As long as fifteen years ago it was possible for a good and quite uncontroversial text on social psychology (Middlebrook 1974) to say things like ‘Thus the individual is not a single self, but many selves, which change somewhat as the individual shifts from situation to situation and person to person.

The question of whether there are parts of a person which can be talked to and worked with as if they were separate little personalities with a will of their own is one which has fascinated nearly everyone who has had to work with people in any depth.

It is an extraordinary fact that there is at present no systematic book on subpersonalities, and the word does not appear in any text on personality theory known to me. It is not in the dictionaries of psychology nor in the dictionaries of psychotherapy. Yet the thing itself is used by virtually every clinician who has ever written about working with people, and by more and more psychologists paying attention to what is there as opposed to what is supposed to be there.

My own working definition of a Subpersonality is a semipermanent and semi-autonomous region of the personality capable of acting as a person

I could give you a whole list of the persons I can be. I am an old peasant woman who thinks of cooking and of the house. I am a scholar who thinks about deciphering manuscripts. I am a psychotherapist who thinks about how to interpret people’s dreams. I am a mischievous little boy who enjoys the company of a ten-year-old and playing mischievous tricks on adults, and so on. I could give you twenty more such characters.

It is not easy to discover the history of the idea of subpersonalities. It is one of those concepts, so common in the field of the human sciences, which has a long past but a short history.

Origen: ‘You will see that a man who seems to be one is not one, but as many different persons appear in him as he has attitudes’ (Jung 1946: 197).

The first psychologist I can find who dealt with the idea in a way which seems to throw off the historical distortions was William James, but he really said very little about exactly where subpersonalities came from or how they worked.

In the field of philosophy, too, there was an interest at this time, and one of the most eloquent statements was made by the British Hegelian William Wallace, when he said: ‘We have hardly formed our resolve when we regret it: the voices of our other selves, of that manifold pack of half-formed personalities within us, none of which we dare honestly disown, are raised in protest against the usurping monarchy of our overt resolve’(Wallace 1898: 109).

In 1924 the concept had been taken into psychiatry by Bleuler, who said in his Textbook of Psychiatry: ‘the complexes can actually acquire sub-personalities with some sort of independence of the psyche’. But psychiatry apparently did not retain this insight, and in today’s texts no such admission is found.

In 1936 Kurt Lewin published his book Topological Psychology, in which he described how regions of the personality could become relatively independent. He said: The degree of dynamical connectedness of the different parts of the person can be nearly equal within the whole region of the person, or certain regions can separate themselves to an especially high degree from the others and develop relatively independently.

Not only is a theory of subpersonalities a concept which is important for psychology and psychotherapy: it is also one which is crucial for political theory. Many theories of politics, both conservative and radical, say that the person is conditioned early into conformity with society. The pressures of society are so great, they say, that the person is bludgeoned willy-nilly into being the kind of personality which a particular society needs. It is difficult, with all such theories, to see any place for radical social change.

Different roles bring out different subpersonalities, as William James (1890) urged long ago. So do different social frames, as Goffman (1959) has outlined.

Two or more sides arguing within us (on the one hand I want to – but on the other hand …) may become repetitive enough and frequent and vivid enough to require an identity each before they can be worked out.

We may identify with a hero or heroine, or with an admired group, and take on their characteristics. For example, in the 1970s I frequently came across hippie and revolutionary subpersonalities in the people I was working with.

Actors of the ‘method’ school work by setting up within themselves a sub-personality corresponding to the character they are playing.

Carl Rogers was known as a great psychotherapist who employed no techniques, and who had very little in the way of theory about internal dynamics. But right at the end of his life even he was forced, by the sheer iogic of his own work, to adopt the use of subpersonalities.

It is a common experience to hear someone talking over the telephone in a certain voice, and when coming off the phone to talk to someone in the room in a quite different voice. Some people actually have a ‘telephone voice’ which they cultivate; others do it without thinking, quite without any intention of changing their voice at all.

In the case of transvestites, for example, the contrasex personality is often very different from the original personality, and the fact that each has a different name helps in the process of differentiation.

Once a Subpersonality has been named, it becomes easier to use it in this kind of way, as a method of raising awareness of what is going on internally.

‘I think your martyr is trying to make me feel guilty. Do you agree?’ We have stopped saying that the whole person is bad – now we are drawing attention to a problem for us that may also turn out to be a problem for the person themselves.

We are responsible for our subpersonalities, just as we’re responsible for our children, our pets and our car.

If I can say, ‘Yes, part of me does feel that, but only part of me’, that does justice to the fact that I do feel it, but does not overwhelm me by suggesting that that is all that I feel.

Most psychologists assume that, no matter how many subpersonalities we find, ultimately all has to be reduced to one. This is the most general conception of mental health. But Watkins, and her mentor James Hillman, ask the question ‘Why?’ Would it not make more sense to live with multiplicity, to recognize more than one centre within ourselves? Hillman suggests that this quarrel is rather like the quarrel between monotheism and Polytheism. Psychology, he says, is secretly monotheistic, and wants everything to be neatly hierarchical or bureaucratic. But could we not envisage a polytheistic psychology, which admitted that there could be many gods and goddesses, many egos, many identities, many selves

Harry consists of a hundred or a thousand selves, not of two. His life oscillates, as everyone’s does, not merely between two poles, such as the body and the spirit, the saint and the sinner, but between thousands, between innumerable poles.

Man, says, Hesse, is not capable of thought in any high degree, and even the most cultivated of men use absurd simplifications in their thought and understanding. People will do anything to make their life easier. And the idea that there is just one simple, single self is the crudest and most oversimplified assumption of all.

Then I took an LSD trip (perhaps more common then than now, but in any case something familiar to me – 1 regarded myself as something of an astronaut of inner space), with the explicit object of getting into each of these personalities in turn, and asking the same eleven questions of each of them. This was an extremely useful exercise, which made a number of things very much clearer to me, and made me feel that here was something quite powerful, which could be pushed quite a long way in terms of self-understanding and self-acceptance.

It is interesting to see exactly what Freud says about the superego. And when we read about the nature of the superego, we find that it again answers very precisely to our description of a Subpersonality.

A portion of the external world has, at least partially, been abandoned as an object and has instead, by identification, been taken into the ego and thus become an integral part of the internal world. This new psychical agency continues to carry on the functions which have hitherto been performed by people in the external world. (Freud 1938: 203)

One of the great pioneers in this field was Jung. Some of his first work, in the first ten years of this century, was with association tests, where the experimenter says a word and the subject comes back with the first word which comes into his or her head. Through the use of this device, Jung became convinced that within the person there were semi-autonomous systems which he called at first the ‘feeling-toned complexes.’ His first thought was that the complex was like a theme or leitmotiv in music, which came back in various forms in different circumstances, but later he saw it more as a Subpersonality.

The tendency to split means that parts of the psyche detach themselves from consciousness to such an extent that they not only appear foreign but lead an autonomous life of their own. It need not be a question of a hysterical multiple personality, or schizophrenic alterations of personality, but merely of so-called ‘complexes’ that come entirely within the scope of the normal. (Jung 1936)

Complexes obviously represent a kind of inferiority in the broadest sense – a statement I must at once qualify by saying that to have complexes does not necessarily indicate inferiority. It only means that something discordant, unassimilated and antagonistic exists, perhaps as an obstacle, but also as an incentive to greater effort, and so, perhaps, to new possibilities of achievement. In this sense, therefore, complexes are focal or nodal points of psychic life which we would not wish to do without; indeed, they should not be missing, for otherwise psychic activity would come to a fatal standstill. (Jung 1928)

This notion, that the complexes are good and to be taken seriously as necessary parts of human development and functioning, has been taken further by James Hillman. Hillman is a modern Jungian who has taken particularly seriously the whole question of the imaginal life – the deep life of the psyche beneath the surface. He sees this as ‘no longer single-centred but polycentric’.

All we have to do is to allow and encourage these semi-autonomous parts to speak their minds, to interact with each other, to change, merge, separate, integrate, and differentiate, to transform. And to do all this the first step is to personify the complex.

One of Jung’s great interests was in ancient alchemy, seen as a way of talking about the psyche and its processes of transformation in symbolic form. And in alchemy there is a process called the separatio, where the different ingredients are set side by side. ‘Only separated things can unite’, said the alchemists.

This process of naming is very important. We must not fall into the trap of going immediately into theory and saying that the person who arises in our dream must be one of the Jungian archetypes already discovered and named by Jung or someone else.

It was in 1935 (Collected Works, vol. 6) that Jung first used the concept of ‘active imagination’. In active imagination we fix upon a particular point, mood, picture, or event, and then allow a fantasy to develop in which certain images become personified.

A classic in this area came from Barbara Hannah, who worked with Jung, and then taught at the C. G. Jung Institute in Zurich. She says that this method enables conversations to take place with contents of the unconscious that appear personified. It does not matter how the image may come, but the essential thing is to hang on to it and not let it go until it has revealed its message through dialogue. She emphasizes over and over again the distinction between passive imagination, where we merely experience a scene as if looking at it on a screen, and active imagination, where we enter into an interactive discourse which goes back and forth between the personified image and ourselves.

I am convinced that it is nearly impossible to produce anything in the imagination that is not an authentic representation of something in the unconscious. The whole function of the imagination is to draw up the material from the unconscious, clothe it in images, and transmit it to the conscious mind. (Johnson 1986)

I very seldom find, in my own work, that people spontaneously come up with gods and goddesses, so to impose this would be a kind of interpretation on the part of the therapist which could enhance any narcissistic tendencies on the part of the client.

Where I have found the idea of a goddess to be useful is when encouraging women to see strong aspects of themselves as female rather than male, along the lines suggested by Jean Shinoda Bolen (1984), who suggests that Athena, for example, is very warlike and very intellectual but also totally female. So a warlike and intellectual woman client can be encouraged to see herself in this light, rather than as an imitation man, as the general culture of today would tend to suggest

It often makes sense to ask a Subpersonality, ‘How old are you?’ or ‘What do you look like?’

I shall place before you the view that imaginal dialogues do not merely reflect or distort reality, but create reality; that the real is not necessarily antithetical to the imaginal, but can be conceived of more broadly to include the imaginal; and that personifying is not an activity symptomatic of the primitivity of mind, but is expressive of its dramatic and poetic nature. (Watkins 1986)

The word ‘spontaneous’ is very important in all this. These subpersonalities are not things we invent or choose. Rather, we discover them and work with them as a result of following the normal process of psychotherapy, where we take current incidents and uncover the hidden meanings behind them. The phrase ‘configurations of existence’ is important too: these are things which exist already, and have a right to exist.

I remember one working class woman client of mine who said in a puzzled voice, ‘What I can’t understand is how I take myself apart into all these pieces in here, and yet when I go out I feel more whole!’

Kees says: I looked at this sneaky animal, and my whole image of myself as a decent, good person collapsed. It was a filthy creature, with green eyes and a mean disposition. Without a sense of personal centeredness [derived from earlier work] I wouldn’t have been able to confront it. I would have been afraid that it was all I was, that it would overpower me, and so on.

As time went on, he began to see The Golem in quite a different light. If I didn’t possess that personality, I later realized, I would have gotten myself into all kinds of trouble. Though he first emerged as nasty and sneaky, at bottom he was only trying to say ‘no’ when I was always and automatically inclined to say ‘yes’ to everyone and everything.

‘All subpersonalities are expressions of vital elements of our being, however negative they may seem to us at first.’

Subpersonalities become harmful only when they control us, and this usually happens when we are unaware of them.

We have to get to know the subpersonalities from the inside, by playing them, and not from the outside by observing or describing them.

The topdog is righteous and authoritarian; he knows best He is sometimes right, but always righteous …. He manipulates with demands and threats of catastrophe …. The underdog manipulates with being defensive, apologetic, wheedling, playing the crybaby, and so on.... The underdog is the Mickey Mouse. The topdog is the Super Mouse …. This is the basis for the famous self-torture game. (Perls 1969)

The self-torture game referred to here is simply the way in which we very often entertain ‘shoulds’ which we do not really intend to honour. We carry them round with us, and every now and then they beat us over the head with the thought – ‘I still haven’t written to my grandmother’ – or whatever it may be. The reply is something like – T will do it, but I haven’t got time at the moment’. In Perls’s terms this is the topdog and underdog at work.

It seems to us at first that the person must be single because the body is single. But if the body is itself multiple, that puts another complexion on things.

To proponents of singular self-identity nothing would seem simpler than the human body as clear and irrefutable evidence for locating the structure of subjectivity in persons alone: one self per body. Yet appearances can be deceiving even if, like the peels of an onion, appearances constitute the whole of so-called reality. The body is both complex and ambiguous. Even its apparendy obvious unity was less than obvious to the early Greeks. Bruno Snell notes, ‘the early Greeks did not, either in their language or in the visual arts, grasp the body as a unit’ Both the use of plurals to refer to the physical nature of the body as well as the crimped joints in the early vase paintings suggest that for the early Greeks, ‘the physical body of man was comprehended, not as a unit but as an aggregate.’ (Ogilvy 1977: 108)

As we develop greater complexity of living, our own personality separates into myriad functions, related to and isolated from each other, in thousands of ways. We divide our self into patterns of behaviour and experience each of which is appropriate for various situations, and we thrust from activation and awareness reactions that would not be adaptive. Who wants to worry about a mathematics test at a party, or plan one’s budget at a football game? We call these patterns of behaviour and experience ‘ego states’, and they are normally a part of us all. Even as society must separate its merchants from its musicians, its teachers from its builders, etc., so also does the ‘society of self within a single human divide itself into segments for the accomplishing of its various adjustive goals. Thus, ego-state theory holds that normal personalities are characterized by organizational patterns of behaviour and experience that have been partially dissociated from each other for purposes of adaptation and defence. (Watkins and Johnson 1982)

An ego state may be described phenomenologically as a coherent system of feelings related to a given subject, and operationally as a set of coherent behaviour patterns; or pragmatically, as a system of feelings which motivates a related set of behaviour patterns …. Repression of traumatic memories of conflicts is possible in many cases, according to Federn, only through repression of the whole pertinent ego state. Early ego states remain preserved in a latent stage, waiting to be re-cathected. (Berne 1961)

Shapiro has his own version of the typology of subselves. This seems to be a regular temptation of people working in this field, to try to classify the subpersonalities in some way.

Perhaps the most flexible of all the approaches to subpersonalities comes from the psycho-imagination therapy of Joseph Shorr (1983). It is curious that he makes no mention of using different chairs in his work, and he seems to be able to do without all the apparatus the rest of us find so useful. He uses polarities, groups of three, groups of four, and obviously there is no limit to the imagination in his work. I find his approach a continual inspiration and in spite of his frequent superficiality he has much to offer in the understanding of how to work with imaginary persons.

In a sense, each of us is a multiple personality. We exhibit one personality at home, another at the office, and another on vacation. But usually these various ‘sub-personalities’ are governed by a generalized ‘Federal jurisdiction’, even as the States of Montana and Idaho must submit to control from the United States government We sense a continuity of selfness at all times. It is generally unrecognized that each of us is not a unity, but more like a confederation of component segments. Sometimes these parts cooperate and smooth the way of adaptation to the world. At other times they exhibit relative autonomy from each other and, through internal conflict, give us our anxieties, depressions, headaches, phobias, and other painful symptoms. (Watkins and Johnson 1982: 129-30)

‘Fusion, or the merging of two or more personalities into a ‘oneness’, is generally neither essential nor desirable except in special circumstances’ (Watkins and Johnson 1982).

We have left it ambiguous as to whether there is such a thing as the Real Self, or the Higher Self (Transpersonal Self, Greater Self, Deeper Self, Inner Self, Self with a big S, and so on), or whether there is such a thing as the soul. In a way this is not a particularly important issue. There is plenty of work to be done, as we saw in the previous chapter, regardless of how we may wish to answer this question. But at the same time it would be cowardly not to admit that there is a real question here, even though it may be difficult to answer.

One of Wilber’s most insistent themes is that we tend to suffer from the pre/trans fallacy – that is, we confuse what is prepersonal with what is transpersonal.

We have already gone from symbiosis with the mother to separation, and from body-self to membership-self, and from there to the mental ego. At each of these transitions we had to revise our whole notion of who we were, and even what kind of self we were. So we know what it is like to revise our self-definition. The move from mental ego to the next stage is just another such change, and peak experiences are a very common harbinger of this particular transition.

Wilber uses the image of the centaur to refer to the difference between the horse-and-rider separation of the previous stage and the unification of the present stage.

Now certainly the subjective experience here is an experience of unity. It seems to us as if we are for the first time one and single, and as if the subpersonalities are no more than aspects as seen from outside. It feels as if we are real, and as if there is one person taking responsibility for all that we say and do. All the stirring phrases of existential philosophy seem to have come true: ‘Free and alone, without assistance and without excuse’; ‘There are no limits – our own choice is everything’; ‘To be that self which one truly is’; ‘The encounter is here and now’; ‘I create my world’. So what has happened to the subpersonalities? After a while, I believe, we find that they are still there, even though they may have been transformed in various ways.

That is all very well, but Wilber says that it is not the end of the road. What we now come on to, if we continue with the process of development – and we now have some choice over this too – are the transpersonal bands.

It is in the subtle stage that we get in touch with the higher self, just as it was in the centaur stage that we got in touch with the real self.

What the ego, the superego, and the real self have in common is that they all have strict boundaries. All the humanistic writers are very clear about this: ‘I am I and you are you’, and so forth. But at the transpersonal stage all this changes quite radically.

At this stage the separative egoic and existential identifications dissolve. We can take down the barriers which divide us from other people, and experience our common identity.

it is true to say that we are separate individuals: it is also true to say that we are members one of another, that we are all part of something greater; and this is what we realize and experience beyond the transpersonal bands.

This transition will possibly take just as long to carry through as any of those met before. Here we have to move from the subtle realm into the causal realm. We have to differentiate between ourselves as symbolic explorers in the foothills of the transpersonal realm, and people who can do without symbols altogether.

The main sign that we have entered into this new phase is that we do not find symbols of any use any more.

It seems to me that working with subpersonalities is for the most part a good method of crossing the biosocial bands and moving from the mental ego stage into the centaur stage. The reason for this is that it is a very good method of healing the splits in the personality, enabling us to work out the splits in their own terms, doing justice to what is positive about them as well as what is negative. It is a very good path to self-knowledge. When we have done all the work with our subpersonalities we really know ourselves very well, and do not need neurotic projections and other psychological defence mechanisms to carry us through. We have done our therapy, so to speak.

This is one of the most important findings in the best book we have had yet about the whole question of spiritual development. Dick Anthony and his co-workers tell us that genuine glimpses of spiritual reality are quite common and very important when they do occur. But they can lead to spiritual inflation, a quite disastrous occurrence which makes us think we have got much further along the road than we really have.

The term ‘glimpse experience’, which we are introducing in this volume, is intended specifically to be a counter-inflationary term, emphasizing that the great majority of mystical or transpersonal experiences are only temporary glimpses beyond mundane ego-consciousness and do not involve true transformation to a more transcendent, encompassing state. (Anthony et al 1987: 188)

And so we come to the final question: in the end, are we one or many? There is a great temptation to assume that the Wilber story says that the answer must be One. The whole thrust of his approach, we perhaps assume, is to say that we have to talk ultimately about the single Atman, the single Void, the single Ground. But when we come to look at the small print, this is not so. Wilber (1980) says that the final stage, even supposing that we reach it, and even supposing that reaching it is a meaningful phrase, is ‘both One and Many, Only and All, Source and Suchness, Cause and Condition’, and in his earlier book, which goes into this phase in much more detail, he says: ‘Reality is actually neither one nor many, singular nor plural, transcendent nor immanent – it is a non-dual experience’ (Wilber 1977: 64).

Further back still, at the centaur stage, it seems all important to have the sense of being just one single self, and I think this must be allowed at this stage as being a very important and quite necessary illusion.

So our conclusion must be that, while at certain stages unity may be very tempting and even apparently necessary, in the end multiplicity is just as real and just as important, all the way down the line. There never comes a time when we can simply abandon our multiplicity and lie down in a perfect and final unity. We may not have subpersonalities in the sense that they fight with one another, but we shall still have many angles, many colours, many quirks. We shall still be human.