This book is accurately titled; the account of ethics that Kamm presents is indeed intricate. Moreover, Kamm discusses the views of so many contemporary philosophers that the book provides a fairly detailed critical introduction to contemporary moral philosophy. Kamm presents cogent criticisms of most consequentialist systems and even of some non-consequentialist systems such as Scanlon's contractualism. Kamm's criticisms suggest that she holds that all systems are too simple to account for the intricate ethics that she derives from examining her intuitions about particular cases. She describes her approach to deriving moral principles as follows:

Kamm distinguishes her method from that of ‘others [who] use variants of the method known as reflective equilibrium’. She differs from users of reflective equilibrium in that she thinks one should consider one's own

She refers to Marc Hauser, whose work provides psychological evidence for the view that ‘the same “deep structure” is present in all persons’, to support her approach, which assumes that ‘each person would give the same response’ (p. 8).

Kamm's approach assumes that if there is complete agreement on all of the morally relevant facts there will be agreement on how one should act in any moral situation, that is, if one has to choose between doing A and doing B, all equally informed moral agents will agree that one should do A or that one should B or that it is indifferent whether one does A or B. She does not argue for this claim, and serious consideration of it would quickly make it apparent how implausible it is. This is not to deny that in almost all moral situations all equally informed moral agents will agree, but that is because very few moral situations are controversial. However, in some controversial cases equally informed moral agents will not agree about how one should behave. Given that, within limits, it is rationally allowed to rank the various goods and evils differently, it is surprising that Kamm holds that in every moral situation all equally informed rational persons would give the same response. (For a fuller discussion see my article ‘Moral Arrogance and Moral Theories’, Nous Supplementary Volume, Philosophical Issues 15, Normativity (2005).)

It is difficult to summarize Kamm's positive views, for given her approach she develops about a dozen new principles, e.g. the Principle of Irrelevant Goods and the Principle of Secondary Wrong, in criticizing the views of other philosophers. She does hold that non-consequentialism is committed to the importance of the separateness of persons, but she does not put forward a position that one is even tempted to summarize as one might be tempted to summarize Kant by talking of the Categorical Imperative. She presents a large number of cases with slight variations and shows that other philosophers sometimes fail to use the principles she develops to distinguish between them. The level of care and detail is impressive.

However, Kamm does not always challenge the standard practices of contemporary moral philosophers. For example, she shares the standard practice of talking about ‘the impartial point of view’ without any further specification. She uses this phrase in many chapters of this book, never indicating that she realizes that this phrase does not have a clear reference. She even does this when discussing Peter Singer's theory where the lack of a clear reference is obviously significant. She does not seem to appreciate that it can be very misleading to talk about impartiality without further explicit specification. Arguing about whether the moral point of view involves an impartial point of view is pointless without specifying the group with regard to which morality requires one to be impartial and the respect in which morality requires one to be impartial with regard to members of this group.

The basic concept of impartiality should be defined incorporating these specifications, e.g. A is impartial in respect R with regard to group G if and only if A's actions in respect R are not influenced at all by which member(s) of G are benefited or harmed by these actions. Morality requires more than this basic concept, but that is not relevant to the point I am making now. Saying that morality involves taking the impartial point of view does not settle the question about who belongs in the group toward which morality requires impartiality. Nor does it settle the question about the respect in which one must be impartial with regard to members of this group. One cannot pick this group impartially, for it makes no sense to talk about being impartial before one has specified the group with regard to which one must be impartial.

Peter Singer seems to hold that morality requires impartiality with regard to all sentient beings, whereas Kant seems to hold that morality requires impartiality only with regard to moral agents or what he calls rational beings. This disagreement cannot even be sensibly discussed if one does not realize that the disagreement is about how to specify the group with regard to which morality requires impartiality. Further, Peter Singer and other act-consequentialists seem to hold that morality requires impartiality with respect to every action and inaction that is a possible option, whereas many non-consequentialists, including myself, hold that morality requires impartiality only with regard to moral rules. Singer does not seem to realize that his view also requires that people always be thinking about the consequences of their actions and inactions on all sentient beings, for acting impartially only towards those one happens to be thinking about at the time will obviously not give him the results he want.

It is also important to appreciate that talk about impartiality requires specifying the respect in which one is impartial to the relevant group. A teacher may be impartial with respect to grading the papers of her students, but not impartial with respect to calling on them in class, for she may favor boys over girls in this respect. Kamm's uncritical use of the phrase “the impartial point of view” allows her to take part in the misguided discussion on whether morality requires taking the impartial point of view. Recognizing that impartiality requires specification of group and respect would allow the discussion of whether morality requires impartiality to be far more productive. Those holding different views could explicitly argue for their view about the nature of the group with regard to which morality requires impartiality. It might even come to be realized that it is rationally allowed to hold different views about how to specify the group with regard to which morality requires impartiality.

Another point, but of much less significance, in which Kamm is in agreement with the standard practice of moral philosophers, and not only contemporary ones, is in her use of the word ‘duty’. Following Kant, Mill, and Ross she uses ‘duty’ as if it meant the same as ‘morally ought’. She talks about the duty not to kill and the duty to aid. I morally ought not to kill another person, but I do not have a duty not to kill another person. The use of ‘duty’ in the phrase ‘duty not to kill’ seems to me to be an almost purely philosophical use of ‘duty’. It is probably no longer a purely philosophical use as so many people who have taken philosophy courses in college have accepted this technical use. Nonetheless, it seems to me that Bradley was far closer to ordinary use when he talked about ‘My Station and its Duties’. Use of the word ‘duty’ as equivalent to ‘morally ought’ makes it far more difficult to talk about genuine duties, those that arise from one's job, social role, or special circumstances. Further, talking about two kinds of duties, perfect and imperfect, makes them sound far more alike than they are. It is usually morally good to aid, but usually no one has a duty to aid. If one is going to start with one's natural intuitions about case-based judgments, one would be better off not describing these cases using terms like ‘duty’ in their philosophical use.

My final comment is one that involves Kamm's discussion of my moral theory. I am pleased that she included in this book (chapter 15) the short paper that she wrote for the volume Rationality, Rules, and Ideals: Critical Essays on Bernard Gert's Moral Theory and pleased that she seems to accept the distinctions etc. in my moral theory that she discusses. Here, I shall discuss one claim that she makes that seems to involve an important mistake. She says:

Kamm acknowledges that I deny that morality permits such behavior, but she seems to hold that this is inconsistent with my claim that morality is concerned with reducing evils. In my reply I point out: ‘It is a mistake to say that if the point of a game is to test one's athletic abilities, every move in the game should be determined by which move best tests one's athletic abilities’ (304). I go on to point out that on my account of morality, it is a public system for governing the behavior of all those who can understand it and can control their behavior accordingly. My claim is that the point of this public system, which is the moral system or morality, is to lessen the amount of evil suffered. This is not an endorsement of negative act-consequentialism, which does seem to claim that every action should be determined by how much evil is reduced. Kamm's claim is not true even if I were a negative rule-consequentialist, which I am not. It is surprising that someone who is as careful as Kamm would make such a mistake.

This is a fascinating book, with insightful discussions of a wide range of philosophical topics and contemporary philosophers. All philosophers will gain from reading it.