This, then, is how the cycle of research closes, by returning to the initial problem, the problem of perception. In the two opposing systems by which attempts have been made to solve it, Mr Bergson discovers a common postulate, resulting in a common impotence. From the idealistic point of view we do not succeed in explaining how a world is expressed externally, nor from the realistic point of view how an ego is expressed internally. And this double failure comes again from the underlying hypothesis, according to which the duality of the subject and object is conceived as primitive, radical, and static. Our duty is diametrically opposed. We have to consider this duality as gradually elaborated, and the problem concerning it must be first stated, and then solved as a function of time rather than of space. Our representation begins by being impersonal, and it is only later that it adopts our body as centre. We emerge gradually from universal reality, and our realising roots are always sunk in it. But this reality in itself is already consciousness, and the first moment of perception always puts us back into the initial state previous to the separation of the subject and object. It is by the work of life, and by action, that this separation is effected, created, accentuated, and fixed. And the common mistake of realism and idealism is to believe it effected in advance, whereas it is relatively second to perception. Hence comes the absolute value of immediate intuition. For from what source could an irreducible relativity be produced in it? It would be absurd to make it depend on the constitution of our brain, since our brain itself, so far as it is a group of images, is only a part of the universe, presenting the same characteristics as the whole; and in so far as it is a group of mechanisms become habits, is only a result of the initial action of life, of original perceptive discernment. And, on the other hand, no less absurd would be the fear that the subject can ever be excluded or eliminated from its own knowledge, since, in reality, the subject, like the object, is in perception, not perception in the subject--at least not primitively. So that it is by a trick of speech that the theses of fundamental relativity take root: they vanish when we return to immediacy; that is to say, when we present problems as they ought to be presented, in terms which do not suppose any conceptual analysis yet accomplished. VI. The Problem of Evolution: Life and Matter. After the problem of consciousness Mr Bergson was bound to approach that of evolution, for psychological liberty is only truly conceivable if it begins in some measure with the first pulsation of corporal life. "Either sensation has no raison d'etre or it is a beginning of liberty"; that is what the "Essay on the Immediate Data" (Page 25.) already told us.