Those of us who are what Mr. Russell would call ethical sceptics will be delighted at this way of clearing the ground; it opens before us the prospect of a moral philosophy that should estimate the various values of things known and of things imaginable, showing what combinations of goods are possible in any one rational system, and (if fancy could stretch so far) what different rational systems would be possible in places and times remote enough from one another not to come into physical conflict. Such ethics, since it would express in reflection the dumb but actual interests of men, might have both influence and authority over them; two things which an alien and dogmatic ethics necessarily lacks. The joy of the ethical sceptic in Mr. Russell is destined, however, to be short-lived. Before proceeding to the expression of concrete ideals, he thinks it necessary to ask a preliminary and quite abstract question, to which his essay is chiefly devoted; namely, what is the right definition of the predicate "good," which we hope to apply in the sequel to such a variety of things? And he answers at once: The predicate "good" is indefinable. This answer he shows to be unavoidable, and so evidently unavoidable that we might perhaps have been absolved from asking the question; for, as he says, the so-called definitions of "good"--that it is pleasure, the desired, and so forth--are not definitions of the predicate "good," but designations of the things to which this predicate is applied by different persons. Pleasure, and its rivals, are not synonyms for the abstract quality "good," but names for classes of concrete facts that are supposed to possess that quality. From this correct, if somewhat trifling, observation, however, Mr. Russell, like Mr. Moore before him, evokes a portentous dogma. Not being able to define good, he hypostasises it. "Good and bad," he says, "are qualities which belong to objects independently of our opinions, just as much as round and square do; and when two people differ as to whether a thing is good, only one of them can be right, though it may be very hard to know which is right." "We cannot maintain that for me a thing ought to exist on its own account, while for you it ought not; that would merely mean that one of us is mistaken, since in fact everything either ought to exist, or ought not." Thus we are asked to believe that good attaches to things for no reason or cause, and according to no principles of distribution; that it must be found there by a sort of receptive exploration in each separate case; in other words, that it is an absolute, not a relative thing, a primary and not a secondary quality.