all the images on the same plane, to suppose that they no longer vary for him, but for themselves; and to treat them as though they made part of a system in which every change gives the exact measure of its cause. On this condition alone a science of the universe becomes possible; and, since this science exists, since it succeeds in foreseeing the future, its fundamental hypothesis cannot be arbitrary. The first system alone is given to present experience; but we believe in the second, if only because we affirm the continuity of the past, present, and future. Thus in idealism, as in realism, we posit one of the two systems and seek to deduce the other from it.
But in this deduction neither realism nor idealism can succeed, because neither of the two systems of images is implied in the other, and each of them is sufficient to itself. If you posit the system of images which has no centre, and in which each element possesses its absolute dimensions and value, I see no reason why to this system should accrue a second, in which each image has an undetermined value, subject to all the vicissitudes of a central image. You must then, to engender perception, conjure up some deus ex machina, such as the materialistic hypothesis of the epiphenomenal consciousness, whereby you choose, among all the images that vary absolutely and that you posited to begin with, the one which we term our brain,—conferring on the internal states of this image the singular and inexplicable
