Page:Bergson - Matter and Memory (1911).djvu/327

 in short a relation with consciousness—this is what we concede to idealism by the very fact that we term things 'images.' No philosophical doctrine, moreover, provided that it is consistent with itself, can escape from this conclusion. But if we could assemble all the states of consciousness, past, present, and possible, of all conscious beings, we should still only have gathered a very small part of material reality, because images outrun perception on every side. It is just these images that science and metaphysic seek to reconstitute, thus restoring the whole of a chain of which our perception grasps only a few links. But in order thus to discover between perception and reality the relation of the part to the whole, it is necessary to leave to perception its true office, which is to prepare actions. This is what idealism fails to do. Why is it unable, as we said just now, to pass from the order manifested in perception to the order which is successful in science, that is to say, from the contingency with which our sensations appear to follow each other to the determinism which binds together the phenomena of nature? Precisely because it attributes to consciousness, in perception, a speculative rôle, so that it is impossible to see what interest this consciousness has in allowing to escape, between two sensations for instance, the intermediate links through which the second might be deduced from the first. These intermediaries and their strict order thus