Whoever will contribute any touch of sharpness will help us to make sure of what's what and who is who. Anyone can contribute such a definition, and, without it, no one knows exactly where he stands. If I offer my own provisional definition of humanism[109] now and here, others may improve it, some adversary may be led to define his own creed more sharply by the contrast, and a certain quickening of the crystallization of general opinion may result. I The essential service of humanism, as I conceive the situation, is to have seen that _though one part of our experience may lean upon another part to make it what it is in any one of several aspects in which it may be considered, experience as a whole is self-containing and leans on nothing_. Since this formula also expresses the main contention of transcendental idealism, it needs abundant explication to make it unambiguous. It seems, at first sight, to confine itself to denying theism and pantheism. But, in fact, it need not deny either; everything would depend on the exegesis; and if the formula ever became canonical, it would certainly develop both right-wing and left-wing interpreters. I myself read humanism theistically and pluralistically. If there be a God, he is no absolute all-experiencer, but simply the experiencer of widest actual conscious span. Read thus, humanism is for me a religion susceptible of reasoned defence, though I am well aware how many minds there are to whom it can appeal religiously only when it has been monistically translated. Ethically the pluralistic form of it takes for me a stronger hold on reality than any other philosophy I know of--it being essentially a _social_ philosophy, a philosophy of '_co_,' in which conjunctions do the work. But my primary reason for advocating it is its matchless intellectual economy. It gets rid, not only of the standing 'problems' that monism engenders ('problem of evil,' 'problem of freedom,' and the like), but of other metaphysical mysteries and paradoxes as well. It gets rid, for example, of the whole agnostic controversy, by refusing to entertain the hypothesis of trans-empirical reality at all. It gets rid of any need for an absolute of the Bradleyan type (avowedly sterile for intellectual purposes) by insisting that the conjunctive relations found within experience are faultlessly real. It gets rid of the need of an absolute of the Roycean type (similarly sterile) by its pragmatic treatment of the problem of knowledge [a treatment of which I have already given a version in two very inadequate articles].[110] As the views of knowledge, reality and truth imputed to humanism have been those so far most fiercely attacked, it is in regard to these ideas that a sharpening of focus seems most urgently required. I proceed therefore to bring the views which _I_ impute to humanism in these respects into focus as briefly as I can. II