§ I. The inward impressions of distant events with which I have so far dealt have all been waking impressions. They have visited the percipient in the midst of his daily employments, and have often caused emotions that seemed quite incongruous with the normal current of life in which they mingled. But there is another department of experience which we are accustomed to consider as par excellence the domain of inward impressions, and from which the normal current of life is altogether shut off - the department of dreams. And this department, where inward ideas and images dominate unchecked, is also one which covers so large a period of human existence as to make it a priori probable that a considerable number of "transferred impressions" (supposing such things to exist) would fall within it. It would naturally, therefore, suggest itself as our next field of inquiry.

But dreams not only fall in naturally at this point; they are a means of advance. They comprise in themselves the whole range of transition from ideal and emotional to distinctly sensory affections; and they thus supply a most convenient link between the vaguer sort of transferred impressions and the more concrete and definitely embodied sort. The telepathic communications of the last two chapters, even when connected with recognisable images of persons and things, did not affect sight or hearing in such a way as to suggest the physical presence of the objects. Now many dreams are of just this impalpable kind. The material objects which figure in them are often the very vaguest of images, not localised in any particular spot. It is the general idea, the generalised form, of a person that presents itself, not a figure in a special attitude or clad in a special dress; events pass through the mind clothed in the faintly represented imagery in which a waking train of memory or of reverie will embody its contents.

Such a dream only differs from a waking reverie in that it has not to compete, on the field of attention, with any objective facts; it is not contrasted with the immediate experience which the external world forces at every moment on the waking senses; and it is, therefore, itself accepted as immediate experience. With some persons it is rare for their dreams ever to emerge into more concrete actuality than this; and telepathy often seems to act in dreams of just such a veiled and abstract kind. From this lowest stage the transition is a gradual one up to the most vivid and detailed dream-imagery, the features of which are engraved on the memory as sharply as those of a striking scene in waking life1; and at every step of the transition we find evidence (how far conclusive will be seen later) to the action of telepathy.

It is only, however, when we come to the most distinctly sensory class of all that the full theoretic importance of dreams can be realised. To make this clear, I must ask a moment's indulgence for a statement of some very obvious facts. Vivid dreams present themselves to us in two very different aspects. There is first the standpoint which we occupy when we are dreaming them. From that standpoint, the world with which they present us is often as external as the real one; and our perceptions in that world are perceptions of outward and abiding things, among which we live and move with as much sense of reality as if we had never known the disillusion of waking. To the dreamer, his more vivid and concrete images are actual percepts, calling his senses (in physiological language his sensory centres) into play just as external reality would. But there is of course a totally different standpoint from which to regard dreams, namely, the external and critical one that we habitually assume during waking life.

We then think of them merely as that floating phantasmagoria whose transience and unreality have been the theme of philosophers and poets; which has very singular relations to time, and no real relations at all to our familiar space - unless, indeed, we loosely identify it with its physical conditions, and localise it in the brain. Dreams, then, have this peculiarity: they are distinct affections of the senses, which yet, in reflecting on them, we rarely or never confound with objective facts; waking hallucinations, on the other hand - spectral illusions or ghostly visitants - are often so confounded. The sleeping experiences are marked off from external reality in the minds of all of us by the very fact that we wake from them; our change of condition makes their vanishing seem natural; and thus looked back on, they will often seem to have been mere vague representations, i.e., something less than affections of the senses. The waking experiences cannot be woke from; their vanishing seems unconnected with the percipient and therefore unnatural; and thus looked back on, they will often seem to have been independent realities, i.e., something more than affections of the senses.

Now it is as affections of the senses, and not as independent realities, that our Class A, the externalised sort of "phantasms of the living," are treated in this book. In the theory that the percipient is impressed from a distance, and in the very word phantasm, it is implied that what he sees or hears has no objective basis or existence in that part of the external world where his body is situated; and whether he be asleep or awake, his relation to what he perceives, and of this to reality, is the same. But I shall be proceeding by the easiest route if (so far as the evidence will allow) I first trace the occurrence of the telepathic phantasms in the region of experience where sensory phantasms of some sort are normal and familiar, and are habitually judged of rightly as affections of the inner sense,1 before passing to the region where phantasms of any sort are abnormal and unfamiliar, and are perpetually judged of wrongly as affections of the outer sense. The rapprochement which will thus be established between the sleeping and the waking cases will receive further and interesting illustration in certain intermediate stages which we shall encounter on the way.

We shall find that one set of phenomena merges into the other by gradual steps, and that this "borderland" is itself a region specially rich in the telepathic impressions.

1 Those whose dreams are habitually of the more ideal and impalpable sort have sometimes a difficulty in realising the extreme sensory vividness of dreams at the other end of the scale, dreamt often by persons who have no exceptional power of visualising when awake. I myself lately dreamt that, meeting a stranger in Bond Street, I was arrested by a certain familiarity in the face, which I continued to scrutinise with puzzled eagerness, till I at last identified it with a portrait in the Grosvenor Gallery of the preceding year. Awake, I can scarcely recall the portrait at all.

I have also heard it asserted that (apart from actual external sounds) the sense of hearing is never distinctly exercised in dreams. I never had a more vivid dream than one of a few nights back, where some rifle-shooting, in the midst of which I found myself, caused me again and again precisely the same dread before the sound came, and the same intense irritation when it came, as I associate with the firing of a pistol on the stage. I could distinctly trace this dream to a similar (though less acute) irritation which I had suffered from the cracking of whips in a foreign city on the previous day.