Page:Bergson - Matter and Memory (1911).djvu/14

 prets connexion, which is a fact, as signifying parallelism, which is an hypothesis (and an hypothesis to which it is difficult to attach an intelligible meaning ), it does so, consciously or unconsciously, for reasons of a philosophic order: it is because science has been accustomed by a certain type of philosophy to believe that there is no hypothesis more probable, more in accordance with the interests of scientific inquiry.

Now, as soon as we do, indeed, apply to positive facts for such information as may help us to solve the problem, we find it is with memory that we have to deal. This was to be expected, because memory—we shall try to prove it in the course of this work—is just the intersection of mind and matter. But we may leave out the reason here: no one, at any rate, will deny that, among all the facts capable of throwing light on the psycho-physiological relation, those which concern memory, whether in the normal or in the pathological state, hold a privileged position. Not only is the evidence here extremely abundant (consider the enormous mass of observations collected in regard to the various kinds of aphasia), but nowhere else have anatomy, physiology and psychology been able to lend each other such valuable aid. Any one who approaches, without preconceived idea and on the firm ground of facts, the classical problem of the relations of