The best imitation of creative purpose, therefore, which Naturalism can provide breaks down where it is most required—namely, at the highest levels of value. I have just shown this in connection with our powers of thought, and the beliefs to which they lead. But the failure is not confined to them. It is as wide as Humanism itself. Wherever we find great intrinsic worth, there we are in a region where the direct effect of selection is negligible. The noblest things in speculation, in art, in morals, possess small survival value; and, though the geniuses to whom we owe them have added greatly to the glory of their race, they have added but little to its animal successes. In the language of these lectures, they are “accidental”—due neither to purpose nor to any arrangement of causes by which purpose is successfully copied. III You are now in a position to judge how far the hopes held out to you at the beginning of this course have been fulfilled, and to measure the merits and the demerits, the claims and the limitations, of the scheme I have endeavoured to expound. I disowned, as you remember, any intention of providing you with a philosophical system—not because I despise philosophical systems or those who labour to construct them, but in part because I have none to recommend, and in part because it seems to me doubtful whether at our present stage of development a satisfactory system is possible. But how (you may ask) does my point of view differ from a philosophical system? It may be a bad system, as it certainly is a most imperfect one. Yet, seeing that it touches on everything in heaven and earth, seeing that its very title embraces God and man, why should it repudiate a description which seemingly is not a whit in excess of its pretensions? The question thus raised is more than a merely verbal one, and a few observations upon it may fittingly conclude the course. Note, then, in the first place, that my scheme of beliefs does not show itself unworthy to be considered systematic merely because it is incomplete. All systems are incomplete. All systems, however ambitious, admit their inability to exhaust reality. Nor is its unworthiness due to any mere accident of execution, such as inferior workmanship or defective learning. Its failures are essential and irremediable. They are inseparable from “the point of view.” Let me explain. Every system that deserves to be described as a constructive philosophy—be it dogmatic, critical, empirical, idealist, what you will—conceives itself not merely to be rooted in reason, but to be rationalised throughout. The conceptions with which it works should be sifted, clarified, defined. It should assume nothing which requires proof. It should rest nothing (in the last resort) on faith or probability. It should admit no inexplicable residues.