton, Lucy O'Brien, José Bermúdez, and Jason Stanley, to name just a few – assess Evans' account of first-person thought without heeding his warnings that his theory comprises an information and an action component. By omitting the action component, these critics are able to characterize Evans' theory as a perceptual model theory and reject it on that ground. This paper is an attempt to restore the forgotten element. With this component put back in, the charge of Evans' theory as a perceptual model of such thoughts falls apart, and the theory turns out to have enough merit to project itself as a legitimate contender for a plausible account of ‘I’-thought.

This dialogue on thought and language is a sequel to my dialogue ‘Thought and Thinking’, but can be read independently of it. The five disputants are the same as in the previous dialogue, namely Socrates; an imaginary neuroscientist from California (whose opinions reflect those of contemporary cognitive neuroscientists); an Oxford don from the 1950s (who employs the linguistic analytic techniques of his times); a Scottish post-doctoral student; and John Locke (who speaks for himself). The discussion takes place in Elysium in the early evening after dinner. They discuss the relationship between what one thinks and what one says, examine the reasons for supposing that there must be a language of thought and show why there cannot be one, investigate the supposition that thought must have a medium – that one must think in something (words, ideas, pre-linguistic concepts) – and demonstrate that it cannot have one. They investigate whether there can be non-linguistic thinking and whether non language-using animals can be said to think, and determine the limits of thought and thinking.

This paper is a response to a paper by Marcus Giaquinto in which he argues that lexical meaning is moderately indeterminate and that this poses a problem for the linguistic view of a priori knowledge. I argue that accepting the moderate indeterminacy thesis as he presents it is perfectly compatible with both the linguistic view in general and the specific suggestion that some a priori knowledge can be explained by appealing to synonymy. I also argue that, in fact, Giaquinto's considerations speak in favour of the linguistic view rather than against it. The general lesson is that, contrary to what might be suspected, the linguistic view does not presuppose an implausibly simple and tidy conception of lexical meaning.