Review of “Moral Uncertainty and Its Consequences”

For many years now, Peter Singer has been arguing that we should not eat meat, and that we should give more money to famine relief. Many have been convinced, but many more remain sceptical. However, on one point most of us would agree: the actions that Singer recommends here are certainly morally permissible. One rarely feels a twang of moral doubt when eating tofu curry or writing cheques to Oxfam. Even if we do not find Singer totally convincing, we may still feel this moral doubt when eating sirloin, or spending frivolously rather than charitably. If we accept the main principle in Ted Lockhart’s book Moral Uncertainty and Its Consequences, these twangs of moral doubt should be sufficient to make us amend our behaviour.

The main principle Lockhart endorses is that we should perform actions that we are maximally confident are morally permissible. We might be quite confident that having the sirloin is morally permissible, but if we are not certain, and we are certain the tofu is permissible, we should stick to tofu. Similarly, if we are certain that large donations to famine relief are permissible, and not certain that not making these donations is permissible, the chequebook should come out. The principle is not just for left-wingers. As Lockhart notes, approvingly, it can also be used in anti-abortion arguments. In most cases, not having an abortion is almost certainly permissible. Perhaps there is an exception for cases of extreme fetal deformity, but not in everyday cases. So if the woman considering an abortion wants to do the action that is most probably morally permissible, and has any doubts about the permissibility of the procedure, she should decline the abortion.

The bulk of Lockhart’s book is devoted to case studies where this principle is deployed, and amendments to the principle generated by considerations of these cases are adopted. The cases include abortion, patient confidentiality, Roe v Wade and, briefly, charitable giving. The theme behind the studies is that even if people cannot come to agreement on what is morally right, they can come to agreement on what should be done according to the principle, at least as variously amended, and this should be sufficient to provide recommendations for action. Lockhart stresses that if this line of reasoning is correct, then applied ethicists can provide good advice on practical action without conclusively resolving apparently intractable ethical problems.

There are three main amendments Lockhart suggests to the principle. First, he suggests that if moral rightness comes in degrees, we should maximise the expected moral rightness of our actions, rather than the probability that we are doing the right thing. Secondly, in situations where we cannot work out which action maximises expected rightness, because perhaps we do not have perfect access to the relevant subjective probabilities, we should choose the action which most probably maximises expected rightness, or more generally has the highest expected expected degree of moral rightness. And thirdly, he says that we should maximise the expected rightness of courses of action, rather than of individual actions. One might quibble with these amendments, particularly I think with the second, but they do not seem to affect the core philosophical issues.

The principle has some rather striking consequences, so striking we might fear for its refutation by a quick modus tollens. Lockhart, of course, does not think this is so. He does not discuss the vegetarianism issue, and endorses the anti-abortion implications, but argues that the principle need not have such striking implications concerning charitable giving. He notes that for some people, those who think it probable enough that substantial charitable giving is a very bad thing to do, because we have such strong obligations to ourselves and those nearest and dearest, his principle does not recommend such giving (109).

There is a more direct reason for thinking the principle stands in need of some further clarification and defence. It is rather unclear what kind of norm the principle is stating, and hence what force the should in it is has. Lockhart says it is a norm of rational action, but it seems in practice to be neither that, nor a moral norm. To see this, consider the following case where someone clearly does not follow the principle. While on her way to visit a sick friend in hospital, Jane is convinced by a fellow subway rider that morality requires an impersonal concern for the whole world. She is convinced that morality requires that she not visit her friend, but instead find the patient most in need of a visitor, and see them. But when she gets to the hospital, her new moral belief is not strong enough to overcome her desire to visit her friend in need, which, feeling a little guilty, she does.

Assuming that Jane’s newfound moral beliefs are wrong, and that in fact she did the right thing, what criticisms can we make of her action? Not that it was immoral, because she did the right thing, visiting her sick friend, and she acted for the right reason, acting out of care for her friend. Nor, it seems, that it was prudentially irrational, for she did what she believed would best satisfy her desires. Perhaps the fact that her new moral beliefs were not sufficiently motivating indicates a lack of resolve, or even a weakness of will, but alternatively one might think that Jane displayed commendable, and virtuous, common sense in not abandoning her friend precipitously. In any case, I doubt Jane’s action cannot be criticised, even if her resolve can be. Since Jane clearly violated Lockhart’s principle, she did not act in the way she thought most likely to be morally permissible, but her action seems immune from criticism, that suggests the principle should not be an action guiding norm.

One might argue that Jane has a moral responsibility to desire to do the right thing, and if she had this desire, she would have been rationally required to not visit her friend. If one believes in such a responsibility, then one will think that Jane acted against a desire she should have, that she was, at best, lucky that she did the right thing, and hence she was irrational. Lockhart compares such agents, who do the right thing against their better judgement, to gamblers who bet their life savings on unlikely, but ultimately successful, outcomes. (34)

This line of reasoning, however, ultimately does not provide grounds for criticising Jane. A moral agent may well have a moral responsibility to desire to do the things that happen to be the right things to do. For example, she may well have a responsibility to want to visit her sick friends, and to help those in need, and not cause harm to others. But she does not have a responsibility to want to do the right thing, whatever it turns out to be. Indeed, she would be a worse moral agent if many of her actions were motivated by such a desire. She should want to visit her friend because she cares about her friend, not because it is, in the abstract, the right thing to do. Michael Smith has described the desire to do the right thing, whatever it turns out to be, as a moral fetish, and this often seems appropriate. (The Moral Problem, Oxford: Blackwell, 1994, p. 76)

It is no discredit to Jane that she lacks this general desire, and in some cases it may be a virtue. If Jane has the general desire, if in Smith’s terminology she is a moral fetishist, then she may be prudentially required to follow Lockhart’s principle, but not otherwise, and she is not required, by any normative standard, to be a moral fetishist. If Jane (virtuously) does not have that general desire to do the right thing, whatever it turns out to be, then she is importantly dissimilar to the gambler, who does (and should) desire to bet on the successful outcome, whatever it turns out to be.

Whatever the merits of Lockhart’s main principle, his approach raises several fascinating theoretical questions. For example, there is a substantial literature on what the motivational effects of coming to hold a new moral view are, and what they should be. But what is, and what should be, the motivational effects of coming to hold, say, that it is more probable than not that meat eating is permissible? From a different angle, if moral attitudes are more like desires than like beliefs, as some expressivists suggest, then can we even have the attitude that it is more probable than not that meat eating is permissible? Although in general Lockhart says little directly on these theoretical questions, it is a great service to show how they arise.

If Lockhart’s main principle is correct, it has rather radical implications for how applied ethics is practised. Even if it is not, consideration of the issues Lockhart raises may provide a novel and valuable outlook on some familiar theoretical questions.