Show Summary Details
Page of

(p. 241) The Capacity to Be Alone 

(p. 241) The Capacity to Be Alone
(p. 241) The Capacity to Be Alone

Donald W. Winnicott

Page of

Subscriber: null; date: 21 October 2018

Originally published in International Journal of Psychoanalysis, 1958, 39, 416–420. Also published in Psyche, 1958, 12, as ‘Über die Fähigkeit, allein zu sein’; and in The maturational processes and the facilitating environment: Studies in the theory of emotional development (pp. 29–36). London: Hogarth, 1965.
Based on a paper read at an Extra Scientific Meeting of the British Psychoanalytical Society, 24 July 1957.

I wish to make an examination of the capacity of the individual to be alone, acting on the assumption that this capacity is one of the most important signs of maturity in emotional development.

In almost all our psycho-analytic treatments there come times when the ability to be alone is important to the patient. Clinically this may be represented by a silent phase or a silent session, and this silence, far from being evidence of resistance, turns out to be an achievement on the part of the patient. Perhaps it is here that the patient has been able to be alone for the first time. It is to this aspect of the transference in which the patient is alone in the analytic session that I wish to draw attention.

It is probably true to say that in psycho-analytical literature more has been written on the fear of being alone or the wish to be alone than on the ability to be alone; also a considerable amount of work has been done on the withdrawn state, a defensive organization implying an expectation of persecution. It would seem to me that a discussion on the positive aspects of the capacity to be alone is overdue. In the literature there may be specific attempts to state the capacity to be alone, but I am not aware of these. I wish to make reference to Freud’s (1914) concept of the anaclitic relationship (cf. ‘Primary Maternal Preoccupation’ [CW 5:2:16]).

(p. 242) Three- and Two-Body Relationships

Rickman introduced us to the idea of thinking in terms of three-body and two-body relationships. We often refer to the Oedipus complex as a stage in which three-body relationships dominate the field of experience. Any attempt to describe the Oedipus complex in terms of two people must fail. Nevertheless two-body relationships do exist, and these belong to relatively earlier stages in the history of the individual. The original two-body relationship is that of the infant and the mother or mother-substitute, before any property of the mother has been sorted out and moulded into the idea of a father. The Klein concept of the depressive position can be described in terms of two-body relationships, and it is perhaps true to say that a two-body relationship is an essential feature of the concept.

After thinking in terms of three- and two-body relationships, how natural that one should go a stage further back and speak of a one-body relationship! At first it would seem that narcissism would be the one-body relationship, either an early form of secondary narcissism or primary narcissism itself. I am suggesting that this jump from two-body relationships to a one-body relationship cannot, in fact, be made without violation of a great deal that we know through our analytic work and through direct observation of mothers and infants.

Actually Being Alone

It will be appreciated that actually to be alone is not what I am discussing. A person may be in solitary confinement, and yet not be able to be alone. How greatly he must suffer is beyond imagination. However, many people do become able to enjoy solitude before they are out of childhood, and they may even value solitude as a most precious possession.

The capacity to be alone is either a highly sophisticated phenomenon, one that may arrive in a person’s development after the establishment of three-body relationships, or else it is a phenomenon of early life which deserves special study because it is the foundation on which sophisticated aloneness is built.


The main point of this contribution can now be stated. Although many types of experience go to the establishment of the capacity to be alone, there is one that is basic, and without a sufficiency of it the capacity to be alone does not come about; this experience is that of being alone, as an infant and small child, (p. 243) in the presence of mother. Thus the basis of the capacity to be alone is a paradox; it is the experience of being alone while someone else is present.i

Here is implied a rather special type of relationship, that between the infant or small child who is alone, and the mother or mother-substitute who is in fact reliably present even if represented for the moment by a cot or a pram or the general atmosphere of the immediate environment. I would like to suggest a name for this special type of relationship.

Personally I like to use the term ego-relatedness, which is convenient in that it contrasts rather clearly with the word id-relationship, which is a recurring complication in what might be called ego life. Ego-relatedness refers to the relationship between two people, one of whom at any rate is alone; perhaps both are alone, yet the presence of each is important to the other. I consider that if one compares the meaning of the word ‘like’ with that of the word ‘love’, one can see that liking is a matter of ego-relatedness, whereas loving is more a matter of id-relationships, either crude or in sublimated form.

Before developing these two ideas in my own way I wish to remind you how it would be possible to refer to the capacity to be alone in well-worn psycho-analytic phraseology.

After Intercourse

It is perhaps fair to say that after satisfactory intercourse each partner is alone and is contented to be alone. Being able to enjoy being alone along with another person who is also alone is in itself an experience of health. Lack of id-tension may produce anxiety, but time-integration of the personality enables the individual to wait for the natural return of id-tension, and to enjoy sharing solitude, that is to say, solitude that is relatively free from the property that we call ‘withdrawal’.

Primal Scene

It could be said that an individual’s capacity to be alone depends on his ability to deal with the feelings aroused by the primal scene. In the primal scene an excited relationship between the parents is perceived or imagined, and this is accepted by the child who is healthy and who is able to master the hate and to gather it into the service of masturbation. In masturbation the whole responsibility for the conscious and unconscious fantasy is accepted by the individual child, who is the third person in a three-body or triangular (p. 244) relationship. To be able to be alone in these circumstances implies a maturity of erotic development, a genital potency or the corresponding female acceptance; it implies fusion of the aggressive and erotic impulses and ideas, and it implies a tolerance of ambivalence; along with all this there would naturally be a capacity on the part of the individual to identify with each of the parents.

A statement in these or any terms could become almost infinitely complex, because the capacity to be alone is so nearly synonymous with emotional maturity.

Good Internal Object

I will now attempt to use another language, one that derives from the work of Melanie Klein. The capacity to be alone depends on the existence of a good object in the psychic reality of the individual. The good internal breast or penis or the good internal relationships are well enough set up and defended for the individual (at any rate for the time being) to feel confident about the present and the future. The relationship of the individual to his or her internal objects, along with confidence in regard to internal relationships, provides of itself a sufficiency of living, so that temporarily he or she is able to rest contented even in the absence of external objects and stimuli. Maturity and the capacity to be alone implies that the individual has had the chance through good-enough mothering to build up a belief in a benign environment. This belief is built up through a repetition of satisfactory instinctual gratifications.

In this language one finds oneself referring to an earlier stage in the individual’s development than that at which the classical Oedipus complex holds sway. Nevertheless a considerable degree of ego maturity is being assumed. The integration of the individual into a unit is assumed, otherwise there would be no sense in making reference to the inside and the outside, or in giving special significance to the fantasy of the inside. In negative terms: there must be a relative freedom from persecutory anxiety. In positive terms: the good internal objects are in the individual’s personal inner world, and are available for projection at a suitable moment.

To Be Alone in an Immature State

The question which will be asked at this point is this: Can a child or an infant be alone at a very early stage when ego immaturity makes it impossible for a description of being alone to be given in the phraseology that has just been employed? It is the main part of my thesis that we do need to be able to speak of an unsophisticated form of being alone, and that even if we agree that the capacity to be truly alone is a sophistication, the ability to be truly alone has (p. 245) as its basis the early experience of being alone in the presence of someone. Being alone in the presence of someone can take place at a very early stage, when the ego immaturity is naturally balanced by ego-support from the mother. In the course of time the individual introjects the ego-supportive mother and in this way becomes able to be alone without frequent reference to the mother or mother symbol.

‘I Am Alone’

I would like to take up this subject in a different way by studying the words ‘I am alone’.

First there is the word ‘I’, implying much emotional growth. The individual is established as a unit. Integration is a fact. The external world is repudiated and an internal world has become possible. This is simply a topographical statement of the personality as a thing, as an organization of ego-nuclei. At this point no reference is being made to living.

Next come the words ‘I am’, representing a stage in individual growth. By these words the individual not only has shape but also life. In the beginnings of ‘I am’ the individual is (so to speak) raw, is undefended, vulnerable, potentially paranoid. The individual can only achieve the ‘I am’ stage because there exists an environment which is protective; the protective environment is in fact the mother preoccupied with her own infant and orientated to the infant’s ego requirements through her identification with her own infant. There is no need to postulate an awareness of the mother on the part of the infant at this stage of ‘I am’.

Next I come to the words ‘I am alone’. According to the theory that I am putting forward this further stage does indeed involve an appreciation on the part of the infant of the mother’s continued existence. By this I do not necessarily mean an awareness with the conscious mind. I consider, however, that ‘I am alone’ is a development from ‘I am’, dependent on the infant’s awareness of the continued existence of a reliable mother whose reliability makes it possible for the infant to be alone and to enjoy being alone, for a limited period.

In this way I am trying to justify the paradox that the capacity to be alone is based on the experience of being alone in the presence of someone, and that without a sufficiency of this experience the capacity to be alone cannot develop.


Now, if I am right in the matter of this paradox, it is interesting to examine the nature of the relationship of the infant to the mother, that which for (p. 246) the purposes of this paper I have called ego-relatedness. It will be seen that I attach a great importance to this relationship, as I consider that it is the stuff out of which friendship is made. It may turn out to be the matrix of transference.

There is a further reason why I put a special importance on this matter of ego-relatedness, but in order to make my meaning clear I must digress for a moment.

I think it will be generally agreed that id-impulse is significant only if it is contained in ego living. An id-impulse either disrupts a weak ego or else strengthens a strong one. It is possible to say that id-relationships strengthen the ego when they occur in a framework of ego-relatedness. If this be accepted, then an understanding of the importance of the capacity to be alone follows. It is only when alone (that is to say, in the presence of someone) that the infant can discover his own personal life. The pathological alternative is a false life built on reactions to external stimuli. When alone in the sense that I am using the term, and only when alone, the infant is able to do the equivalent of what in an adult would be called relaxing. The infant is able to become unintegrated, to flounder, to be in a state in which there is no orientation, to be able to exist for a time without being either a reactor to an external impingement or an active person with a direction of interest or movement. The stage is set for an id experience. In the course of time there arrives a sensation or an impulse. In this setting the sensation or impulse will feel real and be truly a personal experience.

It will now be seen why it is important that there is someone available, someone present, although present without making demands; the impulse having arrived, the id experience can be fruitful, and the object can be a part or the whole of the attendant person, namely the mother. It is only under these conditions that the infant can have an experience which feels real. A large number of such experiences form the basis for a life that has reality in it instead of futility. The individual who has developed the capacity to be alone is constantly able to rediscover the personal impulse, and the personal impulse is not wasted because the state of being alone is something which (though paradoxically) always implies that someone else is there.

In the course of time the individual becomes able to forgo the actual presence of a mother or mother-figure. This has been referred to in such terms as the establishment of an ‘internal environment’. It is more primitive than the phenomenon which deserves the term ‘introjected mother’.

Climax in Ego-Relatedness

I would now like to go a little further in speculating in regard to the ego-relatedness and the possibilities of experience within this relationship, and (p. 247) to consider the concept of an ego orgasm. I am of course aware that if there is such a thing as an ego orgasm, those who are inhibited in instinctual experience will tend to specialize in such orgasms, so that there would be a pathology of the tendency to ego orgasm. At the moment I wish to leave out consideration of the pathological, not forgetting identification of the whole body with a part-object (phallus), and to ask only whether there can be a value in thinking of ecstasy as an ego orgasm. In the normal person a highly satisfactory experience such as may be obtained at a concert or at the theatre or in a friendship may deserve a term such as ego orgasm, which draws attention to the climax and the importance of the climax. It may be thought unwise that the word orgasm should be used in this context; I think that even so there is room for a discussion of the climax that may occur in satisfactory ego-relatedness. One may ask: when a child is playing, is the whole of the game a sublimation of id-impulse? Could there not be some value in thinking that there is a difference of quality as well as of quantity of id when one compares the game that is satisfactory with the instinct that crudely underlies the game? The concept of sublimation is fully accepted and has great value, but it is a pity to omit reference to the vast difference that exists between the happy playing of children and the play of children who get compulsively excited and who can be seen to be very near to an instinctual experience. It is true that even in the happy playing of the child everything can be interpreted in terms of id-impulse; this is possible because we talk in terms of symbols, and we are undoubtedly on safe ground in our use of symbolism and our understanding of all play in terms of id-relationships. Nevertheless, we leave out something vital if we do not remember that the play of a child is not happy when complicated by bodily excitements with their physical climaxes.

The so-called normal child is able to play, to get excited while playing, and to feel satisfied with the game, without feeling threatened by a physical orgasm of local excitement. By contrast, a deprived child with antisocial tendency, or any child with marked manic-defence restlessness, is unable to enjoy play because the body becomes physically involved. A physical climax is needed, and most parents know the moment when nothing brings an exciting game to an end except a smack—which provides a false climax, but a very useful one. In my opinion, if we compare the happy play of a child or the experience of an adult at a concert with a sexual experience, the difference is so great that we should do no harm in allowing a different term for the description of the two experiences. Whatever the unconscious symbolism, the quantity of actual physical excitement is minimal in the one type of experience and maximal in the other. We may pay tribute to the importance of ego-relatedness per se without giving up the ideas that underlie the concept of sublimation.

(p. 248) Summary

The capacity to be alone is a highly sophisticated phenomenon and has many contributory factors. It is closely related to emotional maturity.

The basis of the capacity to be alone is the experience of being alone in the presence of someone. In this way an infant with weak ego organization may be alone because of reliable ego-support.

The type of relationship that exists between an infant and the ego-supportive mother deserves special study. Although other terms have been used, I suggest that ego-relatedness might be a good term for temporary use.

In a frame of ego-relatedness, id-relationships occur and strengthen rather than disrupt the immature ego.

Gradually, the ego-supportive environment is introjected and built into the individual’s personality, so that there comes about a capacity actually to be alone. Even so, theoretically, there is always someone present, someone who is equated ultimately and unconsciously with the mother, the person who, in the early days and weeks, was temporarily identified with her infant, and for the time being was interested in nothing else but the care of her own infant.


Editorial Note i The first mention of paradox. For further discussion of this topic see the General Introduction in Volume 1.