¹ Guerrero, Alexander A., ‘Don't Know, Don't Kill: Moral Ignorance, Culpability, and Caution’, Philosophical Studies 136 (2007), pp. 59–97CrossRefGoogle Scholar; Lockhart, T., Moral Uncertainty and its Consequences (New York, 2000)Google Scholar; Oddie, Graham, ‘Moral Uncertainty and Human Embryo Experimentation’, Medicine and Moral Reasoning, ed. Fulford, K. W. M., Gillett, G. and Soskice, J. Martin (Cambridge, 1994), pp. 144–61CrossRefGoogle Scholar; Moller, Dan, ‘Abortion and Moral Risk’, Philosophy 86 (2011), pp. 425–43CrossRefGoogle Scholar.

² See Guerrero, ‘Don't Know, Don't Kill’ and Moller, ‘Abortion and Moral Risk’. Sometimes this and the case against abortion are presented as a dominance argument, where vegetarianism, or having a child, is suggested to be certainly permissible (Lockhart, Moral Uncertainty, ch. 2; Weatherson, Brian, ‘Review of Ted Lockhart, Moral Uncertainty and Its Consequences’, Mind 111 (2002), pp. 693–6)CrossRefGoogle Scholar. However, I think that we should be considering how to make decisions in light of all the possible reasons for action that one has. And if one believes that there is no moral reason against eating meat, whereas there is a prudential reason in favour of eating meat, then eating meat is the most all-things-considered choice-worthy option. So the ‘dominance’ form of the argument will almost never apply.

³ Lockhart, Moral Uncertainty, p. 52.

⁴ Moller: ‘[the moral uncertainty argument] does seem to suggest, however, that there is a moral reason – probably not a weak one – for most agents to avoid abortion’ (‘Abortion and Moral Risk’, p. 443). Lockhart: ‘In the vast majority of situations in which decision-makers decide whether to have abortions, not having an abortion is the reasonable choice of action’ (Moral Uncertainty and its Consequences, p. 52). Pope John Paul II: ‘the mere probability that a human person is involved [in the practice of abortion] would suffice to justify an absolute clear prohibition of any intervention aimed at killing a human embryo’ (http://w2.vatican.va/content/john-paul-ii/en/encyclicals/documents/hf_jp-ii_enc_25031995_evangelium-vitae.html).

⁵ Weatherson: ‘[implications] so striking we might fear for its refutation by a quick modus tollens’ (‘Review of Lockhart’, p. 694). Guerrero: ‘[maximizing expected moral value] is not the reading that I prefer, in part because of cases like [abortion]’ (‘Don't Know, Don't Kill’, p. 91).

⁶ For this account, see Sepielli, Andrew, ‘What to Do When You Don't Know What To Do’, Oxford Studies in Metaethics 4 (2009), pp. 5–28Google Scholar; William MacAskill, ‘Moral Uncertainty’ (DPhil dissertation, Oxford University, 2014).

⁷ Ross, Jacob, ‘Rejecting Ethical Deflationism’, Ethics 116 (2006), sects. 4–5CrossRefGoogle Scholar.

⁸ See MacAskill, William, ‘The Infectiousness of Nihilism’, Ethics 123 (2013), pp. 508–20CrossRefGoogle Scholar.

⁹ In ‘Normative Uncertainty as a Voting Problem’, Mind 125 (2016), pp. 967–1004CrossRefGoogle Scholar, I suggest an extension of maximizing expected choice-worthiness that attempts to deal with some of these problems.

¹⁰ The implications of moral uncertainty have been discussed for abortion (Lockhart, Moral Uncertainty, ch. 3; Moller, ‘Abortion and Moral Risk’), embryo destruction (Oddie, ‘Moral Uncertainty and Human Embryo Experimentation’), vegetarianism (Moller, ‘Abortion and Moral Risk’, pp. 426, 441–3; Guerrero, ‘Don't Know, Don't Kill’, pp. 76–82), the ethics of killing more generally (Guerrero, ‘Don't Know, Don't Kill’) and duties of beneficence (Lockhart, Moral Uncertainty ch. 5; Weatherson, ‘Review of Lockhart’). I don't know of other examples of the practical issues being discussed, so believe that the suggested implications for partiality, egalitarianism, the suffering/happiness trade-off, theories of well-being, welfarism, egoism and populations ethics are novel.

¹¹ Singer, Peter, ‘Famine, Affluence and Morality’, Philosophy & Public Affairs 1 (1972), pp. 229–43Google Scholar.

¹² For a more comprehensive discussion of these different views, see Greaves, Hilary and Ord, Toby, ‘Moral Uncertainty about Population Axiology’, Journal of Ethics and Social Philosophy 12 (2017), pp. 135–67CrossRefGoogle Scholar.

¹³ See Blackorby, Charles and Donaldson, David, ‘Social Criteria for Evaluating Population Change’, Journal of Public Economics 25 (1984), pp. 13–33CrossRefGoogle Scholar; Blackorby, Charles, Bossert, Walter and Donaldson, David, ‘Intertemporal Population Ethics: Critical-Level Utilitarian Principles’, Econometrica 63 (1995), pp. 1303–20CrossRefGoogle ScholarPubMed; Broome, J., Weighing Lives (Oxford, 2004)CrossRefGoogle Scholar; Blackorby, C., Bossert, W. and Donaldson, D., Population Issues in Social Choice Theory, Welfare Economics, and Ethics (Cambridge, 2005)CrossRefGoogle Scholar.

¹⁴ This idea is developed in Greaves and Ord, ‘Moral Uncertainty about Population Axiology’.

¹⁵ See Hurka, Thomas, ‘Value and Population Size’, Ethics 93 (1982), pp. 496–507CrossRefGoogle Scholar; Ng, Yew-Kwang, ‘What Should We Do About Future Generations?’, Economics and Philosophy 5 (1989), pp. 235–53CrossRefGoogle Scholar.

¹⁶ See Narveson, Jan, ‘Moral Problems of Population’, The Monist 57 (1973), pp. 62–86CrossRefGoogle Scholar.

¹⁷ For example, Lockhart, Moral Uncertainty, p. 52.

¹⁸ Moller, ‘Abortion and Moral Risk’, p. 441.

¹⁹ Weatherson, ‘Review of Lockhart’, p. 693.

²⁰ An assessment of the welfare levels of various farm animals is given in Norwood, F. B. and Lusk, J. L., Compassion, by the Pound: The Economics of Farm Animal Welfare (New York, 2011), p. 229CrossRefGoogle Scholar, and in Cooney, N., Veganomics: The Surprising Science on What Motivates Vegetarians, from the Breakfast Table to the Bedroom (New York, 2013), ch. 1Google Scholar.

²¹ For example, this view can take the form of an endorsement of Christopher Kutz's Complicity Principle (Complicity: Ethics and Law for a Collective Age (Cambridge, 2000), p. 122Google Scholar), according to which ‘I am accountable for the harm or wrong we do together, independently of the actual difference I make’, combined with the claim that we together wrong the animals we eat.

²² According to the latest estimates from GiveWell, it costs about $3,200 to do the equivalent amount of good to saving a life in poor countries (‘GiveWell cost-effectiveness analysis’, November 2016, https://docs.google.com/spreadsheets/d/1KiWfiAGX_QZhRbC9xkzf3I8IqsXC5kkr-nwY_feVlcM). In order for the costs of a strict vegetarian diet to be greater than the cost to save a life, the strict vegetarian diet would only have to cost an additional $1.53 per week over a span of 40 years. One might object that a vegetarian diet is cheaper than an omnivorous diet. This may, typically, be true. However, because one loses options by being vegetarian, a vegetarian diet must be at least as costly as the diet one has if one acts on the maxim ‘eat whatever's cheapest’, and it seems unlikely that such a maxim would never involve eating meat.

²³ Note that we need to include three outcomes in this table because we are interested in making intertheoretic comparisons of choice-worthiness differences rather than intertheoretic comparisons of choice-worthiness levels.

²⁴ One might claim that (i) one ought to have credence in both possible normalizations and that (ii) given this, the theory with the higher-stakes normalization will still be the primary determiner of different options’ expected choice-worthiness. I find this plausible to some extent, but believe it still depends on what exactly one's credences are; if one has a very small credence in the high-stakes normalization, then one might worry that one is entering ‘fanaticism’ territory if one thinks that the recommendation of MEC in this instance is correct.

²⁵ For the argument why total view consequentialists should care almost exclusively about impacts on the long-run future of the human race, and why, on their view, there is a truly vast amount of value at stake, see Bostrom, Nick, ‘Astronomical Waste: The Opportunity Cost of Delayed Technological Development’, Utilitas 15 (2003), pp. 308–14CrossRefGoogle Scholar.

²⁶ For their generous comments and discussion, I would like to thank Frank Artnzenius, Amanda Askell, John Broome, Krister Bykvist, Toby Ord, Peter Singer, Christian Tarsney, Ralph Wedgwood, and audiences at the University of Oxford, The London School of Economics, the Princeton Center for Human Values Ira W. DeCamp Bioethics Seminar, and the Uehiro Centre Applied Ethics Discussion Group.