To change the outer world appropriately by means of motility requires the accumulation of a large total of experiences in the memory-systems, as well as a manifold consolidation of the relations which are evoked in this memory-material by various directing ideas. We will now proceed further with our assumptions. The activity of the second system, groping in many directions, tentatively sending forth cathexes and retracting them, needs on the one hand full command over all memory- material, but on the other hand it would be a superfluous expenditure of energy were it to send along the individual thought-paths large quantities of cathexis, which would then flow away to no purpose and thus diminish the quantity needed for changing the outer world. Out of a regard for purposiveness, therefore, I postulate that the second system succeeds in maintaining the greater part of the energic cathexes in a state of rest, and in using only a small portion for its operations of displacement. The mechanics of these processes is entirely unknown to me; anyone who seriously wishes to follow up these ideas must address himself to the physical analogies, and find some way of getting a picture of the sequence of motions which ensues on the excitation of the neurones. Here I do no more than hold fast to the idea that the activity of the first Psi-system aims at the free outflow of the quantities of excitation, and that the second system, by means of the cathexes emanating from it, effects an inhibition of this outflow, a transformation into dormant cathexis, probably with a rise of potential. I therefore assume that the course taken by any excitation under the control of the second system is bound to quite different mechanical conditions from those which obtain under the control of the first system. After the second system has completed its work of experimental thought, it removes the inhibition and damming up of the excitations and allows them to flow off into motility.

An interesting train of thought now presents itself if we consider the relations of this inhibition of discharge by the second system to the process of regulation by the pain-principle. Let us now seek out the counterpart of the primary experience of gratification, namely, the objective experience of fear. Let a perception- stimulus act on the primitive apparatus and be the source of a pain-excitation. There will then ensue uncoordinated motor manifestations, which will go on until one of these withdraws the apparatus from perception, and at the same time from the pain. On the reappearance of the percept this manifestation will immediately be repeated (perhaps as a movement of flight), until the percept has again disappeared. But in this case no tendency will remain to recathect the perception of the source of pain by hallucination or otherwise. On the contrary, there will be a tendency in the primary apparatus to turn away again from this painful memory-image immediately if it is in any way awakened, since the overflow of its excitation into perception would, of course, evoke (or more precisely, begin to evoke) pain. This turning away from a recollection, which is merely a repetition of the former flight from perception, is also facilitated by the fact that, unlike the perception, the recollection has not enough quality to arouse consciousness, and thereby to attract fresh cathexis. This effortless and regular turning away of the psychic process from the memory of anything that had once been painful gives us the prototype and the first example of psychic repression. We all know how much of this turning away from the painful, the tactics of the ostrich, may still be shown as present even in the normal psychic life of adults.

In obedience to the pain-principle, therefore, the first Psi- system is quite incapable of introducing anything unpleasant into the thought-nexus. The system cannot do anything but wish. If this were to remain so, the activity of thought of the second system, which needs to have at its disposal all the memories stored up by experience, would be obstructed. But two paths are now open: either the work of the second system frees itself completely from the pain-principle, and continues its course, paying no heed to the pain attached to given memories, or it contrives to cathect the memory of the pain in such a manner as to preclude the liberation of pain. We can reject the first possibility, as the pain-principle also proves to act as a regulator of the cycle of excitation in the second system; we are therefore thrown back upon the second possibility, namely, that this system cathects a memory in such a manner as to inhibit any outflow of excitation from it, and hence, also, the outflow, comparable to a motor-innervation, needed for the development of pain. And thus, setting out from two different starting-points, i.e., from regard for the pain-principle, and from the principle of the least expenditure of innervation, we are led to the hypothesis that cathexis through the second system is at the same time an inhibition of the discharge of excitation. Let us, however, keep a close hold on the fact- for this is the key to the theory of repression- that the second system can only cathect an idea when it is in a position to inhibit any pain emanating from this idea. Anything