If we suggest to a hypnotic subject, first of all, that he is Napoleon, then, shortly afterwards, Frederick the Great, and, finally, restore his own personality, also by suggestion, we find that each of the different chains of memories goes on independently within him; thus Frederick the Great knows nothing about Napoleon, the latter nothing about the reality, and the real person himself is quite unaware of the other two states. In short, we must invariably bear in mind that the theory of the double ego is only a diagram.

As regards double consciousness in relation to the hypnotic state, Max Dessoir thinks that hypnosis represents experimentally that half of our mental life that is usually hidden; the part which is called secondary consciousness, something of which association occasionally enables us to observe in ordinary life, but which in abnormal states appears as a connected whole held together by its own chain of memories. According to Max Dessoir's theory, double consciousness as it appears in hypnosis is no absolutely new phenomenon, but is the experimental representation of a psychic faculty latent in man. Considered within these limits the theory is intelligible, although it does not explain everything. Max Dessoir, whose lieves is capable of inducing the corresponding psychological and physiological effects; if, as I say, we remember these facts, then post-hypnotic suggestion cannot be so enigmatical. And it must appear even less so if we bear in mind that an idea which is accepted in hypnosis has, by means of the association of ideas, a natural tendency to create a state of consciousness and will analogous to that which obtained when the idea was implanted.

I have now considered why post-hypnotic suggestions are carried out without or in spite of the will. I supposed a case in which the subject remembered the order given him in hypnosis after he woke. It is a more enigmatical question why post-hypnotic suggestions are carried out when the subject after waking has no recollection of having received the command.

For explanation let us return to the case of waking life, where X. was to post a letter. Now X. did not keep the request continually in his consciousness, for we certainly saw that he apparently posted the letter unconsciously; yet he would not have performed the action at all if he had not really remembered my request. It is the same in post-hypnotic suggestion. All post-hypnotic suggestions really remain in the memory, and are merely apparently forgotten between waking and fulfilment. Here, again, we must remember that our mental processes are divided into two groups, that of the primary consciousness in which they are subjectively perceived, and that of the secondary consciousness in which they are unperceived. We must further bear in mind that the state of the primary consciousness is not uniform, but, on the contrary, subject to constant changes. In one period we are conscious of ideas which are wanting in others. One period comprises more than another. Now, if we call the sum of mental processes subjectively perceived at a certain time the sphere of primary consciousness, we may suppose a number of such spheres.

But not to complicate the subject too much, we will only suppose two spheres.

We saw, when discussing the memory, that the hypnotic who forgot the events of hypnosis in waking life remembered them in later hypnosis. But he remembered the events of waking life also in hypnosis, though in waking life he was only conscious of the events of that life. We have, then, two different spheres of primary consciousness here; one comprises the events of hypnosis and of waking life, the other only those of waking life. We must further remember that the two consciousnesses are not schematically separated. Impressions made on the secondary consciousness very often rise to the primary, and vice versa. The restoration of memory through the association of ideas, which I spoke of on p. 122, is an example of this, and alone suffices to show that the events of hypnosis are firmly established in the mind, even when there is loss of memory after waking; otherwise it would be impossible for the association of ideas to call up recollection. The events 0/ the hypnosis were unperceived until raised to the primary consciousness by the association of ideas.

But, in addition to this, there are other ways in which it can be proved that the events of hypnosis are really only dormant in the secondary consciousness; and it is automatic writing, of which I have already spoken, and which has been investigated by Gurney, F. Myers, Pierre Janet, Binet, Patrick, Max Dessoir, Flournoy and others, that provides a special proof. For example, I hypnotize X., make him go through all kinds of scenes by suggestion and then wake him. When I ask him what happened during the hypnosis, he declares he does not know. No matter how much I press him to recollect what he experienced during the hypnosis and tell me of it, he is unable to do so. I now order him to write down automatically the events of the hypnosis. X. does so, and writes down everything that was suggested to him during the hypnosis. He could not do this if the events of the hypnosis were banished from his mind. Hence recollection was dormant, as the automatic writing proved.

We have thus seen that it can be proved by automatic writing and other methods that post-hypnotic commands are firmly lodged in the secondary consciousness, and it is now easy for us to show why the post-hypnotic suggestion is carried out in spite of loss of memory after waking. As we have seen, the command lies quiescent in the secondary consciousness, and the loss of memory is so far only apparent. But much goes on in the secondary consciousness often of a very exact kind, and there is no confusion in its thoughts; this explains why the subject carries out a suggested order correctly, even when after waking he has no recollection of having received it.