642 C. Hallucinations And Illusions Of Memory

Another important objection was raised on the score of possible hallucinations of memory as to the incidents alleged. Now, illusions of memory - inaccurate recollections of actual events - are as familiar as illusions of the senses perceptions based on misinterpretations of material objects. It is also possible that there may be hallucinations of memory - or recollections of events that have never really occurred at all - analogous to sensory hallucinations, or perceptions based on no material objects. In the case of memory, it is very difficult to draw the line in any concrete instance between illusion and hallucination; but it is important to recognise the theoretical distinction between the two types and not to assume without proof that what can be predicated of one type necessarily applies also to the other.

The hypothetical "pseudo-presentiments" by which Professor J. Royce attempted to explain a large number of impressions reported as veridical belong to the class of hallucinations of memory. To quote his own description (see Proceedings of the American Society for Psychical Research, vol. i. p. 366): "This hypothesis is that in certain people, under certain exciting circumstances, there occur what I shall henceforth call pseudo-presentiments, i.e. more or less instantaneous and irresistible hallucinations of memory, which make it seem to one that something which now excites or astonishes him has been prefigured in a recent dream, or in the form of some other warning, although this seeming is wholly unfounded, and although the supposed prophecy really succeeds its own fulfilment." Thus, on hearing of a death, one might think that one had just dreamt or had a presentiment of it. " Members of the same family would," he suggests, " be especially apt to be similarly subject to this form of illusion, and then the same news would show them all the same mirage of memory, with startling results in the way of 'telepathic 'evidence." He compares this with the familiar illusion of "double memory" or the déjà vu, - the feeling that an experience which is being passed through is an exact repetition of something that has happened before, - though he admits that this latter illusion generally corrects itself at once.1 The only positive evidence brought forward in support of the supposed new type of illusion is afforded by two cases quoted from Professor Emil Kræpelin (Archiv fur Psychiatric, vol. xviii. p. 397) of insane patients who, among other delusions, constantly fancied that whatever happened to them, or attracted their attention, had already been predicted to them.

1 For exact references to all these cases, see Report on the Census of Hallucinations, p. 223.

In reply, Gurney pointed out (1) that in recent cases where we have evidence that the percipient's experience was related to some one else before the arrival of the news, the hypothesis of collective illusion fails to account for the difference in the recollections of the witnesses, - one remembering that he has had a certain impression, and the other that this impression has been related to him. (2) Professor Royce omitted altogether to consider the type of case most important evidentially, namely, veridical sensory hallucinations. Supposing that the news of an exciting event has some tendency to produce the impression that one had known of it before, this would not show that the receipt of news, say, of the death of a friend, had any tendency to produce the impression of having recently seen an apparition of the friend, which did not announce, or even suggest his death. Further, the assignment of the supposed recollection to a particular time constitutes an essential difference from the familiar illusions of "double memory," which are quite unlocalised in time.

Mr. Hodgson (see Proceedings of the American Society for Psychical Research, p. 542) points out further that the hypothesis can only: - with any show of reason - be applied to cases of certain kinds, namely, remote cases, and those depending only on the evidence of the percipient, especially where the experience is merely a dream or mental impression, still more if it is supposed to foreshadow future events, or if the percipient has forgotten it and is only reminded of it by receiving the news.

The conception of "pseudo-presentiments " has been adopted, greatly extended, and applied in a wholesale manner to the telepathic evidence by E. Parish. 2 He fails altogether to grasp the distinction between illusions of memory, for which there is ample evidence, and hallucinations of memory, for which the evidence is exceedingly meagre - resting, in fact, almost entirely on the rare morbid cases quoted again from Kræpelin. By premising both positive and negative defects of memory, Parish disposes at one blow of the argument that apparitions at the time of death are proportionately too numerous to be attributed to chance; he simply supposes that non-coincidental apparitions are very frequent, but that almost all of them are forgotten; while coincidental ones hardly ever really occur, but are often falsely remembered.

1 For a discussion of this phenomenon, to which I have given the name of fromnesia, see Proceedings S.P.R., vol. xi. pp. 340-347.

2 See Ueber die Trugwahrnehmung, by E. Parish (Leipzig, 1894), and Reviews of it in Proceedings S.P.R., vol. xi. p. 162, and in Mr. A. Lang's The Making of Religion, p. 337 (Appendix on "Oppositions of Science"). Also Zur Kritik des telepathisehcn Beweis-Materiels, by E. Parish (Leipzig, 1897), and Review of it in Proceedings S.P.R., vol. xiii. p. 589.

A priori speculations as to the working of memory will obviously lead to whatever conclusion it is desired to support. But positive evidence is more worthy of serious consideration, and it is possible to obtain this by comparing the recollections of the same person at different times, and also by comparing the recollections of different persons. The great mass of records collected by the S. P. R. have been examined by their investigators with this special end in view. In particular, the study of one important type of cases (see "The Possibilities of Mal-observation and Lapse of Memory from a Practical Point of View: Introduction," by R. Hodgson; "Experimental Investigation," by S. J. Davey; Proceedings S. P. R., vol. iv. pp. 381 - 495) has shown what kinds of error occur in actual experience. Many events are found to be absolutely forgotten; the blank in memory may be filled by conjectured details, or may close up, so that events really separated by an interval of time may be remembered as occurring close together; a detail afterwards remembered may be interpolated in a record; or an event occurring at one time may be remembered as occurring at another.

All these are errors of omission, exaggeration, or distortion of a generally - though more or less vaguely - recognised kind; but few persons who have not made a special study of the subject realise the extent to which testimony is vitiated by them. Such false memories are illusions, which develop and consolidate gradually as time goes on. But investigation of the experiences of sane persons has not revealed any instances of Professor Roycés hypothetical hallucinations of memory which arise instantaneously in a complex and fully elaborated form.

Illusions of memory have also been studied from the point of view of their bearing on the subject of "hypnotic crimes" (see 555 A), stress having been laid especially on the facility with which hysterical or weak-minded persons may be made to imagine, or may spontaneously imagine, that they have witnessed fictitious occurrences, and may thus be brought to bear false witness. Bernheim also found that by hypnotic suggestion he could produce actual hallucinations of memory in a few specially susceptible subjects. But these cases, again, are obviously not comparable with Professor Roycés "pseudo-presentiments," which are supposed to arise suddenly and spontaneously in persons in a normal condition.