C. D. Broad, Mind and Its Place in Nature , 1925



Various Meanings of the Term "Unconscious"

      I will first clear out of the way two not very important senses in which we use the words "conscious" and "unconscious".

      (1) in the first place, we often apply them to distinguish one kind of persistent substance from another kind. We call a stone an unconscious being, and a man or a dog or an oyster a conscious being. By calling a stone an unconscious being I mean that it is incapable of being aware of anything. By calling a man a conscious being I mean that he is capable of being aware of something, even if it should happen that at the present moment he is not aware of anything. So "conscious" and "unconscious", in this sense, mean "capable (or incapable) of being aware of something at some time". I think it would be wise to substitute the words "animate" and "inanimate" for the words "conscious" and "unconscious" when the latter are used in this meaning and with this application .

      (2) We must next notice that the words "conscious" and "unconscious" are often used to distinguish two possible conditions in which an animate being may be at different times. A being which is conscious, in the sense of animate, may from time to time be unconscious in the present sense. A man awake and a man in a deep sleep are both "conscious beings", in the sense of animate beings. But we should say that the former is now "in a conscious condition" and that the latter is now "in an unconsdous condition".

      "Conscious" and "unconscious" in this sense, apply to the temporary conditions of animate beings and to nothing else. We might be tempted to say that an animate being is in a conscious condition provided that it is actually aware of something, and that it is in an unconscious condition provided that it is not actually aware of anything. A little reflection will show that this definition would not be satisfactory as it stands. Many people hold that there is something which is called "unconscious awareness",and that an animate being can be "unconsciously aware" of certain things. Now they would count a man as being in an unconscious condition, even though he were aware of many things, if his awareness of all these things was "unconscious awareness". To meet the possibility of "unconscious awareness" we must say that an animate being is in a conscious condition when it is "consciously aware" of something; and that it is in an unconscious condition when it is either not aware of anything, or, if aware of something, only "unconsciously aware" of it. The amended definitions are now verbally circular. They are not really circular, because a new sense of "conscious" and "unconscious" has turned up. We are in fact defining "conscious" and "unconscious", as applied to the temporary condition of animate beings, in terms of "conscious" and "unconscious", as applied to the process of awareness. But, although the definitions are thus non-circular, they do not tell us much until we know what is meant by "conscious" and "unconscious" as applied to mental events. This is the really important question to which we must now turn.

"Concious" and "Unconscious" as applied to Mental Events.

      When a man talks of "unconscious mental events" or "unconscious experiences" he generally assumes that everyone understands what is denoted by the phrase "conscious mental events or experiences." I do not of course mean that he assumes that every one would agree about the right definition or analysis of either the adjective "conscious" or the substantive "mental event" or "experience". We all know that people differ violently on these subjects. What I do mean is that we assume at the outset that it is easy to give examples to which every one admits that the name "conscious mental events" or "conscious experiences" can be appropriately and quite literally applied. To feel a toothache which is so acute as to make one seriously contemplate going to the dentist is an event which every one would agree could be literally called "an experience" and could be literally called "conscious". Although there are border-line cases which some people would call "conscious experiences" and other people would refuse to call by that name, there seems little doubt that there are thousands of events which every one would agree to deserve the name of "conscious experiences" in a perfectly literal sense.

      Now a literally unconscious experience would be one which differs in a certain respect from these literally conscious experiences and agrees with them in a certain other respect. But, as I have said, the phrase "unconscious experience" is constantly used in a number of senses which are not literal but are highly figurative. By this I mean that a great many events, commonly called "unconscious experiences", are probably not experiences at all in the sense in which every one admits that the feeling of an acute pang of toothache is an experience. And I also mean that a great many events, commonly called "unconscious experiences", are certainly conscious in precisely the sense in which it is admitted that feeling the acute pang of toothache is conscious. My ultimate object is to try to define, or sufficiently describe, literally unconscious mental events. But, before doing this, I want to enumerate and dismiss the various non-literal senses in which the word "unconscious" and the phrase "unconscious experiences" are used. At present the only criterion which we shall be able to employ when we are presented with an alleged case of an unconscious experience is the following: "Is there any reason to think that it resembles admittedly conscious experiences so far as to deserve the name of 'experience'? And is there any reason to suppose that it differs from admittedly conscious experiences so far as not to deserve the name of 'conscious'?"

(1) Traces and Dispositions.

      Far the commonest use of the phrase "unconsmous states" in psychology is in reference to traces and dispositions. It is found that, in order to account for many everyday facts about our ordinary conscious experiences, it is necessary to refer to certain conscious experiences which we had in the remote past. Memory is the most obvious example of such a fact. I remember now something which I saw or heard last year and of which I have not consciously thought in the interval. And of course there are numberless other facts about our present experiences which can be explained only by reference to experiences which we had long ago. We may sum up this whole mass of facts under the name of "Mnemic Phenomena", borrowing this phrase primarily from Mr Russell's Analysis of Mind and ultimately from Semon. Now, either we must assume a wholly new kind of causation, in which one part of the total cause is separated from the rest and from the effect by a considerable gap which contains no relevant events; or we must fill in this temporal gap with some hypothetical persistent entity which we call "traces". I propose to discuss the alternative of "Mnemic Causation", suggested tentatively by Mr Russell, in a later chapter of this section. For the present we will assume the trace theory, as practically all psychologists have done. It is supposed that experiences leave these traces; that the latter persist; and that, when suitable stimuli excite them, they either give rise to new states of mind, such as memories, or else modify states of mind which are in the main due to other causes.

      Along with these traces we must include innate "dispositions". These are assumed in order to explain those differences between the experiences and the behaviour of individuals which cannot be accounted for by differences in their past experiences and present external circumstances. They differ from traces in their origin; for they are supposed to be innate, whilst traces are due to experiences which happened to the individual during his present life. They also differ, in one respect at least, from traces in their consequences. Traces may lead, among other consequences, to memories of the experiences which left the traces. Dispositions cannot do this; for, even if they be ultimately due to experiences, these experiences took place in the minds of our remote ancestors. Apart from these differences, traces and dispositions would seem to be very much alike; and, as both are purely hypothetical and are known only by their effects, there seems to be no harm in lumping them together.

      Now it is common to call traces and dispositions "unconscious states". Many people go further and call them "unconscious mental states" or even "unconscious experiences". They are certainly not conscious, in the sense in which feeling an acute pang of toothache is conscious. And they are no doubt states of something or other. But we have no right whatever to assume that they are "mental states" or "experiences", in the sense in which feeling this pang of toothache is a mental state or an experience. The fact is that we know nothing whatever about the intrinsic nature of traces and dispositions; they are simply the hypothetical cause of certain observable effects, and the hypothetical effects of certain observable causes. True, these observable causes and effects are experiences; but this is no ground for supposing that the traces themselves are of the nature of experiences. This is disguised by the silly metaphor that past experiences are "stored up in the unconscious". I may have had a certain conscious experience which lasted for five minutes and ceased twenty years ago. If we say that this is "stored up in the unconscious", and mean this statement to be taken literally, we must be understood to assert that this same experience has been going on steadily for the last twenty years. Perhaps the original experience was seeing a certain dog for five minutes twenty years ago; if this experience be literally "stored up in the unconscious", I have been literally seeing the same dog in the same situation ever since, though "unconsciously", in spite of the fact that the dog has been dead and buried for the last fifteen years. Of course it will be said that no one does mean to assert anything of this kind when he talks of experiences "persisting in the unconscious". It is quite true that most people hasten to disclaim such preposterous consequences when once they are pointed out. But I think there is no doubt that many people do hold views which, if they could be induced to state them clearly, would be found to lead to these consequences. For instance, Rivers in his Instinct and the Unconscious asserts that the content of the Unconscious is "suppressed experiences", and he gives as an example of such an experience a fright which one of his patients had had many years before with a dog in a passage. Of course, if anything literally persists, it is not the experience itself but the trace of the experience. And there is no more positive reason to suppose that the trace of an experience resembles it or any other experience than to suppose that persistent deafness resembles the attack of scarlet-fever which left it in the patient.

      The plain fact is that we know nothing with certainty about the intrinsic nature of traces, and we ought therefore studiously to avoid all phrases which suggest some particular view of their intrinsic nature. I propose to call traces and dispositions by the innocent name of "mnemic persistents". The reason for calling them "mnemic" is obvious. Our ordinary states of mind may be called "transients"; for they happen from time to time, last for a little while, and then cease. In contrast with these we can call traces and dispositions "persistents"; because they are supposed to last for a long time, and to fill the gaps between our transient states of mind. (I avoid Mr Johnson's terms "occurrents" and "continuants", because they have certain implications which I do not at present wish to assert or deny of traces; and it is a pity to spoil two valuable technical terms by using them loosely in senses which their inventor might not admit.) The phrase "mnemic persistents" has the twin advantage that it does express all that we know about traces and dispositions, and that it does not tacitly imply anything that we do not know about them.

(2) Inaccessible Experiences.

      There is another important non-literal sense in which the phrase "unconscious experiences" has been used. To explain it I will take an example from that excellent book Instinct and the Unsonscious by the late Dr Rivers.

      Rivers quotes the case of a patient who had suffered from claustrophobia for many years. By analysing the patient's dreams Rivers was able to show that the claustrophobia had been started by a terrifying experience which the man had had as a small boy in a narrow passage with a fierce dog. This experience the patient was quite unable to remember by normal means. Now Rivers gives this as a typical example of an unconscious experience; and practically defines "The Unconscious", for his own purposes, as consisting of such experiences. It is clear that this is an entirely new meaning of the phrase "unconscious experience". When the experience originally happened it was in all probability an ordinary conscious experience owned by the patient. There is no reason whatever to suppose that, at the time, the boy was unaware of seeing the dog or of feeling frightened, or at any rate that he could not have been aware of them if he had chosen to introspect at the time. In this the experience contrasts strongly with the case of Livingstone and the lion, which is also quoted by Rivers. Livingstone noticed at the time that he was not aware of any pain or fear while in the jaws of the lion; and the circumstances were such that, if he had been feeling pain or fear, he could hardly have failed to notice the fact. Here we may conclude, either that there was no experience of pain or fear connected with this situation; or that, if there were, it was not a conscious experience of the mind known as "Livingstone".

      The case of Rivers' patient is quite different. To say that his experience is unconscious means only that he cannot now remember it by normal means; and it does not mean that it was not an ordinary conscious experience which belonged to the boy at the time when it happened. It seems to me to be misleading in the highest degree to use the phrase "unconscious experience" in these two utterly different senses. Rivers would no doubt say that the experience "was conscious" when it happened, and that it "became unconscious" afterwards. This, however, does not alter the fact that the words "conscious" and "unconscious" are being used in two senses which are quite disconnected with each other. In the first sense an experience either is conscious or it is not; and, if it is one, it can never become the other. In the second sense one and the same experience may sometimes be conscious and at other times unconscious. For there might be times when a person could remember it normally, and other times when he could be got to remember it only by technical methods, if at all.

      The situation which Rivers is describing is a real and an important one; but the terminology which he uses to describe it is hopeless. I shall substitute for the words "conscious" and "unconscious", when used in this sense, the words "accessible" and "inaccessible" respectively.

accessible inaccessible

An experience is accessible when it can be remembered by normal means. It is inaccessible when it can be remembered only, if at all, by special technical methods. One and the same experience may be accessible at some times and inaccessible at others. Also there will probably be degrees of accessibility. Even when an experience can eventually be remembered by normal means it is sometimes harder and sometimes easier to do this. And I suppose that, when technical methods have to be applied, they sometimes succeed easily and sometimes only with difficulty.

      Corresponding to this distinction between accessible and inaccessible experiences there will be a distinction between mnemic persistents. Some of these can never by any means be made to give rise to memories of the experiences which originated them. If innate dispositions originated, as some think, in the experiences of our remote ancestors, they fall into this class. And probably some traces fall into it too. Other mnemic persistents will give rise to memories if special technical methods be applied, but not otherwise. And a third class give rise to memories without needing the application of special technical methods. Probably there is no sharp line between the second and third classes.

      The work of the psycho-analysts enables us to state one at least of the causes which tend to make certain experiences inaccessible if the memory of a past experience would be specially painful or shocking to the present self there is a tendency for this experience to become inaccessible. It is sometimes said that the painfulness or shockingness of the original experience is the operative factor; but I think that this is true only in a derivative way. The essential factor is the emotional effect which the memory of the experience would have if it arose now. The memory of many experiences which were quite enjoyable when they happened might be shocking or painful to tbe present self. Such experiences will tend to become inaccessible in spite of their originally pleasant character. Again, the memories of some experiences which were painful or shocking when they happened might be neutral, or even pleasant and amusing, to my present self. I see no reason to think that such experiences would be specially likely to become inaccessible. All that we can say is that, in a good many cases, the memory of an experience which was painful or shocking when it happened is likely to be itself painful or shocking now. So far, and only so far, as this is true painful or shocking experiences will tend to become inaccessible.

(3) Ignored, Misdescribed, or Dislocated Desires and Emotions.

      There is another non-literal sense of "unconscious experiences", which applies specially to desires and emotions. It is rather closely connected with the sense which we have just been discussing, but it must be distinguished from this. There is no doubt that we have a general undiscriminating awareness of many of our experiences without introspectively analysing and discriminating them. Introspective analysis and discrimination involve a special act of attention which we can make or not as we like. And, if we choose to make it at all, we may take more or less trouble over it and can perform it more or less thoroughly. Even if we choose to make the attempt, and perform the discrimination and analysis to the best of our ability, we can make mistakes about the right analysis of our experiences, just as we can make mistakes in trying to analyse and describe external objects whicb are presented together in a confused jumble in our field of view. Introspective discrimination is a difficult, tiresome and unwonted process; and no one who is not used to it is likely to avoid mistakes.

      Now there are two classes of experience about which we are specially and systematically liable to make mistakes; and these mistakes may have several different forms. The two classes in question are desires and emotions. Desires and emotions are the experiences par excellence about which we pass judgments of praise or blame on ourselves and others. If we find that we have certain desires and emotions we are obliged to think badly of ourselves; and, if we confess such desires and emotion to others, they will think badly of us. We thus have a strong tendency not to discriminate these desires and emotions; or, if we do discriminate them, to misdescribe them to ourselves; or if we discriminate them and describe them rightly to ourselves, to refuse to acknowledge them to others.

      Now, in the case of emotions, we can go wrong either about the mental attitude itself or about its epistemological object. There is perhaps hardly any emotional attitude which is regarded as intrinsically bad; i.e., as bad, no matter what kind of epistemological object it may be directed to. The rule seems to be that the same emotional attitude is good when directed on to one kind of epistemological object and bad when directed on to an object of another kind. Conversely, of two emotional attitudes which may be directed on to the same epistemological object, one may be good and the other bad. In fact we apply ethical predicates to the whole situation composed of such and such an attitude directed to such and such an epistemological object, and not to the attitude taken in abstraction. It is, e.g., considered virtuous to hate sin, but wicked to hate even sinful people. And it is considered virtuous to feel emulation towards one's rivals, but wicked to feel envy towards them. There are thus three methods of saving one's self-respect when one feels a certain emotion towards a certain object and believes that this kind of emotion ought not to be felt towards this kind of object. One method is to ignore the existence of the emotion altogether; i.e., to refuse to turn our introspective attention in this dangerous direction. A second method is to discriminate the emotional attitude properly, but to substitute for its actual object another pretended object of such a kind that it would be respectable to take up this emotional attitude towards this object. E.g., I may really hate Germans or capitalists, and may recognise that I am feeling the emotion of hatred. But I may persuade myself, and try to persuade others, that what I hate is not Germans or capitalists as such but is the supposed special wickedness of these classes. In order to do this I shall very often have to make up a myth about them, and refuse to contemplate any of the perfectly obvious facts which show that Germans or capitalists are neither much better nor much worse than Englishmen or trades unionists. A third method is to make no mistake about the object of my emotion, and to recognise that I am feeling an emotion towards this object; but to substitute for the emotion which I actually feel, and which I believe that it is not respectable to feel towards that kind of object, another pretended emotional attitude which I believe it would be respectable to feel towards this object. I may recognise, e.g., that I feel a certain emotion towards the success of a fellow philosopher's book; and I may pretend to myself and others that this is the respectable emotion of healthy rivalry when it is really the disreputable emotion of disappointed envy. This third method is easiest when the real and the pretended emotion do resemble each other or contain certain common constituents, as envy and rivalry do. Of course the second and third methods may be, and often are, combined, with the happiest results. The two emotions of malice and of righteous indignation are different; but they certainly contain common factors, for both involve satisfaction at the thought of another's pain. And their appropriate objects are different, but have something in common. If now I actually feel malice towards Smith, I can easily keep my self-respect and the respect of others by persuading myself and them that I am feeling an exalted kind of satisfaction at the thought of Smith's moral purification through suffering. One of the reasons for the extreme popularity of war with childless women and others who are in no immediate personal or family danger is that it renders such substitutions easy, and enables quite ordinary people to go about swelling with pretensions to moral superiority which would be exploded at once in a more normal atmosphere.

      The case of desires is, in one way, simpler than that of emotions. There do not seem to be intrinsically different kinds of conative attitude, as there are intrinsically different kinds of emotional attitude, such as fearing and hating. So far as I can see, desires differ from each other only in their intensity and in their epistemologic objects; and the goodness or badness of a desire depends almost wholly on the nature of its object. (It no doubt depends partly also on the intensity of the desire. It would be considered that a very intense desire for knowledge is good, and that a moderate desire for bodily pleasure is good; but a very intense desire for bodily pleasure would be regarded as bad by many people.) If I entertain a desire for some object which it is considered wrong to desire, there are two courses open to me in order to keep my present high opinion of my moral character and to confirm other people in their high opinion of it. One is to ignore the existence of the desire altogether. Another is to recognise the existence of the desire, but to pretend to myself and to others that it is for some object which it is considered respectable to desire. As our motives are nearly always mixed, this process is childishly simple. It is only necessary to emphasise that part of the desired object which it is considered respectable to want, and to slur over that part of it which it is considered disreputable to want. It is needless to give examples of a process which we are all doing continually.

      Such emotions and desires as we have been considering are often given as examples of unconscious experiences. It seems to me that they are quite literally conscious. They are in fact quite ordinary desires and emotions about whose existence, nature, and objects we need make no mistake if we introspect honestly and carefully enough. But, as a matter of fact, we do not do this. We ignore them altogether; or we "dislocate" them, i.e., ascribe to them a different object from that which they really have; or we misdescribe them, i.e., put them into a certain class of mental attitudes when we ought to put them into a certain other class. If there be anything literally unconscious in the whole business, it is not the desire or the emotion itself, but the process of ignoring, dislocating, or misdescribing it. We must therefore consider this process in rather more detail.

      If I am going to ignore, misdescribe, or dislocate a certain desire or emotion which I own, I must in some sense know that it is there and that there is a reason for treating it in this way. Now we have simultaneous undiscriminating awareness of many experiences which we do not attentively and deliberately introspect. I suggest that this kind of knowledge suffices to warn us that the ice is thin in certain places, and that we had better not turn our introspective attention in these particular directions. The question might then be raised: "How far is this aversion of discriminating introspection from certain desires and emotions a deliberate and conscious process?" In answer to this I think that the following considerations are important. (a) If we have a conscious desire to ignore certain experiences, because we think that they would turn out to be unflattering to our self-respect, this desire is itself an experience which we shall tend to ignore. For it is not flattering to our self-respect to have to acknowledge that we can keep it only by averting our attention from certain of our desires and emotions. It follows that, even if we deliberately and consciously ignore certain desires and emotions, we shall almost certainly refuse to acknowledge this fact to ourselves, and still more so to others. Thus I think that the aversion of our discriminative introspection from certain of our experiences is much oftener a deliberate and literally conscious process than is commonly admitted. I believe that we generally know when we are doing this, and that the extreme "touchiness" which we are liable to display when taxed with it is a sign that we do.

      (b) An aversion of introspective attention, which begins by being deliberate, will quickly become habitual. An analogy will make this plain. If I have a tender tooth I shall at first deliberately try to avoid biting on it, and shall sometimes make mistakes and hurt myself. But very soon I shall automatically avoid biting on it. Now emotions and desires tend to recur; and, if I at first deliberately avert my attention from some of them, I shall very soon come to do so habitually. This habit, like any other, may eventually become so strong that it cannot be overcome by deliberate volition.

      (c) A method which we very commonly use is to put a ring-fence round a certain region, to label it as dangerous, and to avert our attention from the whole of it. All patriots do this with the whole subject of the virtues of their enemies and the faults of their fellow-countrymen; many scientists put such a fence round all the subjects which are investigated by Psychical Researchers; and the minds of most clergymen appear to be full of regions guarded with barbed wire and a notice that "Trespassers will be Prosecuted". Once this has been done it becomes perfectly easy to assert with complete good faith that we are not deliberately turning our attention away from any assigned desire or emotion which falls within such a region. We can truthfully say that we never thought for a moment of this particular experience, and therefore cannot have deliberately ignored it; just as a thief might truly say that he had never touched a certain necklace if he had merely pocketed the case which in fact contains it.

      Now I think it is certain that what are called "unconscious" desires and emotions are often simply desires and emotions which we habitually ignore, misdescribe, or dislocate. An experience which is "unconscious" only in this sense is not literally an unconscious experience. And the process of ignoring, misdescribing, or dislocating it is not literally an unconscious mental process. Sometimes it is a conscious and deliberate process which is itself ignored or misdescribed. Sometimes it is habitual. In the first case it is not literally unconscious; and in the second there is no positive reason for thinking that it is literally an experience or series of experiences.

      Ignored experiences cannot be identified with inaccessible experiences. Many experiences which have become inaccessible were not ignored when they happened; and many which were ignored when they happened have not become inaccessible. Nevertheless, there probably is a close connexion between ignored and inaccessible experiences. Experiences which it would be painful or shocking to discriminate are generally those which it would be painful or shocking to remember; and these, as we know, tend to become inaccessible. Moreover, the mere fact that an experience is habitually ignored probably tends to make its trace less definite and more isolated, and therefore to increase the difficulty of remembering it by normal means.

      I have discussed this subject mainly in connexion with the ignoring of experiences. But exactly the same remarks apply to misdescribed and dislocated experiences in themselves these experiences, though often called "unconscious", are literally conscious. And the process of misdescribing them or dislocating them is either a deliberate process which we choose to ignore or misdescribe, or it is an habitual process which is not literally an experience at all.

(4) Unrerognised Needs.

      There is another sense in which the phrase "unconscious desires" has been used, in which it does not denote a literally unconscious experience. This has been brought out very clearly by Mr Russell in his Analysis of Mind; though I do not agree with his apparent opinion that it covers all that is meant by the phrase "unconscious desire", and I agree still less with the arguments and the conclusions which he bases on it. When we have the experience of desiring something we present to ourselves in imagination some possible future state of affairs to which we take up the conative attitude. And it is an essential part of this attitude that we believe that this state of affairs, if realised, will satisfy us and bring the conative situation to an end. Now, of course, what I now believe would satisfy me may be extremely different both in outline and in detail from what would really satisfy me. I have no infallible revelation about what state of affairs will bring a certain kind of uneasiness to rest. I cannot learn about this by introspection, however careful and thorough; for this will tell me only about the elements and the structure of my present conative situation. The recorded experiences of others may provide me with the basis for a more or less probable inference on the subject; but, in the main, the only available policy is to "wait and see."

      Now sometimes it is said that what I "really desire" is what would in fact satisfy my present conation. With this terminology it is certain that I am often not conscious of what I really desire. And this fact is expressed by saying that I have an "unconscious desire" for what would in fact satisfy me. I think that this is a most unfortunate and misleading terminology. It is much better to begin by distinguishing between what I "desire" or "want" and what I "need". I may set before myself the idea of a large fortune, and spend most of my life trying to gain it. If so, it is preposterous to say that I only think that I desire money; I really do desire it. I have a conative attitude, and the epistemological object to which it is directed is my future wealth. It is true to say that I desire money in precisely the same sense in which it is true to say that the drunkard sees pink rats; and to deny this is to confuse an epistemological object with an ontological object. Now I may find that, when I have made a great deal of money, the same kind of dissatisfaction still persists. And it may be true that this dissatisfaction would in fact have been removed if I had acquired fame instead of money. If so, I needed fame. But it is preposterous to say that I desired fame, if I never put the idea of fame before myself, or felt any attraction for it, or strove after it. To say that I "unconsciously desired" fame, is like saying that the drunkard "unconsciously saw" the alcohol in his stomach.

      What is true then is that needs often give rise to desires, and that the desire which is caused by a certain need may have an epistemological object which fails to agree with the ontological object which would satisfy that need. But needs are not desires, nor are they experiences at all; hence a need of which I am unaware cannot properly be called an unconscious desire, or an unconscious experience of any kind. Still, there is no doubt that one of the meanings which is given to the phrase "unconscious desires" is "needs of which a person is unaware". I shall call "unconscious desires", in this sense, by the much less misleading name of "unrecognised needs".

      I have now pointed out four non-literal senses in which psychologists use the phrase "unconscious experiences" or at any rate "unconscious mental states".

  1. In the sense of traces and dispositions, they seem to have no claim to be called "experiences", and no obvious claim to be called "mental" unless it can be shown that they cannot be simply modifications of the brain and nervous system.
  2. In the sense of inaccessible experiences, "unconscious mental states" were literally experiences when they happened. But they were also literally conscious; and no subsequent facts about memory or the lack of it can make them literally unconscious.
  3. "Unconscious desires and emotions", in the sense of desires and emotions which we ignore, misdescribe, or dislocate, are certainly experiences. But they are literally conscious.
  4. "Unconscious desires", in the sense of unrecognised needs, are, so far as one can see, not experiences at all. And the desires for objects which will not in fact satisfy us, which are often caused by unrecognised needs, are ordinary conscious experlences.

      It now remains to try to see what is meant by literally conscious and literally unconscious experiences or mental states. The existence of "unconscious mental states", in the four non-literal senses which we have enumerated, is so obvious that people would not have thought of quarrelling about unconscious mental states unless they had had some other and more literal meaning of this phrase at the back of their minds.

Literally Unconscious Mental Events.

      I will first try to point out what seem to me to be the characteristic marks of a conscious mental state, and I will then describe a literally unconscious mental state as one which lacks some of these marks.

      In order that a mental state of mine may be conscious it is certainly not necessary that I should be "conscious of it" when it happens, in the sense of making it the object of an act of introspective attention. I have no doubt that I have been seeing the words of this page as I wrote them down; and I am sure that my perceptions of the words have been instances of what every one would call "conscious experiences". But I most certainly did not make these perceptual states into objects of introspective attention while they were happening. My attention was taken up with my argument and with the words themselves, and I was not attending at all to the process of seeing the words. No doubt all processes which I introspectively discriminate are conscious, but the converse of this is certainly not true. A conscious experience of mine cannot therefore be defined as an experience of mine of which I was conscious at the time when it happened, if by "being conscious of" you mean "making an object of introspective attention".

      Nevertheless, it might be possible to mark off conscious experiences from all other mental events by means of some more hypothetical references to introspective discrimination. Might we not say that every conscious experience of mine is one that I should have succeeded in discriminating if I had introspected carefully enough while it was happening or immediately afterwards? I think that we are inclined to believe this about all mental events which we should be prepared to count as "conscious experiences" of ours, and that we are not inclined to believe it about anything else. It is therefore plausible to take it as a sufficient description, if not as a definition, of a "conscious experience" of mine.

      But, even as a description, it needs some further elaboration. There are certain experiences which probably could have been introspectively discriminated while they were happening, and which would yet be called "unconscious", in a sense which does not fall under any of the four non-literal headings already mentioned. Take dreams, for instance. From one point of view all dreams would be called "unconscious experience". Yet, from another point of view, to see my friend in a dream is as much a "conscious experience" as to see him in waking life. It is certain that many dream-experiences could have been introspected by the dreamer while they were happening; for I have quite often introspected my dream-experiences while dreaming, and I do not suppose that this is at all exceptional in people who are given to introspection. Another example is provided by alleged cases of co-consciousness. Sally Beauchamp, in Dr Morton Prince's Dissociation of a Personality, claimed to be aware of most of the things of which B1 was aware when the latter was in control of the body and the former was not. From Sally's point of view these states of awareness were just as "conscious" as the contemporary states of awareness in B1; in fact this is precisely what the claim to co-consciousness, as distinct from alternating consciousness, amounts to. It seems to me that the only way to deal with such cases as these is to introduce a distinction between "relatively" and "absolutely" unconscious mental events. We shall then have to distinguish relatively unconscious experiences from strictly conscious experiences by considering who precisely could have introspected them when they were happening. We call the vivid dream of a normal man and the alleged co-conscious experiences of Sally "conscious", because there was some mind, viz., my sleeping self in the one case and Sally in the other, which could have introspectively discriminated them if it had tried at the time when they were happening. We call the same experiences "unconscious" simply because the only mind which could have introspectively discriminated them at the time when they were happening was a mind which was not then in control of the body concerned in the experience. Such experiences as these I shall call "relatively unconscious". An "absolutely unconscious" mental event would be one that could not have been introspectively discriminated at the time of its occurrence by any mind, whether in control at the time of a body or not.

      I am well aware that even these amended descriptions are open to serious objections. What do we mean by "controlling a body"? And when is a certain mind in control of a certain body, and when is it not? If you press me with these questions, I doubt whether I can give a perfectly satisfactory answer to them. Still, it seems to me that the kind of fact which I refer to under the name of "control" is pretty obvious. There is an important sense in which the mind which is known as "C.D.B." is at present in control of my body, and in which it will not be in control of my body later in the evening when I am in bed and asleep. At that time no mind will, in this sense, be "in control of" it; though it will, no doubt, still be behaving in a somewhat different way from that in which it would behave if it were no longer animated by a mind at all. In the case of alternating personality a recognisably different mind is at different times "in control of" the same body, even if there be reason to suppose that all these minds are in a sense parts of a single "mind" which continues to animate this body throughout life.

      There is unfortunately one other highly debatable conception which I must introduce before giving a description of literally unconscious mental events which can make any claim to be satisfactory. This is the notion of the "ownership" of a mental event by a mind. To own an experience is evidently not the same as to discriminate it introspectively. It is commonly believed that only mental events which are owned by some mind can be introspectively discriminated, and that the only mind which can introspectively discrimhlate an experience is the mind which owns it. If this be true, both absolutely conscious and relatively unconscious mental events must be owned by minds. But theoretically there would be two quite different kinds of absolutely unconscious mental events. The first would be owned by a mind which, for some reason, could not have introspectively discriminated them when they happened even if it had tried. The second would have been incapable of being introspectively discriminated by any mind simply because they were not owned by any mind. I have so far drawn no distinction between "experiences" and "mental events". This seems to be a convenient place to do so. All experiences are mental events; but I think it would be in accordance with usage to say that, if there be unowned mental events, they should not be called "experiences". So I will define an "experience" as a mental event which is owned by some mind. Thus absolutely unconscious mental events will divide in theory into (a) absolutely unconscious experiences; i.e., mental events which are owned, but could not have been introspectively discriminated when they happened; and (b) unowned mental events.

      Of course many people would deny off-hand the possibility of unowned mental events. They may be right. On the other hand, they may be defining "mental event" in some way which includes ownership by a mind as part of the definition. In that case their denial is merely an analytic proposition. Again, anyone who holds that a mind does not require a Pure Ego, but consists simply of a mass of suitably interconnected mental events, can hardly deny the possibility of masses of mental events so poor in content and so loosely interconnected as not to deserve the name of "minds". If so, he can hardly deny the possibility that some mental events are not owned by minds, even if he denies the possibility of completely isolated mental events. For these reasons it is wise to introduce the class of "unowned mental events", even though it may turn out to be a mere blank window.

      I must now state more clearly what I mean by "ownership". I do not think that it is definable, in the sense in which I am using it. But it is a highly ambiguous word; and, by pointing out the senses in which I am not using it, I may be able to indicate to the reader what I want him to think about. (1) In the very widest sense we call a mental event "an experience of Smith's" if it is specially connected with the stimulation of Smith's body. And by "Smith's body" we mean the body which is normally controlled by a certain recognisable mind known as "Smith". In this sense a mental event might be called "an experience of Smith's" even if it were not owned by his mind or by any other mind. (2) In a slightly narrower sense a mind M might be said to be still owning a mental event if this has left some trace which still affects M's conscious experiences from time to time. In this sense Rivers' claustrophobic patient was still "owning" the experience with the dog in the passage, both before and after Rivers had cured him by enabling him to remember the incident. (3) In a still narrower sense of "ownership" we should say that a mind still owns those and only those experiences which it can remember at will. In this sense Rivers' patient had ceased to "own" the dog-experience for many years, and began to "own" it again only after Rivers had cured him. We have two senses of "ownership" in connexion with personal property, which correspond to the last two senses mentioned above. In the wider sense of "ownership" I am still the owner of a certain umbrella even after it has been lost or stolen. It remains my umbrella, though it now rests permanently in your hat-stand. In the narrower sense I own it only so long as I can lay hands on it at will. Experiences which are owned in senses (2) or (3) may be said to be "mnemically owned"; because, strictly speaking, their continued ownership by me means only that they continue to affect my conscious experiences. The two senses may be distinguished from each other by calling the former "mnemic ownership de jure" and the latter "mnemic ownership de facto et de jure".

      Now I am not using "ownership" in any of these three senses in my attempts to give sufficient descriptions of literally conscious and unconscious mental events. It is in fact evident that there is another sense of "ownership", which is not mnemic. When I look out of the window and see a man passing, or when I feel a twinge of toothache, these experiences are owned by me in a fundamental and probably indefinable sense. This is a sense in which the visual experience ceases to be owned by me as soon as the man passes out of sight, and the other experience ceases as soon as I cease I to feel my tooth aching. They may still continue to be owned by me both de jure and de facto. I may from time to time remember seeing the man or feeling the pain. And, even if I cannot do this, these experiences may continue to modify my later experiences from time to time in some way or other. Let us call this third, non-mnemic, sense of "ownership" by the name of "literal ownership". In my attempts to give sufficient descriptions of literally conscious and unconscious mental events it must be understood that, wherever I use the word "ownership", I mean "literal ownership". It is probable that many mental events which have been literally owned are not mnemically owned de facto for more than a negligibly short time; and it is probable that many of them are not mnemically owned de jure for long. And it is quite possible that mental events which have never been literally owned by me may be mnemically owned by me de jure if not de facto.

      I will now repeat the descriptions which we have reached. These must be interpreted in the light of the remarks which I have just been making.

  1. Mental events are either owned or unowned. All unowned events are incapable of being introspectively discriminated by any mind, and are therefore absolutely unconscious.
  2. Events which are mental and are owned by some mind are called "experiences". Mental events which are unowned are not to be called "experiences", and therefore not "unconscious expenences".
  3. Mental events which are owned may be such that the mind which owns them could, or such that it could not, have introspectively discriminated them at the time of their occurrence if it had tried. In the latter case they are absolutely unconscious experiences of that mind.
  4. Mental events which are owned and could have been introspected may be owned by a mind which is at the time of their occurrence in control of a body, or by a mind which is not in control of a body at that time. In the former case they are absolutely conscious experiences . In the latter case they are relatively unconscious experiences , i.e. , they are conscious experiences of what would commonly be called an "unconscious mind".
These results are exhibited synoptically in the table which follows:

Mental Events Owned
Introspectible By controlling mind (Absolutely Conscious)
By non-controlling mind (Relatively Unconscious)
Non-introspectible Absolutely Unconscious

      The next point to notice is that I do not pretend to have given either definitions of literally conscious or unconscious mental events, or tests for them. At most I claim to have given descriptions which suffice to distinguish the two in theory. I will now say somethmg more about this first point. I think that almost everyone would be inclined to say: "It may be that all my conscious experiences are mental events which I owned and which I could have introspectively discriminated at the time; and it may be that there is nothing else of which this is true. But this purely hypothetical proposition about what would have happened if I had introspected cannot be ultimate. There must be some intrinsic difference between those experiences of mine which I could have introspected and those which I could not have introspectively discriminated even if I had tried my hardest. And this intrinsic difference, whatever it may be, is what we mean by the difference between a conscious and an unconscious experience."

      This may very well be true. One can think of at least three circumstances which might tend to make it impossible to discriminate an experience introspectively. (1) The difficulty might arise through the experience being an event all of whose characteristics have an extremely weak intensive magnitude. (2) The characteristics of the experience might be reasonably intense, but it might be part of a larger mass of experiences which were extremely like it both qualitatively and quantitatively. (3) The experience might have considerable intensity, and might differ from other co-existing experiences both qualitatively and quantitatively to a marked extent; but it might stand in certain special relations to other contents of the mind which prevent it from being introspectively discriminated. This third possibility would seem to split into two. (a) it might take a merely negative form. It might be that this experience is relatively isolated, and stands in but few relations to the other contents of the mind. The limiting case of this arises when a mental event is not owned by a mind at all, in the literal sense of "ownership". (b) It might take a positive form. There might be some positive relation between this experience and the rest of the mind which positively averts introspective attention from the former. And this peculiar relation might depend on some intrinsic quality in the experience. E.g., the experience may be such that, whenever it begins to be introspected, an intolerably painful feeling begins to arise in the mind which owns it. It seems likely that all these various possibilities are realised in practice in various cases of literally unconscious experiences.

"Simultaneous Undiscriminated Awareness" and "Ownership".

      I will end this section by raising a ratherdifficult question which is very closely connected with what we have just been discussing. In our descriptions of literally conscious and unconscious mental events we have used two obviously different relations which may hold between a mind and a mental event, viz., literal ownership and introspective discrimination. The latter seems to imply the former, but the former can evidently hold without the latter. Now there seems to be a third possible relation between a mind and a mental event, which might be called "Simultaneous Undiscriminating Awareness". I want to say something about this, and to consider how it is related to literal ownership.

      Let us consider the case of looking for one's spectacles in a certain drawer, and failing to find them though they were staring one in the face all the time. If I were asked whether I was at the time aware of seeing the drawer and most of its contents, I should answer "Yes", in one sense, and "No", in another. Certainly I was aware of seeing the drawer and most of its contents in a sense in which I was not aware of seeing the spectacles. On the olher hand, I was almost certainly not introspectively discriminating the process of seeing the drawer; for my whole attention was devoted at the time to the drawer itself and its contents, and not to my own mental states. It is evident that, in the vast majority of cases of conscious perception, I am not aware of my perception, in the sense of introspectively discriminating it. Nevertheless, I should certainly refuse to entertain the suggestion that I am not aware, in any sense, of my conscious perceptions while they are taking place. I shall say then that the person in our example was aware of his act of seeing the drawer and most of its contents, in the sense that he had "simultaneous undiscriminating awareness" of this mental event. It is also true that he "literally owned " this mental event; it was literally a part of his mental history.

      It would seem then that literally conscious experiences are always literally owned by a mind, and that the mind which owns them has always simultaneous undiscriminating awareness of them. But, in ninety-nine cases out of a hundred, it does not also introspectively discriminate them. The question now arises: "Do we ever literally own mental events of which we do not have at least simultaneous undiscriminating awareness?"

      It will first be necessary to modify the question in order to remove the danger of an infinite regress. If "literal ownership" and "simultaneous undiscriminating awareness" be just two different names for a single relation, it will of course follow that I must have simultaneous undiscriminating awareness of any mental event which I own. And there will be no infinite regress in this assertion. But, suppose that the two names stand for different relations. Suppose I own a certain state s, and that I have simultaneous undiscriminating awareness of s. This awareness of s will also be a state which is owned by me. And, if I must have simultaneous undiscriminating awareness of every mental event that I own, I must have it of this state too. We should thus be launched on an infinite regress of awarenesses of awarenesses of awarenesses of . . . . So, unless we take "literal ownership" and "simultaneous undiscriminating awareness" to be simply two names for the same relation, or else modify the proposition which we are investigating, we can be sure that this proposition involves an infinite multiplication of states of mind. I do not maintain that such an infinite series of mental states belonging to a single mind involves any contradiction. But there is not the faintest reason to believe that it is a fact; and it certainly would be undesirable to accept a proposition which has this implication, unless there were the strongest grounds for doing so.

      It is of course quite easy to modify the proposition which we are considering, so that it shall not entail the existence of an infinite series of contemporary mental states belonging to the same mind. We can begin by distinguishing "orders" of mental events. We might call my awareness of the drawer a state of the first order; my simultaneous undiscriminating awareness of my awareness of the drawer a state of the second order; and so on. And we might put the proposition into the milder form that I have simultaneous undiscriminating awareness of every mental event which I own, provided that its order does not exceed some finite number n. The state of n+1th order, which is my awareness of a state of nth order, would then be owned by me but would not be an object of simultaneous undiscriminating awareness to me. In particular it might be that I necessarily have simultaneous undiscriminating awareness of all mental events of the first order which I own; but that I do not necessarily have simultaneous undiscriminating awareness of any of my experiences whose order is greater than one. Stated in this form the proposition is intrinsically unobjectionable, whether it be in fact true or not.

      We have now to consider whether there is any reason to believe it. For this purpose we had better return to our example about the spectacles. It is plausible to hold that I have simultaneous undiscriminating awareness of all my conscious experiences of the first order. The question is whether there may not be unconscious experiences of the first order which I own but of which I do not have simultaneous undiscriminating awareness. We must notice that I should not normally use the words "conscious" and "unconscious" at all in describing my experience with the drawer and the spectacles. I should simply say: "I saw the drawer and most of its contents, but I did not see the spectacles." The adjectives "conscious" and "unconscious" are added later, as a result of reflection and inference. I find that the spectacles must have been physically affecting my retina just as much as the drawer and the rest of its contents did. I then perhaps persuade myself that I must have seen the spectacles. And I express the obvious difference between the way in which I must have seen the spectacles, if I saw them at all, and the way in which I certainly did see the drawer and the rest of its contents, by saying that I saw the drawer "consciously" and that I must have seen the spectacles "unconsciously", if at all. Now this phraseology does suggest the possibility of first-order experiences which are owned by me, but of which I am not aware even in the sense of simultaneous undiscriminating awareness. When I say: "I saw the spectacles unconsciously" or "My seeing of them was unconscious", I imply that this experience was owned by me. And, when I say that it was unconscious, I do imply that I was not aware of it even in the sense in which I was aware of seeing the drawer and the rest of its contents. That is, I do imply that I did not have even simultaneous undiscriminating awareness of seeing the spectacles.

      But it seems to me very doubtful whether we have any right to accept the verbal implications of this phraseology. The natural thing for me to say is simply: "I did not see the spectacles". And the plain, straightforward meaning of this is that either there was no mental event at all called "seeing the spectacles", or that, if there were, it was not literally owned by me. Now it does not seem to me that the facts which are taken into consideration on later reflection give us any ground for reversing this view, even if they do give us some ground for accepting the view that a mental event called "seeing the spectacles" did exist at the time. The facts which are adduced in favour of the view that a mental event, called "seeing the spectacles", must have existed at the time fall into two main groups. (i) It is argued that the spectacles and my retina were in such relative positions that light trom the former must have affected the latter in a way which might reasonably have been expected to produce such a mental event. (ii) It may be that in dreams, or by hypnosis or psycho-analysis or some other technical method, I come to have experiences or to do or say things which are hard to explain except on the assumption that a certain mental event existed in the past and that it is affecting my present experiences or actions. Even if we admit that such arguments make it probable that a mental event of "seeing the spectacles" existed while I was searching in the drawer, there seems no reason to believe that they make it probable that this mental event was literally owned by me. No doubt, if it existed at all, its occurrence depended on the stimulation of my body. It is also true that it is a mental event which afterwards affects experiences in my mind. But this does not suffice to prove that, when it happened, it was my experience, in the plain straightforward sense in which the experience of seeing the drawer and the rest of its contents was an experience of mine.

      Now, if this be granted, there would seem to be no very good ground for distinguishing between the first-order mental states which I own and the first-order mental states of which I have simultaneous undiscriminating awareness. The only ground for distinguishing between the two was that certain common phrases do seem to suggest that there are mental events which I literally own but of which I do not have even simultaneous undiscriminating awareness. But we have now seen that, even if there be mental events which arise through the stimulation of my body and subsequently affect my experiences, and of which I have not simultaneous undiscriminating awareness, there is no good reason to think that they are literally owned by me. Hence I think that it is quite likely that all first-order experiences which I literally own are also experiences of which I have at least simultaneous undiscriminating awareness; and that all mental events of which I have simultaneous undiscriminating awareness are literally owned by me. This of course leaves it quite possible that literal ownership and simultaneous undiscriminating awareness are different relations; just as size and shape are different qualities, though anything which has either must have both. And I think it is pretty certain that they are different relations, for the following reason. If they were just two names for a single relation it would be quite certain that every mental event which I own, no matter what its order might be, would be an object of simultaneous undiscriminating awareness to me. This would be an identical proposition. Now it does not seem to be in the least certain that there could not be mental events which I own but of which I do not have even simultaneous undiscriminating awareness. I think that a person who felt quite certain that he had simultaneous undiscriminating awareness of all the first-order events which he owns might feel very doubtful indeed whether he had simultaneous undiscriminating awareness of his simultaneous undiscriminating awareness of his first-order experiences. And, if we press the question on him for experiences of higher orders, I think there will certainly come a stage at which he will feel pretty certain that he could own a mental event without having even this kind of awareness of it. If this be so, "literal ownership" and "simultaneous undiscriminating awareness" can hardly be two names for a single relation. Thus, on the whole, I think the most probable conclusion is that we are concerned with two different relations which, in the case of first-order experiences, always go together. In the case of experiences of a sufficiently high order literal ownership holds without simultaneous undiscriminating awareness.

The Notion of "The Unconscious."

      We are now in a position to deal with the substantive "The Unconscious", after clearing up the meanings of the adjective "unconscious". Here again we find that there are great ambiguities. We must first notice a systematic ambiguity in all such phrases as this. When we talk of "the State" or "the Internal Combustion Engine" we generally mean a typical idealised state or internal combustion engine. We use such phrases in this way when we are pretty certain that we are dealing with a class, such as states and internal combustion engines, having several quite distinct members which do not combine to form a single complex whole which is itself a state or an internal combustion engine. Each of these members is supposed to be a more or less imperfect approximation to that ideal limit which we call "the State" or "the Internal Combustion Engine", and which Plato would consider to be "laid up in Heaven". On the other hand, when we talk of "the Sea", we do not as a rule mean a typical ideal sea, but just the whole mass of salt water on earth of which the various seas are so many different parts. This ambiguity is inconvenient even when we are talking about "the State" (except to Idealistic Metaphysicians whose more exciting results all depend on juggling with the defects of lauguage instead of trying to correct them). It would be still more so if there were a single international world-state, as there ought to be. For, in that case "the State" might mean an ideal typical state, or it might mean the actual Super-State of which all other states would be constituents.

      Now this kind of ambiguity is specially dangerous when we are dealing with something about which we know so little as we do about the Unconscious. It may be that there is only this and that Unconscious, just as there is this and that internal combustion engine; and that the totality of all unconscious mental events has as little unity and individuality as the totality of all internal combustion engines. On the other hand, it is possible that the total Unconscious is not divisible into Smith's Unconscious, Brown's Unconscious, and so on; so that it is only the Unconscious taken as a whole, and without reference to the various origins of various parts of it, which can be treated as an individual unit. Lastly, there is the much more likely alternative that the total contents of the Unconscious do form an important unity, and that they also fall into various sub-groups, each of which has a greater internal unity than the Unconscious as a whole. The Unconscious as a whole might be like the United States; and Smith's Unconscious and Brown's Unconscious might be like the State of New York and the State of Nebraska. It is most undesirable that we should use phrases which tacitly prejudge these questions, and tie us down to one or other of the extreme alternatives. It will be wise to introduce at once certain technical terms to avoid these dangers. I will call the whole contents of the Unconscious, taken collectively as a single mass and without regard to the various origins of its various parts, the "Total Unconscious". It is then open to anyone to raise the questions: (a) "What are the contents of the Total Unconscious?"; and (b) "Does the Total Unconscious possess anything worth calling a 'structure'; and, if so, what kind of structure does it possess?" If the Total Unconscious should contain organised sub-groups, each having an important degree of unity and individuality, these may be called "Unconscious Sub-groups''. If, and only if, the Total Unconscious proved to have that kind of structure which characterises minds like our own, we could talk of the "Total Unconscious Mind". If any of the unconscious sub-groups proved to have this kind of structure, it could be called a "Special Unconscious Mind".

The Total Unconscious.

      One part of the contents of the Total Unconscious will be all mnemic persistents, i.e., all traces and dispositions, no matter whose experiences left the traces or whose experiences these traces and dispositions may subsequently modify. This will be divisible into an accessible and an inaccessible part, in the sense defined earlier in this Chapter. I will call this part of the Total Unconscious the "Total Mnemic Mass". I see no reason to suppose that there is any fundamental intrinsic difference between the accessible and the inaccessible parts of the Total Mnemic Mass. The inaccessible part is mainly dealt with by abnormal psychologists and by psycho-analysts; the accessible part has long been recognised in normal psychology. The importance of the work of the psycho-analysts is not that they have revealed anything absolutely new and unheard of. It is only the extreme ignorance of most of them about all subjects except their own which causes them to make such claims. The real importance of their work is in the following points. (a) They have shown that many inaccessible traces or groups of traces do not rest idly. In so far as these fail to produce their normal effects, e.g., memories, they are liable to produce various bodily and mental disorders. (b) They have devised several new technical methods for making inaccessible traces accessible. (c) They have shown that, when this has been done, the mental and bodily disorders are often (for a time, at least) alleviated. (d) They have stated some of the probable causes which tend to make certain experiences become inaccessible. These are great achievements; and it is a pity to create prejudice against them by ignorant pontifications about "the New Psychology". The psychologists of instinct (such as M'Dougall) also deal with the inaccessible part of the Total Mnemic Mass; but I cannot pretend to believe that they have accomplished anything except to revive the faculty-psychology in an extreme form and with an amusingly pretentious parade of "science."

      Now, if the whole content of the Total Unconscious be mnemic persistents, there is no reason to suppose that it contains mental events at all. For there is no known reason to believe that traces and dispositiens are sufficiently like the only mental events which we know directly (viz., our own conscious experiences) to be called "mental events". And, if the Total Unconscious does not contain mental events, it cannot possibly be a mind or comprise sub-groups which are minds, no matter how complex its structure may be or how definitely it is divided into sub-groups having their own unity and individuality. Thus, on this hypothesis, there would be no Total Unconscious Mind, and there would be no Special Unconscious Minds.

      If, however, there be literally unconscious mental events, viz. (a) unowned mental events, or (b) mental events which were literally owned by a mind but which could not have been introspected by it, or (c) mental events which were owned and could be introspected only by a mind which was not then in control of a body, these ought to be counted as part of the contents of the Total Unconscious. I propose to call the whole mass of such mental events, supposing that they exist, the "Total Subconscious Mass". In theory then the Total Unconscious will consist of the Total Mnemic Mass and the Total Subconscious Mass. The former consists of traces and dispositions; the latter, if it exists at all, will consist of literally unconscious mental events. If the Total Subconscious Mass exists, and if the mental events which belong to it leave traces, these will presumably belong to the inaccessible part of the Total Mnemic Mass.

      Now if, and only if, the Total Subconscious Mass exists, there is a possibility of a Total Unconscious Mind and of Special Unconscious Minds. I do not know how to define a "mind ", but I think it is evident that a thing could not be called a "mind" unless it had a peculiar kind of content and a peculiar kind of structure. Its content must be the kind of events which we call "mental" and observe when we choose to introspect. And these mental events must be interconnected in a very peculiar way. It is possible that mental events can exist only as factors in those peculiar complex wholes which we call "minds", but I do not see any very good reason to believe this. It is also possible, and much more likely, that nothing but mental events can be interconnected in the peculiar way which is characteristic of the structure of a mind. Now, among the relations which are characteristic of the structure of minds, a most important place must be given to mnemic relations. It seems essential to the notion of a mind that its contents at one moment shall be largely dependent on its contents at other moments in the remote past, and shall not be completely explicable by reference to events within or without it which have happened in the immediate past. I do not suggest for an instant that it is a sufficient description of a mind to say that it consists of mental events mnemically interconnected; but there is little doubt that this is an essential part of what we understand by a "mind". Now, unless we assume a quite new kind of causation, the mnemic relations between transient mental events depend on the existence of mnemic persistents. Thus a mind requires a Mnemic Mass. On the other hand, a Mnemic Mass by itself does not suffice to constitute a mind; for a mind must contain experiences, whilst a Mnemic Mass consists solely of traces and dispositions.

      In order to prove the existence of anything that could reasonably be called the "Total Unconscious Mind" it would therefore be necessary to establish the following points. (1) That there are literally unconscious experiences, in the sense defined above. If so, there is a Total Subconscious Mass. (2) That these literally unconscious experiences leave traces; i.e., that there is a part of the Total Mnemic Mass which consists of the traces of the Total Subconscious Mass. (3) That, given these two indispensable prerequisites, the mental events which make up the Total Subconscious Mass do in fact have to each other such relations as to form a single individual whole analogous to the minds which we know by introspection. It seems to me most important that people should recognise that the Total Unconscious may contain mental events, and may form a very important unity taken as a whole; and yet that it may be absolutely misleading to call it a "mind". At present anyone who thinks that there is reason to hold that the Total Unconscious has a unity which stretches beyond and between recognised individual human beings, is at once liable to be accused of believing in a Total Unconscious Mind. It is therefore important to point out how much more than this is needed to constitute a belief in the Total Unconscious Mind.

Unconscious Sub-groups.

      It is commonly assumed that the Total Unconscious falls quite definitely into well marked sub-groups, one specially associated with one human being and another with another. People constantly talk of "My Unconscious" and "Your Unconscious"; and such phraseology would seem to imply the belief just mentioned. So we must now consider the alleged subdivision of the Total Unconscious into one part which is Smith's, another which is Brown's, and a third which is Robinson's.

      We will begin by considering the subdivisions of the Total Mnemic Mass. All that we know about traces is that certain experiences leave them, and that they produce or modify certain later experiences. We might therefore classify the contents of the Total Mnemic Mass on either of two principles, viz., by their place of origin or by the place where they produce their effects. Each of these principles can be used in two different ways, thus giving four different methods of subdivision. We might class together

  1. all traces left by experiences of Smith's mind; or
  2. all traces left by events which happened to Smith's body, or
  3. all traces and dispositions which produce or modify experiences in Smith's mind; or
  4. all traces and dispositions which cause or modify behaviour of Smith's body.
Now it is quite certain that these four equally sensible ways of subdividing the Total Mnemic Mass will lead to different results as soon as we leave completely normal and commonplace phenomena. I will now show this in detail.

      (i) If innate dispositions be classified by origin they must be assigned to that part of the Total Mnemic Mass which belonged to our remote ancestors. For, if such dispositions originated in experiences or in bodily processes, it was in those of our remote ancestors and not of ourselves that they must have originated. If, on the other hand, we classify innate dispositions by the minds whose experiences they modify or the bodies whose activities they determine, we must assign them to the Mnemic Masses of contemporary men. Let us then, for the future, confine the discussion to traces.

      (ii) If there be literally unconscious mental events, and we classify traces by origin, we shall often reach a different result according as we classify by the mind or the body in which they originated. By "Smith's body" we mean that body which is most usually controlled by a mind with certain marked characteristics, whom we know as "Smith". Now, even if Smith be the most normal person in the world, he is often asleep and sometimes in a swoon. At such times the mind known as "Smith" is not in control of the organism known as "Smith's body", even though it be still animating the latter. If stimuli act on the body at such times and leave traces, these traces cannot be counted as belonging to Smith's Mnemic Mass, if we mean by this the set of traces left by experiences owned by Smith's mind. For these stimuli, if they produced mental events at all, did not produce experiences which were literally owned by Smith's mind. On the other hand, if by "Smith's Mnemic Mass" we mean traces left by events that happened to Smith's body, these traces will belong to Smith's Mnemic Mass.

      (iii) Generally, when a body is not controlled by the mind which normally controls it, it is not controlled by any mind at all. But in cases of multiple personality the same body may be controlled successively by several recognisably different minds. The phrase "Smith's body" then means the body of which the mind called "Smith" is one of the controlling minds. The fact that it is called "Smith's body" is then largely a matter of chance; it will depend on which of these minds is most often in control or was earliest in control. If we classify traces by the bodies in which they originate there will be one group connected with Smith's body. If we classify traces by the minds from whose experiences they originate, this Mnemic Mass will split up into S1's Mnemic Mass, S2's Mnemic Mass, and so on. This subdivision may very well not be exhaustive; there may be traces, due to events which happened in Smith's body, which were not originated by experiences in any of the minds which successively control Smith's body.

      (iv) I have so far confined myself to the classification of traces by their place of origin; and have shown that those which originate from experiences belonging to a certain mind will be contained in, but will not exhaust, the group originated by events which happen to the body which is said to be controlled by this mind. I will now point out that classification by place of origin and classification by results will lead to different groupings. Suppose that a certain body B is controlled alternately by minds B1 and B2. An experience which is owned by B1 may leave a trace which afterwards modifies experiences which are owned by B2. If we classify by origin, this trace will belong to B,'s Mnemic Mass; if we classify by results, it will belong to B2's Mnemic Mass. It is of course quite possible that a trace left by one of B,'s experiences may modify the later experiences of both Bl and B2. If we classify by origin, this trace will belong to the Mnemic Mass of B1 and not to that of B2; if we classify by results, it will be common to the Mnemic Masses of B1 and B2. There are examples of such facts as these in Dr Prince's account of the Beauchamp case.

      (v) So far it has appeared that, even in abnormal cases, all the traces which produce effects in any of the minds which control a certain body were started by events which happened to that body. This breaks down in the phenomena of telepathy. So far as I can judge from my own experiences with Mrs Leonard and from what I have read of the experiences of others, telepathy from a sitter to an entranced medium most often concerns past experiences of the sitter which he is not at the moment thinking about. We must therefore suppose that traces of some of the sitter's past experiences now affect the mind of the medium. From the point of view of origin such traces belong to the Mnemic Mass of the sitter and not to that of the medium. From the point of view of effects these traces are common to the Mnemic Masses of sitter and medium; for the sitter can remember the experience, and the medium can have telepathic knowledge of it.

      (vi) Lastly, we have the rare, but reasonably well attested phenomenon of "possession"; where the normal control of an entranced medium ceases to control her body, and the medium begins to speak with a quite different voice, gestures, and mannerisms, which are said to be recognisably characteristic of a certain dead person whom the medium has never met. I have witnessed and taken dictaphone records of one alleged case of this kind. There is no doubt at all of the striking and sudden change which takes place in voice, manner, and subject-matter communicated; but I cannot personally vouch for the resemblance of these characteristics in the so-called "personal control" to those of a certain dead person whom the medium has never met. In the experiments in which I took part the alleged communicator had been known intimately by the other sitter, and not at all by myself; so that I had to take his word for the resemblances between the "personal control" and the alleged communicator. I see, however, no special reason to doubt that the phenomenon in question sometimes does happen. Let us take it as a hypothesis that it does. The most plausible way to explain such phenomena would be to suppose that a set of traces, which originated in the mind or body of a dead person, can persist for a while after the destruction of all that is recognisable of the body; and that this set of traces is capable of affecting in a marked way the speech and bodily behaviour of an entranced medium under specially favourable conditions. Such traces would have to be counted as belonging to the Mnemic Mass of the dead person, if we count by origin; and as belonging to the Mnemic Mass of the medium, if we count by effects.

      The upshot of this discussion is that the Total Mnemic Mass almost certainly does contain important subgroups specially correlated with various recognisably different human minds and bodies. But we must recognise that, since there are several equally reasonable ways of grouping which lead to different results, any such phrase as "Smith's Unconscious" is highly ambiguous until the precise method of selection adopted has been clearly stated. And we must beware of assuming either (a) that all the contents of the Total Mnemic Mass fall into one or other of such groups; or (b) that every pair of such groups are mutually exclusive, so as to have no traces in common; or (c) that there may not be important bigger groups which include several of those smaller masses which are specially correlated with each individual human mind or body. For there are abnormal phenomena, which cannot safely be ignored, which, between them, cast doubt on all these assumptions.


      We know that certain personalities claim, not merely to alternate with others in controlling a single human organism, but also to be co-conscious with the rest. It is no part of my present business to discuss the alleged evidence for co-consciousness; but, for the sake of completeness, I must try to define what "co-consciousness" would mean. Let us suppose that a certain body B is controlled in turn by two personalities B1 and B2; and that B1 claims to be co-conscious with B2. (So far as I know, a claim to reciprocal co-consciousness is never made.) This involves making one or more of the three following demands on our belief.

  1. That certain stimuli which act on the body B when the mind B2 is in control of it produce mental events which are not literally owned by B2 and are literally owned by B1. Of such events B2 will not have even simultaneous undiscriminating awareness, and B1 will have at least this kind of awareness.
  2. That certain stimuli which act on the common body when B2 is in control produce either
    1. two very similar experiences, one of which is owned by B1 and the other by B2; or
    2. produce a single experience which is owned by both of them. If we call these "common experiences", it may be that B1 introspectively discriminates some of the common experiences which B2 does not, and conversely.
  3. That B, has some kind of discriminating awareness of certain mental events which are owned by B2 and not by B1 itself.
Sally Beauchamp seems to have made all these claims.

      I do not think that there is anything wildly paradoxical in the notion of co-consciousness as such. No doubt it is extremely hard to prove that it is a fact; but it does not seem to me to be antecedently improbable, as many people think. We commonly identify the mind which sleeps and dreams with the waking mind. In so far as this is legitimate it is certain that what I call "my mind" exists and has experiences at times when it is not controlling my body. Hence there is nothing extravagantly unfamiliar in the notion of a mind, which sometimes controls a body, literally existing and having experiences at times when it is not doing this. Of course one important difference between a co-conscious personality, like Sally, and our minds when asleep and dreaming is that, when Sally is not controlling the body, another mind is doing so; whilst, when we are asleep and dreaming, no mind is controlling our body. But, if this were the only difference, it would not be very important. For it is admitted that there are several distinct personalities which control the Beauchamp body alternately; and the only question is whether one of these persists and goes on having experiences when the control of the body has been taken over by one of the other personalities. Granted the plurality of personalities, the analogy to the normal dreaming self is enough to make this possibility quite intelligible.

      The real difficulties are over the second and third claims. We have very little analogy in ordinary life to the alleged common ownership of certain experiences by two minds. And we have very little analogy in normal life to the alleged direct knowledge by one mind of experiences which belong to another mind and not to itself. Let us first consider the claim to common ownership of certain experiences. I do not think that there is any insuperable a priori objection to this, for it can easily be reconciled with either of the three standard theories of the structure of minds. These are the Pure Ego theory; the view that a mind is a peculiar complex of interrelated simultaneous and successive mental events; and the view that the mind is a peculiar complex of interrelated non-mental objects. On the Pure Ego theory an experience would be a complex consisting of a Pure Ego and an object related in some characteristic way. Now, if there be Pure Egos at all, I know of no reason why there should be one and only one of them connected with each human body at a time. Suppose that there were two connected with the Beauchamp body. To say that Sally and Miss Beauchamp had certain experiences in common would simply mean that there are certain objects to which the Sally ego and the Miss Beauchamp ego stood at the same time in the same kind of relation. If the Pure Ego theory be true, it must sometimes happen that two Pure Egos, connected with diffierent bodies, stand at once in the same kind of relation to the same object. And, if this be so, there is no antecedent improbability that the same thing should happen when the two Egos are connected with the same body.

      It is still easier to reconcile the claim to common experiences with the view that a mind is a complex of suitably interrelated mental events. A simple geometrical analogy will make this perfectly clear. Let us represent mental events by points. And let us represent the relations which bind a number of mental events into a single mind by putting a number of points on the same ellipse. A pair of ellipses can cut each other at four points. If one of these ellipses represents Sally's mind and the other represents Miss Beauchamp's mind, the four points in which the two ellipses cut each other will represent four experiences which are owned in common by Sally and Miss Beauchamp. It is evident that the claim to common experiences can be reconciled by tbe same method with the view that the mind is a complex of characteristically interrelated objects. We have merely to let the points stand for objects, instead of letting them stand for experiences; and the same diagram will represent the alleged facts on this theory about the structure of the mind. (Of course it is also possible that the "common experiences" are really two different experiences, one belonging to one mind and one to the other, and that their "community" merely consists in the fact that they are very much alike and are produced simultaneously in the two minds by a common external cause. On that hypothesis there is even less diffficulty in admitting the possibility of "common experiences".)

      It remains to consider Sally's claim to some kind of direct knowledge of experiences which were owned by other personalities and not by herself. Really, she professed to have two very different kinds of knowledge of such experiences, both of which fall under the present heading. (a) She alleged that she could get to know isolated experiences of the personality B4 by a special method which resembled crystal-gazing. Of the rest of B4's experiences she was wholly ignorant. (b) On the other hand, Sally claimed to be aware, without any special effort except that of attention, of whole masses of experience which belonged to Miss Beauchamp and not to herself. Thus, when Miss Beauchamp was ill and somewhat light-headed, Sally claims to have been aware of illusory perceptions in Miss Beauchamp's mind and of the fear which these engendered in Miss Beauchamp. Sally herself did not share the illusions or the fear, though she was aware of the objects in the room, which Miss Beauchamp was misperceiving and consequently fearing.

      Now I feel no kind of a priori difficulty about the first kind of knowledge. It is closely analogous to ordinary telepathy; and, on her own showing, Sally was throwing herself into a state which is known to be favourable to telepathy when she got this kind of know ledge of isolated events in B4's mental history. There is no doubt of the reality of telepathy between minds which animate different bodies; and it is easier rather than harder to conceive of telepathy taking place between two minds which animate the same body. The real difficulty is over the second kind of alleged knowledge. With us introspective discrimination can be applied only to mental events which we own and of which we have simultaneous undiscrimhlating awareness. Sally's claim practically amounts to asserting that she could introspect experiences which she did not own and which Miss Beauchamp did own, just as we can introspect our own experiences. As we have no analogy to this power anywhere in normal mental life, it must be regarded as antecedently improbable that it exists; though this does not of course prove that it cannot exist.

      The upshot of the matter is that, whether there be enough empirical evidence for co-consciousness or not, it ought not to be dismissed a priori as antecedently too improbable to be worth consideration. Three out of four of Sally's claims have enough analogy to admitted facts, and are easy enough to reconcile with current theories, to give them a reasonable antecedent probability.