Human rights debates neglect social rights. This paper defends one fundamentally important, but largely unacknowledged social human right. The right is both a condition for and a constitutive part of a minimally decent human life. Indeed, protection of this right is necessary to secure many less controversial human rights. The right in question is the human right against social deprivation. In this context, ‘social deprivation’ refers not to poverty, but to genuine, interpersonal, social deprivation as a persisting lack of minimally adequate opportunities for decent human contact and social inclusion. Such deprivation is endured not only in arenas of institutional segregation by prisoners and patients held in long-term solitary confinement and quarantine, but also by persons who suffer less organised forms of persistent social deprivation. The human right against social deprivation can be fleshed out both as a civil and political right and as a socio-economic right. The defence for it faces objections familiar to human rights theory such as undue burdensomeness, unclaimability, and infeasibility, as well as some less familiar objections such as illiberality, intolerability, and ideals of the family. All of these objections can be answered.

What is it about a person's becoming an adult that makes it generally inappropriate to treat that person paternalistically any longer? The Standard View holds that a mere difference in age or stage of life cannot in itself be morally relevant, but only matters insofar as it is correlated with the development of capacities for mature practical reasoning. This paper defends the contrary view: two people can have all the same general psychological attributes and yet the mere fact that one person is at the beginning of a life and another in the middle of one can justify treating the younger person more paternalistically than the older one. Recognising the moral relevance of age, moreover, is crucial if one is to accommodate both the liberal moral ideal of respect for autonomy and our demanding educational aims, given that these otherwise come into conflict with one another.

Some particularists have argued that even virtue properties can exhibit a form of holism or context variance, e.g. sometimes an act is worse for being kind, say. But, on a common conception of virtuous acts, one derived from Aristotle, claims of virtue holism will be shown to be false. I argue, perhaps surprisingly, that on this conception the virtuousness of an act is not a reason to do it, and hence this conception of virtuous acts presents no challenge to particularist claims about the context variance of reasons. Still, I argue that the virtues nevertheless have important implications for our understanding of the particularism debate. Specifically, we can accept the particularist claim that reasons do not need to be principled in order to have the normative status that they do have, while still maintaining that sound moral thought and judgement has a principled structure understood in terms of the virtues.

Anslem raised a puzzle about mercy: How can anyone (God, say, or a judge) be both just and merciful at the same time? For it seemed to Anselm that justice requires giving people what they deserve, while being merciful involves treating people less harshly than they deserve. This puzzle has led to a number of analyses of mercy. But a strange thing emerges from discussions of this topic: people seem to have wildly divergent intuitions about putative cases of mercy. Examples that are taken by some to be paradigm cases of mercy are seen by others to be clear instances of non-mercy. This phenomenon gives rise to a new puzzle: Why is it that people seem to have such wildly divergent intuitions about putative examples of mercy? This paper proposes a new analysis of mercy that provides solutions to both of these puzzles about mercy.

Epistemic relationism in the theory of assertion is the view that an assertion's epistemic propriety depends purely on the relation between the asserter and the proposition asserted. Many accounts of assertion are relationist in this sense, including the familiar knowledge, belief, and justification accounts. A notable feature of such accounts is that they give no direct importance to the role of hearer: as far as such accounts are concerned, we need make no mention of hearers in characterising an assertion's propriety conditions. This paper develops an account which rejects relationism, by giving central importance to the role of hearer. The paper introduces the knowledge provision account, according to which an assertion that p is proper only if it is fit to give a hearer knowledge that p. The paper aims to show: (i) that we can understand this account in a way which does not leave it open to obvious counterexamples, (ii) that it does not reduce to any familiar relationist account, and (iii) that it carries certain advantages over familiar relationist accounts.

C endorses the doctrine of Composition as Identity, which holds that a composite object is (plurally) identical to its many parts, and entails that one object can be (plurally) identical to several others. In this dialogue, N argues that many-one identity, and thus composition as identity, is conceptually confused. In particular, N claims it violates two conceptual truths: that existence facts fix identity facts, and that identity is no addition to being. In response to pressure from C, N considers several candidate interpretations of the first slogan. Each interpretation imagines God issuing a divine existential decree that counts as ‘fixing the existence facts’, and then considers what questions about identity are left open by it. N eventually hits on an appealing interpretation of the slogan that C can neither undermine nor accommodate, and uses it to interpret the second slogan as well.

It is a familiar point that many ordinary dispositions are multi-track, that is, not fully and adequately characterisable by a single conditional. In this paper, I argue that both the extent and the implications of this point have been severely underestimated. First, I provide new arguments to show that every disposition whose stimulus condition is a determinable quantity must be infinitely multi-track. Secondly, I argue that this result should incline us to move away from the standard assumption that dispositions are in some way importantly linked to conditionals, as presupposed by the debate about various versions of the ‘conditional analysis’ of dispositions. I introduce an alternative conception of dispositionality, which is motivated by linguistic observations about dispositional adjectives and links dispositions to possibility instead of conditionals. I argue that, because of the multi-track nature of dispositions, the possibility-based conception of dispositions is to be preferred.

I discuss Turri's puzzle about withholding. I argue that attention to the way in which evidence can justify withholding dissolves the puzzle.