The recognised phantasms have, moreover, usually represented persons whom the percipient was habitually seeing in real life - often a relative or a servant living in the same house - so that the delusion may still be explained by the analogy of after-images; or else they have represented dead persons whose memory was dear to the percipient, and whose images might readily be evoked by memory.1 Of the unrecognised phantasms we have no reason to suppose that they represented any one at all, even in those cases - the majority of the whole number - where the non-recognition was not due to any indistinctness in the phantasm, but was of the same kind as when in real life we see a face or hear a voice which we perceive at once to be strange to us. But while as a rule so clearly subjective, these human and well-developed forms of hallucination include also nearly the whole of the instances which will be presented in the following chapters as telepathic. It is naturally among them, therefore, that the ambiguous cases to which I have referred are principally to be found.

1Some of these may possibly have been real noises whoso cause was not ascertained. But I have included no ease where my informant did not himself hold this explanation to be excluded.

2We have had, however, a telepathic example of this type, in case 101, p. 204.

The main grounds of ambiguity (apart from mere uncertainties as to matters of fact, which are not now in question) are, I think, four in number, and are summed up in the following types2: -

(1) Coincidentally with a "recognised" hallucination, the person whose presence it suggested has been in a condition more or less unusual; but we may fairly doubt whether it was unusual enough to justify even a provisional supposition that it affected the percipient.

(2) At the time of a "recognised" hallucination, the person whose presence it suggested has been apparently in a perfectly normal condition; but the same hallucination has been repeated, and repeated to different percipients; and the improbability that the independent hallucinations of several persons should by accident represent the same person suggests some special power of influence in the latter.3

(3) The person whose presence the hallucination presented has been in a decidedly abnormal condition at a time sufficiently near the time of the hallucination for the correspondence to be observed; but the correspondence has not been exact; and the doubt is whether the amount of discrepancy is such as to admit, or such as to preclude, the hypothesis of a causal connection. The question here is the one which was early raised, when we were considering the elements of a telepathic case (pp. 112-13), and was then left undecided.

1See above, p. 314. The term "after-image" in connection with the apparitions of dead people was suggested in a paper privately communicated to me by Mr. J. Jacobs.

2To be quite complete, this list ought perhaps to include the ambiguity which is involved in the percipient's failure to connect his impression, at the time, with the supposed agent; but this point is sufficiently noticed elsewhere (p. 1G2, and Chap. xii., § 5).

3For illustrations of this very rare and isolated type, see Chap. xiv., § 6.

(4) The case is so far suggestive of telepathy that, at or very near the time of the hallucination, some absent person has been in circumstances which might seem to identify him as the agent; but there is also evidence of a condition in the percipient's own mind which might be regarded as the independent cause of his experience.

Let us begin with this last head, and inquire of what nature an independent subjective cause of the phenomena now in question is likely to be.

Though for the most part, as I have said, the predisposing conditions of these rare and transient delusions entirely elude us, just three emotional states can be named which are present often enough to warrant a suspicion that they may be truly efficacious. These are special forms of anxiety; of awe; and of expectancy. We will consider them in order.