On both of these questions there have been full and repeated discussions elsewhere. I need not re-argue them at length here, but will refer the reader to the "Report on the Census of Hallucinations," Proceedings S.P.R., vol. x., where every source of error as yet discovered has been pretty fully considered.
To that volume also I must refer him for a thorough discussion of the arguments for and against chance-coincidence. The conclusion to which the Committee unanimously came is expressed in the closing words: "Between deaths and apparitions of the dying person a connection exists which is not due to chance alone."
We have a right, I think, to say that only by another census of hallucinations, equally careful, more extensive, and yielding absolutely different results, could this conclusion be overthrown.
In forming this conclusion, apparitions at death are of course selected, because, death being an unique event in man's earthly existence, the coincidences between death and apparitions afford a favourable case for statistical treatment. But the coincidences between apparitions and crises other than death, although not susceptible of the same arithmetical precision of estimate, are, as will be seen, quite equally convincing. To this great mass of spontaneous cases we must now turn.
The arrangement of these cases is not easy; nor are they capable of being presented in one logically consequent series.
But the conception of psychical invasion or excursion on which I have already dwelt has at any rate this advantage, that it is sufficiently fundamental to allow of our arrangement of all our recorded cases—perhaps of all possible cases of apparition—in accordance with its own lines.
Our scheme will include all observable telepathic action, from the faint currents which we may imagine to be continually passing between man and man, up to the point—reserved for the following chapter—where one of the parties to the telepathic intercourse has definitely quitted the flesh. The first term in our series must be conveniently vague: the last must lead us to the threshold of the spiritual world.
I must begin with cases where the action of the excursive fragment of the personality is of the weakest kind—the least capable of affecting other observers, or of being recalled into the agent's own waking memory.
Such cases, naturally enough, will be hard to bring up to evidential level. It must depend on mere chance whether these weak and aimless psychical excursions are observed at all; or are observed in such a way as to lead us to attribute them to anything more than the subjective fancy of the observers.
How can a casual vision—say, of a lady sitting in her drawing-room,—of a man returning home at six o'clock—be distinguished from memory-images on the one hand and from what I may term "expectation-images" on the other? The picture of the lady may be a slightly modified and externalised reminiscence; the picture of the man walking up to the door may be a mere projection of what the observer was hoping to see.