These hypotheses—none of which, as Mrs. Sidgwick expressly states (op. cit., p. 145), seemed to herself satisfactory—did nevertheless, I think, comprise all the deductions which could reasonably be made from the evidence as it at that time stood. A few modifications, which the experience of subsequent years has led me to introduce, can hardly be said to afford further explanation, although they state the difficulties in what now seems to me a more hopeful way.

In the first place then—as already explained in Chapter VI.—I in some sense fuse into one Mrs. Sidgwick's two first hypotheses by my own hypothesis of actual presence, actual spatial changes induced in the metetherial, but not in the material world. I hold that when the phantasm is discerned by more than one person at once (and on some other, but not all other occasions) it is actually effecting a change in that portion of space where it is perceived, although not, as a rule, in the matter which occupies that place. It is, therefore, not optically nor acoustically perceived; perhaps no rays of light are reflected nor waves of air set in motion; but an unknown form of supernormal perception, not necessarily acting through the sensory end-organs, comes into play. In the next place, I am inclined to lay stress on the parallel between these narratives of haunting and those phantasms of the living which I have already classed as psychorrhagic. In each case, as it seems to me, there is an involuntary detachment of some element of the spirit, probably with no knowledge thereof at the main centre of consciousness. Those "haunts by the living," as they may be called, where, for instance, a man is seen phantasmally standing before his own fireplace, seem to me to be repeated, perhaps more readily, after the spirit is freed from the flesh.

Again, I think that the curious question as to the influence of certain houses in generating apparitions may be included under the broader heading of Retrocognition. That is to say, we are not here dealing with a special condition of certain houses, but with a branch of the wide problem as to the relation of supernormal phenomena to time. Manifestations which occur in haunted houses depend, let us say, on something which has taken place a long time ago. In what way do they depend on that past event? Are they a sequel, or only a residue? Is there fresh operation going on, or only fresh perception of something already accomplished? Or can we in such a case draw any real distinction between a continued action and a continued perception of a past action? The closest parallel, as it seems to me, although not at first sight an obvious one, lies between these phenomena of haunting, these persistent sights and sounds, and certain phenomena of crystal-vision and of automatic script, which also seem to depend somehow upon long-past events,—to be their sequel or their residue. One specimen case I give in Appendix (VII. G), where the connection of the haunting apparition with a certain person long deceased may be maintained with more than usual plausibility. From that level the traceable connections get weaker and weaker, until we come to phantasmal scenes where there is no longer any even apparent claim to the contemporary agency of human spirits. Such a vision, for instance, as that of a line of spectral deer crossing a ford, may indeed, if seen in the same place by several independent observers, be held to be something more than a mere subjective fancy; but what in reality such a picture signifies is a question which brings us at once to theories of the permanence or simultaneity of all phenomena in a timeless Universal Soul.

Such conceptions, however difficult, are among the highest to which our mind can reach. Could we approach them more nearly, they might deeply influence our view, even of our own remote individual destiny. So, perhaps, shall it some day be; at present we may be well satisfied if we can push our knowledge of that destiny one step further than of old, even just behind that veil which has so long hung impenetrably before the eyes of men.

Here, then, is a natural place of pause in our inquiry.

The discussion of the ethical aspect of these questions I have postponed to my concluding chapter. But one point already stands out from the evidence—at once so important and so manifest that it seems well to call attention to it at once—as a solvent more potent than any Lucretius could apply to human superstition and human fears.

In this long string of narratives, complex and bizarre though their details may be, we yet observe that the character of the appearance varies in a definite manner with their distinctness and individuality. Haunting phantoms, incoherent and unintelligent, may seem restless and unhappy. But as they rise into definiteness, intelligence, individuality, the phantoms rise also into love and joy. I cannot recall one single case of a proved posthumous combination of intelligence with wickedness. Such evil as our evidence will show us—we have as yet hardly come across it in this book—is scarcely more than monkeyish mischief, childish folly. In dealing with automatic script, for instance, we shall have to wonder whence come the occasional vulgar jokes or silly mystifications. We shall discuss whether they are a kind of dream of the automatist's own, or whether they indicate the existence of unembodied intelligences on the level of the dog or the ape. But, on the other hand, all that world-old conception of Evil Spirits, of malevolent Powers, which has been the basis of so much of actual devil-worship and of so much more of vague supernatural fear;—all this insensibly melts from the mind as we study the evidence before us.

Hunc igitur terrorem animi tenebrasque necessest
Non radii solis, neque lucida tela diei
Discutiant sed, naturæ species ratioque.

Here surely is a fact of no little meaning. Our narratives have been collected from men and women of many types, holding all varieties of ordinary opinion. Yet the upshot of all these narratives is to emphasise a point which profoundly differentiates the scientific from the superstitious view of spiritual phenomena. The terror which shaped primitive theologies still tinges for the populace every hint of intercourse with disembodied souls. The transmutation of savage fear into scientific curiosity is of the essence of civilisation. Towards that transmutation each separate fragment of our evidence, with undesigned concordance, indisputably tends. In that faintly opening world of spirit I can find nothing worse than living men; I seem to discern not an intensification but a disintegration of selfishness, malevolence, pride. And is not this a natural result of any cosmic moral evolution? If the selfish man (as Marcus Antoninus has it) "is a kind of boil or imposthume upon the universe," must not his egoistic impulses suffer in that wider world a sure, even if a painful, decay; finding no support or sustenance among those permanent forces which maintain the stream of things?

I have thus indicated one point of primary importance on which the undesignedly coincident testimony of hundreds of first-hand narratives supports a conclusion, not yet popularly accepted, but in harmony with the evolutionary conceptions which rule our modern thought. Nor does this point stand alone. I can find, indeed, no guarantee of absolute and idle bliss; no triumph in any exclusive salvation. But the student of these narratives will, I think, discover throughout them uncontradicted indications of the persistence of Love, the growth of Joy, the willing submission to Law.