projection from the centre to the periphery a certain force is necessary, which must be attributed to a consciousness that is to some extent active. Therefore, he must go further; and, after having made sensations converge towards the cerebral centre, must push them out of the brain, and thereby out of space. So he has to imagine on the one hand sensations that are absolutely unextended, and on the other hand an empty space indifferent to the sensations which are projected into it: henceforth he will exhaust himself in efforts of every kind to make us understand how unextended sensations acquire extensity, and why they choose for their abode this or that point of space rather than any other. But this doctrine is not only incapable of showing us clearly how the unextended takes on extension; it renders affection, extension, and representation equally inexplicable. It must assume affective states as so many absolutes, of which it is impossible to say why they appear in or disappear from consciousness at definite moments. The passage from affection to representation remains wrapt in an equally impenetrable mystery, because, once again, you will never find in internal states, which are supposed to be simple and unextended, any reason why they should prefer this or that particular order in space. And, finally, representation itself must be posited as an absolute: we cannot guess either its origin or its goal.
Everything becomes clearer, on the other hand,
