One of the crucial steps in the method of moral reasoning advocated by R. M. Hare in his book Moral Thinking involves one mirroring within oneself the preferences of the other people affected by one's action.' Hare argues that only by acquiring the hypothetical preference that, were I in the other person's situation, x should happen (where x is what the other person prefers to happen) can I satisfy the demands of universalizability (of my ought judgment). This is because universalizability requires that I understand the other person's situation in such a way that if I were in his situation I would prefer what he in fact prefers. And I can only reach this understanding, Hare thinks, by mirroring his preferences. That is, Hare argues that only by its being the case that (1) I now prefer with strength S that if I were in that situation X should happen rather than can I come to know that (2) If I were in that situation, I would prefer with strength S that X should happen rather than not.2 As Ingmar Persson has pointed out this argument seems unsound since it is possible to think of cases where one might perfectly well know that if one were in a certain situation one would have a certain preference without its being the case that one now prefers that if one were in that situation this thing should happen.3 For instance, a reformed alcoholic, now sober, might know that if he were on a binge he would prefer another drink. At the same time he might clearly see the consequences of his drinking and so not now prefer that if he were on a binge he should take another drink. There is, however, a further difficulty with this point in Hare's method of moral reasoning. Hare's idea is that once I have, so to speak, mirrored in myself the preferences of the other people affected by my proposed action, I can simply weigh up these preferences exactly as I do when I have an intrapersonal conflict of preferences. Speaking of a case in which someone has parked his bicycle in the spot in which I need to park my