The Idealisation of the Twin Relationship

The Idealisation of the Twin Relationship

Vivienne Lewin


Siblings have an important place in our inner world. Twins are siblings of a distinctive kind and while their relationship is at least as important as those of other siblings, the twin relationship also poses problems for the developing twins, difficulties that are unique to twins. The early dyadic and triadic relationships of the infant with its parents are disrupted by the presence of a twin. For all parties, the presence of two infants of the same age complicates the dynamics by creating an extra set of relationships to negotiate. Twin infants have to negotiate the triangular relationship between infant, mother and father; and also the relationship with the other twin and the awareness of the other twin’s relationship with mother, father, and with itself. For twins the early dyadic stage is essentially another triangle from the start.

Twins seem to fascinate us and I hope to explore the reasons for this. I have found it interesting that on many occasions, as soon as I have mentioned in conversation that I was writing about twins, the person I was talking to seemed to feel compelled to tell me an anecdote or story about twins, excited by some aspect of the twin relationship. Most often this interest has related to the uncanniness of two people who look so similar, or to the extraordinary, even apparently telepathic, communication between twins. Rank (1971) has written about our sense of the uncanny when we see two people who look the same. He explores the significance of “the double” in mirrors, shadows and phantasies. He suggests that the concept of the double is linked with death – that by duplicating oneself we hope to avoid the inevitability of death. Fanthorpe (2000) refers to “the strangeness of the other who looks the same”.

As an recent example of my being offered stories about our fascination with the sameness of twins, a neighbour (anonymously) put through my door a newspaper article from the Evening Standard about a pair of twins who each underwent extensive plastic surgery to make them look more alike (Prigg, M., 2004). These monozygotic (same egg) twins disliked the differences between them, particularly as they were frequently referred to as the “ugly twin” and the “pretty twin”. The surgical remodelling minimised these differences, as a result of which the twins claim to have been brought “closer together than ever”. The twins subsequently moved into the same home. Rather than being able to enjoy the difference that might have brought each twin sister a sense of individuality within the twin relationship, they sought sameness. This demonstrates the narcissistic aspect of a twin relationship, the seeking of other as self, and of self in the other. The idealisation of sameness obliterates difference and the value of difference, and denies the need for an individual sense of self. Perhaps the need for sameness in these twin sisters was exacerbated by the splitting that seemed to define each twin. If the “ugly twin” was deemed to carry the “bad” aspects, while the “pretty twin” had the “good” ones, they might have felt that only by bringing the two selves together as two halves of a unit could the split be lessened. They enacted this by creating superficial sameness through surgery, and by moving in together. It seems they did not feel it was possible to redistribute the projected qualities of goodness and badness between them in a way that could be integrated into each individual personality.

When I first started my research on twins, I came across an account of the extraordinary similarities that had been observed between twins who had been separated at birth and reared apart, and then re-united as adults. This research was initially conducted by Tom Bouchard Jr. (Wright, 1997). Like so many others I was fascinated to read of these uncanny accounts, and pondered about the underlying implication that the way we are, our identity, all lies in our genes. The study was called the Minnesota Study of Twins Reared Apart. (I will say more about this below.) This excited research on twins is the sort of material that really feeds our fascination with twins. We want to find an uncanny sameness in twins, particularly in monozygotic twins. It seems to satisfy a deep longing in us.

I want to give another example of our fascination with twins and how this blends with the reality of twin development. Elvis Presley was a twin. Apparently his twin brother Jesse was born dead and the family have recounted stories about their surprise at the emergence of a second live baby. Dundy (1985) has written a biography of Elvis and his mother. I have used her book to illustrate both the idealisation of twins (and, of course, the idealisation of Elvis) and the reality of the enduring nature of the twin relationship in the internal world and its manifestations in all aspects of life. Dundy writes:

“Elvis was born a twin….The fact that the mystery of death was attendant at his birth, that the very beginning of his life marked the end of his brother’s, affected him throughout his life in a way that people who are not twins would find hard to understand.” (Dundy, 1985: 67)

It is true that the death of Elvis’s twin would most likely have been an important issue in his developing sense of self. Whatever residual sensory memory Elvis would have had of his twin in utero, the birth stories told by his parents and family would certainly have enhanced his sense of himself as a lone twin. However, Dundy then goes on to discuss the mystery of such twin bonding, citing the above-mentioned studies of twins reared apart and the apparent extraordinary similarities in the details and particularities of the lives of the separated twins. These studies have now been largely discredited. Joseph (2004) has examined twin studies and the erroneous conclusions based on some of the research. I mention it here because I think our fascination with twins might lead us to make assumptions about the genetic aspects of development based on unsound experimentation using the twin method and what Joseph calls the “gene illusion”. Joseph suggests that these studies are skewed by many factors, including researcher bias; the possible omission of pairs of twins separated at birth who might turn out to have no greater similarities than any other people; the amount and quality of contact between the twins before the study was conducted; and any vested interest the twins themselves might have had in appearing so alike. The fundamental issue that is not addressed by these studies is whether factors other than genetic identity have contributed to the observed similarities.

Twins are usually divided into two distinct groupings: monozygous or same-egg twins, and dizygous or different-egg twins. Thus, monozygous twins originate from one egg that splits at a very early stage into two embryos. Dizygous twins are the result of the simultaneous fertilisation of two eggs. The twin method of research relies on the assumption that monozygous twins have identical genes, while dizygous twins have a genetic similarity no closer than that of other siblings. The suggestion that monozygotic twins have identical genes has been somewhat overemphasised in the nature-nurture debate. Gringras and Chen (2001) have reviewed the scientific research on twins. They conclude that monozygotic twins do not have exactly identical genes. Although both twins develop from the splitting of one egg, both genetic and physical differences occur in monozygous twins as a result of various factors, including intrauterine effects, mutations in the genes, and epigenetic modifications within the chromosomes (chromosomes that are chemically altered after formation). The development of each twin will be affected according to which genes are activated, and this is altered by both environmental and hormonal factors. Our genes provide us with the potential for development within certain parameters, but the direction that development takes would depend on other factors. Hence the differences we observe in monozygous twins in both appearance and personality.

It is evident to anyone who is a twin or has known or worked with twins that no two twins are truly “identical” and I will therefore refer to twins according to their zygosity rather than using the terms “identical” and “fraternal”. (I think these terms are useful only as regards the nature of the twinning processes that occur between the twins, either seeking sameness or allowing difference.) Whether mono- or dizygotic, the processes of twinning occur between twins to variable degrees. The nature and degree of the twinning processes within each pair of twins will affect the development and the nature of the twin relationship. It is important to note that twinning processes are not unique to twins. They also occur between infant and mother; and between singletons and other siblings, younger or older. I will talk later about our earliest twinning experiences.

To go back to Elvis, as Dundy (1985) notes, twin bonding is indeed an important factor in the twin relationship. Twin bonding is the result of various developmental and emotional factors including the closeness of the twins in age and the way in which this affects the relationship between the twins and their relationships with the mother and father. Where twinning processes between the twins is extensive, the twin bond will become both central to development, and would also limit the development of a separate identity for each twin, leading to a more enmeshed twinship. The twin relationship is powerfully represented in the internal worlds of the twins and the internal twinship is enduring. It is not, however, a mystical bond of an order unknown in other relationships. As I will discuss later, I believe the tenaciousness of the twin bond has its roots in the infant’s earliest relationship with its mother.

Dundy (1985) links Elvis’s preoccupation with gazing at himself in mirrors as a teenager, with Narcissus’s gazing into the stream looking for the image of his dead twin. Pausanias tells a story about Narcissus that is contrary to the popular myth of Narcissus dying at the stream unable to relinquish the reflection of his own beauty. Instead, Pausanias suggests that Narcissus gazed into the stream to comfort himself for the death of his twin sister. Narcissus knew that the image he saw was a reflection of himself, but it was so similar to his lost twin’s image that seeing it comforted him. Here we see again the linking of mirrors, reflections and doubles with death, as suggested by Rank (1971). It may be that Elvis, like Narcissus, looked and longed for his dead twin; but we also know that teenagers have a predilection for gazing at themselves in mirrors.

Dundy notes that Elvis’s middle name was Aron, while that of his dead twin was Garon. (It is another feature of our fascination with twins that they are so often given very similar or rhyming names.) The death of a twin at birth generates a confusion of feelings in an immature infant. These feelings would include a sense of loss, guilt at survival when the other twin has died, but also feelings of triumph in survival and triumph over the dead twin. The task of dealing with this loss is made more difficult for the infant by the fact of its emotional immaturity. The infant has not yet developed the capacity to mourn its lost twin. Woodward (1998) suggests that the ability to mourn is linked with the capacity to verbalise, and that before this capacity develops, the infant twin survivor will have difficulty dealing with its loss.  Apparently Elvis always signed his name as Elvis aron Presley, that is, his middle name had a lower case “a” instead of a capital letter. Dundy suggests that he did so to leave space for the missing G in his twin’s middle name “indicating how strongly Elvis felt his twin’s absence” (Dundy, 1985: 69).

Dundy goes on to develop this theme. She suggests that Elvis’s survival in the face of his twin’s death created in him a sense of power, and that all his life he was engaged in a conflict between powerfulness and powerlessness. “Elvis might relate to friends and lovers with the intimate dependency of a twin looking for his other half – but he would always be the dominant one” (Dundy, 1985: 69). Thus Dundy is illustrating the way in which the enduring nature of the internal twinship is expressed in all other relationships. The death of his twin and the internal representation of this twinship would of course, also have been affected by Elvis’s relationship with his mother. A mother faced with the need to welcome a live infant at the same time as she mourns a dead one has a difficult task. The surviving twin would always remind her of the dead baby and so the life and death twinning would endure in the minds of both mother (and father) and the surviving twin.

Elvis was apparently obsessed with comic books in which, like so many boys, he identified with the heroes. He kept all his comics in meticulous order, and even retained them as an adult. The comic book characters that most interested Elvis were those that had a dual personality. In these heroes there is a twinship that resides within one person (as it does in a twin whose other twin has died), the one side of the twinship being powerful, the other powerless. For these comic characters, the powerful self was always a secret self and the powerless one dull, crippled or weak.

Amongst these comics were Captain Marvel and his alter ego Billy Batson; Batman and Bruce Wayne; Superman and Clark Kent; Spirit and Dandy Colt; and Plastic Man and The Eel. Apparently Billy Batson (Captain Marvel’s alter ego) even had a twin sister, Mary, who also had a twinned powerful alter ego called Mary Marvel. Dundy claims that Elvis modelled himself primarily on another comic book twinned hero, Captain Marvel Jr., who had a twin self called Freddy – a poor, crippled newsboy. She refers to Elvis’s “twin-fusion” (Dundy, 1985: 71) with Captain Marvel Jr. and wonders whether Elvis felt crippled by his twin’s death. She also notes that the powerless twin, Freddy’s, surname was Freeman, echoing Captain Marvel Jr.’s ability to fly high over the earth. I have found that it is not uncommon when a twin has died at birth for the surviving twin to believe that the dead twin has been freed by death, leaving the surviving twin to suffer both the difficulties in mourning the loss of the other twin and the exigencies of life. Dundy suggests that Elvis so identified with his comic hero that he adopted Captain Marvel Jr.’s hairstyle, black hair with a lock falling forward, and his stance, standing with legs astride. Elvis even used the striking lightning emblem that appeared each time that Captain Marvel Jr emerged in the comic, in his personal jewellery and as the emblem on his private plane. Again, we need to be wary of making links that are too simple. The lightning sign is very common in all sorts of graphic displays, and is not necessarily connected with Captain Marvel Jr..

What about Elvis’s relationship with his mother? It is clear from the book that Elvis’s mother, Gladys, came from a very poor and deprived background, and was intensely possessive of her surviving twin son throughout his life. Dundy describes Gladys’ “passionate concentration which deepened into a powerful intensity when her son was not there” (Dundy, 1985: 73). Gladys could not bear to be parted from her son even when he was an adult. We discover that Gladys’ mother and grandmother had also died in the year of Elvis’s birth and his twin’s death. All these losses would have affected Gladys’s relationship with her surviving son. It does seem that powerful twinning processes developed and persisted between mother and son, based not only on the death of Elvis’s twin brother, but also on circumstances that made the ordinary separateness between infant and mother more difficult to attain.

Where twinning processes are extensive, whether between twins or as seem to have occurred between Elvis and his mother, separateness poses a threat to the identity of each “twin”. The twins feel themselves to be inseparably bound to each other and feel that their psychic wholeness would be damaged or destroyed by separateness. Dundy notes that at his mother’s graveside, Elvis cried out “Goodbye, darling, goodbye. I love you so much. You know how much. I lived my whole life for you….Oh God, everything I have is gone.” (Dundy, 1985: 347). The unconscious narcissistic elements of the twinning are very powerful.

In my recently published book, The Twin in the Transference (Lewin, 2004: 68), I wrote about an account by George Engel of a twin relationship and the surviving twin’s anniversary reaction to his twin’s death, based on unconscious factors (Engel, 1975):

“Engel’s anniversary reaction to his twin’s death after a heart attack is an expression of how powerful this unconscious narcissistic entanglement in twins can be, a persistent element at the core of a more healthy twin relationship. Although Engel seems to believe that he and his twin had dealt with their rivalry, had tamed it to manageable proportions by establishing an equivalent mutual aggression and developing a complementary relationship, it seems that unconscious elements prevailed. After his twin died, Engel waited with a sense of prescience for his own equivalent heart attack. This occurred one day short of 11 months after his twin had died. His immediate reaction was one of relief – he no longer had to anticipate the heart attack, “the other shoe had fallen” (Engel, 1975: 25). He could now exonerate himself of the phantasied crime of killing his brother and the associated guilt (indicated in a phantasy that he experienced while in hospital). Engel had recognised his murderous wishes towards his twin brother, but he had not escaped the twinship. In the Judaic tradition, the period of mourning is exactly one day short of 11 months, and Engel’s heart attack thus did indeed occur on the anniversary of his twin’s death. In so doing, it phantastically united him with his twin again, recreating their narcissistic bond. The power of the unconscious phantasy of oneness with his twin showed itself with force and accuracy.”

Along the same lines, it is interesting to note that Elvis was 42 when he died, the same that age his mother claimed to be at her death. (She was actually 48, but had hidden her true age.) Elvis’s twinning with his mother would have made her death all the more unbearable to him. We might also interpret his extreme distress at her death as the reliving of the death of his twin, an event that apparently haunted him all his life.

I have found the material about Elvis interesting not just because it is about a man who is something of an idol for millions of people, but also because of the recurring twin theme, both the idealisation of twins and the reality of the enduring nature of a twin relationship. It is an example of our fascination with twins, and the ways in which we seek to emphasise the idealised twinning between them. Dundy positively fishes out twinning themes in Elvis’s life. As an example of Elvis’s eternal twinning, Dundy traces the number of twins that are featured in Elvis’s films. She writes with reverence and excitement not only about Elvis, but also about his twinship with both his dead twin and his mother. It is true that a twin relationship endures throughout the lives of the twins, even after one twin dies. The twin relationship is a primary relationship, on a par with the parental relationships in terms of its developmental importance. However, I think that the idealisation of twins and the wish to attribute magical qualities to the twin relationship has other roots, as I will now explore.

The perceived closeness of the relationship in some twins and the intimate understanding between them is something we all long for. Twins have frequently been observed to use a private language in their communications with each other (crytophasia). The private language of twins is usually a distortion of ordinary language, and it may also consist of made up words, signals and gestures. Here is a quote from a twin writing in a student magazine:

“Growing up alongside a peer causes such mysteries as “twin babble,’ a private language created between twins that are learning to talk. One of these words was “picky,” which was our term for soda. Our disposition to talk in our own private language was so strong in fact that our mother had to refuse us our requests until we used the correct, English word.

“Since our birth it has been evident that my sister and I share our own private world that has never needed to include anyone else. Even though the days of twin babble are long gone, we still have conversations consisting of two words, none finished sentences, that only [we understand]. When we are alone, we are content with one another’s company. Silence is comfortable, and the topics of conversation are endless. We rarely talk about other people’s lives because they have never interested us since we became accustomed to talking about topics that pertain to life from an early age. It’s difficult to understand what you will never experience, but people are always trying to uncover the mysteries of twins.” (Hillary S., 2004)

Thus, Hillary S. describes very clearly the closed system created by some twins in their twinship, one in which their secret language is used not only to communicate intimately with the other twin, but also to exclude anyone external to the twinship. The intimate communication between twins is sometimes attributed to telepathy. However, I understand that tests to establish whether or not twins communicate telepathically have concluded that while they do sometimes have an unusually close degree of understanding, this is due to non-verbal signals and their private language, rather than to any telepathic contact (Wright, 1997). I believe that the closeness of communication between twins resonates with us for a reason that is central to our idealisation of twins – that is, a longing for perfect understanding as is perceived to happen between twins.

The longing for perfect understanding and our fascination with twins originate in an infant’s earliest experiences with its mother. Klein (1963) suggests that the close preverbal contact between the unconscious of the mother and that of the infant creates an experience of complete understanding without words. However, this “perfect understanding” is intermittent. Mother is not always there, and she does not always get it right. She would at times be tired or pre-occupied. Even at times when the infant has been satisfied, persecutory/paranoid anxieties would arise. The inevitable and irretrievable loss of an understanding without words creates an unsatisfied longing and leads to a sense of internal loneliness. As Klein notes, the integration of the good and bad breast, and the good and bad aspects of the self can never be securely established. Some polarity between the life and death instincts persists. As a result, “complete understanding and acceptance of ones own emotions, phantasies and anxieties is not possible and this continues as an important factor in loneliness. The longing to understand oneself is also bound up with the need to be understood by the internalised good object.” (Klein, 1963: 302).

This longing is expressed in the universal phantasy of having a twin. The infant creates a phantasy of the breast (the maternal function) using the processes of splitting and projective identification, in the hope of finding again the understanding without words. Thus the phantasy breast is a twin of the infant, created by the projection of aspects of the infant itself. The phantasy twin represents those parts of the infant that have not been understood, and which the infant is longing to regain. Bion (1967) suggests that the breast is the infant’s first imaginary twin. The creation of this phantasy twin-breast provides the infant with the illusion of attaining the desired perfect state of mind. Thus, the experience of being understood without words is at the heart of the ubiquitous longing for a twin.

The processes of twinning, that is the creation of the other as self, may be used in the search for understanding, as in the relationship of the infant and the breast, described above. However, twinning may also be used to try and rid oneself of unwanted aspects of the self, splitting off these undesirable parts and disowning them as belonging to the other. The twinning processes between the infant and the breast are echoed in the twinning processes between twins themselves in both the above modes. Thus, twins engage in twinning processes between them as the infant does with its mother, the other twin being perceived as either providing perfect understanding (as exemplified by Castor and Pollux in an idealised twinship) or as carrying the unwanted dissociated aspects of the self.  In this latter situation, the other twin represents the hated aspects of the self and is felt to be alien and undesirable. (The rivalry and hatred between Jacob and Esau are examples of this.)

Thus, in addition to the companionable twinship that is so important to the twins, each twin may also perceive the other twin as the embodiment of the breast-twin. Two factors thus account for the enduring nature of the twin relationship and for the idealisation of twin relationship both by the twins themselves and by outsiders. The overlay of the breast-twin and the  actual twin, and the importance of the actual twin as a primary object, together create an indelible internal twinship.

In childhood, many children “find” an imaginary companion to whom they relate as if they were actual friends. The imaginary companion essentially represents aspects of the child, and is therefore a phantasy twin of the child. Twinning and the processes of splitting and projective identification are central to the creation of a phantasy twin or imaginary companion in childhood, as they are in the infant. However, the creation of an imaginary childhood twin or companion is a more conscious affair. The child knows it has created its imaginary companion, and may use it in various ways. The imaginary twin may help the child combat loneliness at the loss of mother to a younger sibling (Burlingham, 1945). It may provide the child with the illusion of greater strength and power. In being a repository for the split-off parts of the child, the imaginary companion may serve either a defensive or an organising function. In the organising mode, the imaginary twin may be used as a temporary alter ego that will later be integrated into the child’s psyche. In defensive mode, the imaginary twin would remain a split off part of the child and would cause developmental problems (Nagera, 1969). Thus, the twinning processes in the infant are unconscious and persist to a greater or lesser extent throughout life whether the infant is a singleton or a twin. In contrast, an imaginary childhood companion usually fades when the child feels more able to tolerate the split-off aspects of the self and integrate them into the developing ego. Occasionally an imaginary childhood twin or companion persists into adulthood, and this would be indicative of a developmental difficulty.

It is important in considering twin relationships to distinguish between the two aspects of the twinship. On the one hand there are the ‘special’ aspects of the relationship between twins that are the result of the unparalleled closeness and companionship of the twins. On the other hand, the more narcissistic elements of the twinship may result in the idealisation of a twin relationship that seems to exemplify and embody an understanding without words. In the former, companionable type of relationship, the loss of the twin-breast is acknowledged, and the lost ideal object is mourned and relinquished (although never completely, hence our ubiquitous longing for perfect understanding). In the latter, the ideal twin-breast becomes concretely identified with the other twin, and the recognition of the loss of perfect understanding is evaded. This may lead not only to an enmeshed twinship in which each twin feels dependent on the other twin for its identity, indeed for its survival; it also creates a relationship in which the twins feel trapped.

Both mother and a twin may become primary objects for twin infants. Where the mother is used predominantly in development, the infant will internalise a good integrating object by introjecting the maternal function. This good internal object will serve as the core of the developing ego. Alongside this will be the internalised twin relationship as an important and indelible internal object relationship, but one that does not replace the maternal object relationship. However, in other situations, whether due to maternal neglect or absence, or to the infant’s limitations in its capacity to use mother’s help, the infant twins may turn to each other for their developmental needs. The internal twinship would then be the predominant object relationship. However, as the twins are equally immature, they would not be able to provide each other with a sufficiently containing object, one that would enable and allow them to grow towards separateness and individuality, as would a maternal object relationship. While older siblings may provide some greater degree of containment, the lack of generational or even sibling-age difference in twins would be a limiting factor in this regard. The result would be a lasting twinship of a narcissistic kind.

A degree of narcissistic twinning remains as a residue in all twin relationships, including those where there has been a greater degree of development of individual identities in the twins. There is also another source of confusion in the development of a sense of self for twins. The closeness in age of the twins will affect not only their relationship with each other, but also mother’s (and father’s) relationship with each of them. The narcissistic twinning and confusion of identity between the twins may be reinforced by each twin’s experience with mother and father, especially where the twins are so similar that it is difficult for the parents to distinguish between them. The parents’ feelings in relation to each baby might, so to speak, be directed at the ‘wrong’ infant. The twins may then have the experience of having to take in something that belongs to the other twin, creating a sense of a sharing of self with the other twin. This interference in, or a lack of, containment, would lead to a further confusion of identity as the infant twin’s communication with mother seems to have been misunderstood or misdirected.

The parental attitude towards each twin would have powerful effects on it. Bettelheim (1955) describes the way in which twins he was treating in his school were “turned into” the “good” or “bad” twin as a result of “neurotic” projections from their parents. This transfer of projections between one child and the other occurred only with twins, not with other children in the family.  He notes that when the delinquent child had been rehabilitated during his stay at the school, the other, previously “good” twin started to display the symptoms of neurosis projected by the parents.

“In both these situations, as the “bad” delinquent twin became rehabilitated during his stay at the School, the previously “good” sibling slowly turned into the “bad” one. In these cases, where both children probably had very similar life histories, neurotic attachments and needs were seemingly more readily transferred from one to the other. Even then, however, the neurosis or delinquency of the “good” twin who turned “bad”, though of the same type, was relatively mild compared with that of the twin placed in the School. Apparently the original “bad” twin shielded his twin sibling during their earlier years from the full impact of parental neurotic involvement. When, on removal of the “bad” twin, these neurotic feelings were directed against the one remaining at home, the latter met this crisis with a personality much better able to withstand it, because of his previous more fortunate life experience.” (Bettelheim, 1955: 492).

Although Bettelheim refers to the “very similar life histories” of the twins concerned, he goes on to point out that the twins’ experiences in relation to the parents were actually different, and furthermore that this had an enduring effect on the capacities of the children. The twin that had had the better start with his parents was more able to tolerate the parental projections without too great a disturbance. So again we see that the idea that twins are exposed to exactly the same developmental factors through their shared environment does not hold true. Each twin has its own unique experiences within the family environment and as a consequence, each will have an individual pattern of psychic development, whatever the degree of overlap described above. The twins may distribute qualities between themselves either within an enmeshed system, keeping the twins bound to each other, or in a more separate twinship, sometimes through the development of complementarities to each other.

The creation of a phantasy twin as described above is a narcissistic affair. The processes of splitting and projective identification may operate between parts of the self, creating an internal twin pair, or between the self and an external object – either the mother/breast or the other twin. The creation of a phantasy twin would be part of normal development as the infant engages with its mother. As a result of maternal containment the infant would gradually relinquish this idealised twin and develop towards an individual and separate identity. In the presence of a twin, the process is more complex and may be more problematic. The interaction with the other twin who is always present in the mind of both mother and baby adds an additional dimension to the development and the internal world of the infants. Where the twins engage with each other in extensive mutual projective and introjective identification there will be a blurring of individual boundaries between them and the twinship itself may become a narcissistic system encompassing both twins. The other twin may come to be perceived as an embodiment of the phantasy twin as a narcissistic twin system evolves.

An omnipotently fused narcissistic relationship of the kind described by Rosenfeld (1964) may exist between twins. The self becomes so identified with the incorporated other twin that all sense of a boundary between self and twin, and of an identity separate from the other twin, is denied. The twins may believe that the twinship offers them a self-sufficient system and that they have no need of any other objects. The enmeshed narcissistic twinship may then be used as a psychic retreat (Steiner, 1993). The twinship retreat may offer apparent safety from unbearable anxiety associated with development towards the depressive position and separateness for each twin within the twin relationship. However, alongside the apparent safety offered by the psychic retreat, the twins would feel trapped in the narcissistic twinship. Emerging from the psychic retreat would confront the individual twin with anxieties of an intense and possibly psychotic nature. In the extreme this may result in a sense of fragmentation of the self. The lack of maternal containment, whatever the cause, would be a potent factor promoting the use of the twinship as a psychic retreat.

The use of such a twinship retreat would enable the twins to avoid an experience of dependence on a maternal object rather than the twin. The inter-twin dependency would predominate, frequently excluding the mother. However, it is dependence on a maternal object that would enable the twins to develop separately within the twin relationship and free them from the enmeshed twinship. In therapy or analysis, the analyst would be perceived as a twin in all the manifestations of the twin relationship. This provides the analyst with vital information about the twin patient’s internal world of relationships and formative experiences.  An analysis of the twin transference relationship is essential to the successful resolution of the transference relationship and the analysis. While the patient may perceive the analyst as a twin in the transference relationship, the capacity of the analyst to use her understanding of the twin transference is a parental function. However, the patient may avoid any recognition of a parental transference relationship with the analyst and may instead use the analytic twinship as a retreat to avoid a developmental (parental) relationship with the analyst.

The idealisation of the twinship may then be enacted in the analytic relationship, as the analyst becomes the transference twin instead of analysing it. The sense of the “specialness” of the twinship will become manifest in the enacted analytic twinship, as the analyst colludes with the twin patient to avoid the painful developmental realisations of difference and discord, smallness and dependency; and the loss of the ideal twin-breast.

The overvaluation of the twin relationship would lead to the creation of a narcissistic system in which the twins may be bound by either loving or hating feelings (Rosenfeld, 1971). In narcissism, there is withdrawal from external object relationships to an identification with an idealised internal object, either in love or hatred. For twins the loving internal object may be the internal twin. Castor and Pollux represent the loving or libidinal aspects of the narcissistic twin system. They were twins born to Leda after she was seduced by Zeus disguised as a swan, and she was also pregnant by her husband, Tyndareus. Pollux was a god and Castor mortal, thus embodying the splits between twins. They formed an impregnable “loving” twinship, acting together to inflict terrible injuries on their enemies. They remained inseparable, even after the death of Castor, and lived forever alternately in the heavens and the netherworld. The loving narcissistic element of twin relationships is commonly encountered after one twin has died. The surviving twin feels incomplete, and longs to be reunited with the other twin rather than mourn the loss of the twin.

Where the destructive narcissistic forces in the twinship are idealised, rather than the loving ones (Rosenfeld, 1971), the twins would become bound together in mutual hatred. The internal twinship would be a destructive narcissistic organisation that opposes any contact with a loving external object relationship. The murderous feelings between the twins would be played out between them, rather than directed outwards as it was with Castor and Pollux. A twinship based on these processes would be developmentally damaging to both twins. It is likely that one twin would not only be dominant, but would also keep the other twin in thrall using threats and propaganda to maintain this order. The Gibbons twins described by Wallace (1996), and Bill and Bert described by Burlingham (1963) are both examples of a destructive narcissistic twinship in which the twins can neither live harmoniously together, nor apart. When together, their murderous aggression is played out. When apart, they feel incomplete and fear disintegration.

The narcissistic twinship may be psychotic in nature. Potash and Brunell (2004) describe monozygotic twin sisters who were locked in a folie a deux of a shared delusional world. The authors suggest the twin sisters used this delusional system in order to cope with abusive parents, but that the psychotic twinship rendered them inseparable. They were visually paranoid, highly disturbed, and entertained suicidal thoughts. They supported, believed and identified with each other’s delusions. The twins were hospitalised in order to both separate them and to keep them safe while they received multiple-conjoint psychotherapy with a male and a female therapist. Through this work, each twin discovered a safe setting for the expression of “dangerous” emotions, and began to discover that her anger did not harm her dependent twin. As a result they no longer felt the need to deflect their hostility into a delusional system. The twins were able to separate as they discovered that it was safe to be an individual, and “each learned that she had her own ideas, beliefs, and interests that could be shared without the other having to accept it as her own reality.”

We might think of the sense of self as cohering within a psychic membrane/skin (Bick, 1968), so that an individual has a more or less cogent idea that “this is me”. This process is more complex for twins for whom there is likely to be a greater or lesser degree of overlap with the other twin in terms of a sense of identity both within and outside the twinship. As described earlier, the sense of self develops within the intimate relationship with mother and father. Where a twin is always present in the minds of both infant and parents, there is likely to be some overlap between the twins in which the sense of self is to a greater or lesser extent shared, or the boundaries less clear.

The issue of separateness and separation from the other twin is central to both development and analytic work with a twin. By this I do not mean separation leading to the ending of the twinship. The twin relationship is an enduring, indeed a valuable and an ineradicable object relationship. Instead, I believe that it is possible to enable a twin or twin patient to find separateness and individuality within the twin relationship without either denying the importance of the twinship or psychically murdering the other twin. This development might be characterised as the twinning processes changing from those of an “identical” twin to a “non-identical” twin, a change from a more narcissistic twinship to one that is more object-related.

Following Rosenfeld’s (1987) ideas about thick and thin-skinned narcissism, Britton (2003) writes about the thickness or thinness of the psychic membrane in narcissistic individuals and the way in which this affects the nature of their relationships with others, particularly in his counter-transference experiences in analysis. In intimate relationships such as marriage and analysis an individual negotiates the shared psychic space. The experience of being in that psychic space with that individual and the quality of the negotiation of the space is determined by the nature of the psychic skins of the participants.

I have suggested (Lewin, 2004) that twins who are enmeshed jointly occupy a psychic skin. It is as if the psychic membrane between them is thin and easily penetrated so that they are highly responsive to each other, leading to the impression that they share an identity. The psychic envelope around the twins is thick, isolating the twins from the outside world. The twinship becomes the most highly charged and important relationship. Within this narcissistic system, the twins negotiate the shared psychic space according to the way they have formed an individual identity in relation to the other twin. As Piontelli (2002) has observed of the behaviour of twins, starting from their time in utero, the pattern of relating and the nature of the relationship between twins are enduring.

The birth stories for twins are redolent with ideas about the shared space – fighting for space in the womb, one twin taking too much space, the approach/avoidance patterns of the foetuses in relation to each other. The emotional space twins have to share by virtue of their identical age and the sharing of the environment in the crucial early years in a way that no other siblings do, will be re-enacted within all their intimate relationships. This is particularly so within the consulting room where the analyst becomes the analytic twin in all its many aspects.

It is my experience that a twin patient at the more narcissistic end of the continuum may occupy two overall positions.

  1. The analyst may be excluded from the twinship, kept outside the thick psychic twin envelope, and regarded as irrelevant. This would fit with what Britton (2003) refers to as a thick-skinned (schizoid) narcissist. The patient appears to be solely and intensely preoccupied with the other twin, in loving or hating mode, while the analyst is treated as an unnecessary parental figure. In this situation, it is not uncommon for the patient to end treatment when the twinship membrane is penetrated by the analyst.
  2. The analyst may be over-included in the twin patient’s world, as if she is an analytic twin in an enmeshed twinship with the patient. The patient fights to occupy the psychic space in what at times seems to be a life and death struggle. There is no room for separateness of any degree and the analytic space is obliterated by the patient. This may take the form of outpourings that fill the session, leaving no space for thinking, or what appears to be an emotionally charged fight expressed either in silence or verbal aggression. The patient seems to experience any independent thought by the analyst as an attempt by the analyst to intrude and take over his/her mind, to occupy all the space. Likewise the analyst feels grossly intruded upon in a way that may paralyse analytic activity.

Typically for these thin-skinned twin patients, the ordinary difficulties of everyday life seem to take on giant proportions that threaten the patient. It is as if the attacking twin is all around them. It is interesting how frequently these patients arrive late for sessions because in all sorts of ways, life has made it too difficult for them to get there on time. The patient will anxiously and angrily describe the details of their apparently horrendous journey and the unfairness of the world. There is a pervasive sense of someone being to blame, leading either to profuse apologies by the patient or to a sense of the analyst being to blame for the difficulty of the journey. The persistent complaint is that the patient feels that the analyst is useless in removing her pain and/or that the analyst is a hostile invader causing her more pain. The patient fills the consulting room and the analyst’s mind with her angry pain and distress about her life, past and present. The analyst feels obliged to listen to the painful story, but feels pinned down, invaded, overwhelmed, and sometimes suffocated by the awfulness the patient is describing. The patient’s mood seems larger than life and all pervasive. It is as if there is no helpful internal parent, only a twin-analyst who functions as a receptacle for the expulsion of unbearable experiences.

These modes of relating may vary for each patient at times, but it seems that one or the other tends to predominate.

In conclusion, the nature of the twin relationship will be determined by many factors including heredity and the way in which each twin of a particular genetic disposition negotiates the primary relationships with mother, father and the other twin. Each twin has a dual task of forging a personal identity through individual psychic development, and of negotiating the twin relationship, maturing from the more narcissistic end of the spectrum to a twinship of both separateness and togetherness. In a twinship at the more narcissistic end of the spectrum, intense twinning processes hamper the recognition and experience of the “otherness” of the other twin. Instead, the unity of the twinship is idealised in denial of the recognition of “otherness” of each individual twin. The idealisation of the twin relationship is based on essential internal loneliness that leads to a ubiquitous longing for a twin, a longing that emanates from the infant’s earliest preverbal experiences with its mother.



Bettelheim, B. (1955) Truants from Life (The Rehabilitation of Emotionally Disturbed Children). The Free Press, Glencoe, Ill. (Collier. Macmillan, London. (Paperback version 1964, Simon and Schuster Inc.)


Bick, E. (1968) “The experience of the skin in early object-relations.” International Journal of Psycho-Analysis, 49:484-486


Bion, W. R. (1967) “The imaginary twin” in W.R. Bion (1967) Second Thoughts. Selected Papers on Psycho-analysis Jason Aronson: New York pp. 3-22


Britton, R. (2003) Sex, death and the Superego. Experiences in Psychoanalysis. Karnac, London and New York.


Burlingham, D. T. (1945) “The fantasy of having a twin.”  Psychoanalytic Study of the Child, 1:205-210


Burlingham, D. T. (1963) “A study of identical twins—their analytic material compared with existing observation data of their early childhood.”  Psychoanalytic Study of the Child, 18:367-423


Dundy, E. (1985) Elvis and Gladys. The Genesis of the King. Weidenfeld and Nicolson, London


Fanthorpe, U. A. (2000) “Sightings.” in Consequences. Peterloo Poets.


Gringras, P. and Chen, W. (2001) “Mechanisms for differences in monozygous twins.” Early Human Development, 64: 105-117


Hillary S. (and Heather S.), 2004, from the Fall Mountain’s Student Magazine, “Mountain Expressions”:, 11/05/04, 11pm.


Joseph, J. (2004)”The gene Illusion.” Human Givens Journal: 10,4,12-19.


Klein, M. (1963) “On the sense of loneliness.” in Envy and Gratitude and Other Works.  pp. 300-313.  The Hogarth press, London, 1980.


Lewin, V (2004) The Twin in the Transference. Whurr Publishers Ltd., London and Philadelphia.


Piontelli, A. (2002) Twins: From Foetus to Child. Routledge: London and New York


Potash, H. and Brunell L. (1974) “Multiple-conjoint psychotherapy with folie a deux.” Psychotherapy: Theory, Practice, Research; 270-6; quoted by Sharon, I., Sharon, R., Eliyahu, Y. (2004) “Shared Psychotic Disorder.”, Inc. 16 February, 2004


Prigg, M. (2004). “Twins in £40.000 makeover…to look more alike.” Evening Standard, Thursday 6 May 2004.


Rank, O. (1971) The Double. A Psychoanalytic Study. Translated and edited by H. Tucker, The University of North Carolina Press, Chapel Hill


Rosenfeld, H. (1964) “On the Psychopathology of Narcissism a Clinical Approach.” International Journal of Psycho-Analysis, 45:332-337


Rosenfeld, H. (1971) “A clinical approach to the psychoanalytic theory of the life and death instincts: An investigation into the aggressive aspects of narcissism.” International Journal of Psycho-Analysis, 52:169-178


Rosenfeld, H. (1987) Impasse and Interpretation. Tavistock: London and New York


Steiner, J. (1993) Psychic Retreats. Pathological Organisations in Psychotic, Neurotic and Borderline Patients. Routledge: London and New York


Woodward, J. (1998) The Lone Twin. A Study in Bereavement and Loss. Free Association Books: London and New York.


Wright, L. (1997) Twins. Genes, Environment and the Mystery of Human Identity. Phoenix: London.