In Chapter 17 of his magnificent Reasons and Persons, Derek Parfit asks what he describes as an ‘awesome question’: ‘How many people should there ever be?’ For a utilitarian like me, the answer seems simple: there should be however many people it takes to make the world best. Unfortunately, if I answer Parfit's awesome question in this way, I may sink myself in a quagmire of axiological confusion. In this paper, I first describe certain aspects of the quagmire. Then I introduce and explain some of the elements of a novel axiological view – ‘justicism’. Justicism is derived from some ideas originally suggested by Franz Brentano. It was developed in an effort to solve certain other problems confronting utilitarianism – problems explicitly about justice. I think, however, that as a sort of happy by-product, justicism also generates a plausible answer to Parfit's awesome question. This may come as a bit of surprise, since justicism is a form of totalism, and it is widely thought that no totalistic theory can provide a satisfactory answer to Parfit's question. After presenting and explaining my proposed solution, I address some objections.

One of the standard objections to traditional act utilitarianism is that it is insensitive to issues of justice and desert. Traditional act utilitarianism holds, for example, that it is morally obligatory to torture or kill an innocent person, when doing so increases the happiness of others more than it decreases the happiness of the innocent person. Utilitarianism is, of course, sensitive to what people believe about justice (for example, people might riot, if they believe a gross injustice has been done), but it is not sensitive to justice itself.

There are many disagreements about what people have moral obligations to do, but almost everyone believes that some people have some moral obligations. Moreover, there are some moral obligations in which almost everyone believes. For example, if I promise to give a talk at this conference, I have a moral obligation to do so. Of course, my obligation might be overridden. Moreover, even if my obligation were overridden, I would still have a moral obligation to give a talk at this conference.

Sinnott-Armstrong's paper principally defends our inability to justify, philosophically, normal moral claims. In particular, we cannot justify them against other claims, especially the claim of moral nihilism. Moral nihilism is the doctrine that there are no moral obligations (for anybody to do anything). This thesis ‘does not lie in meta-ethics. It is a universally quantified substantive moral claim’ (p. 219). Sinnott-Annstrong makes it clear that he does not actually believe this doctrine (p. 217), but he believes that it is coherent, and that a variety of strategies philosophers might attempt all fail to disprove it. And because of this, ordinary claims to obligation are philosophically unjustified.

You face two buttons. Pushing one will destroy Greensboro. Pushing the other will save it. (Sorry, doing nothing also destroys Greenboro.) There is no way for you to know which button saves and which destroys. What ought you to do? Answer: You ought to make the correct guess and push the button that saves Greensboro. Second question: Do you have an obligation to push the correct button?

Vagueness in the phrase ‘can know’ aside, the principle of (Access) An act is obligatory only if its agent can know that it is obligatory is an important principle, worthy of serious scrutiny. Its truth or falsity bears on the question of whether moral rightness, obligatoriness, etc., are a matter of factors ‘internal’ to an agent (such as motives and beliefs), or whether ‘external’ factors (such as consequences) are relevant to determining the moral normative status of acts. Moreover, Access enjoys considerable intuitive support. If I destroy Greensboro in Professor Sorensen's example by pushing the wrong button, I seem to have a good excuse to give to anyone who would accuse me of wrongdoing: ‘I had no way of knowing that this action would be wrong!’ But if I have a good excuse, then pushing the button does not seem wrong; and if that action is not wrong, then I had no obligation to refrain from pushing that button.

Virtue ethics has generated a great deal of excitement among ethicists largely because it is seen as an alternative to the traditional theories – utilitarianism and Kantian ethics – which have come under considerable scrutiny and criticism in the past 30 years. Rather than give up the enterprise of doing moral theory altogether, as some have suggested, others have opted to develop an alternative that would hopefully avoid the shortcomings of both utilitarianism and Kantian ethics. Several writers, such as Jorge Garcia and Michael Slote, have tried to develop this alternative of virtue ethics, or at least sketch out ways such a theory could be developed.

Moral evaluation is concerned with the attribution (to its various objects – actions, character, attitudes, states of affairs, institutions) of values whose distinction into two broad groups has become familiar. On the one hand, there are the most general moral values of lightness, wrongness, goodness, badness, and what ought to be or to be done. On the other, there is a great diversity of more specific moral values which these objects can have: of being a theft, for instance, or a thief; of honesty, reliability or callousness. Within the recent body of work attempting to restore to the virtues a central place in ethical thinking, two claims stand out. One is that, of these two kinds of values, the specific ones are explanatorily prior to the general – that if an action is wrong, it is because it is wrong in one of those specific respects. A second claim, though, is now standardly made definitive of ‘Virtue ethics’: that amongst the specific values, the value of character is explanatorily prior to that of action – that if an action is callous, say, it is because it expresses callousness of character – and that in this sense, the moral value of action derives from that of character. This second claim has been widely attacked; in what follows, I present a reason for believing that, at least in the case of callousness, it is right.

Near the beginning of the last chapter of Life's Dominion, Ronald Dworkin expounds the following problem. Margo has Alzheimer's disease. She suffers from ‘serious and permanent dementia’ (p. 220). It transpires that some years ago, at a time when she was mentally fully competent, Margo executed an advance directive. In this formal document she expressed her wishes concerning what should happen to her if she were to develop Alzheimer's. Should those wishes now be acceded to? For instance, suppose that in her document Margo directed that she should not receive treatment for any life-threatening illness she might contract. Should a doctor therefore now refrain from such treatment? What if, more than this, Margo indicated in her will that after the definitive onset of Alzheimer's ‘she should be killed as soon and as painlessly as possible’ (p. 226)? Could it possibly be right to grant that request?