I certainly do not think we are justified in exclusively attributing all the various states observed to the influence of the experimenter, or to training.

In what precedes I have discussed the state of the subject during the carrying out of the post-hypnotic suggestion. It will not take long to consider the state between waking and the execution of the suggestion. The subject is then nearly always fully awake, and the state is, in fact, as if he had been wakened without previous post-hypnotic suggestion. However, there are some cases in which the wakening is not complete so long as the effect of the suggestion lasts; this occurs particularly when the suggestion is repugnant to the subject's character. Such subjects look tired and sleepy, and when questioned, often say that they are not quite awake but still half-asleep. I have had cases in which I was obliged to cancel the suggestion before I could completely awaken the subject.

In other cases I have observed a subjective discomfort instead of a feeling of fatigue, until the suggestion was executed. This subjective discomfort is sometimes felt without the suggestion being carried out. One lady to whom I had suggested that she should put a book on the floor woke in great discomfort, but it did not occur to her to put the book on the floor. She recovered herself, however, when, at my request, she had put the book on the floor in the waking state. Another subject complained of twitching in the arm after waking; I had suggested to him to give me his hand when he woke. He did not do it till I asked him again in the waking state; before that he was aware of nothing but the twitching. I have, however, never observed evidences of discomfort when the post-hypnotic suggestion had a therapeutic aim. It is obvious that the discomfort is caused by the conflict waging in the subject's mind, even when at the threshold of consciousness; and, of course, there can be no question of such a struggle when the suggestion has a therapeutic aim.

Kohler declares that between waking and the execution of the post-hypnotic suggestion, the subject is invariably in one or other of two abnormal states - the one an obvious, but exceptional, form of hypnosis, the other a peculiar state of intermittent hypnosis. By which Kohler means that the patient remains hypnotized from the beginning of the experiment to the execution of the suggestion, or else in an exceptional condition in which he can carry on a rational conversation until he suddenly falls into a fresh hypnosis. The abnormal state terminates directly the posthypnotic suggestion is fulfilled. Nevertheless, I consider Kohler's views on this question erroneous - at least they as little agree with the results of my own observations as they do with those of Lcewenfeld and others.

There are, of course, numerous cases of post-hypnotic suggestion in which the suggested act is not performed, many persons being able to offer a more or less successful resistance. It sometimes even happens that a hypnotic rejects a suggestion during hypnosis. Many carry out only the suggestions to which they have assented (Pierre Janet). Scripture reports a case observed at Brown University. The post-hypnotic suggestion was given to a person to pronounce a always like ee - e.g., "feether" instead of "father"; on being awakened she was often asked about her parents, but always avoided using the word "father," speaking instead of "the husband of my mother." Pitres relates an interesting case of a girl who would not allow him to wake her, because he had suggested that on waking she would not be able to speak. She positively declared that she would not wake till he gave up his suggestion: But even when the suggestion is accepted as such, a decided resistance is often expressed during its post-hypnotic execution. This shows itself as often in slow and lingering movements as in a decided refusal to perform the act at all.

The more repugnant the action, the more likely is it to be omitted.

As in all the above cases of post-hypnotic suggestion the command was not remembered, it is particularly interesting to observe how the subjects try to account for their execution of the suggestions. Naturally, we shall only here consider those cases in which the action is not immediately forgotten; in the others, subjects do not try to find reasons for actions which they have forgotten.

Let us take an example. I say to a hypnotized woman, "After you wake you will take a book from the table and put it on the bookshelf." She wakes and does what I told her. When I ask her what she has been doing, she answers that she has moved the book from the table to the shelf. When asked for her reason, she answers, " I do not like to see things so untidy; the shelf is the place for the book, and that is why I put it there."

In this case my command led to a definite action being carried out; but the subject does not remember my suggestion, believing that she has so acted of her own accord, from love of order.

Let us go on with our experiment. I suggest to the re-hypnotized subject to take the book from the shelf and lay it under the table, which she does. I ask her why she did it; she can give no reason. "It came into my head," she answers. I repeat the experiment several times. To a new request for her reason she finally replies, "Something made me feel as if I must put the book there."

In this case the subject, who at first believed she was acting freely, came by degrees to recognize the restraint put upon her; she, perhaps, suspected the suggestion, but was not sure of it.

Another case. I suggest to a hypnotized man to use an insulting expression to me when he wakes. He wakes, and after a pause of a few seconds, during which his face expresses an inward struggle, he calls out, "Donkey!" When asked why he so insults me, he makes many excuses, and explains, " I felt as if I must say 'Donkey!' "

Here we have to do with a paradoxical action; the man knew at once that constraint was being put upon him; the woman who performed the simpler act above only perceived the constraint after several experiments. However, in a great number of cases the result is different.