180 CREATIVE EVOLUTION [chap. 



If the first is true, consciousness must express exactly, 

 at each instant, the state of the brain; there is strict 

 parallelism (so far as intelligible) between the psychical 

 and the cerebral state. On the second hypothesis, on 

 the contrary, there is indeed solidarity and interdependence 

 between the brain and consciousness, but not parallelism: 

 the more complicated the brain becomes, thus giving the 

 organism greater choice of possible actions, the more 

 does consciousness outrun its physical concomitant. Thus, 

 the recollection of the same spectacle probably modifies 

 in the same way a dog's brain and a man's brain, if the 

 perception has been the same; yet the recollection must 

 be very different in the man's consciousness from what 

 it is in the dog's. In the dog, the recollection remains 

 the captive of perception; it is brought back to conscious- 

 ness only when an analogous perception recalls it by re- 

 producing the same spectacle, and then it is manifested 

 by the recognition, acted rather than thought, of the present 

 perception much more than by an actual reappearance 

 of the recollection itself. Man, on the contrary, is capable 

 of calling up the recollection at will, at any moment, in- 

 dependently of the present perception. He is not limited 

 to playing his past life again; he represents and dreams it. 

 The local modification of the brain to which the recollection 

 is attached being the same in each case, the psychological 

 difference between the two recollections cannot have its 

 ground in a particular difference of detail between the two 

 cerebral mechanisms, but in the difference between the 

 two brains taken each as a whole. The more complex 

 of the two, in putting a greater number of mechanisms in 

 opposition to one another, has enabled consciousness 

 to disengage itself from the restraint of one and all and to 

 reach independence. That things do happen in this way, 

 that the second of the two hypotheses is that which must 



