Midterm
THROUGH PAEDIATRICS
PSYCHO-ANALYSIS Collected Papers
da9 - 2 Y 2 rn by
D. W. WINNICOTT //
With an Introduction by M. MASUD R. KHAN
BRUNNER/MAZEL, Publishers NEW YORK
f Chapter XVIII
Transitional Objects and Transitional
A STUDY OF THE FIRST NOT-ME POSSESSION'
Introduction IT IS W E L L K N O W N that infants as soon as they are born tend to use fist, fingers, thumbs in stimulation of the oral erotogenic zone, in satisfaction of the instincts a t that zone, and also in quiet union. I t is also well known that after a few months infants of either sex become fond of playing with dolls, and that most mothers allow their infants some special object and expect them t o become, as it were, addicted to such objects.
There is a relationship between these two sets of phenomena that are separated by a time interval, and a study of the development from the earlier into the later can be profitable, and can make use of important clinical material that has been some-vhat neglected.
B e First Possession Those who happen to be in close touch with mothers' interests and prob-
lems will be already aware of the very rich patterns ordinarily displayed by babies in their use of the &st Not-Me possession. These patterns, being dis- played, can be subjected todirect observation.
There is a wide variation t o be found.in a sequence of events which starts with the new-born infant's fist-in-mouth activities, and that leads eventually on t o an attachment to a teddy, a doll o r soft toy, o r to a hard toy.
Based on a paper read before the British Psycho-Analytical Society on 30th May, 1951. Int. I . Psycho-Awl., Vol. XXXIV, 1953.
i I t is ngzssaly to stress that the word used h m is 'possession' and not 'object'. In the typed version distributed to members I did in fact use the word 'object' (instead of 'posses- sion') in one place by mistake, and this led to confusion in the discussion. It was pointed out
1 that the first Not-Me object is usually taken to be the breast. The reader's attention is drawn to the use of the word 'transitional' in many places by Fairbairn (1952, p. 35.).
D . W . W I N N I C O T T : COLLECTED P A P E R S
It is clear that something is important here other than oral excitement and satisfaction, although this may be the basis of everything else. Many other important things can be studied, and they include:
The nature of the object. The infant's capacity to recognize the object as Not-Me. The place of the object - outside, inside, a t the border. The infant's capacity to create, think up, devise, originate, produce an object. The initiation of an affectionate type of object relationship.
I have introduced the terms 'transitional object' and 'transitional pheno- mena' for designation of the intermediate area of experience, between the thumb and the teddy bear, between the oral erotism and true object relation- ship, between primary creative activity and projection of what has already been introjected, between primary unawareness of indebtedness and the acknowledgement of indebtedness ('Say: ta!').
By this definition an infant's babbling or the way an older child goes over a repertoire of songs and tunes while preparing for sleep come within the inter- mediate area as transitional phenomena, along with the use made of objects that are not part of the infant's body yet are not fully recognized as belonging to external reality.
It is generally acknowledged that a statement of human nature is inadequate when given in terms of interpersonal relationships, even when the imaginative elaboration of function, the whole of fantasy both conscious and uncon- scious, including the repressed unconscious, is allowed for. There is another way of describing persons that comes out of the researches of the past two decades, that suggests that of every individual who has reached to the stage of being a unit (with a limiting mzmbrane and an ou&de and a n inside) it can be said that there is an inner reality to that individual, an inner world which can be rich or poor and can be at peace o r in a state of war.
My claim is that if there is a need for this double statement, there is need for a triple one; there is the third part of $bclTGb$uman being, a part that we cannot ignore, an intermediate area of experiencing, t o which inner reality and external life both contribute. It is an a d is not challenged, because no claim is made on its behalf except that it shall exist as a resting-place for the individual engaged in the perpetual human task of keeping inner and outer reality separate yet inter-related.
It is usual to refer to 'reality-testing', and to make a clear distinction b e tween apperception and perception. I am here staking a claim for a n inter- mediate state between a baby's inability and growing ability to recognize and accept reality. I a m therefore studying the substance of illusion, that which is allowed to the infant, and which in adult life is inherent in art and religion.
230
T R A N S I T I O N A L OBJECTS A N D T R A N S I T I O N A L P H E N O M E N A
We can share a respect for ilhsory experience, and if we wish we may collect together and form a group on the basis of the similarity of our illusory experiences. This is a natural root of grouping among human beings. Yet it is a hall-mark of madness when an adult puts too powerful a claim on the credu- lity of others, forcing them to acknowledge a sharing of illusion that is not their own.
I hope it will be understood that I am not referring exactly to the little child's teddy bear nor to the infant's first use of the fist (thumb, fingers). I a m not specifically studying the first object of object relationships. 1 am concerned with the first possession, and with the intermediate area between the subjec- tive and that which is objectively perceived.
Development of a Personal Pattern
There is plenty of reference in psycho-analytic literature to the progress from 'hand to mouth' to 'hand to genital', but perhaps less t o progress leading to the handling of truly 'Not-Me' objects. Sooner or later in an infant's develop ment there comes a tendency on the part of the infant to weave other-than- me objects into the personal pattern. T o some extent these objects stand for the breast, but it is not especially this point that is under discussion.
In the case of some infants the thumb is placed in the mouth while fingers are made to caress the face by pronation and supination movements of the forearm. The mouth is then active in relation to the thumb, but not in relation to the fingers. The fingers caressing the upper lip, or some other part, may be or may become more important than the thumb engaging the mouth. More- over this caressing activity may be found alone, without the more direct thumb-mouth union. (Freud, 1905. Hoffer, 1949.)
In common experience one of the following occurs, complicating an auto- erotic experience such as thumb-sucking :
1. with the other hand the baby take; an external object, say a part of a sheet or blanket, into the mouth along with the fingers; or
2. somehow or other the bit of cloth1 is held and sucked, o r not actually sucked. Thq objects used naturally include napkins and (later) handker- chiefs, and this depends on what is readily and reliably available ; or
3. the baby starts from early months to pluck wool and to collect it and to use it for the caressing part of the a c t i ~ i t y . ~ Less commonly, the wool is swallowed, even causing trouble ; or
' A recent example is the blanket-doll of the child in the film A Two-year-old Goes to Hospital by James Robertson Ravistock Clinic). cf. also Robertson el al. (1952).
*Here there could possibly be an explanation for the use of the term 'wool-gathering', which means ; inhabiting the transitional or intermediate area.
23 1
D. W. W I N N I C O T T : COLLECTED P A P E R S
4. mouthing, accompanied by sounds of 'mum-mum', babbling.1 anal noises, the first musical notes and so on.
One may suppose that thinking, or fantasying, gets linked up with these functional experiences.
All these things I am calling transitionalphenomena. Also, out of all this (if we study any one infant) there may emerge some thing or some phenomenon -perhaps a bundle of wool or the corner of a blanket or eiderdown, or a word or tune, or a mannerism, which becomes vitally important to the infant for use at the time of going to sleep, and is a defence against anxiety, especially anxiety of depressive type. (Illingworth, 1951.) Perhaps some soft object or cot cover has been found and used by the infant, and this then becomes what I am calling a transitional object. This object goes on being important. The parents get to know its value and carry it round when travelling. The mother lets it get dirty and even smelly, knowing that by washing it she intro- duces a break in continuity in the infant's experience, a break that may destroy the meaning and value of the object to the infant.
I suggest that the pattern of transitional phenomena begins to show at about 6 8 - 1 2 months. Purposely I leave room for wide variations.
Patterns set in infancy may persist into childhood, so that the original soft object continues to be absolutely necessary at bed-time or at time of loneliness or when a depressed mood threatens. In health, however, there is a gradual extension of range of interest, and eventually the extended range is main- tained, even when depressive anxiety is near. A need for a specific object or a behaviour pattern that started at a very early date may reappear a t a later age when deprivation threatens.
This first possession is used in conjunction with special techniques derived from very early infancy, which can include or e%st apart from the more direct autoerotic activities. Gradually i n the life of an infant teddies and dolls and hard toys are acquired. Boys to some extent tend to go over to use hard objects, whereas girls tend to proceed right ahead to the acquisition of a family. It is important to note, however, that there is no noticeable d~flerence between boy and girl in their use of the original Not-Meposscssion, which I am calling the transitional object.
As the infant starts to use organized sounds (mum, ta, da) there may appear a 'word' for the transitional object. The name given by the infant to these earliest objects is often significant, and it usually has a word used by the adults partly incorporated in it. For instance, 'baa' may be the name, and the 'b' may have come from the adult's use of the word 'baby' or 'bear'.
I should mention that sometimes there is no transitional object except the mother herself. Or an infant may be so disturbed in emotional development
cf. Scott (1955).
232
TRANSITIONAL OBJECTS A N D T R A N S I T I O N A L PHENOMENA
that the transition state cannot be enjoyed, or the sequence of objects used is broken. The sequence may nevertheless be maintained in a hidden way.
S ~ n m a r y of Special Qualities in the Relationship 1. The infant assumes rights over the object. and we agree t o this assump
tion. Nevertheless some abrogation of omnipotence is a feature from the start.
2. The object is affectionately cuddled as well as excitedly 4oved and mutilated.
3. It must never change, unless changed by the infant. 4. It must survive instinctual loving, and also hating, and, if it be a feature,
pure aggression. 5. Yet it must seem t o the infant to give warmth, o r to move, o r to have
texture, o r to do something that seems t o show it has vitality or reality of its own.
6. I t comes from without from our point of view, but not so from the point of view of the baby. Neither does it come from within; it is not a n hallucination.
7. Its fate is t o be gradually allowed to be decathected, s o that in the course of years it becomes not so much forgotten as relegated to limbo. By this I mean that in health the transitional object does not 'go inside' nor does the feeling about it necessarily undergo repression. It is not forgotten and it is not mourned. Jtloses meaning, and this is because the transi- tional phenomena have become diffused. have become spread out over the whole intermediate territory between 'inner psychic reality' and 'the external world as perceived by two persons in common', that is to say, over the whole cultural field.
At this point my subject widens out into that of play, and of artistic crea- tivity and appreciation, and of religious feeling, and of dreaming, and also of fetishism, lying and stealing. the origin and loss of affectionate feeling, drug addiction, the talisman of obsessional rituals, etc.
Relationship of the Transitional Object to Symbolism I t is true that the piece of blanket (or whatever it is) is symbolical of some
part-object, such as the breast. Nevertheless the point of it is not its symbolic value so much as its actuality. Its not being the breast (or the mother) is as important as the fact that it stands for the breast (or mother).
When symbolism is employed the infant is already clearly distinguishing between fantasy and fact, between inner objects and external objects, between primary creativity and perception. But the term transitional object, according to my suggestion, gives room for the process of becoming able to accept
233
D . W . W I N N I C O T T : COLLECTED P A P E R S
difference and similarity. I think there is use for a term for the root of sym- bolism in time, a term that describes the infant's journey from the purely sub- jective to objectivity ; and it seems to me that the transitional object (piece of blanket, etc.) is what we see of this journey of progress towards experiencing.
It would be possible to understand the transitional object while not fully understanding the nature of symbolism. I t seems that symbolism can only be properly studied in the process of the growth of an individual, and that it has at the very best a variable meaning. For instance, if we consider the wafer of the Blessed Sacrament, which is symbolic of the body of Christ, I think I a m right in saying that for the Roman Catholic community it iE the body, and for the Protestant community it is a substitute, a reminder, and is essentially not, in fact, actually the body itself. Yet in both cases it is a symbol.
A schizoid patient asked me, after Christmas, had I enjoyed eating her at the feast. And then, had I really eaten her or only in fantary. I knew that she could not be satisfied with either alternative. Her split needed the double answer.
C L I N I C A L D E S C R I P T I O N OF A T R A N S I T I O N A L OBJECT
For anyone in touch with parents and children, there is an infinite quantity and variety of illustrative clinical material.' The following illustrations are given merely to remind readers of similar material in their own experiences.
Two Brothers: Contrart in early Use of Possessions Distortion in use of transitional object. X, now a healthy man, has had to fight his way towards maturity. The mother 'learned how to be a mother' in her management of X when he was an infant and she was able to avoid certain mistakes with the other children because of what she learned with him. There were also external reasons why she was anxious at the time of her rather lonely management of X when he was born. She t o o t h e r job as a mother very seriously and she breast-fed X for seven months. She feels that in his case this was too long and he was very difficult to wean. He never sucked his thumb or his fingers and when she weaned him 'he had nothing to fall back on'. H e had never had the bottle or a dummy or any other form of feeding. He had a very strong and early attachment to the mother herself, as a person, and it was her actual person that he needed.
From twelve months he adopted a rabbit which he would cuddle and his There are excellent examples in the one article I have found on this same subject. Wulff
('Fetishism and Object Choice in Early Childhood', Psychoonol. Quart., 1946, 15, p. 450) is clearly studying this same phenomenon, but he calls the objects 'fetish objects'. It is not clear to,me that this term is correct, and I discuss this below. I did not actually know o f Wulff's paper until I had written my own, but it gave me great pleasure and support t o find the subject had already been considered worthy of discussion by a colleague. See also Abraham (1916) and Lindner (1879).
234
T R A N S I T I O N A L OBJECTS A N D T R A N S I T I O N A L P H E N O M E N A
affectionate regard for the rabbit eventually transferred to real rabbits. This particular rabbit lasted till he was five or six years old. I t could be described as a comforter, but it never had the true quality of a transitional object. I t was never, as a true transitional object would have been, more important than the mother, a n almost inseparable part of the infant. In the case of this particular boy the kind of anxieties which were brought to a head by the weaning a t seven months later produced asthma, and only gradually did he conquer this. I t was important for him that he found employment far away from the home town. His attachment to his mother is still verygowerful. He comes within the wide definition of the term normal, or healthy. This man has not married. Typical use of transitional object. X's younger brother, Y, has developed in quite a straightforward way throughout. He now has three healthy children of his own. H e was fed a t the breast for four months and then weaned without ~ c u 1 t y . l Y sucked his thumb in the early weeks and this again 'made wean- ing easier for him than for his older brother'. Soon after weaning at five to six months he adopted the end of the blanket where the stitching finished. He was pleased if a little bit of the wool stuck out at the corner and with this he would tickle his nose. This very early became his 'Baa' ; he invented this word for it himself as soon as he could use organized sounds. From the time when he was about a year old he was able to substitute for the end of the blanket a soft green jersey with a red tie. This was not a 'comforter' as in the case of the depressive older brother. but a 'soother'. I t was a sedative which always worked. This is a typical example of what I am calling a Transitional Object. When Y was a little boy it was always certain that if anyone gave him his 'Baa' he would immediately suck it and lose anxiety, and in fact he would go to sleep within a few minutes if the time for sleep were at all near. The thumb- sucking continued at the same time, lasting until he was three or four years old, and he remembers thumb-sucking and a hard place on one thumb which resulted from it. H e is now interested (as a father) in the thumb-sucking of his own children and their use of 'Baas'.
The story of seven ordinary children in this family brings out the points, arranged for comparison in the table onipage 236.
I n consultation with a parent it is often valuable to get information about the early techniques and possessions of all the children of the family. This starts the mother on a comparison of her children one with another, and enables her to remember and compare their characteristics a t an early age.
Information can often be obtained from a child in regard to transitional objects; for instance, Angus (1 1 years 9 months) told me that his brother 'has tons of teddies and things' and 'before that he had little bears', and he fol-
The mothex had 'learned from her first child that it was a good idea to give one bottle feed while b m t feeding', that is, t o allow for the positive value of substitutes for herself. and by this means she achieved easier weaning than with X.
235
D . W . W I N N I C O T T : COLLECTED P A P E R S
lowed this u p with a talk about his own history. He said he never had teddies. There was a bell rope which hung down, a tag end of which he would go on hitting, and so go off to sleep. Probably in the end it fell, and that was the end of it. There was, however, something else. He was very shy about this. It was a purple rabbit with red eyes. 'I wasn't fond of it. I used to throw it around. Jeremy has it now. I gave it t o him. I gave it to Jeremy because it was naughty. I t would fall off the chest of drawers. It still visits me. I like it to visit me.' He surprised himself when he drew the purple rabbit. It will be noted that
Innumerable similar soft objects distinguished by colour, length, width, and early subjected t o sorting and classification.
x BOY Y Boy K F g l t e r )
wins {gt! Dummy Donkey (friend) Ee (protective)
Chil- Girl 'Baa' Blanket (reassurance) dren [ Girl Thumb Thumb (satisfaction) o f Y Boy Mimi<* Objects (sorting)
this eleven-year-old boy with the ordinary good reality-sense of his age spoke as if lacking in reality-sense when describing the transitional object's qualities and activities. When I saw the mother later she expressed surprise that Angus remembered the purple rabbit. She easily recognized it from the coloured drawing.
I deliberately refrain from giving more case material here, particularly a s I wish to avoid giving the impression that what I am reporting is rare. In practically every case history there is something to ke found that is interesting in the trans~tional phenomena, o r in their absence (cf. Stevenson, Olive, 1954).
Type of Child
Free Mother-fixated
Late maturity Latent psychopathic
Developing well
THEORETICAL S T U D Y
There are certain comments that can be made on the basis of accepted psycho-analytic theory.
1. The transitional object stands for the breast, or the object of the f i s t relationship.
2. The transitional object antedates established reality-testing. , 3. In relation t o the transitional object the infant passes from (magical)
omnipotent control to control by manipulation (involving muscle erotism and coordination pleasure).
4. The transitional object may eventually develop into a fetish object and
236
T R A N S I T I O N A L OBJECTS A N D T R A N S I T I O N A L PHENOMENA
so persist a s a characteristic of the adult sexual life. (See Wulff's development of the theme.)
5. The transitional object may, because of anal-erotic organization, stand for faeces (but it is not for this reason that it may become smelly and remain unwashed).
Relationship to Internal Object (Klein) It is interesting to compare the transitional object concept with Melanie
Klein's concept of the internal objeqt. The transitional object is not an internal object (which is a mental concept) - it is a possession. Yet it is not (for the infant) an external object either.
The following complex statement has t o be made. The infant can employ a transitional object when the internal object is alive and real and good enough (not too persecutory). But this internal object depends for its qualities on the existence and aliveness and behaviour of the external object (breast, mother figure, general environmental care). Badness or failure of the latter indirectly leads to deadness or to a persecutory quality of internal object. After a persistence of failure of the external object the internal object fails to have meaning to the infant, and then, and then only, does the transitional ob- ject become meaningless too. The transitional object may therefore stand for the 'external' breast, but indirectly so, through standing for a n 'internal' breast.
The transitional object is never under magical control like the internal object, nor is it outside control as the real mother is.
Illusion- Disillusionmen t In order to prepare the ground for my own positive contribution to this
subject I must put into words some of the things that I think are taken too easily for granted in many psycho-analytic writings on infantile emotional development, although they may be understood in practice.
There is no possibility whatever for an infant to proceed from the pleasure- principle to the reality principle or towar$ and beyond primary identification (see Freud, 1923, p. 14),' unless there is a good enough m ~ t h e r . ~ The good enough 'mother' (not necessarily the infant's own mother) is one who makes active adaptation t o the infant's needs, an active adaptation that gradually
.l See also Freud (1921). p. 65. ' One effect, and the main effect, o f failure o f the mother in this respect at the start of an
infant's life, is discussed clearly (in my view) by Marion Milner (1952, p. 181). She shows that because o f the mother's failure there is brought about a premature ego development, with precocious sorting out o f a bad from a good object. The period o f illusion (or my Transitional Phase) is disturbed. In analysis or in various activities in ordinary life an individual can be seen t o be going on seeking the valuable resting-place o f illusion. Illusion in this way has its positive value. See also Freud (1950).
237
D . W . W I N N I C O T T : COLLECTED P A P E R S T R A N S I T I O N A L OBJECTS A N D T R A N S I T I O N A L P H E N O M E N A
lessens, according t o the infant's growing ability t o account for failure of adaptation and t o tolerate the results of frustration. Naturally the infant's own mother is more likely to be good enough than some other person, since this active adaptation demands an easy and unresented preoccupation with the one infant; in fact, success in infant care depends on the fact of devotion, not o n cleverness o r intellectual enlightenment.
The good-enough mother, as I have stated, starts off with an almost com- plete adaptation t o her infant's needs, and as time proceeds she adapts less and less completely, gradually, according to the infant's growing ability t o deal with her failure.
The infant's means of dealing with this maternal failure include the fol- lowing :
The infant's experience, often repeated, that there is a time limit to frustration. At first, naturally, this time limit must be short. A growing sense of process. The beginnings of mental activity. The employment of auto-erotic satisfactions. Remembering, reliving, fantasying, dreaming; the integrating of past, present, and future.
Ifall goes well the infant can actually come t o gain from the experience of frustration, since incomplete adaptation t o need makes objects real, that is t o say hated as well as loved. The consequence of this is that i f a l l goes well the infant can be disturbed by a close adaptation t o need that is continued too long, not allowed its natural decrease, since exact adaptation resembles magic and the object that behaves perfectly becomes no better than an hallucina- tion. Nevertheless at the start adaptation needs t o be almost exact, and unless this is so it is not possible for the infant t o begin t o develop a capacity t o ex- perience a relationship to external reality, or eve,-to form a conception of external reality.
Illusion and the Value of Illusion The mother, at the beginning, by an almost 100 per cent adaptation affords
the infant the opportunity for the illusion that her breast is part of the infant. It is, as it were, under magical control. The same can be said in terms of infant care in general, in the quiet times between excitements. Omnipotence is nearly a fact of experience. The mother's eventual task is gradually t o disillusion the infant, but she has no hope of success unless at first she has been able t o give
' sufficient opportunity for illusion. In another language, the breast is created by the infant over and over again
out of the infant's capacity t o love or (one can say) out of need. A subjective
238
phenomenon develops in the baby which we call the mother's breast1 The mother places the actual breast just there where the infant is ready t o create. and a t the right moment.
From birth, therefore, the human being is concerned with the problem of the relationship between what is objectively perceived and what is subjectively conceived of, and in the solution of this problem there is no health for the human being who has not been started off well enough by the mother. The intermediate area to which I amyeferring is the area that is allowed to the infant between primary creativity and objective perception based on reality-testing. The transitional phenomena represent the early stages of the use of illusion, with- out which there is no meaning for the human being in the idea of a relation- ship with an object that is perceived by others as external t o that being.
The idea illustrated in Fig. 19 is this: that at some theoretical point early in the development of every human individual an infant in a certain setting pro- vided by the mother is capable of conceiving of the idea of something which would meet the growing need which arises out of instinctual tension. The in- fant cannot be said to know at first what is t o be created. At this point in time the mother presents herself. In the ordinary way she gives her breast and her potential feeding urge. The mother's adaptation to the infant's needs, when good enough, gives the infant the illusion that there is an external reality that corresponds t o the infant's own capacity t o create. In other words, there is an overlap between what the mother supplies and what the child might conceive of. T o the observer the child perceives what the mother actually presents, but this is not the whole truth. The infant perceives the breast only in so far as a breast could be created just there and then. There is no interchange between the mother and the infant. Psychologically the infant takes from a breast that is part of the infant, and the mother gives milk t o a n infant that is part of her- self. In psychology, the idea of interchange is based on an illusion.
In Fig. 20 a shape is given t o the area of illusion, t o illustrate what I consider t o be the main function of the transitional object and of transitional pheno- mena. The transitional object and the transitional phenomena start each human being off with what will always be important for them, i.e. a neutral area of experience which will not be challenged. Of the transitionalobject it can be said that it is a matter of agreement between us and the baby that we will never ask the question 'Did you conceive of this or was it presented to youfrom
I include the whole technique of mothering. When it is said that the first object is the breast, the word 'breast' is used, I believe, to stand for the technique of mothering as well as for the actual flesh. It is not impossible for a mother to be a good-enough mother (in my way of putting it) with a bottle for the actual feeding. If this wide meaning of the word 'breast' is kept in mind, and maternal technique is seen to be included in the total meaning of the term, then there is a bridge forming between the wording of Melanie Klein's state- ment of early history and that of Anna Freud. The only difference left is one of dates, which is in fact an unimportant difference which will automatically disappear in the c o w of time.
239
F I G . 19 FIG. 20
without?' nip important point is that no decision on this point is expected. Z7ae question is not to be formulated.
This problem, which undoubtedly concerns the human infant in a hidden way at the beginning, gradually becomes an obvious problem on account of the fact that the mother's main task (next to providing opportunity for illu- sion) is disillusionment. This is preliminary to the task of weaning, and it also continues a s one of the tasks of parents and educators. In other words, this matter of illusion is one which belongs inherently to human beings and which no individual finally solves for himself or herself, although a theoretical under- standing of it may provide a tl~eoretical solution. If things go well, in this gradual disillusionment process, the stage is set for the frustrations that we gather together under the word weaning; but it should be remembered that when we talk about the phenomena (which Klein has specifically illumin- ated) that cluster round weaning we are assuming the underlying process, the process by which opportunity for illusion and gradual disillusionment is pro- vided. If illusiondisillusionment has goneastraytke infant cannot attain t o s o normal a thing a s weaning, nor to a reaction to weaning, and it is then absurd to refer to weaning a t all. The mere termination of breast feeding is not a weaning.
We can see the tremendous significance of weaning in the case of the normal child. When we witness the complex reaction that is set going in a certain child by the weaning process we know that this is able to take place in that child because the illusion-disillusionment process is being carried through so well that we can ignore it while discussing actual weaning.
It is assumed here that the task of reality-acceptance is never completed, that no human being is free from the strain of relating inner and outer reality, and that relief from this strain is provided by an intermediate area of experi- ence which is not challenged (arts, religion, etc.). (cf. Riviere. 1936). This
240
T R A N S I T I O N A L OBJECTS A N D T R A N S I T I O N A L PHENOMENA
intermediate area is in direct continuity with the play area of the small child who is 'lost' in play.
In infancy this intermediate area is necessary for the initiation of a relation- I 1 ship between the child and the world, and is made possible by good enough I mothering at the early critical phase. Essential t o all this is continuity (in I time) of the external emotional environment and of particular elements in the
physical environment such as the transitional object o r objects. The transitional phenomena are allowable t o the infant because of the
parents' intuitive recognition of the strain inherent in objective perception, and we d o not challenge the infant in regard to subjectivity o r objectivity just here where there is the transitional object.
Should an adult make claims on us for our acceptance of the objectivity of his subjective phenomena we discern or diagnose madness. If, however, the adult can manage to enjoy the personal intermediate area without making claims, then we can acknowledge our own corresponding intermediate areas, and are pleased to find examples of overlapping, that is t o say common ex- perience between members of a group in art o r religion o r philosophy.
I wish t o draw particular attention t o the paper by Wulff, referred t o above, in wifich clinical material is given illustrating exactly that which I am referring to under the heading of transitional objects and transitional phenomena. There is a difference between my point of view and that of Wulff which is reflected in my use of this special term and his use of the term 'fetish object'. A study of ~ u l f f ' s paper seems t6 show that in using the word fetish he~has taken back to infancy something that belongs in ordinary theory t o the sexual perversions. I am not able to find in his article sufficient room for the consideration of the child's transitional object as a healthy early experience. Yet I d o consider that transitional phenomena are healthy and universal.
I Moreover if we extend the use of the word fetish t o cover normal phenomena we shall perhaps be losing some of the value of the term.
I would prefer to retain the word fetish to describe the object that is em- ployed on account of a delusion of a maternal phallus. I would then go further and say that we must keep a place for the dlusion of a maternal phallus, that is t o say, an idea that is universal and not pathological. If we shift the accent now from the object on t o the word illusion we get near t o the infant's transi- tional object ; the importance lies in the concept of illusion, a universal in the field of experience.
Following this, we can allow the transitional object t o be potentially a maternal phallus but originally the breast, that is t o say, the thing created by the infant and at the same time provided from the environment. In this way I
b think that a study of the infant's use of the transitional object and of transi- tional phenomena in general may throw light on the origin of the fetish object
b and of fetishism. There is something t o be lost, however, in working
D . W . WINNICOTT: COLLECTED P A P E R S
backwards from the psychopathology of fetishism to the transitional phenomena which belong to the beginnings of experience and which are inherent in healthy emotional development.
S U M M A R Y
Attention is drawn to the rich field for observation provided by the earliest experiences of the healthy infant as expressed principally in the relationship to the first possession.
This first possession is related backwards in time to auto-erotic phenomena and fist- and thumb-sucking, and also forwards t o the first soft animal or doll and to hard toys. It is related both to the external object (mother's breast) and to internal objects (magically introjected breast), but is distinct from each.
The transitional objects and transitional phenomena belong to the realm of illusion which is at the basis of initiation of experience. This early stage in development is made possible by the mother's special capacity for making adaptation to the needs of her infant, thus allowing the infant the illusion that what the infant creates realljexists.
This intermediate area of experience, unchallenged in respect of its belong- ing to inner or external (shared) reality, constitutes the greater part of the in- fant's experience and throughout life is retained in the intense experiencing that belongs to the arts and to religion and t o imaginative living, and to creative scientific work.
A positive value of illusion can therefore be stated. An infant's transitional object ordinarily becomes gradually decathected,
especially as cultural interests develop. In psychopathology : Addiction can be stated in terms of regression to the early stage at which the
transitional phenomena are unchallenged. Fetishism can be described in terms of a persistence of a specific object or
type of object dating from infantile experience in the transitional field, linked with the delusion of a maternal phallus.
Pseudologia fmfastica and thieving can be described in terms of a n indi- vidual's unconscious urge to bridge a gap in continuity of experience in respect of a transitional object.