The Utilitarian writers of last century were of course conscious of this problem, conscious that there was a possible discrepancy between egoistic conduct and altruistic conduct; but they agreed to lay stress upon altruistic results as determining moral quality. Their tendency was to minimise the difference between the egoistic and the altruistic effects of action, and in so far as a difference had to be allowed to emphasise the importance of the claims of the community at large, that is, roughly speaking, to take the altruistic standpoint. Recent and more careful investigators have brought out more exactly the extent and significance of the divergence. In particular this was done with perfect clearness and precision by the late Professor Sidgwick. He showed that the difference--although it might be easily exaggerated--was yet real and important, that the two systems did not mean the same thing, that we could not rely upon altruistic conduct always being for individual benefit, that there was no 'natural identity' between egoism and altruism. He held that morality, to save it from an unsolved dualism, required a principle capable of reconciling the discrepancy between the conduct in accordance with the axiom of Benevolence and the conduct in accordance with the equally rational axiom of Self-love.[1] [Footnote 1: Professor Sidgwick's last words on the question are as follows: "If then the reconciliation of duty and self-interest is to be regarded as a hypothesis logically necessary to avoid a fundamental contradiction in one chief department of our thought, it remains to ask how far this necessity constitutes a sufficient reason for accepting this hypothesis.... Those who hold that the edifice of physical science is really constructed of conclusions logically inferred from self-evident premises, may reasonably demand that any practical judgments claiming philosophic certainty should be based on an equally firm foundation. If, on the other hand, we find that in our supposed knowledge of the world of nature propositions are commonly taken to be universally true, which yet seem to rest on no other grounds than that we have a strong disposition to accept them, and that they are indispensable to the systematic coherence of our beliefs,--it will be more difficult to reject a similarly supported assumption in ethics, without opening the door to universal scepticism" ('Methods of Ethics,' 6th ed., pp. 506, 507).]