Page:Bergson - Matter and Memory (1911).djvu/259

 ation between the unextended and the extended, between quality and quantity.

To take pure perception first. When we make the cerebral state the beginning of an action, and in no sense the condition of a perception we place the perceived images of things outside the image of our body, and thus replace perception within the things themselves. But then, our perception being a part of things, things participate in the nature of our perception. Material extensity is not, cannot any longer be, that composite extensity which is considered in geometry; it indeed resembles rather the undivided extension of our own representation. That is to say that the analysis of pure perception allows us to foreshadow in the idea of extension the possible approach to each other of the extended and the unextended.

But our conception of pure memory should lead us, by a parallel road, to attenuate the second opposition, that of quality and quantity. For we have radically separated pure recollection from the cerebral state which continues it and renders it efficacious. Memory is, then, in no degree an emanation of matter; on the contrary, matter, as grasped in concrete perception which always occupies a certain duration, is in great part the work of memory. Now where is, precisely, the difference between the heterogeneous