of habit, we prefer to set this kind of memory in the foreground, to erect it into the model memory, and to see in spontaneous recollection only the same phenomenon in a nascent state, the beginning of a lesson learnt by heart. But how can we overlook the radical difference between that which must be built up by repetition and that which is essentially incapable of being repeated? Spontaneous recollection is perfect from the outset; time can add nothing to its image without disfiguring it; it retains in memory its place and date. On the contrary, a learnt recollection passes out of time in the measure that the lesson is better known; it becomes more and more impersonal, more and more foreign to our past life. Repetition, therefore, in no sense effects the conversion of the first into the last; its office is merely to utilize more and more the movements by which the first was continued, in order to organize them together and, by setting up a mechanism, to create a bodily habit. Indeed, this habit could not be called a remembrance, were it not that I remember that I have acquired it; and I remember its acquisition only because I appeal to that memory which is spontaneous, which dates events and records them but once. Of the two memories, then, which we have just distinguished, the first appears to be memory par excellence. The second, that generally studied by psychologists, is habit interpreted by memory rather than memory itself.
It is true that the example of a lesson learnt
