Although the belief that killing is normally wrong is as universal and uncontroversial a moral belief as we are likely to find, no one, to my knowledge, has ever offered an account of why killing is wrong that even begins to do justice to the full range of common sense beliefs about the morality of killing. Yet such an account would be of considerable practical significance, since understanding why some killings are wrong should help us to determine the conditions in which killing is not wrong. For, in those cases in which the reasons why killing is wrong do not apply, killing may be permissible or, if there are positive reasons that favour it, even morally required.

In recent years the distinction between agent-relative and agent-neutral reasons has been taken by many to play a key role in distinguishing deontology from consequentialism. It is central to all universalist consequentialist theories that value is determined impersonally; the real value of any state of affairs does not depend on the point of view of the agent. No reference, therefore, to the agent or to his or her position in the world need enter into a consequentialist understanding of what makes an action right or wrong or morally permissible. Consequentialism thus provides an agent-neutral account of both the right and the good.

It seems common at the moment to make agent-neutrality a necessary condition of ‘consequentialism” and to hold that deontological ethics are agent-relative. This note argues that both these tendencies regrettably obscure useful terms and distinctions. It concludes by considering what it would be best, now, to mean by ‘utilitarianism” and making a proposal.

In the preface to the second edition of his deservedly popular Practical Ethics, Peter Singer notes that one of the ‘two significant changes” of his ‘underlying ethical views” consists in dropping the tentative suggestion that ‘one might try to combine both the “total” and the “prior existence” versions of utilitarianism, applying the former to sentient beings who are not self-conscious and the latter to those who are” (pp. x–xi). On the total view our aim is ‘to increase the total amount of pleasure (and reduce the total amount of pain)” regardless of ‘whether this is done by increasing the pleasure of existing beings, or increasing the number of beings who exist”, whereas on the prior existence view we ‘count only beings who already exist, prior to the decision we are taking, or at least will exist independently of that decision” (p. 103). Instead he proposes ‘that preference utilitarianism draws a sufficiently sharp distinction between these two categories of being to enable us to apply one version of utilitarianism to all sentient beings” (p. xi).

Impartialism in ethics has been said to be the common ground shared by both Kantian and utilitarian approaches to ethics. Lawrence Blum describes this common ground as follows:

Both views identify morality with a perspective of impartiality, impersonality, objectivity and universality. Both views imply the ‘ubiquity of impartiality” – that our commitments and projects derive their legitimacy only by reference to this impartial perspective.

In the paper I offer a brief sketch of one of the sources of utilitarianism.

Our biological ancestry is a matter of fact that is not altered by the way we describe ourselves. With philosophical theories it is otherwise. Utilitarianism can be described in ways that make it look as if it is as old as moral philosophy – as J. S. Mill thought it was. For my historical purposes, it is more useful to have an account that brings out what is specific about Benthamism and its descendants. Let us try to make do with the following. First, utilitarianism asserts that the fundamental requirement of morality is that we are to maximize good, for everyone and not just for the agent. This basic principle presupposes that it makes sense to think of aggregating goods to make a total, and of comparing amounts of good thus aggregated. Second, the good to be brought about is located in feelings of pleasure, and the evil to be avoided in feelings of pain. These feelings have inherent value or disvalue regardless of how they are caused to exist and regardless of their own consequences. Third, all moral principles can be derived from the requirement that good be maximized. The principles involved in evaluating agents as well as in giving moral direction to action are nothing but applications of the basic principle.

Hume's theory of government and allegiance falls into two parts. In its better known segment Hume explains the conjectural origin of government in general as a convention necessary to enforce the rules of justice and provide other public goods, and he grounds the general duty of allegiance on the utility of government in making stable social life possible. To his credit, however, Hume goes on to give separate treatment to the topic of what he terms the ‘objects of allegiance”, or rules for assessing the legitimacy of particular political regimes or rulers. Given the general desirability of government and of obedience to it, we might nevertheless ask what entitles the particular government in existence to rule over us; more pressingly, there might be competing claimants to this position. What standards should guide our decision about where our allegiance is due when there is more than one alternative?

Was Green a utilitarian? At least two studies suggest that he was, at least in some sense. One claim is inspired by Macpherson's association of nineteenth-century liberalism with utilitarianism. Drawing on this argument, Greengarten and Hansen claim that Green's departure from utilitarianism is only partial. His commitment to capitalism indicates a subscription to utilitarianism since the latter is the justificatory force of capitalist institutions.