Classics in the History of Psychology

An internet resource developed by
Christopher D. Green
York University, Toronto, Ontario

ISSN 1492-3173

(Return to index)


By William H.R. Rivers (1920)

Posted April 2000



In the last chapter I have made no reference to an aspect of hypnotism which is of much importance and must now be considered. I have described how it is possible in the hypnotic state to apply the power of suggestion so as to produce acts of various degrees of complexity as well as the cruder paralyses and anæsthesias, but I have not mentioned a state which is so frequently produced by means of hypnotism, or so often follows its application, that it may be regarded as an almost necessary part of its nature. Not only is the suggestion to sleep one of the most frequent measures which are used when it is desired to produce the hypnotic state, but when this state has been produced, it is often difficult to distinguish it from normal sleep, the chief difference being that the time-limit of the sleep may be determined by the suggestion of the hypnotist and has not the free character possessed by the customary process of sleep. We may assume with some confidence that sleep is a state which is allied to hypnotism, and therefore stands in some relation to the process of suggestion. This raises the problem of the nature of sleep, which I must now consider.

Sleep is an aspect of life which has been strangely neglected by the psychologist. Students of psychology are profoundly interested in the dreams which occur during sleep and in their more abnormal varieties, such as nightmares and somnambulism, but they seem to have been content to regard sleep itself as a process which belongs to the realm of physiology. Even those who have come fully to recognise that external stimuli to the organs of sense or internal visceral disturbances are insufficient as explanations of the dream, are yet content to refer sleep to [p. 111] such purely physiological causes as alterations of blood-supply to the brain, accumulation of the toxic products of activity, or other similar material agencies. They fail to recognise that sleep is much more than a mere negation of psychological activity, and that, quite apart from the occurrence of dreams, sleep has characters of a positive kind which must be fitted into any scheme of mental happenings which seeks to be consistent and complete.

I have already mentioned the resemblance of the hypnotic state to sleep. I have now to consider other points in which sleep resembles the state which follows the hypnotic suggestion of another person. Several features of sleep have led students to believe that it is the result of suggestion exerted by the conditions of time and space with which sleep is habitually associated. There is no question that such conditions as fatigue within certain limits, warmth, the toxic influence of the products of respiration, etc., act as agents in the production of sleep, but they cannot explain the occurrence of sleep under any conditions of time and space of which certain persons are capable, or the immediate occurrence of sleep, even in the absence of sleepiness, which in many people immediately follows going to bed. In such cases it is customary to speak of auto-suggestion. This term may seem in some degree appropriate when applied to the sleep of persons who are able to sleep at will. In the case of those who sleep immediately after going to bed, on the other hand, we should have to regard the suggestion as coming from the pillow, the darkness, the silence, or other habitual feature of the surroundings in which a person is accustomed to sleep.

The conditions of awaking from sleep point more definitely to the influence of suggestion. The selective action of certain conditions of awaking cannot be explained on physiological lines, but demands some kind of discriminative and selective activity on the part of the sleeping person. As examples of such selective activity I may cite the doctor who is awakened even by the movements of wires which precede the ringing of his night-bell, while he is undisturbed by the crying of his [p. 112] child to whose slightest sound the mother immediately responds. Awaking is determined, not by physiological demands but by conditions the efficacy of which is determined almost entirely by predispositions of a psychological kind. We may regard the conditions which awake a person as suggestions determined by special systems within the personality of the sleeper. It may be noted that such awaking stimuli seem to be especially effective when they stand in a close relation to affective states of the sleeper. The awakening of the mother at the slightest sound or movement of her child may be regarded as a reaction of the parental instinct. The reaction of the doctor to his night-bell bears a less direct relation to instinctive process, but in spite of its more complex character can usually be traced to affective factors connected with instinctive means.

An even closer relation of sleep to suggestion appears in cases in which a sleeping person responds to the questions or commands of another person, especially one with whom he is habitually in intimate relation. A sleeping person may converse with another and may yet be completely unaware of what he has said and done when he awakes. The absence of memory is altogether comparable with that which accompanies the suppression of the post-hypnotic state when no suggestion has been given that the hypnotic experience shall be remembered.

An interesting point of resemblance between hypnotism and sleep may be noted here. Probably everyone has at some time or another experienced how, when sleepless, the difficulty in sleeping is greatly enhanced if doubt arises in the mind whether sleep will come. Going to sleep is an act which normally takes place unwittingly. If we once begin to think whether we are going to sleep, a state of mind is induced which works strongly against the occurrence of the sleep so greatly desired. Sleep is essentially a process which, like forgetting, takes place unwittingly. It is, therefore, interesting to note that those with much experience in the practice of hypnotism have found that if a person who wishes to be hypnotised has doubts whether his wish will be realised, success is definitely prejudiced.[1] Instead [p. 113] of falling into the receptive and passive state which forms the best background for the efforts of the hypnotiser, the doubter becomes agitated and the hypnotic process fails or is delayed. Forel remarks that the more frequently and the more energetically a person endeavours to become passive, the more certainly will he fail, and he compares this state, not only with sleep, but also with affective states. Thus, we cannot force ourselves to become pleased. Pleasure and other affective states must arrive spontaneously. They come without invitation or conscious direction, and resemble in this respect sleep, hypnotism, and active forgetting.

I have so far considered the relationship of sleep and suggestion. I have now to deal with the relation of sleep to other processes of the unconscious. Sleep affords a striking example of suppression. Not only does the conscious activity of the waking life disappear, but any experience acquired in sleep is forgotten, or tends to be forgotten, when we awake. The sleeping experience which is suppressed may be of the kind I have already mentioned in which a person may be wholly unaware of what he had said or done in sleep. Dreams provide an equally characteristic example of forgetting. One of their most definite features is the ease with which they are forgotten. It often happens that we only become aware that we have dreamed owing to some event of the day which recalls an image or incident of a dream which has occurred during the preceding night, or even on some more remote occasion. This has often raised the question whether sleep is not habitually accompanied by dreams, only a small proportion of which are remembered. Whether this be so or not, we can be confident that dream-experience bulks largely in the content of the unconscious. We all know that dreams differ greatly in the ease and completeness with which they are forgotten. The memory of the special variety, which is known as the nightmare, may be as vivid and persistent as that of the most poignant and striking events of the waking life, and all degrees occur between this realistic persistence and the complete forgetting of a dream which is present only for a few moments after awaking. It is significant [p. 114] that the dream which persists in memory is one which has been accompanied by a definite affect, especially of a painful kind.

The process of disassociation is also definitely present in sleep. This is especially obvious in those dreams which are accompanied by acts ranging in complexity from the elaborate behaviour of the sleepwalker to the apparently disjointed utterances of one who talks in his sleep. The most elaborate of these performances may be regarded as a pattern of dissociation, but the difference between it and the slightest movement or utterance in a dream is one only in degree and not in kind.

Somnambulism is of especial interest as an example of dissociation on account of its very close resemblance to a fugue. One who is walking in his sleep is carrying out a series of activities, often of the most varied and complicated kind, which are wholly independent of the activities of his normal life. In some cases the sleepwalker is aware of these activities in the form of a dream when he awakes, but more often any consciousness which may have accompanied the somnambulistic acts becomes inaccessible as soon as the sleeper awakes. Its recovery in the hypnotic or hypnoidal states, however, shows that we have to do with independent consciousness as well as with independent activity, so that the state answers completely to my definition of dissociation. There is, in fact, no difference between a fugue and a somnambulistic attack except that one occurs in sleep and the other in the waking state.

A point which is of the greatest interest in the light of the present argument is that the somnambulistic state is much more frequent than the fugue. In other words, the state of sleep predisposes to the occurrence of dissociation. Moreover, sleep-walking is especially frequent in childhood. The sleep of childhood is especially prone to be disturbed by activities accompanied by consciousness cut off from the activities and consciousness of the ordinary life.

I have supposed dissociation to be an instinctive process. I regard it as a process necessary for the welfare of some of the ancestors of Man which still comes into action in Man himself [p. 115] under certain special circumstances. The special tendency of somnambulism, as a special kind of dissociation, to occur in childhood accords thoroughly with its instinctive character, for it is one of the main theses of this book that instinctive reactions are especially liable to occur, or are liable to occur in an especially pure form, in childhood. In the same way the special liability of dissociation, in the form of somnambulism, to occur in sleep points to the instinctive nature of sleep already brought into view by its close relation to suppression.

At this stage of the argument I should like to call attention to two distinct varieties of somnambulism. In one, with which we have become very familiar during the war, a sleeper reproduces the activity of some experience, while in other cases the actions of the sleep-walker have no such definite relation to one another, but have rather the apparent inconsequence and incoherence of the dream. If, however, the somnambulistic behaviour is an acted dream, we should expect to find it showing varieties of the same kind as those shown by the dream itself.

If, now, we consider the relation of sleep to the three processes of suppression, dissociation and suggestion, it is clear that the primary position belongs to suppression. Sleep is pre-eminently a process in which certain mental processes are put in abeyance for the purpose of recuperation and restoration of the sleeper to full mental as well as bodily efficiency. The dissociation of sleep is the result of the action of lower activities which are released by the suppression of the higher controlling processes, while the enhancement of suggestibility which accompanies sleep is not a necessary, and may even be only an occasional, feature.

Sleep, therefore, may be regarded as primarily an example of the instinctive process of suppression coming into action for the purpose of affording rest to those parts of the mind and body which, being less organised and less stable than the lower instinctive processes, are more liable to fatigue and more in need of the opportunity of rest. Sleep may be regarded as of essentially the same order as the instinct of immobility, but as having through the wide scope of its beneficent action become [p. 116] an almost universal attribute of animal life. Sleep is, in fact, an instinct, allied to the instinct of immobility, which, instead of coming into action only in the presence of danger, is normally of daily occurrence. When it takes place under the influence of certain external surroundings, it is no more appropriate to regard their action as an example of suggestion than it is appropriate to speak of a danger as "suggesting" the special kind of behaviour which is dictated by the instinct of immobility. I do not propose, therefore, to adopt "auto-suggestion" as a term for the process by which external objects induce or help to induce sleep. Nor is it necessary to use the term for the cases in which a person sleeps at will, or apparently at will. This power depends on the ease with which the person can fall into the passive attitude which forms the best opportunity for sleep or hypnotism, but there is no reason to connect the power itself with the process of suggestion which yet undoubtedly takes a part both in sleep and hypnotism. Because, however, I exclude from the category of suggestion certain features of sleep which have frequently been ascribed to it, we must not blind ourselves to the large part which is taken by suggestion in sleep. The selective nature of awakening especially brings into notice features which are closely related to this process.

It is not difficult to see why there should be this relation. The conditions which induce sleep in the individual will also induce it in the group, but there is no special reason why going to sleep should be influenced by factors promoting the welfare of the group as distinguished from that of the individual. With waking, however, the case is different. Here it is essential to the safety of the individual that he shall respond in sleep, not merely to sounds or movements which threaten danger, but also to the sounds or movements of the other members of the group. Moreover, it is necessary that this response shall be discriminative and selective. If each member of the group awakened in response to any kind of stimulus, it would conflict seriously with the recuperation which is the special function of sleep. It is essential that each species of animal shall react in sleep, as in the waking life, to those stimuli which indicate danger, [p. 117] and shall not react to stimuli of an indifferent kind. The power of discrimination and selection which is shown in the process of awakening in Man may be regarded as the direct descendant of the similar power which is essential to the safety of gregarious animals.

Before I leave the subject of sleep, I must consider whether, and if so to what extent, it is subject to the "all-or-none" principle. A moment's consideration will show that the principle does not apply. Sleep is a definitely, and even finely, graded process. It is often difficult to say whether a person, including oneself, is or is not, or has or has not been, asleep. Moreover, the process of waking, as I have just said, implies the presence of the processes of discrimination and selection. I have regarded sleep as allied to the instinct of immobility, but we can now see the essential difference between the two. If I am right, sleep is an example of the instinctive process of suppression in which this process has become capable to a very high degree of being graded and of reacting to delicately-discriminated and selected stimuli. Since this power of gradation and selection appear especially in relation to awaking, and since the process of awaking stands in a close relation to the welfare of the group, it is no great assumption to suppose that the grading stands in a definite relation to gregarious needs and to the process of suggestion by means of which these needs are satisfied. I suggest, therefore, that sleep is an example of instinctive behaviour in which the process of suppression, originally subject to the "all-or-none" principle, has become capable of gradation in a high degree, not through the action of intelligence, but through the working of the power of suggestion which is itself an instinctive process of a graded and discriminative kind. If we are to classify instinctive processes into protopathic and epicritic varieties, sleep belongs pre-eminently, and in a high degree, to the latter group.

There is some reason to believe that one phenomenon of sleep must be excepted from this generalisation, and must be regarded as an example of the "all-or-none " reaction. I refer to the nightmare. In this form of dream, the dreamer is [p. 118] liable to experience affects of extreme intensity. The fear which forms its usual content is more intense and is accompanied by more pronounced physical manifestations than are ever known in the waking state. We seem to have here a form of reaction in which the suppressed experience which underlies the dream reappears with all the force of which it is capable. While some subjects of psycho-neurosis are liable to intense reactions of this kind, others whose general history is very similar show complete suppression of similar experience which may, however, underlie somnambulistic attacks. The completeness of the suppression in some cases, side by side with the extremity of affective disturbance in others, suggests that in the instinctive state to which persons are reduced in psycho-neurosis, the suppressed experience either manifests itself with all its available force or undergoes complete suppression, thus exhibiting a feature which reminds us of the "all-or-none" reaction.


[1] See A. Forel, Hypnotism, London and New York (1906), p. 67.