Page:Bergson - Matter and Memory (1911).djvu/156

 words. And, lastly, acoustic verbal images are not destroyed without a serious lesion of certain determined convolutions of the cortex: so that we are here provided with an undisputed example of localization, in regard to which we can enquire whether the brain is really capable of storing up memories. We have, then, to show in the auditory recognition of words: first, an automatic sensori-motor process; secondly, an active and, so to speak, excentric projection of memory-images.

1. I listen to two people speaking in a language which is unknown to me. Do I therefore hear them talk? The vibrations which reach my ears are the same as those which strike theirs. Yet I perceive only a confused noise, in which all sounds are alike. I distinguish nothing, and could not repeat anything. In this same sonorous mass, however, the two interlocutors distinguish consonants, vowels and syllables which are not at all alike, in short, separate words. Between them and me where is the difference?

The question is, how can the knowledge of a language, which is only memory, modify the material content of a present perception, and cause some listeners actually to hear what others, in the same physical conditions, do not hear. It is alleged, indeed, that the auditory