Page:Bergson - Matter and Memory (1911).djvu/95

 versely, if recollection is regarded as a weakened perception, perception must be regarded as a stronger recollection. We are driven to argue as though it was given to us after the manner of a memory, as an internal state, a mere modification of our personality; and our eyes are closed to the primordial and fundamental act of perception,—the act, constituting pure perception, whereby we place ourselves in the very heart of things. And thus the same error, which manifests itself in psychology by a radical incapacity to explain the mechanism of memory, will in metaphysics profoundly influence the idealistic and realistic conceptions of matter.

For realism, in fact, the invariable order of the phenomena of nature lies in a cause distinct from our perceptions, whether this cause must remain unknowable, or whether we can reach it by an effort (always more or less arbitrary) of metaphysical construction. For the idealist, on the contrary, these perceptions are the whole of reality, and the invariable order of the phenomena of nature is but the symbol whereby we express, alongside of real perceptions, perceptions that are possible. But, for realism as for idealism, perceptions are 'veridical hallucinations,' states of the subject, projected outside himself; and the two doctrines differ merely in this: that in the one these states constitute reality, in the other they are sent forth to unite with it.

But behind this illusion lurks yet another that