Dr. Winslow gives some cases of extraordinary memory, characterising incipient brain-disease, and he also records many curious instances of "double consciousness", in which the patient practically lives a double life, remembering in each state only those incidents which occurred in it. Here, again, we seem to be confronted with the shifting threshold as the only tenable hypothesis.

Persons under hypnotism frequently exhibit an extreme exaltation of memory, repeating long passages read to them but once, recalling with accuracy long past and trivial events, describing minutely the insignificant occurrences of many successive days. Many instances of this kind will be found by the student in Binet and Féré's Animal Magnetism, and in Dr. Richer's Etudes sur la grande Hystérie.

With this rough survey of the field of memory in our minds, we must seek for some hypothesis which will resume the facts, and which, tested by fresh experiment, will explain other memory phenomena. I put Hume's hypothesis out of court, and proceed to consider the Materialistic and Theosophical theories of memory, to answer the question whether memory is a function of matter in motion, or a faculty of the Self, the Ego, functioning through matter, but not resultant from it. [Page 12]

The Materialistic Theory of Memory. — According to this theory, memory, like all other mental functions, is the result of the vibrations of nerve-cells, and may be expressed in terms of matter and motion. When a stimulus from the Object World sets up a vibration in a sense-organ, that vibration is propagated as a wave from cell to cell of the nervous chain till it reaches its appropriate centre in the cerebrum. There arises the perception, the outcome of mental activity. This nervous action, once set up, tends to repeat itself more easily with each similar stimulus, the nervous energy following the path of least resistance, and each recurrence of the similar vibration making easier further repetition. Such a vibration having once been set up, it may recur in the absence of the external stimulus, and we have the idea in lieu of the sensation-perception. Whenever the nerve-cells vibrate as they vibrated under the first stimulus, the idea recurs, and this recurrence is termed memory. Now when the vibration is first set up it is at its strongest, and it is argued that this intensity of vibration lessens, until it is not sufficient to affect the consciousness. Mr. James Ward writes (Journal of Speculative Philosophy, vol. xvii, No. 2, quoted by Sully):

What, now, do we know concerning this central image in the intervals when it is not consciously presented?. Manifestly our knowledge in this case can only be inferential at the best. But there are two facts, the importance of which Herbart was the first to see, from which we may learn something: I refer to what he calls the rising and falling of presentations. All presentations having more than a liminal intensity rise gradually to a maximum and gradually decline; and when they have fallen below the threshold of consciousness altogether, the process seems to continue; for the longer the time that elapses before their "revival", the fainter they appear when revived, and the more slowly they rise. This [Page 13] evanescence in most rapid at first, becoming less as the intensity of the presentation diminishes. It is too much to say that this holds with mathematical accuracy, although Herbart has gone this length. Still it is true enough to suggest the notion that an object, even when it is no longer able to influence attention, continues to be presented, though with even less and less absolute intensity, till at length this intensity declines to an almost dead level just above zero.

In order to understand more precisely what is meant by the Law of Contiguous Association, we may let A and B stand for two impressions (percepts) occurring together, and a and b for the two representations answering to these. Then the Law asserts that when A (or a) recurs it will tend to excite or call up b; and similarly that the recurrence of B (or b) will tend to excite a . ... The physiological basis of this contiguous association seems to be the fact that two nerve structures which have repeatedly acted together acquire a disposition to act in combination in the same way. This fact is explained by the hypothesis that such a conjoint action of two nerve centres somehow tends to fix the line of nervous excitation or nervous discharge when one centre is again stimulated in the direction of the other. In other words the paths of connection are formed between the two regions. But it may be doubted whether [Page 14] physiologists can as yet give a satisfactory account of the nervous concomitants of the associative process.

If the association of each feeling with its general class answers to the localisation of the corresponding nervous action within the great nervous mass in which all feelings of that class arise — if the association of this feeling with its sub-class answers to the localisation of the nervous action within that part of this great nervous mass in which feelings of this sub-class arise, and so on to the end with the smallest groups of feelings and smallest clusters of nerve-vesicles; then to what answers the association of each feeling with predecessors identical in kind ? It answers to the re-excitation of the particular vesicle or vesicles which, when before excited, yielded the like feeling before experienced; the appropriate stimulus having set up in certain vesicles the molecular changes which they undergo when disturbed, there is aroused a feeling of the same quality with feelings previously aroused when such stimuli set up such changes in these vesicles. And the association of the feeling with preceding like feelings, corresponds to the physical re-excitation of the same structures. (Principles of Psychology, vol. 1.)

We are then to regard memory as the result of the re-excitation of vesicles in the brain — the theory is clear and definite enough. Is it true ?

The first difficulty that arises is the limited space available for the containment of these vesicles, and the consequent limitation of their number. It is true that their possible combinations may be practically infinite in number, but this does not much help us; for they are to vibrate continually, however feebly, [Page 15] so long as an idea is capable of revival, and a vesicle vibrating simultaneously in some thousands of combinations would be in a parlous molecular condition. For all these combinations must exist simultaneously, and each must maintain its interrelated vibrations without cessation. Now, is this possible ? It is true that from the vibrating strings of a piano you may get myriads of combinations of notes; but you cannot have all these combinations sounding from the strings at the same time, some loud and some soft, some forcible and some feeble. By keeping the loud pedal down you may keep some combinations going for a short while, while you produce fresh vibrations; but what is the effect ? A blurred confusion of sounds, causing an intolerable discord. If we are to explain memory under the laws of matter in motion, we must accept the consequences deductible from those laws, and these consequences are inconsistent with the facts of memory as we know them. Any attempt to represent clearly in consciousness the physical concomitants of memory as merely the outcome of vibrating nervous elements will prove to the student the impossibility of this hypothesis. The brain is a sufficiently wonderful mechanism as the organ of mind; as the creator of mind, it is inconceivable.

Du Prel (Philosophy of Mysticism) helps us to realise the difficulties enveloping the Materialistic hypothesis. On this hypothesis "memory would depend on material brain-traces, left behind by impressions; by the act of memory such traces are continually renewed, re-chiselled as it were, and so there arise well-worn tracks", (Herbert Spencer's "lines of least resistance") "in which the coach of memory is conducted with especial facility ". And he adds;[Page 16]

The deductions from this view had already been drawn by the materialists of the last century. Hook and others reckoned that, since one third of a second sufficed for the production of an impression, in 100 years a man must have collected in his brain 9,467,280,000 traces or copies of impressions, or, reduced by one third for the period of sleep, 3,155,760,000; thus in fifty years, 1,577,880,000; further, that allowing a weight of four pounds to the brain, and subtracting one pound for blood and vessels, and another for the external integument, a single grain of brain substance must contain 205,542 traces . .. Moreover our intellectual life does not consist in mere impressions; these form only the material of our judgment. These brain atoms do not help us to judgment, notwithstanding their magical properties, so that, we must suppose that whenever we form a sentence or a judgment, the impressions are combined, like the letters in a compositor's box, these atoms, however, being at the same time, compositor and box.

There is another result that would follow from Memory being only the outcome of vibrating cells. Memory is the faculty which receives the impress of our experiences and preserves them; many of these impressions fade away, and we say we have forgotten. Yet it is clear that these impressions may be revived. They are therefore not destroyed, but they are so faint that they sink below the threshold of consciousness, and so no longer form part of its normal content. If thought be but a "mode of motion", memory must be similarly regarded: but it is not possible to conceive that each impression of our past life, recorded in consciousness, is still vibrating in some group of brain cells, only so feebly that it does not rise over the threshold. For these same cells are continually being thrown into new groupings for new vibrations, and these cannot all coexist, and the fainter ones be each capable of receiving fresh impulses which may so intensify their motion as to [Page 17] raise them again into consciousness. Now if these vibrations=memory, if we have only matter in motion, we know the laws of Dynamics sufficiently well to say that if a body be set vibrating, and new forces be successively brought to act upon it and set up new vibrations, there will not be in that body the coexistence of each, separate set of vibrations successively impressed upon it, but it will vibrate in a way differing from each single set and compounded of all. So that memory, as a mode of motion, would not give us the record of the past, but would present us with a new story, the resultant of all those past vibrations, and this would be ever changing, as new impressions, causing new vibrations, come in to modify the resultant of the whole. If the reader have in mind the phenomena of memory given in the earlier part of this essay; if he note that these seem to imply that we forget nothing, i.e., that every vibration caused throughout life persists; if, remembering this, he once more attempts to represent clearly in consciousness the brain-condition required by this theory, is it too much to say that he will be compelled to admit that it is inconceivable ?

Nor can we forget that there is a certain race-memory, wrought into our physical organisms, which still further complicates the work accomplished by these overburdened vesicles. This unconscious memory of the body, derived through physical inheritance, cannot be wholly thrown out of account when we deal with cell-vibrations.

The Theosophical Theory of Memory. — Here I must guard myself; I cannot really put the Theosophical theory, for I do not find it set out in any work that I have read. I can only suggest a theory, which seems to me, as a student of Theosophy, to be fairly deducible from the constitution of man as laid down [Page 18] in Theosophical treatises. We learn to distinguish between true individuality, the Ego, and the temporary personality that clothes it. The Ego is the conscious, the thinking, agent. It is this Ego of whom the mind forms part, one of whose functions is Memory. Every event that occurs passes into the consciousness of the Ego and is there stored up: the Past is thus, to it, ever the Present, since all is present in consciousness. [All is present in eternal ideation, Alaya the universal soul and consciousness — we are taught; and the Higher Ego (Manas) is the first-born of Alaya or Mahat, being called Mãnasaputra “Son of the Mind”] But how far this Ego can impress its knowledge on the brain of the physical organism with which it is connected, and thus cause this knowledge to enter the consciousness of the person concerned, must, in the nature of the case, depend on the condition of the organism at the moment, and the laws within which it works. What we call the threshold of consciousness divides what is "remembered" from what is "forgotten ". All above the threshold is within the personal consciousness, while all below this threshold is outside it. But this threshold belongs to the personal consciousness, and — here is the significant point — varies with the material conditions of the moment. It is movable, not fixed, and the contents of consciousness vary with the movement of the threshold. Thus:

[Page 19] Let A B represent the consciousness of the Ego; let C D represent the threshold of consciousness of the person: of all above C D the person will be conscious, and it will be impressed on the material brain: of all below C D he will be unconscious. But if C D be movable upwards and downwards, the contents of his consciousness will vary with its movement, and he will remember or forget according as the idea is above or below this dividing line. [We have to exclude from this the impressions of a purely physical nature, such as enter into the category of animal perception and memory. Such impressions reach the Human Ego, and it cannot fail to note them; but they do not impress themselves indelibly on its consciousness, and can never, therefore, follow the Ego to Devachan]

Now the condition of the organism is constantly varying, but there are two states of consciousness which occur in every one and are clearly distinguishable — the waking consciousness and the dream consciousness. The contents of these differ to a remarkable extent, and they work under curiously different conditions. The waking consciousness works under conditions of time and space: the dream consciousness is free from them; it can live through years in a second of time, it can annihilate space in its movements. In the dream, the place of the dreamer depends on his thought; he is where he thinks himself. Not only so, but the dream consciousness often retains events erased from the waking. Let the reader run back a few pages and note the curious phenomena of reproduction without recollection in the dream state. Is it an impossible theory that when the senses are closed to the Object World, when the bodily functions have touched their lowest activity, then the Ego may be able to impress on this negative organism far more of its own contents than it can impress upon it in its more vigorous [Page 20] state ? Does not it seem as though that which is below the threshold of the waking consciousness becomes that which is above the threshold of the dream consciousness, and as though the double life of waking and sleeping is but the activity of the one Ego working under contrasted physical conditions ?

If this be not so, we seem to be driven to the conception of a duality at the very centre of our being: each man is not one, but twain, in the innermost recesses of consciousness.

On the other hand, the theory for which I contend, leaves the individuality single, varying in its manifestations according to the physical conditions through which it works; and all the strange cases of double consciousness, which have so perplexed the physiologist and the psychologist, together with the phenomena of somnambulism, mesmerism, hypnotism, and similar conditions, fall into line as severally belonging to one of the two states of consciousness, the dream and the waking, the Ego working equally in either but conditioned in turn by each.

"Ordinary sleep", as Du Prel says, is "a condition intermediate between waking and somnambulism, the latter being only its exaltation". In this connection these facts are to be noted: if we sleep lightly and dream, we remember our dreams; if we sleep more soundly, we sometimes remember the dream vividly on waking, but in an hour or two we have completely forgotten it and cannot revive the memory, try as we may; in deep sleep we dream, as has often been discovered by closely watching a person wrapped in profound slumber, but no trace remains on our waking memory. In somnambulism, which is closely allied to this deep sleep, no memory persists, as a rule, into the waking state. A person who is a somnambulist lives a double life: sleeping, he remembers his sleep [Page 21] experiences and sometimes his waking ones; waking he remembers only his waking life. Occasionally, but comparatively rarely, the golden bridge of memory spans the gulf between the waking and the somnambulic consciousness, dream sometimes interposing as connecting link between the two. It must be remembered that a somnambule, left to himself, will pass into ordinary sleep before awaking, and when this is the case dream may carry on memory of the somnambulic into the waking state.

Du Prel puts very clearly the existence of what he calls the "transcendental consciousness", which has much in common, though it is not identical, with the Theosophical Ego.