Many political libertarians argue, or assume, that negative moral duties (duties not to harm others) prevail over positive moral duties (duties to aid others), and that the legal system ought to reflect such pre-eminence. I call into question this strategy for defending a libertarian order. I start by arguing that a successful account of the well-known case of a runaway trolley that is about to kill five innocents unless a passer-by diverts it onto one innocent, killing him, should point to (i) the ex ante advantage to all six of being subject to a policy of redirection of runaway trolleys, and (ii) the causal structure of killing vs. letting-die choices. I then argue that this account of the trolley case entails that legal systems reflecting the relative stringency of negative and positive moral duties should uphold redistributive measures at odds with libertarianism. The assumption that the legal system ought to reflect, through non-causal routes, moral principles and their relative weights leads to either an ideal-theory (in Rawls's sense) assessment of libertarianism or a symbolic account of the relationships between morality and law. Libertarians should undermine this assumption if they hope to offer an all-things-considered case for free markets.

The non-identity problem is really a collection of problems having distinct logical features. For that reason, non-identity problems can be typed. This article focuses on just one type of non-identity problem, the ‘can't-expect-better’ problem, which includes Derek Parfit's depletion example and many others. The can't-expect-better problem uses an assessment about the low probability of any particular person's coming into existence to reason that an earlier wrong act does not harm that person. This article argues that that line of reasoning is unusually treacherous in that it makes not just one hard-to-detect error in what is done with the relevant probability assessments but rather alternates between two. We sort out one fallacy only to fall, against all odds (as it were), into a second. By avoiding both errors, we become able to discern harm in cases in which the can't-expect-better problem argues there is none. We will then be in a position to set aside the can't-expect-better problem as an objection against the person-based intuition that acts that are ‘bad’ must be ‘bad for’ at least some existing or future person.

It is sometimes suggested that if a moral theory implies that infanticide can sometimes be permissible, that is sufficient to discredit the theory. I argue in this article that the common-sense belief that infanticide is wrong, and perhaps even worse than the killing of an adult, is challenged not so much by theoretical considerations as by common-sense beliefs about abortion, the killing of non-human animals, and so on. Because there are no intrinsic differences between premature infants and viable fetuses, it is difficult to accept that an abortion performed after the point of viability can be permissible while denying that infanticide can be permissible for a comparably important reason. This and other challenges to the consistency of our intuitions exert pressure on us either to accept the occasional permissibility of infanticide or to reject liberal beliefs about abortion.

I argue that pregnant women have a duty to refrain from behaviours (e.g. taking illicit drugs) or to allow certain acts to be done to them (e.g. caesarean section) for the sake of their foetus if the foetus has a reasonable chance of living and being in a harmed state if the woman does not refrain from those behaviours or allow those things to be done to her. There is a proviso: that her refraining from acting or allowing acts to be performed upon her does not significantly harm her. This duty does not presuppose that the foetus is a person. It is grounded on principles of respect for the interests of sentient beings and prevention of harm to future individuals. I give an argument for a general duty of easy rescue.

In this article I maintain that the anti-theoretical spirit which pervades Williams's ethics is close to the Humean project of developing and defending an ethics based on sentiments which has its main focus in the virtues. In particular, I argue that there are similar underlying themes which run through the philosophies of Hume and Williams, such as the view that a correct ethical perspective cannot avoid dealing with a broader theory of human nature; the conviction that this inquiry cannot be developed in abstraction from the contingencies which are distinctive of the lives of flesh-and-blood human beings driven by passions; and the belief that the notion of character plays a key role in identifying and morally evaluating such lives. Finally, Williams' account of the psychological mechanism of shame in explaining character formation bears a strong resemblance to Hume's treatment of the passion of humility.

In this article, I argue that the state has a prima facie obligation to help its citizens satisfy their autonomous preferences. I argue that this obligation is grounded in the state's obligation to respect its citizens as persons, and that part of what is involved in respecting someone as a person is helping her satisfy her autonomous preferences. I argue that that which makes preferences autonomous is also that which makes them, and not their non-autonomous counterparts, worthy of respect. In addition, I reject other views of what makes preferences worthy of respect, in particular Ronald Dworkin's view that only preferences for one's own enjoyment of some goods or opportunities deserve political consideration. Finally, I consider the state's obligation towards immoral autonomous preferences, and I argue that the state's prima facie obligation to promote the satisfaction of autonomous preferences is quite strong.

Philosophers have recently argued that we should revise our understanding of character. An individual's behaviour is governed not by a set of ‘global’ traits, each elicited by a certain kind of situational feature, they argue, but by a much larger array of ‘local’ traits, each elicited by a certain combination of situational features. But the data cited by these philosophers support their theory only if we conceive of traits purely in terms of stimulus and response, rather than in the more traditional terms of inner mental items such as inclinations. We should not adopt the former conception, moreover, since doing so would impede pursuit of the ethical aims for which we need a theory of character, whereas retaining the latter conception will facilitate this pursuit. So we should not revise our understanding of character in the way proposed.

The contrast typically made between utilitarianism and virtue theory is overdrawn. Utilitarianism is a universal emulator: it implies that we should lie, cheat, steal, even appropriate Aristotle, when that is what brings about the best outcomes. In some cases and in some worlds it is best for us to focus as precisely as possible on individual acts. In other cases and worlds it is best for us to be concerned with character traits. Global environmental change leads to concerns about character because the best results will be produced by generally uncoupling my behavior from that of others. Thus, in this case and in this world, utilitarians should be virtue theorists.

This article concerns two themes in Bart Schultz's recent biography of Henry Sidgwick, Henry Sidgwick: Eye of the Universe. The first is the importance of Sidgwick's conflict over his religious beliefs to the development of his thinking in The Methods of Ethics. I suggest that, in addition to the characteristics of Methods that Schulz highlights, the work's epistemology, specifically, Sidgwick's program of presenting ethics as an axiomatic system on the traditional understanding of such systems, is due to the conflict. The second is the relative neglect into which Methods fell in the first part of the twentieth century, neglect Schultz attributes to changes in philosophical fashions and to the undue influence of the Bloomsbury literati on British intellectual culture. I suggest that there is a deeper explanation, which lies in Sidgwick's program of presenting ethics as an axiomatic system on the traditional understanding of such systems. Such programs, I argue, became obsolete in analytic philosophy owing to changes in how axiomatization in mathematics was understood that resulted initially from the rise of non-Euclidean geometries and ultimately from the collapse of Frege's and Russell's logicism.

I very much enjoyed reading and thinking about Krister Bykvist's interesting and carefully written paper. My comments will not be criticisms. I will not challenge the conclusions the paper draws about the complicated examples of conflicts between preferences that it discusses. For example, I will not attempt to defend any of the views criticized in section IV of the paper. And I agree with the positive solution presented in section V, when that solution is characterized in the broadest possible way. In cases where a choice that I must make now will create one or the other of two possible future ‘selves’ whose preferences would conflict over the question of which choice I should make, prudence tells me to create the self that will experience a higher level of well-being or welfare.

T. M. Scanlon has revived a venerable tradition according to which something's being good consists in its being such that there is a reason to respond positively towards it. He has presented novel arguments for this thesis. In this article, I first develop some refinements of the thesis with a view to focusing on intrinsic value in particular, then discuss the relation between the thesis and consequentialism, then critically examine Scanlon's arguments for the thesis, and finally turn to the question whether we should reject the thesis on the grounds that, when there is a reason to respond positively towards something, this is so because the thing in question is good. Two appendices follow. In the first, I discuss whether it is good to do right. In the second, I discuss whether an act's being wrong provides a reason not to do it.

Philippa Foot has recently argued that non-cognitivism rests on a mistake. According to Foot, non-cognitivism cannot properly account for the role of reasons in moral thinking. Furthermore, Foot argues that moral judgements share a conceptual structure with the kind of evaluations that we make about plants and animals, which cannot be couched in non-cognitivist terms. In this article I argue that, in the form of expressivism, non-cognitivism is capable of accommodating most of what Foot says about reasons and morality. I then argue that the kind of evaluative judgements Foot suggests that we make about plants and animals, does not constitute a plausible alternative to an expressivist understanding of moral judgements. Finally I consider an account similar to Foot's, defended by Rosalind Hursthouse, which, I argue, suffers from an inconsistency, the avoidance of which leaves Hursthouse with a view that is either compatible with expressivism or shares the same problems as Foot's.

Sometimes we make decisions which affect our lives at times when we will hold values that are different from our values at the time the decision is made. What is the reasonable way to make such a choice? Some think we should accept a requirement of temporal neutrality and take both sets of values into account, others think we should decide on the strength of our present values, yet others think that in evaluating what will happen at that other time we should use the values that we will endorse at that time instead of our present values. These views see the problem as one about finding some attitude towards time itself that is distinctively rational. This article argues that these views are subject to serious objections. It suggests that instead we should think in terms of well-being. If a person approves of, or positively responds to, the way their life is going they will experience more well-being than if there is no positive response. The article explores the implications of a positive response condition on well-being for deciding what it is rational to do in cases involving changing goals.

Consequentialist theories of virtue and vice, such as the theories of Jeremy Bentham and Julia Driver, characterize virtue and vice in terms of the consequential, or instrumental, properties of these character traits. There are two problems with theories of this sort. First they imply that, under the right circumstances, paradigmatic virtues, such as benevolence, are vices and paradigmatic vices, such as maliciousness, are virtues. This is conceptually problematic. Second, they say nothing about the intrinsic nature of the virtues and vices, which is less than we could hope for from a theory of virtue and vice. Thus, we have reason to reject consequentialist theories in favour of theories that characterize virtue and vice in terms of the intrinsic properties of these character traits. Aristotle and Thomas Hurka have theories this sort.

In Considerations on Representative Government, John Stuart Mill concedes that secrecy in voting is sometimes justified but, nonetheless, maintains that it should be the exception rather than the rule. This article critically examines Mill's arguments. It shows that Mill's idea of voting depends on a sharp public/private distinction which is difficult to square with democratic ideas about the different powers and responsibilities of voters and their representatives, or with legitimate differences of belief and interest amongst voters themselves. Hence, it concludes, we should reject the assumption, which many of us share with Mill, that the secret ballot is justified only on prudential grounds and recognize how central privacy is to any democratic conception of citizenship and politics.

Are torture and torturers unforgivable? The article examines this question in the light of a Humean account of forgiveness. Initially, the Humean account appears to suggest that torturers are unforgivable. However, in the end, I argue it provides us with good reasons to think that even torturers may be forgiven.

Principle monists believe that our moral duties, such as fidelity and non-maleficence, can be justified in terms of one basic moral principle. Principle pluralists disagree, some suggesting that only an excessive taste for simplicity or a desire to mimic natural science could lead one to endorse monism. In Ideal Code, Real World (Oxford, 2000), Brad Hooker defends a monist theory, employing the method of reflective equilibrium to unify the moral duties under a version of rule consequentialism. Hooker's arguments have drawn powerful criticisms from pluralists such as Alan Thomas, Phillip Montague and Philip Stratton-Lake. Against these critics, I argue that Hooker's monism enjoys certain practical advantages associated with the simplicity of a single basic principle. These advantages are often overlooked because they appear primarily in cases of second-order deliberation, in which one must decide whether our basic moral duties support a certain derivative duty. I argue that these advantages of monism over pluralism are analogous to the advantages that generalists claim over moral particularism. Because pluralists are generalists, I conclude that they are in an awkward dialectical position to dismiss Hooker's monism for the reasons they usually offer.

In these remarks on Feldman's recent book, Pleasure and the Good Life, I concentrate on Feldman's account of pleasure as attitudinal. I argue that an account of pleasure according to which pleasure need not have any feel is implausible. I suggest that Feldman could avoid this problem but retain the advantages of his attitudinal hedonism by giving an account of the attitude ‘taking pleasure in’ such that the attitude has a feel.

In Uneasy Virtue, Julia Driver advocates a consequentialist account of the virtues. In so far as her view is ‘psychologically minimalist’, Driver's account is superior to the psychologically rich theories of virtue offered by Aristotle, Hume and Kant. However, Driver is also committed to ‘instrumentalism’ about virtue: a trait is a virtue only if it has instrumental value. In contrast, I argue for a ‘disjunctive’ form of minimalism, according to which a character trait counts as a virtue if it has either instrumental or intrinsic value. The common intuitions about virtue that Driver takes to support her ‘instrumental minimalism’ actually fit better with disjunctive minimalism. Admittedly, disjunctive minimalism is a messy account of virtue. However, this messiness would be a problem only if we drew a tight connection between virtue and right action, and we have good independent reasons for thinking there is no such tight link.

I am grateful to have been given the opportunity to comment on McKerlie's interesting article, especially since it concerns one of my pet topics and provides many helpful comments on one of my own articles on this topic. My comments will be brief because I agree with most of his points, in particular, his criticisms of the prudential view and the present-aim theory. What I want to do here is just to clarify a couple of things concerning my own theory, concede some of the difficulties that McKerlie raises for my theory, and see to what extent his own proposal fares better than my own theory.