The expressions of a repetition-compulsion which we have described, both in the early activities of infantile psychic life and in the experiences of psycho-analytic treatment, show in a high degree an instinctive character, and, where they come into contrast with the pleasure-principle, a daemonic character. In the play of children we seem to arrive at the conclusion that the child repeats even the unpleasant experiences because through his own activity he gains a far more thorough mastery of the strong impression than was possible by mere passive experience. Every fresh repetition seems to strengthen this mastery for which the child strives; even with pleasurable experiences the child cannot do enough in the way of repetition and will inexorably insist on the identity of the impression. This characteristic is destined later to disappear. A witticism heard for the second time will almost fail of effect; a theatrical performance will never make the same impression the second time that it did on the first occasion; indeed it is hard to persuade the adult to read again at all soon a book he has enjoyed. Novelty is always the necessary condition of enjoyment. The child, however, never gets tired of demanding from a grown-up the repetition of a game he has played with him before or has shown him, till at last the grown-up refuses, utterly worn out; similarly if he has been told a pretty story, he wants always to hear the same story instead of a new one, insists inexorably on exact repetition and corrects each deviation which the narrator lets slip by mistake, which perhaps he even thought to gain new merit by inserting. Here there is no contradiction of the pleasure-principle: it is evident that the repetition, the rediscovery of the identity, is itself a source of pleasure. In the case of a patient in analysis, on the other hand, it is plain that the compulsion to repeat in the transference the occurrences of his infantile life disregards _in every way_ the pleasure-principle. The patient behaves in this respect completely like a child, and thus makes it clear to us that the repressed memory-traces of his primitive experience are not present in a ‘bound’ form, are indeed, in a sense, not capable of the secondary process. To this fact of their not being bound they owe their power to weave a wish-phantasy that will be represented in a dream, by adhering to the residues from waking experiences. We frequently encounter the same repetition-compulsion as a therapeutic obstacle, when at the end of the treatment we wish to bring about complete detachment from the physician; and it may be supposed that the vague dread with which those who are unfamiliar with it view analysis, as though they feared to wake what they think is better left to sleep, is at root a fear of the appearance of this daemonic compulsion.