to the truth than could otherwise be reached; so a system of ideal ethical truths, expressing the absolutely right, will be applicable to the questions of our transitional state in such ways that, allowing for the friction of an incomplete life and the imperfection of existing natures, we may ascertain with approximate correctness what is the relatively right. § 105. In a chapter entitled "Definition of Morality" in _Social Statics_, I contended that the moral law, properly so-called, is the law of the perfect man--is the formula of ideal conduct--is the statement in all cases of that which should be, and cannot recognize in its propositions any elements implying existence of that which should not be. Instancing questions concerning the right course to be taken in cases where wrong has already been done, I alleged that the answers to such questions cannot be given "on purely ethical principles." I argued that: "No conclusions can lay claim to absolute truth, but such as depend upon truths that are themselves absolute. Before there can be exactness in an inference, there must be exactness in the antecedent propositions. A geometrician requires that the straight lines with which he deals shall be veritably straight; and that his circles, and ellipses, and parabolas shall agree with precise definitions--shall perfectly and invariably answer to specified equations. If you put to him a question in which these conditions are not complied with, he tells you that it cannot be answered. So likewise is it with the philosophical moralist. He treats solely of the _straight_ man. He determines the properties of the straight man; describes how the straight man comports himself; shows in what relationship he stands to other straight men; shows how a community of straight men is constituted. Any deviation from strict rectitude he is obliged wholly to ignore. It cannot be admitted into his premises without vitiating all his conclusions. A problem in which a _crooked_ man forms one of the elements is insoluble to him." Referring to this view, specifically in the first edition of the _Methods of Ethics_, but more generally in the second edition, Mr. Sidgwick says: