This dialogue concerns metasemantics and language cognition. It defends a Lewisian conception of languages as abstract entities (Lewis 1975), arguing that semantic facts are necessities (Soames 1984), and therefore not naturalistically reducible. It identifies spoken languages as idiolects, in line roughly with Chomskyan I-languages. It relocates traditional metasemantic indeterminacy arguments as indeterminacies of what language an agent speaks or cognizes. Finally, it aims to provide a theoretical analysis of the cognizing relation in terms of the agent's assigning certain meanings to strings.

What are the earliest beings that have minds in evolutionary order? Two marks of mind are consciousness and representation. I focus on representation. I distinguish a psychologically distinctive notion of representation from a family of notions, often called ‘representation’, that invoke information, causation, and/or function. The psychologically distinctive notion implies that a representational state has veridicality conditions as an aspect of its nature. Perception is the most primitive type of representational state. It is a natural psychological kind, recognized in a mature science: perceptual psychology. This kind involves a type of objectification, and is marked by perceptual constancies. The simplest animals known to exhibit perceptual constancies, perception, and representation in a distinctively psychological sense, are certain arthropods. Representational mind, or representational psychology, begins in the arthropods. We lack scientific knowledge about the beginnings of consciousness. Consciousness is neither necessary nor sufficient for perception. I conclude by reflecting on the kinds mind and psychology.

This dialogue on the mind/body relation is a sequel to my dialogue ‘On the Nature of the Mind’, but can be read independently of it. The five disputants include an imaginary neuroscientist from California, an Oxford don from the 1950s, a lady, Peter Strawson and Alan White. They examine the peculiar idioms of having: having a body, having a mind, having a soul and having a self. To have a mind, it is concluded, is not to own anything, but to be able to do a variety of things: to reason from premises to conclusions; to act, think, and feel things for reasons. To have a body is to possess a variety of somatic characteristics. Hence a distinction is drawn between the body one is, and the body one has. Accordingly, the problem of the relation between my mind and my body simply disintegrates, since there can be no relation between my intellectual abilities and my somatic features.

This article aims to review the standard objections to dualism and to argue that either they will fail to convince someone committed to dualism or are flawed on independent grounds. I begin by presenting the taxonomy of metaphysical positions on concrete particulars as they relate to the dispute between materialists and dualists, and in particular substance dualism is defined. In the first section, several kinds of substance dualism are distinguished and the relevant varieties of this kind of dualism are selected. The remaining sections are analyses of the standard objections to substance dualism: It is uninformative, has troubles accounting for soul individuation, causal pairing and interaction, violates laws of physics, is made implausible by the development of neuroscience and it postulates entities beyond necessity. I conclude that none of these objections is successful.

What do we learn when we focus analysis – not so much on the content of experience – as on its universal features and functioning? Descartes believed that such focus (when exercised by someone employing his first-personal method of inquiry) held the key to the fundamental metaphysics of our universe – that it could reveal fundamental truths about the nature of substance, or at any rate could reveal some fundamental metaphysical categories and their contrasts. He believed such focus could lead to a certain doctrine of dualism. Philosophers now widely hold that Descartes' method was profoundly wrongheaded, in no way a candidate method for illuminating the material universe that is our own. In fact, however, Descartes' method is considerably more serviceable. While unable to do what Descartes thought it could do, nonetheless it is ideal for examining a taxon that this essay will refer to as grammar or structure in the most general sense. Grammar is a contrary of content or materiality, though not the only one since materiality enjoys multiple contraries. Thus Descartes' method can lead us to a certain dualism, but not the one that Descartes imagined.

It has often been claimed that Locke's agnostic remarks in the Essay represent his suspension of philosophical judgment on crucial ontological issues or his hesitation over which metaphysical stance to adopt. Against this often-raised criticism, I argue that Locke actually held a clear position – a type of functionalism about thingness in general, whether macro or micro, or whether mental or physical. What Locke refers to as a ‘nominal essence’, I further argue, represents a set of functional roles that a thing plays in order to be classified as of a kind to which it belongs. Our empirical knowledge about things – confined to their nominal essences – can only tell us about their functional roles but not their intrinsic properties that realize those roles. One remains therefore incurably ignorant about the intrinsic property of things in themselves. I explore the historical and philosophical significance of Locke's functional approach.

In some recent papers I have been arguing that the concept ‘good-for’ is prior to the concept of ‘good’ (in the sense in which final ends are good), and exploring the implications of that claim. One of those implications is that everything that is good is good for someone. That implication seems to fall afoul of our intuitions about certain cases, such as the intuition that a world full of happy people and animals is better than a world full of miserable ones, even if the people and animals are different in the two cases, so that there is no one for whom the second world is better. Such cases tempt people to think that there must be impersonal goods, and that what it means to say that something is good for you is that you are the one who ‘has’ some impersonal good. In this paper, I argue that if we approach things in this way, it is impossible to say what the ‘having’ consists of, what relation it names. This leads me to a discussion of various things we do mean by saying that something is good for someone, how they are related to each other, and what sorts of entities can ‘have a good’. Finally, I explain why we think that a world full of happy people and animals is better than a world full of miserable ones, even if the people and animals are different in the two cases.

Respect, as Kant describes it, has a duality of nature that seems to embody a contradiction – i.e., it is both a moral motive and a feeling, where these are thought to be mutually exclusive. Most solutions involve eliminating one of the two natures, but unfortunately, this also destroys what is unique about respect. So instead, I question the non-cognitive theory of emotion giving rise to the contradiction. In its place, I develop the cognitive theory implicit in Kant's work, one in which emotions take the form of evaluative judgments that determine the will. I then show that, as a purely rational emotion, respect is perfectly suited to be a moral motive.

The object of this essay is to discuss Ludwig Wittgenstein's remarks in Philosophical Investigations and elsewhere in the posthumously published writings concerning the role of therapy in relation to philosophy. Wittgenstein's reflections seem to suggest that there is a kind of philosophy or mode of investigation targeting the philosophical grammar of language uses that gratuitously give rise to philosophical problems, and produce in many thinkers philosophical anxieties for which the proper therapy is intended to offer relief. Two possible objectives of later Wittgensteinian therapy are proposed, for subjective psychological versus objective semantic symptoms of ailments that a therapy might address for the sake of relieving philosophical anxieties. The psychological in its most plausible form is rejected, leaving only the semantic. Semantic therapy in the sense defined and developed is more general and long-lasting, and more in the spirit of Wittgenstein's project on a variety of levels. A semantic approach treats language rather than the thinking, language-using subject as the patient needing therapy, and directs its attention to the treatment of problems in language and the conceptual framework a language game use expresses in its philosophical grammar, rather than to soothing unhappy or socially ill-adjusted individual psychologies.

There are five basic ways to resolve disagreements: keep arguing until capitulation, compromise, locate an ambiguity or contextual factors, accept Pyrrhonian skepticism, and adopt relativism. Relativism is perhaps the most radical and least popular solution to a disagreement, and its defenders generally think the best motivator for relativism is to be found in disputes over predicates of personal taste. I argue that taste predicates do not adequately motivate relativism over the other possible solutions, and argue that relativism looks like the most promising approach when disputants cannot even agree on the meta-evidence for a contested proposition.

Because of the idealizations involved in the ideas of a total state of the world and of all the laws of nature, the thesis of all-encompassing determinism is unverifiable. Our everyday non-scientific talk of causation does not imply determinism; nor is it needed for the Kantian argument for a general causal framework as a condition for experience of an objective world. Determinism is at best a regulative ideal for science, something to be approached but never reached.

Indifference is often described as a vice. Yet who is indifferent; to what; and in what way is poorly understood, and frequently subject to controversy and confusion. This paper proposes a framework for the interpretation and analysis of ethically problematic forms of indifference in terms of how different states of indifference can be either more or less dynamic, or more or less sensitive to the nature and state of their object.

I shall situate Hume's scepticism within a broader philosophical and historical context. Firstly, I shall consider the place of Hume's thought within the early modern break with the almost millennium long metaphysical tradition, a break initiated by Descartes. The framework of being structured by a universal order was replaced by the individual human mind that broke free from any higher authority and became an autonomous cognitive agent. Subsequently, the ontological self-evidence of the world or the possibility of adequate knowledge came under sceptical attack. Hume firmly belongs to this discourse and can be seen as the most consistent exponent of this early modern sceptical line. In this light, the ‘New Hume’ claim that Hume was an ontological realist will be shown to be misplaced in principle. Secondly, the strong influence of Pyrrhonism on Hume's philosophy will be considered, together with his concept of mitigated scepticism. The Pyrrhonian legacy is especially noticeable in Hume's acceptance of the weakness of reason and in his emphasis on the instructive role of philosophy – instead of attempting to make it the foundation of science it becomes a guide to a balanced, happy life. In this respect, Hume stands outside the early modern mainstream in philosophy.

Recent work on dispositions offers a new solution to the long-running dispute about whether explanations of intentional action are causal explanations. The dispute seemed intractable because of a lack of percipience about dispositions and a commitment to Humean orthodoxies about causation on both sides.

Politics has been defined in different and contradictory ways in the last century or so. If politics is to be a single subject of study then contradictory theories should be capable of being related together. In this article I argue that they can be related in terms of what I call the Aristotelian criterion. The article is in four parts. Firstly, I discuss the problem of defining politics; secondly, I introduce the criterion; thirdly, I consider five modern theories of politics (those of Arendt, Oakeshott, Collingwood, Schmitt and Rancière) in relation to the criterion; and fourthly, I use the criterion to put forward an original and capacious definition of politics.1

Some philosophy – Wittgenstein's would be an example – is written in clear sentences, yet most people find it obscure at a first reading. This is because the prime location of clarity in philosophy is not sentences but structures. Only if a reader can relate what he is currently reading to a wider framework does he know where he is. Coherent utterance in all discursive media – not only language but mathematics, for example, or music – possesses two kinds of structure at the same time. In this article these are distinguished, and their radically different relationships with language shown. In the process, the commonest causes of unclarity are also identified.