Now by a simple substitution of terms, such language as this will reveal at once one important fact to us. According to the avowed principles of positive morality, morality has no other test but happiness. Immorality, therefore, can have no conceivable meaning but unhappiness, or at least the means to it, which in this case are hardly distinguishable from the end; and thus, according to the above rigid reasoners, the human race will not have reached the lowest depths of misery so long as it rejects the one thing which _ex hypothesi_ might render it less miserable. Either then all this talk about truth must really be so much irrelevant nonsense, or else, if it be not nonsense, the test of conduct is something distinct from happiness. The question before us is a plain one, which may be answered in one of two ways, but which positivism cannot possibly answer in both. Is truth to be sought only because it conduces to happiness, or is happiness only to be sought for when it is based on truth? In the latter case truth, not happiness, is the test of conduct. Are our positive moralists prepared to admit this? If so, let them explicitly and consistently say so. Let them keep this test and reject the other, for the two cannot be fused together. [Greek: oxos t' aleipha t' egcheas tauto kutei dichostatount an ou philoin prosennepois.] This inconsistency is here, however, only a side point--a passing illustration of the slovenliness of the positivist logic. As far as my present argument goes, we may let this pass altogether, and allow the joint existence of these mutually exclusive ends. What I am about to do is to show that on positive grounds the last of these is more hopelessly inadequate than the first--that truth as a moral end has even more of religion in its composition than happiness, and that when this religion goes, its value will even more hopelessly evaporate.