In his Nicomachean Ethics Aristotle remarks that we must be satisfied with the kind of certainty that each subject allows. That may be relevant here. We have to ask ourselves what kind of certainty is possible in the case of moral principles.

As far as knowledge goes, it may be useful to keep in mind that we already use different concepts of knowledge in different areas. What counts as knowledge in biology is different from what counts as knowledge in mathematics. If we insisted on applying the mathematical concept of knowledge to biology, we would have no biological knowledge at all.

Maybe this can help us when we think about morals. Maybe we need to revise the concept of knowledge in order to apply it to morals.

In any case, from the point of view of the relativist, moral knowledge might be thought of as the results of the approach that we have described, based on the ideas of Rawls and Regan. I'm not saying that Rawls and Regan are relativists. I'm saying that the relativist could use some of their ideas and call it moral knowledge.

If we choose to revise the concept, I think the important thing is to avoid being gratuitous. There must be a clear bridge or connection between our idea of moral knowledge and other ideas of knowledge. That maybe captured in the notion of "good reason to believe or accept" certain principles and the requirements described by Rawls and Regan come into play.

I know we can also drop the concept of ethical knowledge altogether as professor Sidgwick suggested. But I like keeping it. I think Aristotle may be right.