§ 1. We are now approaching the most important division of our subject. So far the impressions, possibly or probably telepathic, that we have considered, have been (1) the non-externalised sort (chiefly ideal or emotional, but sometimes with a physical element) occurring during the hours of normal waking life; and (2) this sort, and also the externalised sort, occurring either in sleep or in a bodily and mental state which, though not that of sleep, is yet to some extent distinguishable from that of ordinary waking life. The class, then, that remains to be considered is the externalised sort - impressions of sight, hearing, or touch - occurring to persons who are quite clearly wide awake. The reader will not now need to be told to what family of natural phenomena I am about to refer this class. Something is presented to the percipient as apparently an independent object (or as due to an independent object) in his material environment; but no such object is really there, and what is presented is a phantasm.

Whatever peculiarities such an experience may present, there can be no mistake as to its generic characteristics: it is a hallucination.

It is naturally only with one particular species of the great family of hallucinations - the veridical species which psychology has so far not recognised - that I am here directly concerned. But it is not easy to treat the single species satisfactorily, without either assuming or supplying a certain amount of information with regard to the family to which it belongs. To assume this information would hardly be safe; for though most educated persons may have a general idea what hallucinations are, the idea is not always the result of very close or critical study. It seems better then, to err, if at all, on the side of excess and to devote one chapter to a brief general sketch of the subject; in attempting which I shall endeavour to avoid side issues, and to confine myself to points that will aid comprehension in the sequel.

Is it possible to treat Hallucinations as a single class of phenomena, marked out by definite characteristics? The popular answer would no doubt be Yes - that the distinguishing characteristic is some sort of false belief. But this is an error: in many of the best known cases of hallucination - that of Nicolai, for instance - the percipient has held, with respect to the figures that he saw or the voices that he heard, not a false but a true belief; to wit, that they did not correspond to any external reality. The only sort of hallucination which is necessarily characterised by false belief is the purely non-sensory sort - as where a person has a fixed idea that everyone is plotting against him, or that he is being secretly mesmerised from a distance. Of hallucinations of the senses, belief in their reality, though a frequent, is by no means an essential feature; a tendency to deceive is all that we can safely predicate of them.

If we seek for some further quality which shall be distinctive of both sensory and non-sensory hallucinations, the most hopeful suggestion would seem to be that both sorts are idiosyncratic and unshared. However false a belief may be, we do not call it a hallucination if it has "been in the air," and has arisen in a natural way in a plurality of minds. This is just what an idee fixe of the kind above mentioned never does: A may imagine that the world is plotting against him; but B, if he spontaneously evolves a similar notion, will imagine that the world is plotting not against A, but against himself. Instances, however, are not wanting where the idee fixe of an insane person has gradually infected an associate;1 and as contact between mind and mind is, after all, the " natural way" of spreading ideas, we can make no scientific distinction between these cases and those where, e.g., the leader of a sect has instilled delusive notions into a number of (technically) sane followers. But again, hallucinations of the senses are also occasionally shared by several persons.

Most of the alleged instances of this phenomenon are, no doubt, merely cases of collective illusion - an agreement in the misinterpretation of sensory signs produced by a real external object; but wide inquiries have brought to light a certain number of instances which I regard as genuine collective hallucinations, neither externally caused nor communicated by suggestion from one spectator to another. If then sensory and non-sensory hallucinations agree in being as a rule unshared, they agree also in presenting marked exceptions to the rule - exceptions easy to account for in the latter class, and peculiarly difficult in the former. The conclusion does not seem favourable to our chance of obtaining a neat general definition which will embrace the two; and in abandoning the search for one, I can only point, with envy, to the convenient way in which French writers are enabled not to combine but to keep them apart, by appropriating to the non-sensory species the words delire and conception delirante.

Let us then try to fix the character of hallucinations of the senses independently. The most comprehensive view is that all our instinctive judgments of visual, auditory and tactile phenomena are hallucinations, inasmuch as what is really nothing more than an affection of ourselves is instantly interpreted by us as an external object. In immediate perception, what we thus objectify is present sensation; in mental pictures, what we objectify is remembered or represented sensation. This is the view which has been worked out very ingeniously, and for psychological purposes very effectively, by M. Taine2; but it is better adapted to a general theory of sensation than to a theory of hallucinations as such.

1 A distinet description of folie a deux scorns to have been first given in 1860, by Baillarger, in the Gazctle des Hopitcux. The subject was developed in 1873 by Lasegne and Falret, in an essay reprinted in Lasegue's Etudes Meaicales. See also E. C. Seguin in the Archives of Medicine (New York, 1870), i., p. 334; Dr. Marandon de Montyel, in the Annales Medico-psychologiques, 6th series, vol. v., p. 28; and G. Lehmann in the Archiv fur Psychiatrie (Berlin), vol. xiv., p. 145.

2 De l'Intelligence, vol. i., p. 408, etc.