The third topic is a particular kind of value, which we can call “ideals.” Foot gives as an example men who opposed the Nazis and were captured; their letters showed they were capable of enjoying life. They didn’t exactly sacrifice their happiness; they instead realized that happiness was no longer an option. Still, they accepted their suffering and persevered, realizing happiness was an ideal no longer realizable. Another example is that we want people we love not to forsake their virtue when life makes virtue difficult for them. This kind of wanting is an “ideal” preference.

Putting the matter in terms of “ideal” preferences allows a connection with RMH. This matters for the present chapter because we are discussing whether we can deduce moral goodness from our nature. If a large part of the goodness of a human life is specified by our ideals and most ideals are not deducible from our nature, this will put the deductivist in a hard position. This point will be made later; for now Hare wants to emphasize how RMH was in a difficulty about ideals, unable to accommodate ideals in his version of utilitarianism. RMH’s view of morality was that of universalized prudence, meaning that moral thinking has a two-step process: we first determine what prudence dictates from each person’s point of view affected by our action, then make a moral decision by giving equal weight to all those points of view including our own. RMH admitted he couldn’t accommodate ideals within his theory. Gibbard argues that RMH’s proof of utilitarianism doesn’t work if we try to make it cover ideal preferences. If our conception of the good human life can’t be deduced either from maximizing basic-preference satisfaction, or from our nature, we need some other standard for discernment.

For both RMH and Foot, religion provided central cases of ideals. Foot’s letter writers were Christians, and RMH’s central example in Freedom and Reason was St. Francis. For Hursthouse, piety to the Judeo-Christian God is a virtue that “undoubtedly brings great joy and serenity to its possessors, [but] no atheist can regard such joy as ‘characteristic of human beings’, that is, as something that reason can endorse.” But Hare says we need to acknowledge how utterly pervasive ideals are even in ordinary non-religious thought about the good life. Morality itself is a universal ideal if the moral agent prefers that she herself and everyone else live morally whether she continues to have that preference or not. The preference to approximate the archangel’s thinking is itself a universal ideal, and the morally good life is not dependent for its value on anyone desiring it.

RMH introduced the figure of the archangel to avoid talk of God. But he sometimes acknowledged that he was in fact talking about God. The fact that God was his model of critical thinking supports the claim that this kind of thinking is an ideal.

Much of our moral thinking, including our commitment to morality itself, is in terms of ideal preferences. We need some way to determine which ideals to try to live by. The archangelic method is insufficient. The next section of this chapter argues that the way of deduction from facts about human nature is insufficient also, and for some of the same reasons. Hare suggest that what we should learn from the references in religion in RMH and Foot and Hursthouse is that faith in God and receptivity to divine command can give us a way to select the ideals that shape our conception of the good human life, even when basic-preference satisfaction and deduction from our nature do not give it to us.