Dr. Mollon is a psychoanalyst and member of the Independent Group of the British Psycho-Analytical Society; a psychotherapist with the Tavistock Society; and a clinical psychologist at Lister Hospital. He has authored a number of works: The Fragile Self (1993); Multiple Selves, Multiple Voices (1996); Remembering Trauma(2001); Shame and Jealousy (2001), that explore, elaborate, and contrast Kohut's theory of the self with other contemporary theorists.

In his recent book, Releasing the Self: The Healing Legacy of Heinz Kohut (2002), he explores the interplay between Kohut and other contemporary psychoanalytic and neurobiological points of view while arguing that Kohut's significant contributions to contemporary psychoanalysis have been widely misunderstood, ignored and lost.

This paper may be quoted provided that the author, Phil Mollon, PhD and this website, www.selfpsychologypsychoanalysis.org are cited as the source.

Some implications are examined of Kohut's formulation of the state of the fragmented self. It is hypothesised that the threat of fragmentation is experienced by human beings (individually and collectively) unconsciously as a continual danger.

Initially the state of fragmentation is kept at bay by the selfobject functions of the caregivers. The availability of the caregiver allows the child to form a selfobject system that regulates his or her psychophysiological state.

Later, the developing child and adult may form selfobject systems with a great variety of human phenomena, drawing upon the available linguistic, cultural, imagistic and behavioural routines and organisations.

This perspective may be linked to some of the insights of Lacan, who also emphasised the original human state of fragmentation. The threat of fragmentation means that the child will form a psychic system in whatever way, and with whomever he or she can.

In optimum development the caregiver will be responsive to the child's own initiatives so that his or her own potential, talents and inherent developmental agenda can be supported.

However, in less good circumstances, the child will be compelled to incorporate the mother's desire as an internalised structure around which to organise the self - or will organise around an idealised abuser, or indeed around any available perceived source of strength.

This inner structure will then be opposed to the child's authentic potential. The work of psychoanalysis involves undoing this alien parasitic structure, thereby releasing the unknown true self. However, such work brings the threat of fragmentation and may result in great anxiety.


For some years I have been puzzled by two related common phenomena. The first is the way that people seem very often to have established within their psyche, alien and oppressive mental structures, based partly upon the characteristics of hostile caregivers - alien structures which they then defend vigorously as their own self. The second is the way in which there is often a rigid clinging to fundamental beliefs and assumptions, even when these are maladaptive and cause pain. For example, a patient often complained that I surely did not really care about him since I saw him only because he paid me. Recently, he referred to this again, using an odd form of words. He said "I continually look to assure myself that you do not care about me". Whilst his conscious meaning had been the familiar one, he agreed that his phrasing was ambiguous and could appear to be saying that he looked to reassure himself that I did not care about him. It then struck him that if he found evidence that I cared about him, this could be very alarming because it would threaten one of his fundamental beliefs - that people do not really care about one another. As he talked about his alarm at the idea of a fundamental belief being challenged, I was reminded of a schizophrenic patient who once remarked that she could not contemplate giving up her persecutory internal voices because they constituted the 'skeleton' that held her psyche together. The preservation of psychic structures appears to be a compelling human need.

In what follows, I will be presenting the following basic paradigm. From a substrate of the fragmented self, the urge to form an organisation is powerfully compelling. In the absence of selfobject organisation based on empathy and respect for the child's developmental needs, the child will form oppressive organisations and internal structures using whatever psychological material is available. The task of analysis is to help free the patient from these oppressive structures so that the authentic unknown self can be released.

One of the unusual features of Kohut's writings was his emphasis upon disturbances of structure and arousal alongside the more common psychoanalytic concern with content (of fantasies, desires etc). He regarded the overt content - even that which is unconscious to the patient but can be discerned by the analyst - as driven by the hidden or invisible breakdown of structure - the structure of the self held by the selfobject functions provided by the caregiver (or later by the spouse, friends, culture etc). The unconscious impulses and fantasies, that typically and historically have been the focus of psychoanalytic work, were seen by Kohut as merely the shrapnel of disintegration and defences against this. Thus he came to view unconscious impulses and conflicts as themselves defences against a deeper and more unspeakable dread - that of disintegration - and, as he commented (1984), "… the attempt to describe disintegration anxiety is the attempt to describe the indescribable …" [p 16].

In his paper on narcissistic rage (1972), Kohut describes certain kinds of patients as follows:

"These patients initially create the impression of a classical neurosis. When their apparent psychopathology is approached by interpretation, however, the immediate result is nearly catastrophic; they act our wildly, overwhelm the analyst with oedipal love demands, threaten suicide - in short, although the content (of symptoms, fantasies, and manifest transference) is all triangular oedipal, the very openness of their infantile wishes, the lack of resistances to their being uncovered, are not in tune with the initial impression…" [p 625-626]

Thus Kohut describes how, in some cases, the apparent analytic material covers, and is driven by, something much less visible. The 'drive' is an absence - a hole where the core self should be. It is perhaps analogous to an astronomical black hole, invisible and discernable only through its secondary effects. He goes on:

"The nuclear psychopathology of these individuals concerns the self. Being threatened in the maintenance of a cohesive self because in early life they were lacking in adequate confirming responses ('mirroring') from the environment, they turned to self-stimulation in order to retain the precarious cohesion of their experiencing and acting self. The oedipal phase, including its conflicts and anxieties, became, paradoxically, a remedial stimulant, its very intensity being used by the psyche to counteract the tendency toward the breakup of the self - just as a small child may attempt to use self-inflicted pain (head banging, for example) in order to retain a sense of aliveness and cohesion." [1972 626-627].

Kohut suggests that psychodynamic conflict is used itself as a defence against the deeper danger - the breakup of the self. He refers to the cohesion of the experiencing and acting self as 'precarious'. Thus the fundamental danger is of fragmentation. Kohut did not have much to say about states of fragmentation per se, appearing to view them as characterising psychosis and inherently beyond the reach of analysis. He was more concerned with the threat of fragmentation and defences against this. To continue the astrophysical analogy, it is as if Kohut began to look beyond the visible and manifest (albeit unconscious) realms of the psyche and found the areas of dark matter and black holes that account for what is observed. The selfobject transferences, of mirroring, idealising and twinship, tend to be invisible until they are disrupted. It is through the breakdown products of their destruction that their existence is discerned.

Disintegration anxiety is difficult to articulate and communicate, compared with other more specific anxieties (such as the series Freud described: loss of the object, loss of love, castration). Kohut (1977) wrote:

"… the expression of the ill-defined yet intense and pervasive anxiety that accompanies a patient's dawning awareness that his self is disintegrating (severe fragmentation, serious loss of initiative, profound drop in self-esteem, sense of utter meaninglessness) … may initially be veiled; the analysand may attempt to express his awareness of the frightening alterations in the state of his self through the medium of verbalisations about circumscribed fears - and it is only gradually and against resistances that his associations will begin to communicate the central content of his anxiety, which, indeed, he can only describe with the aid of analogies and metaphors." [p 103].

A woman whom I have been seeing in psychotherapy for about 12 years, and who is having to face ending because of her geographical move, remarked "the only thing between me and disintegration is you". She had a dream in which the house that she and her husband are moving to was riddled with woodworm and in danger of collapse. The problem for her is that although she can represent fragmentation (in words and symbols) when she is communicating with me, when she is alone she has no means of articulating her anxiety - it reverts to an overwhelming and nameless dread - and in such a state she cannot think nor use words but simply experiences psychic agony that is beyond communication. The loneliness and desolation of such a state of mind is complete. Her childhood was characterised by isolation, rejection, emotional coldness and sexual abuse - perhaps combined with a temperament of ADHD which added to her tendency to fragment. Another feature of this patient is that on superficial meeting she can appear rather normal - intelligent, cultured, polite - but this is a personality veneer that covers her core of fragmentation and psychic agony.

In Kohut's thinking, the selfobject functions provide the bulwark against fragmentation. The infant organises its self around its selfobjects. This organisation provides order, affect regulation (Schore 1994), and emotional meaning - or (to put this in other language) facilitates 'mentalisation' [Fonagy et al 2002]. Whilst the original selfobjects are the organising functions provided by the child's caregivers, I suggest that in later life we tend to form organisations with a wide variety of linguistic, cultural, image-based and behavioural selfobjects. These may include systems of knowledge and belief (political, scientific or religious systems, including psychoanalysis), tribal affiliations (including professional affiliations), daily routines (the obsessional and autistic phenomena of everyday life), addictions of all kinds (behavioural and sexual as well as pharmacological) and all manner of compulsive seeking of stimulation. In all these diverse ways, we may attempt to organise our inner states of mind by linking these to external sources of stimulation, soothing and order. The problem for human beings, as Lacan pointed out, is that we are born prematurely (in order to accommodate the large brain) and are unable to coordinate our movements and maintain bodily homeostasis without continual intervention from the mother. Thus we begin in a state of fragmentation and require an external source of organisation. Even secure attachment is built on a substrate of the fragmented self, since it is, for any human baby, only the ministrations of the mother that hold back the threat of bio-psychological disintegration. This is the background threat of annihilation - which I think corresponds to what the Kleinians viewed as the manifestation of the terrifying death instinct.

I believe it may have been as a result of Kohut's discernment of states of fragmentation that he theorised about structures of the self - the 'nuclear self', the 'bipolar self', the 'cohesive self' and so on. Such terms denote structures (of experience, agency, goals and values) that can hold together or fall into fragmentation according to the availability of empathic caregivers. The bipolar self, containing the person's enduring ambitions and values, expresses an inherent tension, reaching into the past and leading into the future. It is developed through the selfobject functions provided by the object, but is not inherently concerned with the relationship with the object. Kohut was concerned with psychic problems rather different from those addressed by the Kleinians or the object relations school, or perhaps even the current relational schools of psychoanalysis in the U.S.A. Nevertheless, the disturbances, tensions and anxieties relating to the structure of the self may underpin disturbances in relation to others. Pathological structures or patterns of object-relating may be clung to because change may threaten fragmentation of the self. Psychic change is feared because it brings the threat of fragmentation. Thus, internal working models of relationships, as well as systems of belief (religious and political systems) may be tenaciously retained because these structure the person's experience. The emergence of the unknown authentic self may evoke dread - as well as joy.

In some of Kohut's writings he implied the possibility that the threat of fragmentation may be ever-present as a potential - even in relatively healthy personalities. For example, in describing the origin of the nuclear self, he wrote:

"In trying, again and again, in analysis after analysis, to determine the genetic roots of the selves of my analysands, I obtained the impression that during early psychic development a process takes place in which some archaic mental contents that had been experienced as belonging to the self become obliterated or are assigned to the area of the nonself while others are retained within the self or are added to it. As a result of this process a core self - the 'nuclear self' - is established" [1977 p 177]

Drawing upon this idea of the origin of the structure of self, he writes (in discussing the case of Mr. M) of two layers of the diseased self, only one of which participates in the therapeutic process, whilst the other "sinks into the darkness and remains out of sight" [p 176] Then in a footnote, he considers the possible continuing unseen influence of this foreclosed area of fragmentation:

"The possibility cannot be discounted that the untreated layer, surviving outside the cohesive structure of the active and productive self, may serve as a stimulus to the self to persist relentlessly in its activities. I cannot support this claim with detailed clinical data, but the fact that creativity often has a compulsive, involuntary quality and in its absence depression may supervene can be adduced in support of such a hypothesis." [p 176-177]

Thus Kohut implied that even when a cohesive self has been established, the threat of fragmentation may remain, ever ready to invade the self when adverse psychological circumstances are encountered. He considered that persons suffering from psychoses, borderline, or schizoid states were not amenable to psychoanalysis because the underlying state of fragmentation might be exacerbated. Certainly, the phenomenology of many severely disturbed states of mind (such as hallucinatory voices and perceptual distortions) may be understood partly as an expression of the 'disintegration products' of the fragmented mind.

Furthermore, Kohut indicated that fragmentation anxiety may emerge at crucial moments of psychic change, when an existing maladaptive selfobject organisation is about to be given up. For example, in the case of Mr. Z [1979], at the point when the patient was beginning to shift from the archaic tie to the mother and turn instead to a selfobject organisation with the strong father (expressed in the transference and in his emerging positively-toned childhood memories of his father), he experienced "a number of frightening, quasi-psychotic experiences in which he felt himself disintegrating and was beset by intense hypochondriacal concerns. At such times he dreamed of desolate landscapes, burned-out cities, and, most deeply upsetting, of heaps of piled up human bodies." [431] In one of these dreams, Mr. Z.'s mother appeared standing with her back to him. This was accompanied by "the deepest anxiety he had ever experienced" [431] - expressing his realisation of her icy withdrawal from him when he attempted to assert his independence, and his emerging awareness that she had never presented a smiling mirroring face in response to his own developmental initiatives.

Kohut's emphasis upon structure and loss of structure (fragmentation) - and the underlying bedrock of disintegration anxiety - was fundamental to his whole work. Therefore, it seems worth noting that those later theorists who have wished to discard Kohut's concept of the bipolar self (without proposing an alternative structure) are avoiding one of his most important insights about what drives human beings.

The motif of the fragmented body-self occurs in the theorising of many analysts. Kohut himself drew upon Freud's inference of an autoerotic stage of bodily incoherence prior to the stage of narcissism and the formation of the ego - as well as Glover's concept of 'ego nuclei'. In words similar to those used by Winnicott who wrote of 'unthinkable agonies', Kohut (1984 p 8) wrote of "the unspeakable anxieties accompanying … a prepsychological state". Lacan too wrote of the stage of the 'fragmented body image', prior to the engagement with images and language supplied by the prevailing culture. He postulated that the earliest experiences of the baby are of fragmentation due to the prematurity of birth in the human species - "a primordial Discord betrayed by the signs of uneasiness and motor uncoordination of the neo-natal months" (Lacan 1949 p 4). He believed that this infantile hell is displaced by illusions of coherence and completeness offered by an identification with the external image found in a mirror or in the sight of another child - and later in the images offered by pre-existing culture. These are Lacan's version of the selfobject. So for Lacan, the ego, the illusory sense of a coherent self, is rooted in this alienating identification with external images - yet the `threat of fragmentation is ever present, giving rise to what he called a 'narcissistic passion' - "the passionate desire peculiar to man to impress his image in reality …" (Lacan 1948 p 22). Lacan pointed out that we all must construct our 'identities' from the roles, images and language available within the culture into which we are born - especially the microculture of our particular birth family. All human beings are born into a culture, but cultures are highly variable and plastic. The identities - the selves - that human beings prize and protect so vehemently are all illusory. Yet it is the threat of this illusion unravelling and exposing us to fragmentation and dread that can make us cling to it with such passion - as evidenced in fanatical religious and tribal conflicts throughout history.

We find examples of this breakdown of the illusory self in certain forms of reactions to excessive use of recreational drugs, such as marijuana. A young woman of 19 who had used marijuana extensively since the age of 14 described an occasion when she had smoked some particularly potent variety on her own. She had felt overwhelmed with utter dread and terror, believing that she was going to die, but feeling totally unable to communicate the nature of her anxiety. Since then she had often felt prone to anxiety and panic, as well as chronic feelings of depersonalisation. I suspect that such a state of mind, precipitated by marijuana poisoning, involves a partial decathexis of language - by which I mean language in the broadest sense of all the structures of symbol and meaning that underpin the organisation of our phenomenal world. This decathexis means that the net of language and symbol which normally holds us, supporting our sense of self and identity - our location in a particular cultural, social and familial matrix, at a particular point in history - falls away, leaving the subject stranded in an empty psychological space. Without words we cannot adequately reach out to others. Without words - the mother tongue given in lullabies of our earliest interactions - our experience becomes dumb and beyond the reach of empathy.

The identities we take in from the available roles and images in the surrounding culture could be seen as like our psychological clothes. Without them we feel naked, exposed and vulnerable. This was brought home to me when I talking with a young woman patient who is an artist. She attended art school and obtained a decent degree. However, her art is difficult to categorise - but seems to be somewhere in the area of conceptual and performance art. This has not been commercially successful. Sometimes she thinks she would like to do something very conventional. She told me of how she often feels anxious and self-conscious on leaving her house, as if naked, or wondering if she is inappropriately dressed. In some vague way she felt this was connected with her creativity. As she was speaking, it suddenly occurred to me that she was telling me of how it felt for her not to have any familiar or recognised social role and identity - that she felt unclothed, without a category to place herself in and for others to place her in. This idea intrigued her and we went on to look at the dilemma of the artist - the creative urge leads to a suspension of the usual social categorising of self, others and life as a whole - and yet this leaves the artist psychologically naked (without a semiotic covering) and exposed.

The combination of Kohut and Lacan (although these two are very different in style and culture) can be very fruitful. Both view the stage of the fragmented self as fundamental. Putting the two together, we can arrive at the following stark insight. All human beings are born into a state of disorganisation and chaos - Lacan's 'Primordial Discord' - because the infant is unable to coordinate and regulate its own body. The infant must form a system with the caregiver - the selfobject. Through the adequate availability of the empathic selfobject, the baby's mental and physiological being is organised - but the threat of chaos-fragmentation is ever-present. Whilst the original selfobject is usually the mother, the need to form a system with an external organisation remains throughout life. In addition to forming selfobject systems with significant other people, human beings also create organisation through locking into/latching onto all manner of diverse structures and patterned phenomena. These include behavioural routines, addictions of all kinds, belief systems, identities (familial, tribal, racial, cultural or professional), and all the linguistic and quasi-linguistic structures within which we find our place in the social world. Threats to these - for example through confrontation with competing belief systems - bring the dread of disintegration anxiety.

Here is a brief vignette illustrating fragmentation. A woman of thirty presented with a relentless desire for suicide although she did not display any other psychiatric symptoms of depression. She appeared professionally competent and successful, as well as socially and sexually attractive. However, her childhood had been characterised by neglect and chaos at the hands of her alcoholic mother, followed by multiple foster homes. Her adult life had reinforced her general sense of disillusionment regarding the reliability of human beings. She was extremely intelligent and articulate and communicated vividly in psychotherapy - arguing vehemently regarding her right to take her own life. However, she complained that I was insufficiently responsive, did not give her clear guidance regarding solutions to her problems, and that psychotherapy took too long and would have an uncertain outcome. Her concern with control was a prominent feature - as she readily acknowledged. The therapy was characterised by an alternation between periods in which she allowed me to arrive at a deeper understanding and emotional contact with her, when she would indicate increased trust in me, and other periods when she would become hostile, combative and would break off the therapy. A deep fear appeared to be of becoming trusting and vulnerable and of then being hurt - and of feeling utterly helpless. Her failure to kill herself in two very serious suicide attempts left her distraught. On returning to see me after one of her periods of withdrawal, she reported having spent a couple of weeks in a psychotic state - not sleeping, feeling confused, 'thinking every thought in the universe', hallucinating, experiencing paranoia, feeling overwhelmed with meaning and yet not understanding anything - a state she had never experienced before (at least not that she remembered). However, one day she had woken up and this state and her suicidal feelings were gone. She described feeling calmer and more positive. I talked to her about my thoughts regarding the stage of the fragmented self - and how the fear of fragmentation may have been particularly pronounced for her in view of her early experiences. This was meaningful to her and she acknowledged that during this period of confusion and disorder she had longed for the certainty of her previous clinging to her plan for suicide. I put to her that she had, as part of a healing process, relinquished her previous very rigid organisation of her psyche and entered a spontaneous regression to the stage of the fragmented self, from which she had eventually re-emerged in a new less rigid state of organisation. She found this formulation extremely plausible and intriguing. I also had a vivid impression of how close she had been in her state of confusion to the possibility of reorganising on a psychotic basis of delusional structures.

The subject's alienation in language, the paranoiac structure of identity emphasised by Lacan, can be viewed as the broader socio-cultural version of the alienation in false imported images that can occur within the personal relationships of the original family. Here we encounter the false self described by Winnicott - or the 'alien self' as recently described by Fonagy et al. [2002]. Similarly, Kohut described many instances in which the child's natural exhibitionism had been hijacked by the mother's narcissism - her wish to have a child that mirrored her and fulfilled her ambitions. This is shown clearly in his diagram of vertical and horizontal splits in The Analysis of the Self. The authentic self - the child's own initiatives and his or her wishes to be mirrored and affirmed - is oppressed by an alien self based on the mother's narcissistic desires. However, Kohut also argued that if the child's own exhibitionistic initiatives are responded to, and if his or her needs to idealise are accepted, then the lines of development of the bipolar self can progress. Thus for Kohut, an unalienated self is possible. How can this be, in view of what I have indicated of the inevitable structuring of identity around the imported language and images of the culture into which the child happens to be born? Whilst no child can escape culture (and the limitations of geography, technology, social class, educational opportunities etc), there are differences in the degree to which a child is able to choose his or her place within the available roles and language of that culture. These differences must depend in part upon the willingness of the primary caregivers to offer selfobject responses to the child's initiatives. Such selfobject responses respect the child as a centre of initiative - a subject rather than an object of the parent's narcissistic desire. The child can then choose the language, roles and images with which to construct his or her identity rather than have these imposed. It is analogous to the way the body selects the nutrients it needs from the food that is eaten, incorporating these into its own structure. The existence of an 'internal object' or an 'alien self' would indicate an externally derived structure that has been taken in as a foreign body - perhaps to become what Fairbairn termed an 'internal saboteur'.

In optimum development, the mother provides selfobject functions of showing interest and pleasure in the child's natural exhibitionistic display. When this is not the case, the exhibitionism becomes a source of anxiety and shame - the flow of energy outwards is thrown into reverse and the subject is filled with painful tension (shown literally by the response of blushing); in place of the wish to exhibit there is an urgent wish to hide - to disappear. To be overwhelmed with shame is to be in a state of disorganisation; the shame-filled person cannot function coherently.

Even more profound is the toxic shame and fragmentation resulting from abuse by a caregiver. For the purpose of this discussion, I define 'abuse' as the deliberate and knowing withholding, distortion or perversion of the normal selfobject responses for reasons of the caregiver's own sadistic gratification - gratification that might include the projective evocation in the victim of unwanted feelings and images belonging to the perpetrator; a process of emotional violence that I have termed 'imposed identity' is also an aspect of this abuse - the insistence on viewing a person in a particular way irrespective of their own subjective experience, thereby imposing an alien identity. This definition can incorporate physical, emotional and sexual abuse. Toxic shame resulting from abuse is worlds away from the normal shame, embarrassment, chagrin and disappointment in the self that we can all experience from time to time. This toxic shame can be lethal, poisoning the soul and corroding any vestige of self-esteem. Fonagy et al (2002) refer to this as 'ego-destructive shame', which they see as resulting from severe interpersonal trauma that cannot be processed and alleviated by 'mentalisation'. The latter, as I interpret it, involves two components: [1] recognising that people (both child and mother) have minds, with intentions, thoughts and feeling; [2] giving meaning to experience and to the other's behaviour. For the abused child, the recognition of the abuser's mind and hostile intention may be too disturbing and so mentalisation is defensively blocked. Without mentalisation, shame is experienced in a very concrete and absolute way. The desire to disappear - a normal component of shame - is experienced as a compelling annihilation of the self. Fonagy et al (2002) comment:

"Why should the brutalisation of affectional bonds, whether in the context of relationships with parents or with intimate peers, be associated with such an intense and destructive sense of self-disgust verging on self-hatred? Once again, there is a paradox: the shame concerns being treated as a physical object in the very context where special personal recognition is expected. Overwhelming mental pain is associated with experiencing a discrepancy between the representation of an actual self, based on how one is being treated, and the representation of the ideal shape of the self. The expectation of being seen and understood as a feeling and thinking person, which is created by the attachment context, clashes violently with the brutalised person's objectification and dehumanisation. Shame is a higher order derivative of this basic affect of pain. Unbearable shame is generated through the incongruity of having one's humanity negated, exactly when one is legitimately expecting to be cherished. [p 426]

Whereas mirroring and availability for idealisation brings about the emergence and firming of the bipolar self, abuse within a caregiver relationship forces the child back into the state of the fragmented self. In response, the child organises his or her fragmented self around the figure of the abuser - forming an aspect of the alien self. This is a variant of identification with the aggressor. Challenge to the perception of the self as bad and the abuser as good brings the threat of fragmentation. The shame is structured within the personality - resulting in a part experienced as bad and unlovable, and another part experienced as accusing and condemning. A young woman extensively abused in childhood expressed recurrent anxiety about what she might inadvertently blurt out to me if she did not keep very careful control over herself. She also became anxious that her e mails to me might be intercepted and read. Partly this anxiety related to her childhood fear that she might reveal to her mother that her stepfather was abusing her. However, in addition, we came to see that it expressed her feeling that there was something fundamentally bad about her - a fear that I would discover this awful truth, that she was bad in some absolute and irredeemable way, and I would then reject and condemn her. She feared that although I had seemed understanding and accepting of her so far, there would come a point when I would realise my error and turn against her and banish her. When engulfed with this shame and fear of being seen and condemned, she would curl up and hide behind her long hair. Her belief that she was fundamentally bad is typical of the protective strategies of abused children. A view of the parental figure as bad is often literally unthinkable - and so in order to maintain an idealised image of the parent, as strong and good, there must be a profound denigration of the self. This degree of self-hatred may be unsustainable. Dissociation and repression of this bad self may occur - perhaps repeatedly - with continual dissociative reinventions of a new self. Later, when the repressed split-off bad self threatens to re-emerge into consciousness, there may be considerable anxiety.

Another possibility is that the child does become able to perceive the abuser as bad, but then has to organise his or her self around this denigrated figure. Then, in contrast with Kohut's description of the idealising position, in which the child feels 'you are perfect and I am part of you', the formula becomes 'you are bad and I am part of you'. Again this means that shame pervades the core experience of self.

I have encountered other instances in which the patient has not suffered overt abuse in childhood, but the mother's rejection of the child's own communicative initiatives and his or her needs for understanding and empathy have been so profound that the potential authentic self is pervaded by shame and its development is blocked. In such instances the early environment was perceived as fundamentally opposed to the child's actual authentic self and as intent on replacing it with a preferred alternative. I have written about this in a paper on 'psychic murder syndrome', using the metaphor of a 'Stepford Child', derived from the film 'Stepford Wives', in which the wives in a small respectable town were all killed and replaced with identical replicas that behaved in ways more to the liking of the husbands. In these circumstances the child may internalise the psychically murderous environment and the murder of the authentic self then continues internally throughout later life - the process being particularly activated whenever genuine emotional intimacy and attachment threatens.

In what I have presented so far, I have been describing ways in which the stage of the fragmented self - and the threat of regression to this state - are countered by the taking in of alien structures that are mistakenly perceived as 'self'. If offered a therapeutic environment, the patient may seek unconsciously the opportunity to return to the early unmet selfobject needs, discarding the false self structures, but risking a descent into irreversible fragmentation. This is similar to the process described by Balint (1968) as the patient regresses to the position of the 'basic fault', hoping to find a more 'facilitating environment' (Winnicott 1960) in which the 'true self' can be released.

[Detailed clinical material was presented in the spoken version. This is omitted here for reasons of confidentiality. The account described a patient - Jim - who expressed fears that the analyst wanted to impose his 'blueprint' on him, to turn out a good analytic product. This related to his childhood experience of feeling that he must fit into his mother's desired image of him. The analytic work involved his search for the father in the transference. Disintegration anxiety and a terrifying dream occurred at a moment of potential change from his imprisonment in his mother's desired image.]

When he began the analysis, Jim was trapped in an image - a false representation of self, based upon what he perceived his mother to want him to be. This was a kind of old fashioned English gent, very compliant, detached from the rough and tumble of life amongst the uncouth masses. It was difficult for him to pursue relationships with women beyond a certain point because he felt, unconsciously, that he must never leave his mother and must maintain his role and image of being what she wanted. In his childhood, closeness to his mother, and a sense of being preferred by her over his father, contributed to oedipal gratifications, but left him further trapped with her and feeling let down by his father, whom he saw as also dominated by his mother.

The false representation of self, based on an identification with his mother's desire, could also be seen as a form of grandiosity in which his own potential spontaneous expression was hijacked by his mother's narcissism - the kind of situation portrayed in Kohut's diagram of the vertical and horizontal splits in The Analysis of the Self [page 185]. Jim would imprison himself within his role and image and would also be fiercely protective of the image of his parents. [He was allowed to criticise his parents but the analyst was not.] He secretly saw himself as superior and aloof from the common mass of humanity. However, he gradually came to see that this narcissistic grandiosity was not truly his own but was a distortion of his own self based on excessive accommodation to his mother's desire. It could be regarded both as a narcissistic haven - a retreat from the emotional rough and tumble of life in relationship with others - and as a narcissistic prison which prevented him from expressing himself authentically.

In this state of 'gentrification', Jim felt his true spirit and spontaneity had been crushed by his mother. Behind his false image, Jim felt immensely angry and rebellious. Jim repeated these patterns in the transference, and in his relationship with his girlfriend. He expected to be crushed by the analyst, transformed into the analyst's desired image - his aggression and sexuality removed.

It is always easier to describe and point to a 'false self' than it is to define the 'true self'. The latter might best be regarded as a source of spontaneity, inherently hidden and elusive. Bollas writes as follows:

"In our true self we are essentially alone. Though we negotiate our ego with the other and though we people our internal world with selves and others, and though we are spoken to and for by the Other that is speech … the absolute core of one's being is a wordless, imageless solitude. We cannot reach this true self through insight or introspection. Only by living from this authorising idiom do we know something of that person sample that we are." [1989 p 21].

Thus the 'true self' is an unknown self. Psychoanalysis is the work of releasing this unknown self. This was not how Jim consciously perceived it originally. He believed the analyst's task was to have template of how he should be and then to shape him in accord with this pattern of healthy normality. Thus he envisaged that the analyst would remove one false self and substitute another. Any idea that the analyst did not have such a template was met with disbelief and anxiety. He displayed anxieties expressing the state of the fragmentation that he feared would result from a loosening of the alien structures around which he had organised his 'self'. However, in the later stages of analysis, Jim did discover, with increasing pleasure and developmental excitement, new desires, capacities and possibilities within himself. The emergence of these seemed facilitated by a lessening of the historical transference of the oppressive mother, against the backdrop of a quietly consistent mirror transference. I believe this process corresponded to that outlined by Kohut (in The Analysis of the Self), whereby the vertical and horizontal splits are both reduced: the alien (or false) structure based on adaptation to the mother's narcissistic desire is dislodged (work at the horizontal split), whilst the true self emerges tentatively (and against the resistance of shame) from its place of repression (work at the horizontal split).

Prior to the work of analysis, the patient knows neither self nor other. It is only by clearing away the transference effigies, whereby both self and other in the consulting room are perceived in terms of imposed images, that the truly spontaneous and inherently unknown self can begin to emerge. Crucially, this process depends upon the analyst's (and patient's) toleration of not knowing and not understanding. It requires that the analyst be willing to move beyond his or her own illusory images - to be prepared to face the awe and dread inherent in the encounter with the 'ultimate reality' that Bion [1970] designated as 'O' - a perhaps ironically empty signifier. I believe this is the hidden truth inherent in Kohut's emphasis upon transmuting internalisation. The astonishing implication of this concept is that the patient's most important developments take place in the gaps and failures of the analyst's understanding. As analysts we must try to understand, but the patient's growth takes place at the limits of our understanding - where we can indeed allow the unknown self to emerge.

"… to spend time on what has been discovered is to concentrate on an irrelevance. What matters is the unknown and on this the psycho-analyst must focus his attention." [Bion 1970 p 69]

Balint, M. 1968. The Basic Fault. Tavistock. London.

Bion, W.R. 1970 Attention and Interpretation. Tavistock. London. Reprinted in (1977) Seven Servants. New York. Aronson.

Bollas, C. 1989. Forces of Destiny. Psychoanalysis and Human Idiom. London. Free Association Books.

Fonagy, P. György, G., Elliot, J. & Target, M. 2002. Affect Regulation, Mentalization, and the development of the Self. Other Press. New York.

Kohut, H. 1971. The Analysis of the Self. International Universities Press. New York.

——— Thoughts on narcissism and narcissistic rage. In The Search for the Self. Selected Writings of Heinz Kohut. Vol. 2. International Universities Press. New York.

——— 1977. The Restoration of the Self. International Universities Press. New York.

——— 1979. The two analyses of Mr. Z. International Journal of Psychoanalysis. 60. 3-27. Reprinted in P. Ornstein [Ed.] The Search for the Self. Selected Writings of Heinz Kohut. Vol. 4. 1991. International Universities Press. Madison. Connecticut.

——— 1984. How Does Analysis Cure? University of Chicago Press. Chicago.

Lacan, J. 1948. Aggressivity in psychoanalysis. In (1977) Ecrits. Tavistock. London.

——— 1949. The mirror stage as formative of the function of the I as revealed in psychoanalytic experience. In (1977) Ecrits. Tavistock. London.

Mollon, P. 1993. The Fragile Self. The Structure of Narcissistic Disturbance. Whurr. London.

——— 2001. Releasing the Self. The Healing Legacy of Heinz Kohut. Whurr. London.

——— 2002. Shame and Jealousy. The Hidden Turmoils. Whurr. London.

Schore, A.N. 1994. Affect Regulation and the Origin of the Self. Erlbaum. Hillsdale, NJ.

Winnicott, D.W. 1960. Ego distortion in terms of true and false self. In (1965) The Maturational Processes and the Facilitating Environment. Hogarth. London.


End Note

"The subjective medium is the transference effigy, that is, an invisible image of the object that is transposed between the analytic subject and the analyst, one that the subject cannot distinguish from the analyst without the analyst's interpretive intervention." [Grotstein 1997 211-212].

return to Home page