But now we are confronted by another possible element in these vaguer classes of apparitions, harder to evaluate even than the possible action of incarnate minds. I mean the possible _results_ of past mental action, which, for aught we know, may persist in some perceptible manner, without fresh reinforcement, just as the results of past bodily action persist. This question leads to the still wider question of _retrocognition_, and of the relation of psychical phenomena to _time_ generally--a problem whose discussion cannot be attempted here.[155] Yet we must remember that such possibilities exist; they may explain certain phenomena into which little of fresh intelligence seems to enter, as, for instance, the alleged persistence, perhaps for years, of meaningless sounds in a particular room or house. And since we are coming now to cases into which this element of meaningless sound will enter largely, it seems right to begin their discussion with a small group of cases where there is evidence for the definite agency of some dying or deceased person in connection with inarticulate sounds, or I should rather say of the _connection_ of some deceased person with the sounds; since the best explanation may perhaps be that they are _sounds of welcome_--before or after actual death--corresponding to those _apparitions of welcome_ of which we have already had specimens. One of our cases (see _Phantasms of the Living_, vol. ii. p. 639 [§ 747]) is remarkable in that the auditory hallucination--a sound as of female voices gently singing--was heard by five persons, by four of them, as it seems, independently, and in two places, on different sides of the house. At the same time, one person--the Eton master whose mother had just died, and who was therefore presumably in a frame of mind more prone to hallucination than the physician, matron, friend, or servants who actually did hear the singing--himself heard nothing at all. In this case the physician felt no doubt that Mrs. L. was actually dead; and in fact it was during the laying out of the body that the sounds occurred. I have already discussed (Chapter VI.) the nature of these phantasmal sounds;--nor is it contrary to our analogies that the person most deeply concerned in the death should in this case fail to hear them. But the point on which I would here lay stress is that phantasmal sounds--even non-articulate sounds--may be as clear a manifestation of personality as phantasmal figures. Among non-articulate noises music is, of course, the most pleasing; but sounds, for instance, which imitate the work of a carpenter's shop, may be equally human and intelligent. In some of the cases of this class we see apparent attempts of various kinds to simulate sounds such as men and women--or manufactured, as opposed to natural, objects--are accustomed to produce. To claim this humanity, to indicate this intelligence, seems the only motive of sounds of this kind.[156]