Page:Bergson - Matter and Memory (1911).djvu/329

 out space, and, on the other, a mind which can have no point of contact with matter, unless it be, as materialists maintain, the unintelligible epiphenomenon. If we prefer the standpoint of the Kantian realism, we find between the 'thing-in-itself,' that is to say the real, and the 'sensuous manifold' from which we construct our knowledge, no conceivable relation, no common measure. Now, if we get to the bottom of these two extreme forms of realism, we see that they converge towards the same point: both raise homogeneous space as a barrier between the intellect and things. The simpler realism makes of this space a real medium, in which things are in suspension; Kantian realism regards it as an ideal medium, in which the multiplicity of sensations is coördinated; but for both of them this medium is given to begin with, as the necessary condition of what comes to abide in it. And if we try to get to the bottom of this common hypothesis, in its turn, we find that it consists in attributing to homogeneous space a disinterested office: space is supposed either merely to uphold material reality, or to have the function, still purely speculative, of furnishing sensations with means of coördinating themselves. So that the obscurity of realism, like that of idealism, comes from the fact that, in both of them, our conscious perception and the conditions of our conscious perception are assumed to point to pure knowledge, not to action.—But suppose now