Remember, we have already found ourselves in a similar position. It was when we were enquiring into certain errors, and the instance we took was a slip of the tongue. Someone had said: “Then certain things were _refilled_,” and thereupon we asked—no, fortunately it was not _we_ who asked, but other people who had nothing to do with psycho-analysis—_they_ asked what he meant by this enigmatic expression. He answered at once that what he had intended to say was: “That was a filthy business,” but had checked himself and substituted the milder words: “Things were revealed there.” I explained to you then that this enquiry was the model for every psycho-analytic investigation, and you understand now that psycho-analytic technique endeavours as far as possible to let the persons being analysed give the answer to their own problems. The dreamer himself then should interpret his dream for us. That is not so simple with dreams, however, as we all know. Where errors were concerned, this method proved possible in many cases; there were others where the person questioned refused to say anything and even indignantly repudiated the answer suggested to him. With dreams, instances of the first type are entirely lacking; the dreamer always says he knows nothing about it. He cannot very well repudiate our interpretation, since we have none to offer him. Shall we have to give up our attempt then? Since _he_ knows nothing, and _we_ know nothing, and a third person can surely know nothing either, there cannot be any prospect of finding the answer. Well, if you like, give up the attempt. But if you are not so minded, you can accompany me. For I assure you that it is not only quite possible, but highly probable, that the dreamer really does know the meaning of his dream; _only he does not know that he knows, and therefore thinks that he does not_. At this point you will probably call my attention to the fact that I am again introducing an assumption, the second in quite a short context, and that by so doing I greatly detract from the force of my claim to a trustworthy method of procedure. Given the hypothesis that dreams are a mental phenomenon, and given further the hypothesis that there are in the minds of men certain things which they know without knowing that they know them—and so forth! You have only to keep in view the intrinsic improbability of both these hypotheses, and you may with an easy mind abandon all interest in the conclusions to be drawn from them.