And having been led to this choice by the nature of the actual evidence before us, we may recognise that there is some propriety in dealing first with an issue which, complex though it is, is yet simple as compared to other articles of our programme. For the fact, if it be one, of the direct action of mind upon mind has at least a generality which makes it possible that, like the law of atomic combination in chemistry, it may be a generalisation which, though grasped at first in a very simplified and imperfect fashion, may prove to have been the essential pre-requisite of future progress.

§ 16. In a certain sense it may be said that this hidden action of one mind on another comes next in order of psychical discovery to the hidden action of the mind within itself. It will be remembered that the earliest scientific attempts to explain the phenomena of so-called Spiritualism referred them mainly to "unconscious cerebration" (Carpenter), or to what was virtually the same thing, "unconscious muscular action " (Faraday).

Now these theories, in my view, were, so far as they went, not only legitimate, but the most logical which could have been suggested to explain the scanty evidence with which alone Faraday and Carpenter attempted to deal. This unconscious action of the mind was in reality the first thing which it was needful to take into account in approaching -supernormal phenomena. I believe, indeed, that our knowledge of those hidden processes of mentation is still in its infancy, and I have elsewhere endeavoured to assign a wider range than orthodox science has yet admitted to the mind's unconscious operation.¹ But the result of this further analysis has been (as I hold) not to show that ordinary physiological considerations will suffice (as Dr. Carpenter seems to suppose) to explain all the psychical problems involved, but rather to reveal the fact that these unconscious operations of the mind do not follow the familiar channels alone, but are themselves the facilitation or the starting-point of operations which to science are wholly new.

To state the matter broadly, so as to include in a common formula the unremembered utterances of the hypnotic subject, and the involuntary writings of the waking automatist, I would maintain that when the horizon of consciousness is altered, the opening field of view is not always or wholly filled by a mere mirage or refraction of objects already familiar, but does, on rare occasions, include new objects, as real as the old. And amongst the novel energies thus liberated, the power of entering into direct communication with other intelligences seems to stand plainly forth. Among the objects in the new prospect are fragments of the thoughts and feelings of distant minds. It seems, at any rate, that some element of telepathy is perpetually meeting us throughout the whole range of these inquiries. In the first place, thought-transference is the only supernormal phenomenon which we have as yet acquired the power of inducing, even occasionally, in the normal state. It meets us also in the hypnotic trance, under the various forms of " community of sensation," " silent willing," and the like. Among the alleged cases of "mesmeric clairvoyance " the communication of pictures of places from operator to subject seems the least uncertain ground.

And again, among phenomena commonly attributed to "spirits " (but many of which may perhaps be more safely ascribed to the automatic agency of the sensitive himself), communication of thought still furnishes our best clue to " trance-speaking," "clairvoyant vision," answers to mental questions and the like. It need not, therefore, surprise us if, even in a field so apparently remote from all ordinary analogies as that of apparitions and death-wraiths, we still find that telepathy affords our most satisfactory clue.

1 See Proceedings of the S.P.R., Vols. ii. and iii.

§ 17. And here would seem to be the fitting place to explain why we have given the title of "Phantasms of the Living" to a group of records most of which will present themselves to the ordinary reader as narratives of apparitions of the dead.

When we began, in a manner to be presently described, to collect accounts of experiences which our informants regarded as inexplicable by ordinary laws, we were of course ignorant as to what forms these experiences would mainly take. But after printing and considering over two thousand depositions which seemed prima facie to deserve attention, we find that more than half of them are narratives of appearances or other impressions coincident either with the death of the person seen or with some critical moment in his life-history.

The value of the accounts of apparitions after death is lessened, moreover, by a consideration which is obvious enough as soon as these narratives come to be critically considered. The difficulty in dealing with all these hallucinations - with all appearances to which no persistent three-dimensional reality corresponds - is to determine whether they are veridical, or truth-telling - whether, that is, they do in fact correspond to some action which is going on in some other place or on some other plane of being; - or whether, on the other hand, they are merely morbid or casual - the random and meaningless fictions of an over-stimulated eye or brain. Now, in the case of apparitions at the moment of death or crisis, we have at any rate an objective fact to look to. If we can prove that a great number of apparitions coincide with the death of the person seen, we may fairly say, as we do say, that chance alone cannot explain this coincidence, and that there is a causal connection between the two events. But if I have a vision of a friend recently dead, and on whom my thoughts have been dwelling, we cannot be sure that this may not be a merely delusive hallucination - the mere offspring of my own brooding sorrow.

In order to get at all nearly the same degree of evidence for a dead person's appearance that we can get for a dying person's appearance, it seems necessary that the apparition should either communicate some fact known only to the deceased, or should be noted independently by more than one person at once or successively. And our evidence of this kind is at present scarcely sufficient to support any assured conclusion.1