We were led, at the close of the last paper, to the conclusion that the moral judgment lays claim to a knowledge of what is unknowable. It is not merely that our volition is imperfect, that the act of necessity falls short of what we know to be right. This seems bad enough; but the plight in which we actually find ourselves is even worse. The paradox is that we never know, and never can know, in any particular situation, what it is really right to do. We know indeed that it is always right, really and absolutely right, to do what we believe to be right. For a man to act “against his conscience,” after all possible thought has been taken for its enlightenment, we know to be morally wrong. But this knowledge is purely formal and gives no clue to the matter of moral obligation. It tells us what is common alike to any and every case of moral duty; it does not tell us what we ought to do. For the right that we will cannot be merely the rightness of willing it. What I ought to do cannot be merely that I ought to do it. Now our beliefs and judgments as to material rightness, i.e. as to what it is right to do in a concrete situation, are notoriously liable to error. So we seek for a criterion by which to test our variable and fallible judgments, a criterion of what is really right. But the search is doomed to failure; we can only test beliefs by beliefs, our former beliefs by our later, the beliefs of one man or society or age by those of others.