First, it may happen that the dream-work is not wholly successful in creating a wish-fulfilment, so that part of the painful feeling in the latent thoughts is carried over into the manifest dream. Analysis would then have to show that these thoughts were a great deal more painful than the dream which is formed from them; this much can be proved in every instance. We admit then that the dream-work has failed in its purpose, just as a dream of drinking excited by the stimulus of thirst fails to quench that thirst. One is still thirsty after it and has to wake up and drink. Nevertheless, it is a proper dream: it has renounced nothing of its essential nature. We must say: “Ut desint vires, tamen est laudanda voluntas.” The clearly recognizable intention remains a praiseworthy one, at any rate. Such instances of failure in the work are by no means rare, and one reason is that it is so much more difficult for the dream-work to produce the required change in the nature of the affect than to modify the content; affects are often very intractable. So it happens that in the process of the dream-work the painful _content_ of the dream-thoughts is transformed into a wish-fulfilment while the painful _affect_ persists unchanged. When this occurs the affect is quite out of harmony with the content, which gives our critics the opportunity of remarking that the dream is so far from being a wish-fulfilment that even a harmless content may be accompanied in it by painful feelings. Our answer to this rather unintelligent comment will be that it is just in dreams of this sort that the wish-fulfilling tendency of the dream-work is most apparent, because it is there seen in isolation. The mistake in this criticism arises because people who are not familiar with the neuroses imagine a more intimate connection between content and affect than actually exists, and so cannot understand that there may be an alteration in the content while the accompanying affect remains unchanged.