the house of dreams—fallacies in the study of dreams—is it possible to study dreams?—how fallacies may be avoided—do we always dream during sleep?—the two main sources of dreams with their sub-divisions.

when we fall asleep we enter a dim and ancient house of shadow, unillumined by any direct ray from the outer world of waking life. we are borne about through its chambers, without conscious volition of our own; we fall down its mouldy and rotten staircases, we are haunted by strange sounds and odours from its mysterious recesses; we move among phantoms we cannot consciously control. as we emerge into the world of daily life again, for an instant the sunlight seems to flash into the obscure house before the door closes behind us; we catch one vivid glimpse of the chambers we have been wandering in, and a few more or less fragmentary memories come back to us of the life we have led there. but they soon fade away in the light of common day, and if a few hours later we seek to recall the strange experiences we have passed through, it usually happens that the visions of the night have already dissolved in memory into a few shreds of mist we can no longer reconstruct.

[2]

for most of us our whole knowledge ends here. our dreams are real enough while they last, but the interests of waking life absorb us so entirely that we rarely have leisure, and still less inclination, to subject our sleeping adventures, trivial and absurd as they must usually seem, to the careful tests which waking intelligence is accustomed to subject more obviously important matters to. the world of dreams and the mysterious light which prevails there[1] are abandoned entirely to our sleeping activities.

this leading characteristic of dream life—the fact that it takes place in another and more shadowy world and in a different kind of consciousness[2]—has led [3] to the criticism of the study of dreams from the scientific side. we cannot really study our dreams, these objectors say, because we—that is to say, our waking consciousness—cannot come sufficiently closely in contact with them. dreams, it is argued, are inevitably transformed in our hands; what we are studying is not our dreams, but only our waking, and probably altogether false, impressions of our dreams. there is a certain element of truth in this objection. it is very difficult, indeed impossible, to recall exactly, and in their proper order, even the details of a real adventure which has only just happened to us. it is, obviously, incomparably more difficult to recall an experience which took place, under such shadowy conditions, in a world so remote from the world of waking life. there is, further, the very definite difficulty that we only catch our dreams for a moment by the light, as it were, of the open door as we are emerging from sleep. in other words, our waking consciousness is for a moment observing and interpreting a process in another kind of consciousness, or even if we assert that it is the same consciousness it is still a consciousness that has been working under quite different conditions from waking consciousness, and accepting data which in the waking state it would not accept. for the student of dreams it must ever be a serious question how far the facts become inevitably distorted in this process. sleeping or waking, it is probable, our consciousness never embraces the whole of the possible psychic field within us. there are, when we are dreaming[4] as well as when we are awake—as will become clearer in the sequel—subconscious, or imperfectly conscious, states just below our consciousness, and exerting an influence upon it.[3] our latent psychic possessions, among which dreams move, would seem to be by no means always at the same depth; the specific gravity of consciousness, as it were, varies, and these latent elements rise or fall, becoming nearer to the conscious surface or falling further away from it. but the greatest change must take place when the waking surface is reached and the outer world breaks on sleeping consciousness. in that change there is doubtless a process of necessary and automatic transformation and interpretation. we may picture it, perhaps, as somewhat the same process as when a person skilled in both languages takes up a foreign book and reads it out in his own tongue. with practice the reader may become unconscious that he is transforming everything, that the words he utters are different from the words he sees, and that he even transposes their order, sometimes[5] putting in the middle of a sentence the verb he sees at the end.

yet even if we admit that the passage from sleeping to waking consciousness involves a change as complete as this—and it is probable, as we shall see, that some such change sometimes takes place—for a faithful interpreter the sense still remains the same. it is impossible to believe that the witness of waking consciousness to the nature of the visions it has caught at the threshold between sleeping and waking life is false, and the most convincing evidence of this is the utter unlikeness of these visions to the data of ordinary waking life.

but even this conclusion has been subjected to severe criticism which we have to face before we proceed further. foucault, an acute investigator of dream psychology—carrying to its extreme point a position more partially and tentatively stated by delb?uf and tannery—has denied that our dreams, as they finally present themselves to waking consciousness, at all correspond to the psychic process in sleep upon which they are founded, and he especially insists that the logical connections are superadded.[4] he considers that dreaming is an 'observation of memory' made under such conditions that 'it would be very imprudent to regard the remembrance of the dream as reproducing faithfully the mental state of sleep.' during sleep, he believes, our dream ideas proceed, concurrently, it may be, but separately and independently; at the moment when awakening begins, the mind, as an act[6] of immediate memory, grasps the plurality of separate pictures and applies itself spontaneously to the task of organising them according to the rules of logic and the laws of the real world, making a drama of them as like as possible to the dramas of waking life.[5] he agrees with goblot that 'the dream we remember is a waking thought,' and with tannery that 'we do not remember our dreams, but only the reconstructions of them we effected at the moment of waking.' it is after awakening, foucault concludes, that the dream develops, and its final shape depends on the period at which it is noted down; 'the evolution of the dream after awakening is a logical evolution, dominated and directed by the instinctive need to give a reasonable appearance to the ensemble of images and sensations present to the mind, and to assimilate the representation of the dream to the system of representations which constitutes our knowledge of the real world.'[6]

in arguing his thesis, foucault makes much of the modifications which can be proved to take place if any one is asked to repeat a dream at intervals of months. under the influence of time and repetition a dream becomes more coherent and more conformed to reality. in illustration foucault presents two versions of an insignificant dream in which a lady imagines that she is out with her husband for a drive, and in the course of it experiences a natural need which she seeks an opportunity to satisfy; the details of the first version were highly improbable; some months later they had[7] become much more like what might have occurred in real life. such a process, foucault thinks, is taking place from the first in the making of dreams as we know them awake.

there are, undoubtedly, facts which may seem to support foucault's argument that the logic of the dream, as we know it, is not in the original dream, but is introduced afterwards. thus i once dreamed in the morning that i asked my wife if she had been into a certain room, and that she replied, 'can't get in.' i immediately awoke and realised that my wife had actually spoken these words, not to me, but to an approaching servant, in anticipation of a message about entering a neighbouring room of which the door was locked. it is thus evident that although it seemed to me in my dream that the question came first and the answer followed in the ordinary course, in reality the answer came first. the question was a theory, supplied automatically by sleeping intelligence and prefixed to the answer, in which order they both appeared to sleeping consciousness, that is to say, in the only way in which sleeping consciousness can ever be known, as translated into waking consciousness.[7]

it must be borne in mind that such a dream as i have recorded—in which an actual sensory experience is introduced, untransformed, as a foreign body into sleeping consciousness—is not a typical dream. dreams are, however, without doubt of various kinds, and we[8] may well admit that there is a class of dreams formed in this way. that supposition will, indeed, be helpful in explaining several dreams i shall have to record. the process is much the same as when a nervous person receives a telegram, and at once assumes that some dreaded accident has occurred, and that the telegram is the announcement of it. the craving for reasons is instinctive, and the dreamer's sense of logic even dominates his sense of time.

but foucault's argument is that waking consciousness effects this logical construction of the dream. here his position is weak and incapable of proof. it is, indeed, contrary to all the tests we are able to apply to it. if it is the object of the logic of our dreams to make them conformable to our waking experience, that end, we must admit, is in most cases very far from being attained. in their original form, as foucault views the matter, our dreams are simple dissociated images. in that shape they would present nothing whatever to shock the consciousness of waking life. the logic, hypothetically introduced solely to make them conformable to real life, is frequently a preposterous logic such as the consciousness of waking life could not accept or even conceive. this fact alone serves to throw serious doubt upon the theory that it is waking consciousness which impresses its logic upon our dreams.

nor, again, is there any analogy, and still less identity, between the process whereby we grasp a dream when we awake, and the process whereby the memory of a dream is transformed during months of waking life.[9] the latter is part of a general process affecting all our memories in greater or less degree. i visit, for instance, a foreign cathedral, and take careful note of the character and arrangement of buttresses and piers; a few months later, if i have failed to set the facts down, my memory of them will become uncertain, confused, and incorrect. but i need not, therefore, lose faith in the tolerable exactitude of my original impressions. in the same way, we cannot argue that the shifting memory of a dream during a long period of time throws the slightest doubt on the accuracy of our original impression of it. we never catch a dream in course of formation. as it presents itself to consciousness on awakening there may be doubtful points and there may be missing links, but the dream is, once for all, completed, and if there are doubtful points or missing links we recognise them as such. we make no attempt to supply a logic that is not there, and we never see any such process going on involuntarily. i should, indeed, myself be inclined to say that there is always a kind of gap between sleeping consciousness and waking consciousness; the change from the one to the other kind of consciousness seems to be effected by a slight shock, and the perception of the already completed dream is the first effort of waking consciousness. the existence of such a shock is indicated by the fact that, even at the first moment of waking consciousness, we never realise that a moment ago we were asleep. as soon as we realise that we are awake it seems to us that we have already been awake for an uncertain but[10] distinct period of time; some people, indeed, especially old people, on awaking, feel this so strongly that they deny they have been asleep. it once happened to me to be in the neighbourhood of a dynamite factory at the moment when a very disastrous explosion occurred; at the time my back was to the factory, and i am quite unable to say how long an interval occurred between the shock of the explosion and my own action in turning round to observe the straight shaft of smoke and solid material high in the air; there was a gap in consciousness, an interval of unknown and seemingly considerable length, caused by the deafening shock of the explosion, although it is probable that my action in turning round was almost or quite instantaneous. it seems to me that the transition from sleeping consciousness to waking consciousness occurs in a similar manner on a smaller scale.

although the view of foucault that the dream is logically organised after sleep has ended seems, when we examine the evidence in its favour, to be unacceptable, we may still admit that, in some cases at all events, the dream only assumes final shape at the moment when sleeping consciousness is breaking up, that the dream, as we know it, is a final synthetic attempt of sleeping consciousness as it dissolves on the approach of waking consciousness. sleeping consciousness, we may even imagine as saying to itself in effect: 'here comes our master, waking consciousness, who attaches such mighty importance to reason and logic and so forth. quick! gather things up, put them in[11] order—any order will do—before he enters to take possession.' that is to say, in other words, that as sleeping consciousness comes nearer to the threshold of waking consciousness it is possible that the need for the same kind of causation or sequence which is manifested in waking consciousness may begin to make itself felt even to sleeping consciousness. even this assumption seems, however, as regards most dreams, to be extravagant. in any case, and at whatever stage the dream is finally constituted, we are not entitled, it seems to me, to believe that any stage of its constitution falls outside the frontiers of sleep. it is satisfactory to be able to feel justified in reaching this conclusion. for if dreams were chiefly or mainly the product of waking consciousness they would certainly lose a considerable part of their significance and interest.

even, however, when we have reached this conclusion the path of the student is still far from easy. the undoubted fact that in any case the difficulties of observing and recording dreams are very great cannot fail to make us extremely careful. although the dreams of some persons, who may be regarded as themselves of vivid and dramatic temperament, seem to be habitually vivid and dramatic to an extent which, in my own case, is extremely rare, one is usually justified in feeling a certain amount of suspicion in regard to dream-narratives which are at every point clear, coherent, connected, and intelligible. dreams, as i know them on awaking from sleep, occasionally present episodes[12] to which these epithets may be applied, but on the whole they are full of obscurities, of uncertainties, of inexplicable lacunae. the memory of dream events is lost so rapidly that one is constantly obliged to leave the exact nature of a detail in doubt. one seems to be recalling a landscape seen by a lightning flash. it is for this reason that i have made it a rule only to admit dreams which are noted very shortly, and if possible immediately, after the moment of awakening. it is further of importance in recording one's dreams, to note the emotional attitude experienced during the dream as well as any physical sensations felt on awakening. the attitude of dream consciousness towards dream visions usually varies from that of waking consciousness, although the normal extent of the difference is a disputable point. when i read dream narratives of landscapes which, as described, appear at every point as beautiful and impressive to waking consciousness as they appeared to dreaming consciousness, i usually suspect that, granting the good faith and accuracy of the narrator, we are really concerned, not with dreams in the proper sense, but with visions experienced under more abnormal conditions, and especially with drug visions. in the present inquiry i am only concerned to ascertain the most elementary and fundamental laws of the dream world, as they occur in fairly ordinary and normal persons, and therefore it becomes necessary to be very strict as to the conditions under which they were recorded. it is the most ordinary dreams that are most likely to reveal the ordinary laws of[13] dream life, but for this end it is necessary that they should be recorded with the greatest accuracy attainable.

i am myself neither a constant nor, usually, a very vivid dreamer, and in these respects i am probably a fairly ordinary and normal person; the personal material which i have accumulated, though it spreads over twenty years, is not notably copious. nor have i ever directed my attention in any systematic and concentrated manner to my dream life. to do so would be, i believe, to distort the phenomena. i have merely recorded any significant phenomena as they occurred.

to remark that one is not a constant dreamer is not to assert that dreaming is rare, but merely that one's recollection of it is rare. though we may only catch a glimpse of our latest vision of the night as we leave the house of sleep, it may well be that there were many earlier adventures of the night which are beyond the reach of waking consciousness. sometimes, it is curious to note, we become vaguely conscious, during the day, for the first time, of a dream we have had during the night. many psychologists, as well as metaphysicians—fearful to admit that the activity of the soul could ever cease—believe that we dream during the whole period of sleep; this has of recent years been the opinion of vaschide, foucault, n?cke, and sir arthur mitchell, as it formerly was of sir benjamin brodie, sir henry holland, and schaaffhausen. in earlier days hippocrates, descartes, leibnitz, and cabanis seem to have been of the same opinion. on[14] the other hand, locke, macnish, and carpenter held that deep sleep is dreamless; this is also the opinion of wundt, beaunis, strümpell, weygandt, hammond, and jastrow. moreover, there are some people, like lessing, who, so far as they know, never dream at all. my own personal experience scarcely inclines me to accept without qualification the belief that we are always dreaming during sleep. i find that my remembered dreams tend to be correlated with some slight mental or physical disturbance, and therefore it seems to me probable that, if dreams are continuous during sleep, they must, during completely undisturbed sleep, be of an extremely faint and shadowy character. to return to a metaphor i have before used, we may say that sleeping consciousness in its descent from the surface of the waking life may fall to a point at which its specific gravity being practically the same as that of its environment, a state approaching complete repose is attained.[8] it cannot of course be said that the failure to remember dreams is any argument against their occurrence. it is well known that when the psychic activity of sleep assumes a definitely motor shape, as in talking in sleep and in somnambulism, it[15] is very rare for any recollection to remain on awakening, though we cannot doubt that psychic activity has been present. in the same way the dream that we remember when awakened from sound sleep by another person is by no means always due to that awakening. this is shown by the fact that if we were turning round or making other movements just before being thus awakened, the dream we remember—in one such case a dream of making one's way with difficulty between a sofa and a chair—may have no relation to the circumstance of the awakening, but clearly be suggested by the movements made during sleep, though these movements themselves remain unknown to waking consciousness. the movements of dogs during sound sleep—the rhythmical lifting of the paws, the wagging of the tail—point in the same direction.[9]

the fact that failure of memory by no means proves the absence of dreaming may be illustrated, not only by the forgetfulness of what takes place during hypnotic sleep, but by what we sometimes witness during partial anaesthesia maintained by drugs. this was well shown in a case i was once concerned with, where it was necessary to administer chloroform (preceded by the alcohol-chloroform-ether mixture) for a prolonged period during a difficult first confinement. the drug[16] was not given to the point of causing complete abolition of mental activity, and the patient talked, and occasionally sang, throughout, referring to various events in her life, from childhood onwards. the sensation and the expression of pain were not altogether abolished, for slight cries and remarks about the discomfort and constraint imposed upon her were sometimes mingled in the same sentence with quite irrelevant remarks concerning, for instance, trivial details of housekeeping. confusions of incompatible ideas also took place, as during ordinary dreaming. 'where is the three-cornered nurse,' she thus asked, 'who does not mind what she does?' there was also the abnormal suggestibility of dream consciousness. the questions of bystanders were answered but always with a tendency to agree with everything that was said, this tendency even displaying itself with a certain ingenuity as when in reply to the playful random query: 'were you drunk or sleeping last night?' she answered, with some hesitation: 'a little of both, i think.' to the casual observer, it might seem that there was a state of full consciousness on the basis of which a partial delirium had established itself. yet on recovery from the drug there was no recollection of anything whatever that had taken place during its administration, and no sense of the lapse of time.

fantastic and marvellous as our dreams may sometimes be, they are in practically all cases made up of very simple elements. it is desirable that we should at the outset have a provisional notion as to the sources[17] of these simple elements. most writers on dreams hold that there are two great sources from which these elements are drawn: the vast reservoir of memories and the actual physical sensations experienced at the moment of dreaming, and interpreted by sleeping consciousness. various names have been given to these two groups, the recognition of which is at least as old as aristotle.[10] thus sully calls them central and peripheral, tissié, psychic and sensorial, foucault, imaginative and perceptive. fairly convenient names are those adopted by miss calkins, who calls the first group representative, the second group presentative, meaning by representative 'connected through the fact of association with the waking life of the past,' and by presentative 'connected through sense excitation with the immediate present.'[11]

the representative group falls into two subdivisions, according as the memories are of old or of recent date; these subdivisions are often quite distinct, recent dream memories belonging—probably with most people—to the previous day, while old dream memories are usually drawn from the experience of many years past, and frequently from early life. in the same way presentative impressions fall into two subdivisions,[18] according as they refer to external stimuli present to the senses, or to internal disturbances within the organism. it is scarcely necessary to observe that any or all of these four sub-groups, into which the whole of our dream life may be analysed, may become woven together in the same dream.

i have called the classification 'provisional' because, though it is convenient to adopt it for the sake of orderly arrangement, when we come to consider the matter it will be found that the material of dreams is in reality all of the same order, and purely psychic, though it may be differentiated in accordance with the character of the stimulus which evokes the psychic material of which it is made. strictly speaking, the source of the dream as a dream can only be central, and a truly presentative dream is impossible. if our senses receive an impression, external or internal, and we recognise and accept that impression for what we should recognise and accept it when awake, then we cannot be said to be dreaming. the internal and external stimuli which act upon sleeping consciousness are not a part of that consciousness, nor in any real sense its source or its cause. the ray of sunlight that falls on the dreamer, the falling off of his bedclothes, the indigestible supper he ate last night—these things can no more 'account' for his dream than the postman's knock can account for the contents of the letter he delivers. whatever the stimuli from the physical world that may knock at the door of dreaming consciousness, that consciousness is apart from them, and[19] stimuli can only reach it by undergoing transformation. they must put off the character which they wear as phenomena of the waking world; they must put on the character of phenomena of another world, the world of dreams.