The problem I am going to discuss here concerns practical rationality, rationality not in thought but in action. More particularly, I am going to discuss the rationality, or absence of rationality (even, as one might put it, the contra-rationality or irrationality) of moral action. And ‘moral action’ shall mean here something done by someone who (let us suppose rightly) believes that to act otherwise would be contrary to, say, justice or charity; or again not done because it is thought that it would be unjust or uncharitable to do it. The question is whether in so acting, or refusing to act, this person will be acting rationally, even in cases where he or she believes that not only desire but self-interest would argue in favour of the wrongdoing.

I am very happy indeed to contribute to this series of lectures, especially because I owe most of my training in philosophy to Elizabeth Anscombe, whose work has given the series its name. I am deeply indebted to the marvellous generosity of her teaching, to the example she set me of an unrelentingly thorough and serious thinker, to the unobtrusive way she introduced me to Wittgenstein's later philosophy. Through Elizabeth Anscombe I also made the acquaintance of my friend Philippa Foot, whose work in moral philosophy has, over decades and more than anyone else's, influenced my own. I hope it will be possible to recognize in what I am going to say here not indeed the excellence but at least traces of the beneficial influence of both these philosophers.

My immediate aim in this lecture is to contribute something to the apt characterization of our representation and knowledge of the specifically human life form, as I will put it—and, to some extent, of things ‘human’ more generally. In particular I want to argue against an exaggerated empiricism about such cognition. Meditation on these themes might be pursued as having a kind of interest of its own, an epistemological and in the end metaphysical interest, but my own purpose in the matter is practical-philosophical. I want to employ my theses to make room for a certain range of doctrines in ethical theory and the theory of practical rationality—doctrines, namely, of natural normativity or natural goodness, as we may call them. I am not proposing to attempt a positive argument for any such ‘neo-Aristotelian’ position, but merely to defend such views against certain familiar lines of objection; and even here my aims will be limited, as will be seen.

Someone once told me that the average number of readers of a philosophy article is about six. That is a particularly depressing thought when one takes into account the huge influence of certain articles. When I think of, say, Gettier's article on knowledge, or Quine's ‘Two Dogmas’, I begin to wonder whether anyone is ever likely to read anything I write. Usually the arguments of these very influential articles have been subjected to widespread analysis and interpretation. The case of Elizabeth Anscombe's ‘Modern Moral Philosophy’, published in 1958, is something of an exception. That article has played a significant part in the development of so-called ‘virtue ethics’, which has burgeoned over the last three decades in particular. But there has been less close attention to its arguments than one might have expected.

How is it possible to love some particular person for herself, or for himself, alone? Love—especially erotic love—does not typically begin when we love someone ‘for herself alone’. It very often begins with some strikingly superficial feature or property of the beloved: a certain grace of movement, maybe, or a glimpse inside a young man0027;s shirt (Plato, Charmides 155d4), or the colour of Anne Gregory's hair in Yeats's poem. As Yeats sardonically points out, it is quite common for love never to get any further than this. It's hardly news that plenty of love affairs have rested on no deeper foundations than hair coloration.

Different questions generate different forms of practical reasoning. A contextually unrestricted ‘What shall I do?’ is too open to focus reflection. More determinately, an agent may ask, ‘Shall I do X, or Y?’ To answer that, he may need to weigh things up—as fits the derivation of ‘deliberation’ from libra (Latin for ‘scales’). Ubiquitous and indispensable though this is, I mention it only to salute it in passing. Or he may ask how to achieve a proposed end: if his end is to do X, he may ask ‘How shall I do X?’ Or he may ask how to apply a universal rule or particular maxim. Aristotle supplies examples in De Motu Animalium (7.701a7 ff.), whose wording I freely adapt to my own purposes:

A1 reasons to a necessary means to achieving an end:

I will make a cloak.

To make a cloak I must do A.

So, I will do A.

Elizabeth Anscombe's ‘Modern Moral Philosophy’ is read and remembered principally as a critique of the state of ethical theory at the time when she was writing—an account of certain faulty assumptions underlying that theory in its different variants, and rendering trivial the points on which they ostensibly disagree. Not unreasonably, the essay serves as a starting point for the recent Oxford Readings collection on ‘virtue ethics’, and as an authoritative text on the failings of other approaches with which philosophy students have to acquaint themselves. Yet what really commands attention on rereading it is Anscombe's denunciation of the impotence of current moral philosophy to generate resistance to certain quite specific forms of wrongdoing. The question that provides a kind of gold standard here, recurring several times in the course of the discussion, is that of the killing (or judicial execution or other ‘punishment’) of the innocent in order to avoid some putatively greater evil, or to bring about some sufficiently great good.

Moral philosophy characteristically sees moral standards as reasons. That an action would be kind or just or in some way morally admirable is supposed to give us a reason for performing it. And surely there is something right about the thought that moral standards imply reasons for conforming to them. For we offer the morality of an action as a relevant consideration in practical argument— a consideration to support that action's performance. You should provide the help, because it would be kind, or just, and so forth. And an argument, surely, is in the business of offering reasons for what the argument supports.

‘The Russian Revolution and the National Socialist ascendancy in Germany are the two most important sources of evidence of moral philosophy in our time, as the French Revolution was for Hegel and Marx, and later to Tocqueville and for Mill. Although both revolutions produced, both in intention and in effect, a triumph on a gigantic scale, there are often remarked differences between the evil effects planned and achieved.’ This is an observation made by Stuart Hampshire, a keen philosophical connoisseur of the 20th century.

There are a number of ways in which a person can share the guilt of another's wrongdoing. He might advise it, command it or consent to it. He might provoke it, praise it, flatter the wrongdoer, or conceal the wrong. He might stay silent when there is a clear duty to denounce the wrong or its perpetrator; or he might positively defend the wrong done. Finally, he might actively participate or cooperate in the wrongdoing. These various activities, apart from cooperation, typically occur before or after the commission of the wrong itself, only provocation being essentially before the fact. As such they fall into the categories of seduction or comfort, seduction being essentially pre-commission and comfort post-commission. In seduction (mutatis mutandis for comfort), the seducer typically leads another into doing wrong who has not definitely made up his mind. He does not assist in the commission, but he leads to its occurring. If the principal (as I will call the one who commits the wrong) has made up his mind, actions which might otherwise amount to seduction are best characterized as amounting to scandal, since they do not lead to wrong but reinforce the principal in his wrongful intent or provide to third parties a bad example since they connote approval of the principal's action. Closely related to the concept of seduction is that of solicitation, though perhaps these are best thought of as two aspects of the same kind of activity. Seduction can be thought of as a strong form of inducement to wrong, typified by command, counsel (where the seducer knows the advice is likely to be relied upon) and enticement through praise or provocation. Solicitation is a softer form of inducement typically involving requests, appeals, and invitations. Whereas the seducer or solicitor leads another into wrong but does not assist in its commission, the co-operator does not lead the principal into wrong but assists in its commission.

As children, we are often told both what to do and what to think. For a child to learn at all, it must in the first instance simply trust those, such as parents, who teach it things; and this goes for practical as well as theoretical learning. Doubting is necessarily something that comes later, for to be able to doubt one must have some beliefs already, e.g. concerning what sort of reasons count as good reasons, and what count as bad. But in growing up, a person does, or should, develop the capacity for rational doubt, and also the capacity for rational resistance to being told what to do. The first capacity constitutes a critical faculty, and the second is an essential constituent of practical autonomy.

A view as widely endorsed as it is disputed says, formulating it in my own words: The only thing we have reason to do is promote value. This I will call The promotion of value thesis (or principle).

It is the famous first thesis of Anscombe's ‘Modern Moral Philosophy’ that we should lay aside moral philosophy—indeed ‘banish ethics totally from our minds’! (p. 38, paragraph 36)—‘until we have an adequate philosophy of psychology’. By a ‘philosophy of psychology’ I understand Anscombe to mean grammatical investigations into various psychological concepts that hold the key to ethics. Anscombe herself instances ‘action’, ‘intention’, ‘pleasure’, ‘wanting’ (‘more will probably turn up if we start with these’). Without such an understanding, she thinks we will simply go astray.

Anscombe's indictment of modern moral philosophy is full-blooded. She began with three strong claims:

The first is that is not profitable to do moral philosophy… until we have an adequate philosophy of psychology, in which we are conspicuously lacking. The second is that the concepts of obligation and duty… and of the moral sense of ‘ought’, ought to be jettisoned… because they are derivatives… from an earlier conception of ethics… and are only harmful without it. The third thesis is that the differences between the well-known English writers on moral philosophy from Sidgwick to the present are of little importance.