Is it really correct to speak of believing in these contexts, or is something less cognitive demanded in moral faith? Adams chooses to concentrate on some of the concrete features of moral faith that incline him to speak of “belief” here. For no theory on metaethical issues is likely to attain a very high degree of certainty. Moral faith is therefore a stance we will have to take, if we are reasonable, in the face of the recognition that any metaethical theory we may hold could rather easily be mistaken. So it would be good to have an understanding of the stance that does not presuppose very much metaethics.

Adams thinks both will and feeling are involved in moral faith, but he does not think that moral faith is merely will and feeling, or that believing another person’s life is worth living is merely caring about that life. It’s not just a volitional matter; an intention central to moral faith is an intention of respecting something more commanding, more external to the self than mere personal preference and feeling.

This is the most important reason for speaking of moral faith as a sort of “belief,” and it is connected with the possibility of error. Faith confronts a temptation to doubt precisely because such possibilities of error must be recognized, and in a way respected.

Emotions too can be mistaken, but it’s far from clear that we can understand how an emotion of faith in the value of life can be inappropriately related to reality if we can’t understand how faith can be, or involve, a false belief that life is worth living. In any event, the possibility of an objectively appropriate or inappropriate relation to reality is precisely the aspect of moral belief most subject to metaethical doubts, and also the aspect that seems to Adams most important to the nonegocentric character of moral faith.

Connected with the possibility of error is the giving of reasons for and against beliefs. In thinking about items of moral faith one uses logic, one aims at consistency and at coherence with one’s beliefs on other subjects, and one is responsive to one’s sense of “plausibility,” as we sometimes put it. All of that provides grounds for classifying moral faith as a sort of belief.

Particular interest attaches to the question of responsiveness or unresponsiveness of moral faith to the evidence of experience. Our faith in the value of particular human lives, in the value and the possibility of a moral life, and in the possibility of a common good, can be put under strain by particular experiences. Indeed, adverse experience is precisely what gives rise, as Adams has argued, to a problem of evil for moral faith.

There is thus a considerable empirical element in faith in moral ends. But Adams does not believe that science, or social science, could devise a definitive empirical test of the truth of faith in any moral end. Objects of faith have vaguer contours that permit reformulation in the face of adverse experience, so we can’t identify experiences that are unequivocally predicted or excluded by such items of faith as that so-and-so’s life is worth living. So faith is not normally subject to definitive proof or refutation by any specifiable finite set of experiences. And from the perspective of moral faith, this is as it should be, for moral faith is supposed to be resistant to adverse evidence.

Empiricists may take offense at this feature of faith. Faith as such is indeed resistant to adverse experience, and is apt to revise itself before simply accepting refutation. Flew is right that faith is in danger of evacuating itself of content if its resistance is undiscriminating or absolutely unconditional. Nonetheless, Adams believes that resistance to adverse experience, and to refutation in general, is an appropriate feature of faith; and he will argue this with specific reference to moral faith.

Our interest in items of faith is importantly different from our interest in scientific hypotheses. Conclusive falsification of a hypothesis is progress in science. But falsification of an item of faith is not progress—at least not from the perspective within which it is an item of faith. To think that falsification of the belief in morality itself, or of the belief that a moral life is worth living, might be pure progress is already to hold an amoral view, a morally bad view. A loss of moral faith would be the loss of something precious.

When we resist refutation of an item of moral faith, we may, and should, be thinking of the danger of being misled into giving it up while it is true. From a moral point of view that would be a worse mistake to make than the mistake of clinging to moral faith while it is false. Even if an item of moral faith is false, moreover, we are not likely in abandoning it to attain anything corresponding to the moral value of believing it if it is true. This is an important asymmetry.

The balance of potential payoffs is much more equal when it is a question of revising moral faith, rather than abandoning it. We can hope that revision of moral faith will be progress from a moral point of view. This is a further reason for thinking it is good for moral faith to combine a variable, revisable form with a vaguer but more enduring core, so that self-critical growth and development may be combined with constancy of commitment.

Items of faith are not hypotheses to be tested by experience, though we may well want their formulations to be tested by experience. Items of faith may in fact be tested by experience, but we are not trying to refute them. We are trying to live by them. It’s not that moral faith is wholly unempirical, let alone noncognitive, but that it involves a different way of accommodating thought to experience.

Maybe this suggests too stark a contrast between morality and science. Notoriously, the reliability of induction and more broadly of empirical scientific methods has been doubted by reasonable persons and it may not be possible to set the doubts to rest in a completely satisfying way. Yet a refutation of the reliability of induction would not be scientific progress in the same way that a refutation of the meteorite explanation of the extinction of dinosaurs might be. So perhaps there is a place, or even a need, for faith in the highest level beliefs of science, but for now Adams is content to make a point just about moral faith.