The paper considers three possible definitions of what it is for an action to be ‘morally’ good: (1) that it is overall important to do; (2) that it is overall important to do in virtue of a universalisable principle; and (3) that it is overall important to do in virtue of a universalisable principle, belonging to a system of such principles, which includes almost all of certain moral fixed points. I defend (3) and show how we can reach such a system, starting from the basic beliefs with which we find ourselves, through the process of reflective equilibrium. Moral realism is then the doctrine that there is such a system of true moral beliefs. My optimistic view is that all human communities could eventually reach the same such system. But, if they cannot, then there will be two (or more) different such systems, and so two (or more) different senses of ‘moral realism’.
