26   Theory
Klein-influenced cultural critics (e.g. Young 1994) emphasise how psychotic processes of projection, splitting, perverse destructive-ness, hatred and sexual violence lurk not far from the surface of domestic and political life. Others have tried to tease out the moral values implicit in psychoanalysis, in an attempt to link its critique of the micro-society of the family to wider social issues. The moral stance of psychoanalysis includes a number of concerns central to social and political debate. First, there is an overriding valuation of the truth, of the need to face reality however painful, rather than turn a blind eye (Steiner 1993). This leads to scepticism about simplistic solutions in which evil is driven out by love - a balance between the two is a more realistic possibility. Second, the work of Bowlby and the Independent analysts emphasise the link between nurturance and the search for a good society: if we neglect childhood, society will inevitably suffer (Ruštin 1992; Holmes 1993). Thus Winnicott (1971; Meares and Coombes 1994) sees psychoanalysis as a process of 'learning to play' - the 'right to play' would appear on any psychoanalytic political manifesto. Third, psychoanalysis, values autonomy as a good in its own right, independent of freedom from want, a view that is central to the liberal tradition, and demonstrates how the cradle of autonomy is sensitive parenting, the lack of which can, with luck, be remedied by psychoanalysis (Holmes and Lindley 1989, 1994).
Such diverse ideas spring from the metapsychological superstructure of psychoanalysis, but, if psychoanalytic thinking is to have any validity and if it is to be based on more than myths or wishes, it must rest on firm foundations of clinical theory and practice - the exposition of which forms the rationale and purpose of this book.
Chapter 2
Models of the .mind
Such ideas as these are part of a speculative superstructure of psychoanalysis, any portion of which can be abandoned or changed
without loss or regret the moment its inadequacy has been proved.
(Freud 1925a: 32)
The archaeological metaphor can be applied to psychoanalytic theory (see Chapter 1), in which new ideas are built upon Freud's foundations. Some ideas are fully covered over, while others are retained in almost their original form. Freud himself continually reworked his theoretical models and he was not afraid to change them radically when the need arose. His successors have happily followed his example and psychoanalytic models of the mind have undergone many developmental changes. All are mixed models, not necessarily building a coherent whole but forming a complex matrix of ideas containing concepts at different levels of abstraction. Contradictory formulations sometimes sit uncomfortably side-foy-side. This mixing of models not only has arisen from a continual process of replacement and cross-fertilisation but also from the constant interaction of different levels of theory.
Waelder (1962) identified a number of levels of psychoanalytic theory:
1  The level of individual clinical interpretation - which is a theory about particular patients.
2  The level of clinical generalisation in which theoretical ideas relate to specific groups of patients, e.g. 'narcissistic organisations' (q.v. p. 79).
3  The level of clinical theory containing general psychoanalytic concepts such as defence mechanisms or transference - the level
28   Theory
with which we shall be primarily concerned in this chapter. 4 The level of abstract explanatory concepts such as life and death i msti met s,
Underlying different theories are different fundamental assumptions about the world: how much experience is determined by the environment, and how much it is innate; whether a basically optimistic or pessimistic viewpoint is adopted; whether a mechanistic or humanistic view of the mind is taken; on the balance between determinism and freedom; on the emphasis on mental forces as opposed to meanings and language; whether a position of mentalism or realism is adopted.
All these issues have been discussed by philosophical and cultural observers of psychoanalysis (see, for example, Rief f 1960; Wollheim and Hopkins 1982; Greenberg and Mitchell 1982; Holmes and Lindley 1939; and Cavell 1994). From the clinical perspective of this book two dimensions stand out. First, the balance between environmental or intrapsychic factors in the development of the personality; second, whether the emphasis is on causation and mechanism in approaching mental phenomena, or on understanding and meaning. Freud found himself on both sides of these conflicts, initially he placed the emphasis more on environmental factors, especially traumatic external events like childhood seduction, and suggested they led to the formation of symptoms in adult life. In the later drive-structural model Freud saw the internal world as primary. External events were triggers releasing inherent patterns such as the oedipus complex. This difference of emphasis continues to this day: models may be-classified as intrapsychic, interpersonal or mixed. Similarly, although Freud set out to produce a mechanistic picture of abnormal mental life, as his work evolved he became more and more concerned with meaning, narrative and communication.
The dilemma faced by clinicians is that in order to practise effectively, especially at the outset of one's career, a firm theoretical fraroiewvjj/!i ':.-, essential; at the same time it is unlikely that any one model holds the key to the workings of the human mind and human motivation. In practise, most analysts draw on a mixture'of different theories, even if their basic allegiance is to a particular school. This chapter contains a schematic overview of the main psychoanalytic concepts and models, all of which will be elaborated in the course of the book.
Models of the mind   29
The unconscious
The concept of the unconscious is central to psychoanalytic theory. Although Freud did not 'discover' the unconscious (Ellenberger 1970), he was the first to explore systematically its role in normal and abnormal mental life. From a contemporary perspective the unconscious is seen within psychoanalysis in one of four basic ways.
The unconscious as a 'thing in itself
Freud initially saw the unconscious as part of the 'mental apparatus' (Laplanche 1989), a Kantian entity that could not be directly apprehended but explained in a deterministic way irrational mental phenomena such as dreams, neurotic symptoms and slips of the tongue. He postulated that unacceptable memories, phantasies, wishes, thoughts, ideas and aspects of painful events were pushed back by repression into the unconscious, along with their associated emotions. In Freud's unpublished 'Project' he hoped to produce a neurobioiogical account of the role of the unconscious, based on the flow, binding, and discharge of psychic energy, or libido. Although this 'hydraulic' model has largely been superseded, modern neuropsychology has confirmed, via subliminal perception and 'pre-conscious processing', that many aspects of mental life vital to survival take place outside of awareness (Dixon and Henley 1991).
The unconscious as reservoir of latent meaning
With the shift in contemporary psychoanalysis away from mechanism towards meaning, 'the unconscious' becomes a metaphor for the affective meanings of which the patient is unaware, and which emerge through the relationship with the analyst. 'Unconscious' becomes an adjective rather than a noun: 'unconscious processes', rather that "the unconscious*. This links psychoanalysis with the 'post-modern' notion of polysemy or multiple meanings which are to be found in any cultural phenomenon or 'text'. The analyst's task, rather than being an anatomist of the mental apparatus, is collaboratively to explore these latent meanings.
The mystery of the unconscious
Jung (1943) emphasised a less tangible quasi-mystical aspect of the unconscious. He was particularly interested in religious and spiritual aspects of human experience and introduced the concept of the
30   Theory
collective unconscious. He thought this was innate, common to all mankind, and existed at a deeper layer of the mind than the personal unconscious described by Freud. He based this view on the finding gi-that the beliefs, symbols and mythology of widely different religions  . and cultures have had a great deal in common throughout the ages ^ and in different parts of the world.
Pásť unconscious and present unconscious
Sandler and Sandler (1984) made a clinically useful distinction between the 'past unconscious' and the 'present unconscious*. The past unconscious of the adult is the 'child within' that continues in an unmodified form to have a powerful role in determining the adult's responses, wishes and needs as if childhood were still operating. More akin to the preconscious (see below, p. 32), the present unconscious modifies the past unconscious by the use of mechanisms of defence, allowing the past unconscious phantasies some expression, albeit in an attenuated form. The Sandlers argue that, in treatment, the analyst should always work from present unconscious to past unconscious, to attend to the here-and-now of the analytic interaction before proceeding to reconstructions of past traumata.
FREUB'S MODELS
Freud's picture of the mind went through three main phases, which Sandler et al. (1972) have called the affect-trauma model, the topographical model and the structural model. We shall consider each in torn.
Áffect-íraMWfL& model
Freud's earliest psychoanalytic ideas were influenced by experiences , with casualties from the Franco-Prussian war in which hysterical paralyses seemed io be related to traumatic experiences at the battle front, relieved once the sufferer was able to speak about (abreact) the terrifying events he had gone through. Once Freud had noted v^hat he took to be the frequency of childhood sexual abuse among his hysterical patients, he speculated that, by analogy with battle trauma, painful external events such as childhood seduction could overwhelm the "mental apparatus', leaving it unable to deal with the resulting affects. He tried to differentiate between the 'actual neuroses' (neurasthenia and anxiety neurosis) which he believed
Models of the mind   31
were caused by current trauma, and the 'psychoneuroses' (hysteria and obsessional neurosis), which he saw as the result of childhood trauma. This distinction is now obsolete (q.v. p. 212), although current trauma, such as a recent accident or bereavement, may lead to symptoms of Post Traumatic Stress Disorder.
The idea of the release of 'dammed up' affects (i.e. feelings), threatening psychic equilibrium and potentially leading to symptoms, has been called the 'affect-trauma* frame of reference forming part of the first phase of Freud's metapsychology. Affect-trauma plays an important part in contemporary psychoanalytic thinking, especially since the reality of childhood abuse, physical, emotional and sexual, has become apparent. The náture of infantile trauma is viewed differently by different authors, some emphasising intrapsychic factors, others stressing environmental influences. From an intrapsychic perspective, Klein and Kernberg see the intensity of childhood hatred, aggression and envy as inherently traumatic and speculate that potential borderline personality disorder sufferers may experience an excess of such unassimilable negative emotions, leading in turn to excessive splitting and projection (q.v. pp. 81-7). In contrast, for Kohut the primal trauma is interpersonal: a failure of parental empathy, leading to disruption of a coherent sense of self and the emergence of 'disintegration products' in later life such as aggression, or attempts at self-soothing through addiction, compul-ive sexuality and even self-injury.
In both accounts trauma leads to painful affect which, in turn, provides the motivational force behind pathological reactions. The potency of affect as an organiser of mental functioning has received some support from Bower (1981) who found that the recall of memories is affect-dependent. Some memories can only be recalled when the same mood state is present as when they were laid down. The recovery of buried memories of trauma in the transference also enables them to be reworked in a less overwhelming way: loss can be grieved, and anger expressed, leading sometimes to acceptance or even forgiveness.
Tfee topograpMca! model
The term 'topographical' implies a spatial model in which different psychological functions are localised in different places. The division of the mind into the unconscious, the preconscious and conscious systems (Freud 1900), the so-called topographical model, ushered in
32   Theory
the second phase of Freud's work (1897-1923), still containing echoes of cerebral localisation, a reminder of Freud's previous career as a neurologist. Contemporary psychoanalysis still uses topographical concepts but has divested itself of their anatomical overtones, jost as Freud did, by talking about each part of the mind as a 'system* - the 'system unconscious', the 'system preconscious', etc. This enables a smooth transition to the 'structural model' (see below), which is primarily concerned with the functions of the different parts of the mind.
The 'twoprinciples'
A fundamental idea that comes out of this phase of Freud's theoretical development is his contrast between the 'two principles of mental functioning' (Freud 1911a), which he called the primary and secondary processes. Secondary process thinking is rational and follows the ordinary laws of logic, time and space. Primary process thinking is typical of dreaming, phantasy, and infantile life in which the laws of time and space and the distinction between opposites do not apply; the distinction between past, present and future no longer holds and different events may occur simultaneously and in the same place; one symbol may represent a number of different objects, or have several different and even contradictory meanings.
Unconscious and preconscious
Freud realised that many psychological processes are unconscious in
a descriptive sense and the individual is not aware of them, but they are easily brought to'mind, and therefore are neither subject to repression nor operating under the sway of the primary processes. These unconscious but non-repressed phenomena he attributed to 'the system preconscious', whose role in the topographical model is both as a reservoir of accessible thoughts and memories, and as a censor capable of modifying instinctual wishes of the system uimconsci ■;:>.>? and so render them acceptable to the system conscious.
instinct theory
The change from the affect-trauma frame of reference to the topographical model represented a significant shift in the evolution of psychoanalytic theory. The spotlight moved from a focus on
Models of the mind   33
external reality and its impact on psychological processes to the internal world itself. For most of his life, Freud saw the internal world as dominated by man's struggle with his instincts or drives.
instinct or drive theory was put forward by Freud to explain human motivation. Confusion has arisen because the term 'instinct' was used to translate both 'Instinkt' and 'Trieb' from the German. Instinkt refers more to innate behaviour patterns and responses while Trieb implies a pressure or push towards a general goal, such as survival. Unfortunately Freud's translators used both words interchangeably, perhaps reflecting the continuing struggle within psychoanalysis between biological and psychological concepts. Freud considered instincts as basic developmental needs constituted from phantasies which had a peremptory quality and required expression and gratification.
Freud was always an 'obstinate dualist' (Jones 1953). In his initial formulation of instincts Freud (1905a) emphasised the sexual drives both in normal development and in the origins of psychological illness; later he emphasised the aggressive and destructive drives or death instinct (Freud 1920, 1930). This became known as the dual-instinct theory, although Freud had earlier stressed the self-preservative (i.e. death-defying) instincts as well as the sexual. The individual was at the mercy of these drives, or instinctual wishes, with adult symptoms arising from the psychological defences mobilised to deal with their infantile demands. Each instinctual wish forms a component of the system unconscious and has an innate need for 'discharge'. In order to achieve its aim it becomes connected during development to an object.
In this classical schema instinctual wishes have a source and an aim as well as an object. Their infantile source or origin is in the body and may be an 'erogenous zone' such as the mouth, anus or genital; sensations in these bodily areas develop levels of tension which aim towards discharge. The infant experiencing hunger pains starts to suck. Gradually his activity increases until the object of the instinctual wish - the mother's breast - is provided and hunger is satiated. The presence of the object and its quality is remembered and the next time the infant feels hungry the memory of the breast is revived and the phantasy is once again enacted. The source, the aim and the object begin to mesh together into a complex interactional phantasy, part of which is represented in the system unconscious.
Although, in this example, the experience and revived memory
34   Theory
derive from a real event - the knitting together of an instinctual wish - its need for expression and subsequent satisfaction can also develop from an imagined occurrence. If an infantile, wish-fulfilling
daydream is iived' within the self and later repressed, the system unconscious will treat the imagined events according to the principles of primary process and, therefore, as if they were real.
The distinction between the recovery of a traumatic memory and the revival of a wish-fulfilling fantasy, both of which have been repressed, is difficult to make in the clinical situation. This also means that reports by patients of their earlier experiences cannot necessarily be taken at face value. Recent accounts of the 'false memory syndrome', in which imagined traumatic events are experienced as real, might, for example, illustrate this principle.
Limitations of the topographical model
Clinical experience Sed Freud to acknowledge an increasing number of inconsistencies in the topographical model. Foremost among these was the realisation that there was no place within his map of the mind for ideals, values and conscience. Moreover, he recognised that the influence of the external world on mental structures and the unconscious nature of defence both needed more exploration. For example, anxiety had initially been seen as a result of the accumulation of repressed 'somatic sexual excitement', or libido (see Chapter 1, p. 8), which was transformed from a sexual wish into an unpleasant feeling; that is, as arising entirely from within. But it became clear thai anxiety also arises in response to threat, either directly, or as part of a psychological conflict induced by the threat. For example, a child who feels anger towards a neglectful parent, may be fearful of expressing this anger for fear of losing the parent altogether. The internalisation of this parent-child relationship was hard to reconcile with the topographical model:
Similarly, in his papers 'On narcissism' (1914a) and 'Mourning and melancholia' (1917), Freud had begun to stress the interaction betweer. tV internal world and external events, especially in his discussion of internalisation and identification. He began to wonder how an experience of. say, a harsh or punitive parent, comes to form a structural part of an individual's internal world. The advent of the structural model (Freud 1923) was Freud's attempt to answer these questions, heralding the third and final phase of his theorising (1923-39).
Modeis of the mind   35
Structural theory
In the structural model Freud (1923) proposed three parts or 'structural components' of the human personality: translated or, according to Bettíeheim (1985), mistranslated as the familiar id, superego and ego. As with topographical theory, these are nowadays best thought of as functions, rather than as structural entities, as metaphors for psychological configurations showing a slow rate of change and reactive stability (Rapaport 1967; Friedman 1978). The structural model remains firmly embedded within instinct theory. The term 'id', for example, refers to the basic inborn drives and the sexual and aggressive impulses. But the structural model is equally concerned with how the individual's personality structure adapts to the demands of these instinctual wishes and it places a greater emphasis on external reality.
Superego
The earlier notion of the 'ego ideal' (Freud 1914a), an internal model to which the individual aspires or attempts to conform, was subsumed by the wider concept of the superego. The term 'superego' is used to describe conscience and ideals; like the ego ideal these are derived through internalisation of parental or other authority figures, and cultural influences from childhood onwards. From an object relations perspective (see below) the superego represents not so much an internalised parent as a relationship with a parent. The internalised parent of the superego may therefore be a representation of a parent into whom has been projected much of the individual's own aggression and punitiveness.
The internalised parental and other figures of the superego are thus formed from phantasy as well as reality and, as such, contain significant components of external i sations and projections of aspects of the self. The whole structure therefore functions according to these modified internal objects, which explains why clinically, for example, a parent who is experienced as harsh may in reality appear to have been relatively benign. The superego is involved in the experience of guilt, perfectionism, indecision, preoccupation with what is the right or wrong thing to do, and hence plays an important role in the aetiology of some forms of depression, obsessional disorders and sexual problems.
Some effects of the functioning of the superego are descriptively
3S    s heory
conscious whereas others are descriptively unconscious. For example some individuals may be quite clear that what they wish to do goes against accepted values or indeed against their own upbringing; in others there is an unconscious sense of guilt (Freud 1923) - a patient tormented by obsessive-compulsive acts may have no conscious idea of why he is compelled to do something and feels excessively guilty if it is not done.
Ego
The term 'ego' is used to describe the more rational, reality-oriented and executive aspects of the personality, and once again is partly conscious and partly unconscious. The ego's task, as seen by Freud, was to control the more primitive id impulses and to adapt these to outer reality in accordance with the reality principle, as well as to mollify the requirements of the superego: 'the poor ego. . . serves three masters and does what it can to bring their claims and demands into harmony with one another. ... Its three tyrannical masters are the external world, the superego and the iď (Freud 1933).
The structural model cannot easily be superimposed on the topographical theory. The system unconscious and the id are equivalent, functioning according to primary process thinking, but the system conscious and the ego are not equivalent, since part of the ego may be unconscious. Equally, preconscious and superego cannot be equated. The key issue clinically is not just whether the patient is conscious or unconscious of some aspect of himself, but what part of the mind holds sway: is the patient behaving and thinking according to the primary processes; under the dictates of conscience, or adaptively?
Conflict and adaptation
Implicit in the idea of an inner world is that of delay. A wish is shaped, influenced, modified, held back, diverted or disguised by the 'pale cast of thought'. Instinctual wishes cannot obtain direct express ľ;;, within the topographical model they have to traverse the preconscious before they reach consciousness; in structural theory the superego and ego hold them back. By the time they reach consciousness they have been modified to such an extent that they can only be pieced together through dreams, parapraxes and, in the clinical situation, transference. Modification of instinctual wishes is effected through the use of the mechanisms of defence (see pp.
Models of the mind   37
81-93), which are mobilised as a result of internal conflict. Conflict occurs between the instinctual wishes under the sway of the 'pleasure principle' (Freud 1920) and the demands of reality - in simple terms, between past and present, or between the inner child and the functioning adult. In the structural model, conflict is seen as occurring between the three structures themselves and between each one of them and the external world. Through the impingement of reality, gratification of the instinctual wishes is delayed or modified if they threaten the self-preservation of the individual, or contravene his moral and ethical beliefs, or oppose the demands of his social and cultural environment. The introduction of the reality principle by Freud (1911a) heralded the move from a primary focus on the internal world rather than on the external environment to an increasing emphasis on the relationship between the two.
POST-FREUDIAN MODELS
In the classical model, psychoanalysis tends to see the personality as a battlefield: central themes are those of innate division and conflict, internal tension and adaptation. Throughout development a struggle occurs between internal demands and external reality, with internal needs taking a primary motivational role. Repression is viewed as the primary mechanism of defence, ensuring that incompatible wishes remain unconscious or disguised, but because of the intrinsic tendency of the repressed wishes and impulses to return to consciousness, tension remains an innate part of the system..
Ego psychology
Heinz Hartmann (1939, 1964), the founder of ego psychology. questioned this view, stressing instead the non-defensive aspects of the ego (and was himself much criticised by Lacan, see pp. 65-6). Instead of the ego simply being a mediator between the demands of the id and the external world. Hartmann conceived of the ego as, in part, outside this area of conflict and thus able to interact with the external world free from internal influences. This 'conflict-free sphere of the ego' develops independently, and flourishes unimpeded by conflict if environmental influences are reasonably favourable. It contains such functions as thinking, perception, language, learning, memory, and rational planning. Development of these aspects of the personality may influence the experience of pleasure and satisfaction.
38   Theory
Freud had moved the focus from the external world to an emphasis on the internal workings of the mind within the topographical theory, and back again in the structural model. Hartmann moved things a little further towards reality by suggesting that the experience of pleasure did not arise simply from satisfaction of instinctual wishes but also depended on what good experiences the external world was in reality able to provide.
A further contribution of the ego psychologists was their distinction between Freud's view of the ego as a structure, and a more modernist concept of the ego as a representation of the self- a view that was developed by Kohut in self-psychology. Anna Freud (1936) also re-eraphasised the importance of the relationship of the ego to the external world and the normal and adaptive aspects of the personality. She considered defence mechanisms as not only responding to the dangers of the internal world but also - for example, in 'identification with the aggressor' (see pp. 88-9) - to those of the external world. However her approach was less rigorous in its adherence to the structural model than that of the ego psychologists, and she continued to emphasise the usefulness of the topographical model. Jacobson (1964) developed these ideas further, but ego psychology in general has recently become more integrated with object relations theory, for example in the work of Kernberg (1976, 1980), Arlow (1991), Gil! and Hoffman (1982) and Sandler (1987).
Tine KOeJk-Bioni moáel
Although primarily a clinician rather than a theory-builder, it is generally agreed that Melanie Klein was one of the most original and challenging thinkers in the history of psychoanalysis. Klein saw that for all its advantages, something had been lost in the move from the topographical to the structural model, especially the notion of unconscious phantasy within intrapsychic life. By focusing on early pre-oedipal experiences Klein hoped to reconcile the apparent opposition between those that emphasised phantasy on the one hand, and drivo iLtiory on the other.
The Kleinian 'positions'
Klein is perhaps best known for her .account of the two basic 'positions' of mental life, the 'paranoid-schizoid' and the 'depressive', and for her notion of 'projective identification* (see p. 82).
Modefs of trie mind   39
In essence the Kleinian 'positions' are constellations of phantasies, anxieties and defences which are mobilised to protect the individual from internal destructiveness. In the earlier, paranoid-schizoid position the focus of the anxiety is on threats of annihilation and disintegration, and the infant attempts to organise these experiences by the use of splitting and projection (see p. 81). Bad experiences are split off and projected into the object which is then felt to be persecuting and dangerous and especially threatening to the good experiences. In order to protect the good experiences, they too may be projected into the object which then becomes idealised. In the later depressive position, anxiety is not so much about the survival of the self but the survival of the object upon whom one depends. The individual realises that the frustrating and hated object is also the one that satisfies and is loved. Recognition that they are one and the same leads to ambivalence and guilt. Although the positions are set out schematically there is a constant oscillation between the two, and a third 'borderline' position has also been described (see p. 226).
Phantasies and drives
For Freud, libido and aggression were structureless phenomena whose form was dictated by developmental bodily stages as well as by drive gratification or frustration. But for Klein the instincts are inherently attached to objects, as preformed 'primary phantasies' (the 'ph' differentiating them from 'fantasies' in the vernacular sense of daydreams or conscious wishes). The basic unit- of mental life therefore becomes object-related unconscious phantasy itself, rather than instinctual wishes that seek expression through 'self-created' objects.
There is a continuing psychoanalytic debate about the degree of innate knowledge possessed by the infant and the extent to which it is formed throughout development. For Klein destructiveness and unconscious phantasy are innate and 'primary', but for Freud they arise out of frustration, and are therefore 'secondary'. Klein believed that the infant's mind contained complex preformed images of the object. For example, in''Envy and gratitude' (1957) she states that 'the infant has an innate unconscious awareness of the existence of the mother' and that this forms 'the instinctual knowledge that is the basis of the infant's primal relation to the mother'. Thus for Klein the unconscious has specific contents right from the start of mental life,'namely unconscious phantasies, which
40'   Theory
are the menial corollaries or the psychic representations of instincts. An instinctual wish can only be experienced (as opposed to theorised about by psychoanalysts) as an unconscious phantasy (Isaacs 1943). This concept of a preprogrammed thinking infant, more sophisticated than Freud's tabula rasa, has received some confirmation from developmental psychology, although Stern's (1985) picture of happy infant-mother atiunement is a far cry from Klein's model of primary envy and hatred (cf. Chapter 3).
Bion and containment
In Freud's early writings 'the object' only appears as the provider or withholder of gratification. By the time the oedipal stage is reached, objects are folly formed; how they became so is unclear. Klein tried io reconcile drive theory with object-finding in her notion of the primary object, but it was Bion, her analysand, who moved Kleinian theory decisively away from drives and towards relationships. In his concept of 'the container and the contained', Bion (1962) extended the idea of projective identification (see p. 82) and suggested that the mother acts as a container for the infant's projected feelings, such as pain, fear of death, envy and hatred. These feelings are 'detoxified' by the nurturing (or in the case of analysis, 'listening') breast, and then returned in such a way that the infant gets back good feelings of being held and understood rather than the original bad projections. In this way the infant makes sense of his experiences, and introjects an object that is capable of bearing and. allaying anxiety. There is thus an explicitly stated interactional component, even though the 'death instinct' remains as the organising force for the projections.
Oltpjeei relations Äe«wry
Although Bion moved Kleinian thinking into the realm of object relations, Fairbairn (1952) and Guntrip (1961) are usually seen as the faCicrs -jf present-day object relations theory. We have already suggested that models of the mind may be divided into those that focus primarily on the internal world (Freud's structural model -Klein), those that focus more on the external world (the 'Neo-Freudians' - Sullivan, Fromm, Homey, Erikson and Bowlby, see below) and those that lie somewhere between the two (Winnicott, Bion, ICohut - Freud's affect-trauma model). A further distinction
Models of the mind   41
may be made between those object relations models that incorporate drive theory and those that do not, Fairbairn and Guntrip made no such attempt and neither did Sullivan, whereas Mahler (see pp. 60-1), Klein, Kernberg and Kohut tried to do so, although the latter, as we shall see, downplayed aggressive drives. Most of the British writers, such as Winnicott and Balint, have had no difficulty in combining the two, especially in their clinical formulations; and more recently Sandler (1981) has proposed a mixed model. Despite their differences, Fairbairn, Guntrip, Winnicott and Balint have been lumped together asüie British object relations theorists (Sutherland 1980; Greenberg and Mitchell 1983; Phillips 1988). Although their theories take a variety of forms they share a number of pivotal assumptions (Westen 1990).
Object seeking
Central to the theory of object relations is the belief that a person's primary motivational drive is to seek a relationship with others. Humans are primarily object seeking rather than pleasure seeking, and the end goal is the relationship with another person. The infant's early activity is directed towards contact with the mother, and later others: 'pleasure is a signpost to the object, rather than vice versa' (Fairbairn 1952). The method of object seeking varies according to the stage of development: initially it is through feeding (including mutual gazing, Wright 1991) and later through the sharing of activities and interests. This does not completely overthrow the concept of pleasure seeking, since, as Balint (1957) sensibly says, the individual is both object and pleasure seeking. The compulsive quest for pleasure may of course also be a pathological response when object relationships fail.
The representational world
The core notion of object relations theory is that of an internal world populated by the self, its objects and the relationships between them. This is Sandler's 'representational world', which he likened to a proscenium stage upon which the scenes and dramas of inner life are enacted. The relationships between these internal objects act as templates for subsequent relationships, especially when the primary processes are operative. Intimate relationships with partners and with the analyst will be profoundly influenced by the valencies of the
42   Theory
internal world. In contrast to Klein's idea of primary object phantasies, Fairbairn (1952) conceived of internal objects and the phantasies associated with them arising as a consequence of the inevitable failure of external objects. For Fairbairn this leads to a split at the heart of the psyche between the 'libidinal object' which gratifies and the 'anti-libidinal object' which frustrates. These objects are associated with corresponding libidinal and anti-libidinal self-representations. Like Freud, he suggested that the internal world developed as a substitute and compensation for unsatisfying experiences in external relationships, with aggression as an organising factor, secondary to these frustrations. He also stressed that what is internalised is not an object as such but a relationship. This point is often overlooked.
Transitional space
Guntrip was analysed first by Fairbairn, a fairly dour Scot (Sutherland 1989) and then by the benign Devonian, Winnicott (Phillips 1988). The latter, as already described, took a much more positive view of human relationships, seeing creativity and the internal world as natural results of a good-enough mother-infant relationship. Winnicott insisted that object relations theory had to understand not just internal and external objects but their mutual interplay. This he located in a 'potential space' which is experienced as being neither inside nor outside but in between.
Transitional phenomena are the missing link between Freud's pleasure principle and the reality principle. Through his notion of 'transitional space', Winnicott (1965) attempted to reconcile drive theory with an interpersonal perspective. He believed the drive-driven child conjures up in his mind an object suited to his needs, especially when excited. If, at this precise moment, the 'good-enough' mother presents him with just "such a suitable object, complementary to his wish, a moment of 'illusion' is created in the baby, who feels that he has 'made' the object himself. The repetition of these 'hallucinatory' wishes and their embodiment ('realisation') by the mother leads the infant to believe he creates his own world. This omnipotence leads to healthy development of a creative and playful self. Only once this 'true self has been established can omnipotence be abrogated and the reality of pain and loss be faced. Where the mother is not 'good enough', a compliant 'false self arises, concealing frustrated and sequestered instinctual drives.
Moders of the mind   43
Gradually the developing infant comes to differentiate between internal and external and reality and illusion, realising that there is an outside reality that is not simply the result of one's own projections. This results in an experience of contacting other minds and a greater sense of oneself. Guntrip believed that this process could be recreated in analysis through contacting the 'regressed ego', a helpless and vulnerable aspect of an unloved self, defended against by compulsive object seeking, and curable through regression into transitional space within the analytic relationship. Balint also emphasised the importance of regression as a therapeutic tool in the analysis of disturbed or 'basic fault' patients (i.e. those with borderline pathology).
Hate
Just as Winnicott was able to reconcile libido with relationships in his notion of transitional phenomena, so in his concept of positive hatred he tried to relationalise the death instinct. In his paper 'The use of the object' he distinguishes between two types of experience which he calls 'relating to the object' and 'using the object'. Initially the object is related to part of one's own mind and not necessarily experienced as real or separate from the self. Later in development, when the object is experienced as real and independent, a different type of relationship can be entered into in which there is an exchange of a shared reality. Winnicott called this 'using the object' and linked it to the individual's struggle to recognise and be recognised by the other. The driving force behind this struggle was 'hatred': for the object to be recognised and experienced as outside the person's control it has to be destroyed in phantasy but then experienced as surviving in reality. Here destruction becomes not so much a dangerous damaging force, but part of the separation-individuation process.
Winnicotťs creative synthesis of the drive-based and relational models did much to prevent an ossification of theoretical views within the British Psycho-Analytic Society. However, in the USA, many psychoanalysts began to chafe at the rigidity of ego psychology, seeing its exclusive emphasis on the oedipal situation and instinct as limited and limiting. Furthermore, in the 1960s there was a cultural shift leading to an interest in the self, both as a positive arena of personal liberation, and negatively as a withdrawal from relationships into self-aggrandisement and self-gratification. Within psychoanalysis these two factors crystallised in Heinz Kohut's self-
44    i heory
psychology. According to Lasch (1979) and Schäfer {1911) this represented a shift from thwarted instinctual gratification to a concern for self-fulfilment, a move from guilty, oedipal man suffering from internal conflicts within a cohesive self, to tragic man struggling with problems of cohesion and the very integrity of the self. In addition to its reaction against the orthodoxies of ego psycholog;/, self-psychology has its roots in the psychoanalytically heretical interpersonal model put forward by Sullivan (1962) which we shall mention first.
Tihe iim£erpersoraal modleli
The interpersonal model, which was developed by the so-called Neo-Freudians - Sullivan (1962, 1964), Homey (1939), Fromm (1973) and Erikson (1965) - takes a radically interpersonal stance; here, to paraphrase Winnicoft, 'there is no such thing as an individual*. Sullivan was a psychiatrist who became convinced that the strict Kraepelinian view of schizophrenia in the 1920s was wrong (i.e. in its view of the illness as a biologically determined, irreversible deterioration of the personality leading to complete breakdown of mental and emotional functioning). Well ahead of his time, he realised that much of what was seen as 'schizophrenia' was the result of institutionaiisation, rather than disease process. -He argued for stimulating human relationships and not just custodial care. It was in this context that Sullivan developed his interpersonal theory.
In common with other psychoanalytic views, the interpersonal model emphasises early mother-child interactions as central to the subsequent development of the personality, but does not see the child's internal world as being the determining influence. Instead, the drive-structure model put forward by Freud is reversed. Anxiety, rather than arising from within as a result of unconscious instinctual wishes pushing for expression and satisfaction, is seen as being stimulated from without, a response to the state of mind of the other. The child forms specific mental representations according to the anxiety ihst is engendered and imagines that a 'Bad Me' elicited anxiety in the (m)other. In the same way a 'Good Me\ which alleviates anxiety, is also set up along with a 'Not Me'. The 'Not Me' is a response to severe panic and confusion, akin to Guntrip's vulnerable helpless self at the core of the schizoid mode of being, which is a nucleus for subsequent psychotic fragmentation.
Anxiety experiences are elaborated into stable interpersonal strat-
Modeis of the mind   45
egies which are manoeuvres designed to establish a sense of security. Such 'security operations' include avoidance, inattentions, tactical misrepresentations and other interpersonal strategies. Although expressed in rather simplistic terms (there are clear links with Sandler's notion of the 'child within'), interpersonal psychonalysis contributed to a significant technical shift away from 'reconstruction' to a focus on the here-and-now in the analytic situation. The interpersonal approach was clearly a reaction to the esotericism of psychoanalytic theory, and encouraged a simple, less theory-laden, more collaborative relationship between patient and analyst, in which the transference was seen primarily as an intensified slice of life, rather than a distortion of the present by complex phantasies about which only the analyst had expert knowledge.
Selff-psyctsology
Kohut (1971, 1977) challenged psychoanalytic orthodoxy in the USA, claiming that a new approach which went beyond oedipal analysis was needed if patients with narcissistic disorders, who were becoming increasingly common, were to be successfully treated. The focus of his theory became the 'self, and the effect that denial, frustration and fulfilment of washes has on its development. At first Kohut tried to build on both object relations theory and ego psychology and he portrayed the self as arising from mental representations within the ego, an elaboration of the idea of self-representation. Later he came to. depict the self as a supraordinate structure with its own developmental line which subsumed instinctual wishes and defences.
Necessary narcissism
Just as Hartmann had postulated a 'conflict-free' zone of the ego, Kohut built on Freud's notion of 'primary narcissism' (q.v, pp. 55-6), to suggest that self-love was necessary for psychological health, seeing narcissistic disorders as resulting from defects in the self brought about by parental empathic failures. He postulated first a 'bipolar' self and later a 'tripolar' self in which self-assertive ambitions crystallise at one pole, attained ideals and values at another, and talents and skills at the third. Pathology may arise from a disturbance at each pole and may be compensated for by strength in one of the others.
46   Theory
The idea.of Ehe self, or narcissistic developments following a separate developmental pathway, can be seen as an expansion of Freud's view of psychosexual development, and of Anna Freud's (1965) notion of separate developmental lines along drive-, ego-, and object-related pathways. However the view that the self has a supraordinate or unifying, overarching perspective on personality development is more controversial. The main point of contention is Kofauťs view that aggressive drives are secondary, arising from an insufficiently consolidated self brought about by empathic failures. Kernberg (1975, 1984) particularly has called attention to this de-emphasis of the aggressive drives. It is also difficult to see in Kohuťs schema where the superego fits as an organising focus, if ideals and values are seen as parts of the self.
The central building block of self-psychology is the self-object; this is one's subjective sense of a sustaining intimate relationship with another whose security and interest maintains the self. Self-object needs were initially described in the treatment of narcissistic patients, but are now considered to be ubiquitous and enduring and a requirement of the normal psychological functioning of the self. Self-object needs lead to 'self-object transferences' (see pp. 106-7), comprising mirroring, idealising and twinship transferences, each corresponding to a different pole of the tripolar self.
The term 'self-object' has come to be used in a generic fashion to describe the role that others perform for the self in relation to mirroring, idealising and twinship needs. These needs, are never outgrown and self-objects are best viewed as aspects of others which are required to gratify the psychological needs of the self, such as engendering security, soothing, admiration and so on. This viewpoint differs markedly from the drive-structural and object relations views on the importance of separation-individuation, although Bowlby's attachment theory also acknowledges the continuing need for dependency. In self-psychology the focus is on the need for empathic and affirming responses from others throughout life, but with a move from reliance on archaic objects towards mature dependency.
CONCLUSIONS
Each developing analyst has to struggle with the tension between conservatism and innovation within the analytic tradition. At one extreme there is a desire to overthrow parental authority and deine an entirely new territory of discourse; at the other there is a
Ivlodeis of the mind   47
determination to preserve what is good in the old. These extreme "oedipaľ reactions, while necessary in exceptional circumstances and for exceptional thinkers, are not part of 'normal science* (Kuhn 1962). The developing analyst is in the position first of an oedipal child who has to negotiate both healthy identification and separation in the course of his intellectual development, and later is in a parental position, having to reconcile, as far as possible, the different voices of the competing analytic factions.
Self-psychology's emphasis on empathy, environmental failure, positive narcissism, and challenge to a drive-based interpretive analytic stance was a necessary counterweight to the excesses of ego psychology. Indeed virtually ail the theories discussed in this chapter developed because of dissatisfaction with aspects of a prevailing theory. Most psychoanalytic models are incomplete, as indeed are all models of the mind from whatever perspective. Different psychoanalytic models are relevant to different aspects of a complex whole, usually emphasising one aspect at the expense of another. Psychoanalytic disputes neglect this complexity and miss the point that differing views are often an attempt to remedy weaknesses within another theory rather than to overthrow it.
Freud's language, influenced by the physics of his day, lives on: psychoanalysts still speak of object, drive and their mutual 'dynamics'. At the same time, in the contemporary search for a unified theory, three themes stand out: representation, affect and narrative. For Sandler (1981) self and object representations are what guide the individual in his relationship with the external world. Sandler suggests that the primary motivational element is the regulation of feeling states to maintain a sense of security, rather than drives. Similarly, Stolorow et al. (1987) have argued that the notion of endogenous drives should be abandoned and replaced with affects as motivational elements formed within the interaction between self and self-objects.
In affect theory meaning intersects with mechanism; Meanings are a way of organising and 'fixing' problematic emotional experience into coherent narratives which explain the self's relationship to its world (Elliot and Shapiro 1992). The focus of interpretive work is no longer on instinctual conflicts, frustration of wishes, or aggressive drives but on the patient's affective experience, its origins within the analytic relationship, and the translation of that experience into coherent stories or narratives that make sense and act as guides and warnings for future action. Pine's (1981) emphasis on the importance
48   Theory
of "intense moments' during both development and treatment is a good example of this.
However, jest as psychoanalysis sees the personality as 'a precipitate of abandoned object cathexes', so changes in theory and practice often have their precursors. For example, a focus on the affective experience of the patient is not new. Fenichel (1941) summarised the problem of resistance as either an intense affect defending against cognitive awareness of unconscious conflict, or the reverse, intel-lectualisarion as a defence against affective experience. Too great a reliance on cognitive or affective experiences impoverishes the understanding of the complexity of human motivation and interaction and loses the unity of affect and cognition.
Psychoanalytic models of the mind remain in a state of development and intellectual tension. Some (Rycroft personal ■ com-munication) have argued the need for a 'new paradigm' in the psychological sciences arising out of, but going 'beyond, current psychoanalytic thinking. If psychoanalysis is to remain a relevant and living discipline, it musí open itself up to findings in related disciplines such as child development, linguistics and cognitive science. Similarities, differences and contradictions both within and without psychoanalysis must be accepted and, where possible, worked through to a new synthesis.
To our surprise we find the child and the child's impulses still living on . . .
(Freud 1900: 191)
Freud was a Darwinian. He saw in the adult mind vestiges of its evolutionary and developmental history and believed that psychological illness could best be explained by tracing back neurotic symptoms to their childhood origins - 'hysterics suffer mainly from reminiscences' (Breuer and Freud 1895). He was also strongly influenced by the ideas of the British neurologist Hughlings-Jackson (Sulloway 1980), who had demonstrated how in illness the nervous system reverts to more primitive modes of functioning.
For Freud then, as for Wordsworth, 'the child is father of the man'. The .clinical implications of this are twofold. First, it suggests that many of our fears and phantasies, doubts and difficulties, are relics of earlier phases of life, no longer relevant to the adult world in which we find ourselves; and, second, it helps us to be more aware and tolerant of the 'child within' (Sandler 1992) that lives on in the" unconscious mind and continues to influence our adult thoughts and actions.
In this chapter we consider psychoanalytical theories about how the features of healthy adult psychology which we take for granted -a secure sense of self, stable self-object differentiation, the capacity for intimacy and aloneness, a regulated and modulated emotional life, and feelings of safety and self-esteem - emerge from the undifferentiated state of infancy. We shall also look at how adult difficulties have their roots in disturbances of early mental life.
We shall do so in a sequential fashion, but first we must discuss some background issues that influence the overall picture of
50   Theory
development adopted by the differing theoretical approaches discussed in the previous chapter.
PRELIMINARIES
'•Stages* versais phases
Freud, at least in his early writings (Freud 1905a), saw psj'cfaological development as passing through a series of 'stages' - oral, anal, phallic, genital - which individuals traverse on their way to maturity, with pathology arising from 'arrest' at one or another stage. His ideas were based on Ehe embryology of his day in which each part of the organism was thought to be in some way preformed within the embryo, the process of development being an orderly unfolding of these pre-existing parts.
Contemporary psychoanalysis has modified this model in a number of important ways. First, each stage cannot be eqoated with an organ or 'erotogenic zone' in the simplistic way that was orgiiially conceived. Observing the handling and mutual gaze of mother and infant is enough to convince us that the world of the newborn infant is as-much tactile and visual as it is 'oral'; the preoccupations, aspirations and fears of the 'phallic' three-year-old can only be seen in the most tenuous way as centring around pleasure and pride in the possession of, or fear of losing, his genital. Second, therefore, Freud's 'zones' are perhaps better seen as a shorthand or metaphor for different existential themes which preoccupy a child at different stages of his development.' Erikson's (1965) stages (see pp. 71-2) - trust vs mistrust, autonomy vs shame, initiative vs guilt, and so on - attempt to capture the fundamental issue that reflect a child's attitude towards the world rather than the predominance of a particular part of the body.
Third, the idea of 'sensitive periods' associated with particular ages is questionable. For instance, the 'anal' phase is often characterised as occurring in the second year of life and to reflect the child's capacity for control and ability to resist parental impositions. But, as Stern (1985) points out, a child can say ''No!' at four months by gaze aversion, at seven months with gestures, at fourteen months by running away, and at two years verbally.
Thus it is perhaps more useful to think not in terms of stages but of Erikson's (1968) concept of 'epigenesis'. This term harks back to the -nineteenth-century embryological debate between the pre-
Origins of the internal world   51
formationists on the one hand, and those who believed that each generation is created 'epigenetically' by the developmental process itself. Epigenesis implies (a) that development is not predetermined but the result of an interaction between the developing individual and his environment; (b) that there are many different possible developmental pathways, only some of which will unfold in any particular environment; (c) that 'stages' are not superseded or arrested at,-but rather remain active as phases-or 'developmental lines' throughout life which may be activated at times of stress; and (d) that the impact of environmental trauma continues throughout the life cycle - the most obvious example being that of childhood sexual abuse which most commonly occurs in latency or eariy adolescence - although good experiences in early iife have, to some extent, a buffering effect (Westen 1990).
Example: the woman who could not trust
Martha came into therapy when she became deeply distressed on discovering that her husband had been lying to her. He lied not just about bis affairs, but a whole tissue of deceit about his past and present activities - 'a compulsive liar', she called him. Formal and correct in her manner, and slightly distant, she recalled a childhood dominated by waiting for her father to return from the war. 'He will soon be back' she was told as she sat at the window waiting and watching, but he never did return. He was dead, and her mother and grandfather who brought her up knew it. This had never been discussed between her mother and herself, even to the very end of her mother's life. Martha's relationships, including that with her analyst, were typified by a mixture of naive trust and emotional withdrawal.
Here we see how a 'relationship theme' - the inability to trust -can permeate a person's life as a developmental line, forming a focus for analytic work. For Martha the links stretched forward from the 'untrustworthy' father who never came home, to the husband and the" analyst; and backwards, by inference, to her earliest relationship to a mother who, grief-stricken in her denied and therefore unmourned widowhood, was probably unable to respond sensitively and trustworthily to her baby's needs.
Internal objects (or archetypes) versms tabula rasa
Freud (1915) assumed that the infant had early knowledge of parental intercourse ('the primal scene'), castration and oedipal
52   Theory
seduction. Klein and Isaacs extended this idea and claimed that the infant has inherent functional knowledge of the 'breast', 'penis', 'urethra' and so on, .which shapes and guides his early and subsequent experiences. For example, the phantasy of the internal object relation 'breast-in-niouth' becomes a prototype for all incorporative activity, whether nutritive., sexual, or the talcing io of knowledge. In this view, the nature of the internal object is primarily intrapsychic. This prototype is only gradually modified by the behaviour of the actual object in question - quintessentially 'the breast'. These phantasies continue to influence mental life in adults as well as children, every thought having unconscious as well as .conscious ramifications and reverberations.
Example: 'only connect'
Tom was a gifted, but unusual and rather isolated 17-year-old, with a precocious interest in politics. The oldest of ten children, he entered analysis following an episode of hypomania in which he became convinced that he was privy to secret knowledge that a prominent politician was about to 'defect' from one political party to another. He had telephoned various important people and newspapers at all hours with his 'discovery'. In treatment he was argumentative, opinionated and hyper-rational, frequently asking 'intrusive questions. He stated that he liked to argue 'in order to penetrate other people's minds*. This remark could be understood at several levels: oedipally as an expression of his burgeoning but repressed and uncertain sexuality, but also at a 'pre-oedipaľ level as expressing his desperate need to make contact with others, based on a primitive phantasy of penetration and containment. Similarly his preoccupation with 'defection' could have been an oedipal expression of his sense of being, in comparison with his potent father, 'defective' as a male, and also a reproach to his mother for having 'defected' so readily from him into her many subsequent pregnancies.
However, as we discussed in the previous chapter, Freud also realised liiai internal objects were not just inborn but, by identification and interaalisation, representations of the relationships to which an individual was exposed: 'The shadow of the object fell upon the ego' and 'The character of the ego is a precipitate of abandoned object cathexes and it continues the history of those object-choices' (both Freud 1917). His idea of the superego as an internal representative of parental prohibition is the most striking
Origins .of the interna! world   53
example of this. Following this, the object relations school, especially Fáirbairn (1952) and Bowlby (1988), adopt a radically inter-psychic or transpersonal approach, in which the nature of the internal object (in Bowlby's terminology, the 'internal working model') is a representation of the behaviour of the object, which is impressed, as it were, on the blank screen or tabula rasa of the mind of the developing infant. Padel (1991) points out that what is internalised is not an 'object' (a 'mother' or a 'penis') but a relationship, and that the individual, via role reversal, may identify with either 'end' of this relationship - thus in the familial intergenerational transmission of abuse, the abused child in one generation becomes the abuser in the next.
Bion (1962, 1970) takes a middle position, giving weight to both intra- and interpsychic factors, envisaging the behaviour of the caregiver rather like an ethological 'releaser' of inherent mental structures. He sees the nurturant function of the 'breast' as having the capacity, not necessarily realised, to turn a potential thought within the infant (in his terminology a 'preconception') into an actual phantasy or internal object (for Bion, a 'conception'). This model can be compared with Chomsky's (1965) view of the development of language in which a 'deep structure' or potential for speech (the Language Acquisition Device or LAD) is transformed into a particular dialect by the prevailing linguistic world which the infant encounters. The growing child could be said to possess a 'Phantasy Acquisition Device' or PAD. The clinical relevance of this would be, in Kleinian terms, a therapist using his 'breast-mind' to elicit previously unrealised conceptions - i.e. phantasies, creative realisations, or unexplored epigenetic pathways - in the mind of the patient.
The Batare of mem©ry
Also relevant to this discussion are contemporary ideas derived from cognitive science about memory. These enable a classification of different types of 'reminiscences', which form the raw material of psychoanalytic work. Tulving (1985) distinguishes between (a) 'procedural memory', a non-verbal representation of how we were handled as infants, akin to Klein's (1946) notion of 'memories in feeling', (b) 'semantic memory', the patterns of interactive behaviour or 'scripts' (Byng-Hall 1991) with which we were brought up, and (c) 'episodic memory','actual events which we can recall.
54   Theory
Although few people can remember specific events before they were two years old, early events and relationships influence the way we experience and relate to the world generally (semantic memory). These influences are continuously revealed through transference. Jest as, for example, the behavioural manifestations of sexual love contain vestiges of infant behaviour - mutual gazing, touching, sucking, intense separation anxiety, etc. - so phantasy, and thus transference, continues to shape adult mental life, especially at times of emotional intensity. In health this enhances and deepens emotions: pianists may be observed making primitive sucking movements with their lips while playing music of the highest sublimity. In pathology this transference distorts and may even create apparent episodic memories. In summary, as Stern (1985) puts it: 'Development is not a succession of events left behind in history. It is a continuing process, constantly updated.' Indeed, this positive 'updating' process is what makes psychoanalysis possible, based on what Freud called 'deferred action' ('nachtraglichkeit') - a kind of reverse transference in which past memories of all three types are revised in the light of present experience.
TTfee 'cEImikaF Siutfauaít anna the fioB>served:r telami It should also be noted that infancy has itself, until recently, been something of a tabula rasa, a blank screen upon which theorists have projected their preoccupations and ideologies, it is only with the emergence of detailed observational studies of early infant-parent interaction that the observed, as opposed to the clinical, infant (Stern 1985} has been able to 'speak back' to its theorists and to correct some of the misconceptions that have been heaped upon it.
With these preliminaries in mind, we shall now turn to the psychoanalytic account of psychological development, working chronologically from birth. Despite the above caveats we shall follow convention, dividing our account into four phases: 'pre-oedipal' or two person; oedipal or three person; adolescence; and adulthood.
THE 'PRE-OEUHPAL' OE TWO-PERSON PHASE
3Pire-©eátpaH ©r 4wo peirsora?
For Freud (1916/17) the oedipus complex was the 'kernel of the neuroses'. In the early stages of the psychoanalytic movement a
Origins of the interna! world   55
belief in the centrality of the oedipus complex was a hallmark of psychoanalysis, a dogma that could not be questioned, so that when Melanie Klein, building on her analyst Abraham's developmental schema, turned her attention to the early months of life it was natural that she should give respectability to her findings by calling them 'pre-oedípaľ. As we shall see, she found evidence for the existence of 'oedipal' structures even at the earliest stages of life. Balint (1952), following Rickman, introduced the more neutral terms 'two-' and 'three-person stages'.
This debate has important consequences which extend beyond terminological disagreement. As we saw in the previous chapter, Freud's model was essentially based on conflict: neuroses derive from the incompatibility of the demands of civilisation and those of the instincts, and the tension between the desire to love and be loved by one's parents and the fear of inevitable rivalry and the consequences this arouses. For Freud, civilisation was built around the defence against unconscious incest-wishes. Kernberg (1984) and Kleinians such as Segal (1986), extend this conflict model back into infancy where there is still a struggle between the love an infant has for his care-givers and the hate he feels at their absences. For them, the essence of civilisation is the defence not so much against incest but psychosis (Young 1994). Others, notably Kohut (1977) and Winnicott (1965), see the problems arising from infancy, especially those leading to 'borderline' disorders in adult life, as being based on deficiency ~ the lack of vital ingredients needed for heaithy development such as maternal empathy and sensitivity. In reality both are probably right (cf. Chapter 10).
The early weeks of life: anEtísira or symbiosis
Freud saw early infancy in terms of an 'egg' model (Hamilton 1982) in which he compared the state of the newborn to that of chick within its shell: 'a psychical system shut off from the stimuli of the external world, and able to satisfy even its nutritional requirements autistic-ally .. .' (Freud 1911a). This led Mahler et at. (1975) to describe a stage of 'normal autism', and Freud (1914a) himself to his controversial concept of 'primary narcissism', the necessary self-love which precedes the capacity to love others, based on the blissful self-absorption of the newborn.
Mahler describes the first six months of life as proceeding from 'normal autism' through 'symbiosis' to 'hatching' (continuing the
56   Theory
egg metaphor). However, the use of the term 'autism' is questionable (but see Tustin 1986 for a contrary view, later retracted, Testin 1994). The newborn is certainly self-centred, but research shows that he participates actively in the two-person environment, and, as mentioned, is able to distinguish his mother's voice, smell and face from, others from a very early stage. This leads to a picture of mother-infant symbiosis, which is fuelled by the infant's intense need to reach out to a nurturing figure, and the responsiveness of the sensitively attuned mother.
For Freed, primitive feelings of 'goodness' were related to 'primary narcissism', the necessary state of self-love which we all need in order to survive, 'autistic' only in the sense that, as Slavin and Kreigman (1992) put it, 'because we are members of a sexually reproducing evolved species, nobody loves us as much as we love ourselves'. Freud (1914a) saw self-love as developrnentally prior to relatedness ("anaclitic', or object love, leaning on narcissistic, or self-love), expressing this in his famous 'amoeba metaphor' of
an original libidinal cathexis of the ego, from which some is later given off to objects, but which fundamentally persists and is related to object cathexes much as the body of an amoeba is related to the pseudopodia which it puts out.
(Freud 1914a: 75)
Freud visualised this primary narcissism persisting into later life in the form of the 'ego ideal', which informs our aims, values and ambitions, later coalescing into the superego; and in the narcissism of parental love which makes one's own children special above all others: 'His majesty the baby . . . parental love which is so moving and at bottom so childish, is nothing but the parent's narcissism born again' (Freud 1914a).
Tlhe Süefim-Keiratoerg model ©f early Infancy
For Freuü (he newborn baby's mind is pure id, lacking an ego with which to relate to the world. Klein disagreed with Freud in that she saw a primitive ego, and therefore primitive object relations, as being present from birth. The infant encounters and introjects two sets of contradictory experiences: 'good', satiating, nurturing feelings associated with successful feeding, warmth and tactile contact with  the mother;  and  'bad', associated with  separation,  aban-
Origins of the internal world   57
donment, hunger, wetness and cold. These form the nucleus of feelings of love and hate, both to some extent preformed under the influence of eros and thanatos, the life and death instincts.
The primitive ego can only maintain psychic equilibrium by keeping good and bad apart. The infant introjects and identifies with the good 'me' experiences, while splitting off the bad 'not me' feelings and projecting them outwards. 'Bad' is projected into the nurturing parent, whose 'good breast' becomes thereby transformed into a 'bad' persecutory one. Thus the external world is a reflection of the inner world. In this omnipotent and potentially psychotic state all is magically personified: 'everything is the result of the actions of objects' (Etchegoyen 1991). Further anxiety now ensues because in order to get good experiences 'in', the child has to 'cannibalise' (i.e. introject) the parental goodness, but the child now feels he has 'destroyed' the breast and this destructiveness also has to be projected, leading to another spiral of the vicious circle (see pp. 81-7 for further discussion of 'primitive defences').
In this model, 'splitting' is very much a two-edged phenomenon. On the one hand, it is a necessary defence which enables a core sense of self-worth and goodness to hatch, uncontaminated by bad feelings of rage and disappointment. The attuned parent, or receptive breast, allows these bad feelings to lodge safely until they are ready to be reintegrated in the depressive position. On the other hand, splitting can act as a brake on development if this detoxifying function of the breast is unavailable due to neglectful, inconsistent or abusive parenting, and/or if the bad feelings are excessive and cannot be reintegrated. The clinical implications of this mean that the analyst must judge whether splitting needs to be left uninterpreted, at least initially, or whether negative transference must be faced and interpreted from the start. Kernberg (1984) tends to advocate the latter, but Joseph (1989: 76) cautions:
When analysts ask their patients to face their fears, their yearning, their sadness long before they have the resources and imagination to do so, they may be asking too much.
In the 'paranoid-schizoid position', Klein's child trails clouds, not so much of glory, but of envy, rage and disappointment. Klein envisaged this state as balanced by the 'depressive position' in which good and bad, love and hate, come together when the child realises that the very object which he hates and fears is also the one he loves and depends on, the guilt and 'depression' associated with this
58   Theory
realisation being the spur £o reparation and creativity. Klein saw the interplay between these two positions as continuing throughout life and underlying the rhythms of all the major crisis-points of psychological development.
laterpersoiniall ehotMs ©f early iraffamcy
In contrast to the Kleinian model, the interpersonal perspective emphasises the emergence of self and pre-linguistic meanings from the two-person matrix of infant and parent. For Fairbairn the fundamental trauma is maternal indifference: not to be intimately known, which can, for example, arise out of maternal depression. The child deals with this trauma omnipotently by imagining that it is either (a) his love that has destroyed her feelings, a schizoid response, or (b) his hate that has obliterated them, a depressive reaction.
Winnicott (1965) and Stern (1985) see a stable sense of self arising out of parental 'handling' based on maternal sensitivity and attune-rnent. For Winnicott, holding, integration and personalisation (a typically Winnicottian term referring to the development of a sense of self) are the key issues. Holding starts long before the birth with primary maternal preoccupation, in which thinking about the baby comes to assume great salience in the mother's conscious and unconscious mind. The child is thus a recipient of parental projection as well as a source of projection. Fraiberg et al. (1975) evocatively call these projections 'ghosts in the nursery'. The 'environment mother' (a holding function, in contrast to the 'object mother' from whom the child receives satisfaction of needs) accepts and makes sense of the infant's 'spontaneous gesture', leading to the emergence of a 'true self, and a psychosomatic sense of being a person.
Winnicott (1965) and Kohut (1977) bring an interactive, two-person perspective to the notion of 'primary narcissism'. For Kohut, "healthy narcissism' arises out of the 'empathic mirroring' of the nurturing parent (or analyst), leading to an internalised 'nurturing self-objecť which is not outgrown as object love supersedes self love, but which is required in- one form or another throughout life. Self-esteem is based on the mutual absorption and devotion of mother and child. He sees narcissistic disorders - with their defences of role reversal, compliance and 'false self - arising not from the internal threat to primary narcissism posed by excessive aggression
Origins of the internal world   59
(the Klein-Kernberg model), but on disruptions to this first all-important bond.
In this account the responsiveness of the mother's/ace as a mirror, and later as a symbol of relatedness (Wright 1991), is as important as the much vaunted 'breast'. Winnicott tries to retain the omnipotence of narcissism, but couches it in interactional terms by imagining a transitional space between mother and infant (transitional in the sense that it is neither self nor other, subjective nor objective, inner nor outer) in which" the infant's wish is so sensitively responded to by the mother ('his wish is her command') that he has the illusion he has 'created' the object, this being the basis for feelings of self-efficacy and creativity in later life.
These interpersonal authors make a sharper distinction than Klein between normal and abnormal development. For Klein there is always a movement from splitting to integration, from the terrors and precarious security of the paranoid-schizoid position to the sadness but also sanity of the depressive position. The interpersonal authors tend to see these primitive defences not as a normal response to the inevitable unmanageable anxiety of infancy soothed and integrated by adequate parenting, but as a pathological manifestation of the infant's struggle to survive in the face of inept or cumulatively traumatic parenting.
With normal responsive parents, a child can deal with 'bad' experiences - a delayed feed, a cold cot, a respiratory infection, a boring day - by healthy protest which will be accepted and 'transmuted' by the responsive 'breast1. The parent can cope with this healthy hatred, with the infant who says, in effect, 'Hello object, I destroyed you!' (Winnicott 1965). Conversely, the infant can keep an image of the good parent alive long enough for inevitable brief separations to be tolerated without feelings of persecution or abandonment. Healthy aggressive protest is linked to 'metabolising' and so surviving loss. Eventually this 'optimal disillusionment' (Kohut 1977) leads to the acceptance of the reality of loss, which can coexist with a sense of persisting 'ordinary specialness'.
The primitive defences of disturbed adult patients - splitting, projection and projective identification, omnipotent control, narcissistic self-absorption - are not in this interpersonal model regressions or developmental arrests at early 'fixation points', but examples of 'secondary narcissism', that is, a withdrawal into the self in the face of a hostile environment. Kohut calls these 'breakdown products', resulting from failed nurturing.
60   Theory
Example: the lOxo mother'
Mark was a young man in his early twenties suffering from schizophrenia. He heard voices telling him how bad he was, and saw visions of terrifying snakes and other monsters. He tended to be attracted to older, motherly, women who at first liked and took pity on him, especially as he was so 'good' and biddable, but then found his possessiveness claustrophobic and so rejected him. A similar pattern characterised Ms relationship with his analyst. His parents had been 'hippies', both heavy drug users, who had found it hard to care for Mark who had throughout his childhood oscillated between being fostered and living with his mother. His earliest memory was of 'stealing' an Qxo cube from the kitchen ' cupboard and gnawing it, while he heard his mother and her friends laughing helplessly in the next door room in a drugged state. Mark was desperately seeking nurturance from this 'square' breast in the face of drastic environmental failure.
Although each model starts from an opposing premise - for Klein, 1 goodness" comes from without, and 'badness' from within; for the interpersonalists, the innocent child is corrupted by a deficient environment - in practice the two positions are not so far removed. KCleinians envisage good parenting as mitigating inherent splitting tendencies within the infant, while the interpersonalists and the self-psychologists emphasise the effects of bad parenting in engendering splitting and other 'primitive' defences. Since the real world is a mixture of good and bad, and infants can from the start love and hate, both accounts contain valuable insights.
Sepauratíaiis-uBsáSTMiiaaluoim
So far we have considered the infant and his care-giver on their own. But two people are needed to make a baby, and even before the father becomes important, a third term has already arisen between infant and mother - absence. Conceived oedipally, the absent mother is imagined to be with the father as a 'combined parent'. Certainly, at about sevem months' old the child begins to show stranger anxiety and to be much more obviously aware of the mother's comings and goings. This is also the starting point for Mahler et aV% (1975) 'separation-individuatioď and the point at which the infant begins to establish internal 'object constancy", the capacity to retain a mental image or memory trace of the absent mother - and by analogy, the analyst absent during weekends and breaks.
Origins of the interna! world   61
Attachment Theory (Bowlby 1988) emphasises that the way separation is handled is the key to secure bonding, and sees the parents providing a 'secure base', both in reality and in the child's mind, from which he can begin to explore the world, and to which he can immediately return when danger threatens. Mahler et al. (1975) identify a 'rapprochement subphase' of separation-individuation in which, during a toddler's ecstatic exploration or 'love affair with the world', he will momentarily return to his mother for reassurance, as though frightened that the secure base has abandoned him, perhaps in punishment for what Einstein called 'holy curiosity' (Hamilton 1982), but what might be seen oedipally as forbidden fruit.
Kernberg (1984) sees the replacement of splitting by repression as the main defence against unwanted feelings arising around this time. In normal development there is now a much stronger ego, based on internalisation and identification with a stable nurturant parent. The early superego is also beginning to emerge. The child is thus able to cope with some degree of integration of good and bad, which can therefore relegate unwanted feelings to the id within, rather than having to project them outwards.
As separation proceeds the child begins to conceive of a world of autonomous individuals, each with its own vantage point: as he gathers a sense of his own emergent subjective self, so he begins to be aware that others have selves too. This is the cognitive aspect of the 'depressive position', in which 'good' and 'bad' part-objects coalesce into a whole separate being, a source of both nurturance and frustration. Fonagy (1991) depicts this as, the emergence of the child's 'theory of mind'. This concept is especially important in the treatment of abused patients who may have defensively obliterated the horrifying idea that 'care-giving' adults might have minds that could wish to harm or even destroy them. In treatment they can begin gradually to recognise the analyst's mind as a separate 'place' which they can safely inhabit.
THE OEBIPAL OE THREE-PERSON PHASE
Freud's conception of the oedipus complex, first mentioned in a letter to his friend Fiiess in 1897 arising out of his self-analysis following the death of his father, has become part of popular folklore. The little boy, like Oedipus, wishes to kill his father and usurp his place in his mother's bed; this wish to possess her arouses fear of retaliatory castration by his all-powerful father - the ultimate punishment of the
S2   Theory
loss of his genital is symbolically enacted in the ritual of circumcision. The reality of castration is confirmed by the little boy's sexual curiosity, when he discovers that females lack a phallus, and in his mind musí therefore have been castrated - a view that is contained by the flow of menstrual Wood.
Freud seemed to assume that the psychosexual world looked roughly similar to little girls, who therefore love their fathers and wish to replace their mothers, and, lacking the essential organ of power and significance, feel castrated and therefore powerless. Freud saw the oedipus complex as a necessary developmental task to be undergone, of 'resolved', for the boy by identification with the feared father which offers him the promise of true potency in the future, in the little girl by identification with the mother, and by the promise that her lack of a phallus will eventually be compensed by the capacity to attract and so 'have'' a man (since she cannot 'be' one) and to produce something much bigger and better than a penis - a baby.
Although contemporary psychoanalysis, strongly influenced by the feminist response to this phaliocentric account, departs radically from Freud's original view of psychosexual development, the oedipus complex continues to resonate psychoanalytically in the image of a parental couple which the child both desires and feels cast out from, and in the experience of desire, prohibition and ambivalence which every individual must negotiate on his or her pathway from infancy to adulthood. We shall look at four aspects of the oedipus complex - the Kleinian, and Lacanian perspectives, feminist responses, and the effect of social change, especially the impact of absent or abusive fathers. Bur first let us consider the continuing relevance of Freud's original conceptions. As Mitchell (1989) points out, a central project for Freud's was to understand the psychology of male impotence ('where they love they do not desire, and where they desire they cannot love' (Freud 1916/17). This problem was the starting point for the following case.
Example: his father's son, or his own man?
Wfaik :;ot technically 'impotent', Peter's presenting problem was his persistent inability to form stable relationships with women. A businessman in Ms early 30s, he had had numerous liaisons with 'suitable' women, and in the early stages all went well, but whenever the possibility of commitment threatened, he would begin to panic and eventually, as he later put it, the relationship would 'peter' out. A subsidiary problem was his moodiness and
Origins of the internal world   63
irritability at work, which, he said, gave him a reputation for being 'an awkward bugger'.
Peter was the eldest of three boys with a very powerful and domineering father who ran the family business in which Peter worked, and with a retiring and compliant mother. It soon became clear that Peter had modelled himself on his father whom he both revered and feared, but that somehow the identity did not quite fit. Secretly he felt that he was 'like' Ms mother - whom he could talk to and who expected him to come to regular meals with the family even though in theory he was living independently, and with which he complied so as not to offend her. As yet another promising relationship foundered, Peter suddenly saw that the point at which things tended to go wrong was when he introduced girlfriends to his parents, and that seeing his father being charming to them filled him with feelings of inadequacy and envy, which he then projected onto the girls whom he saw as faithless and cheap. Expressing these thoughts in analysis created enormous difficulties for him, since he maintained a rather brusque and 'tough' manner, and tended to dismiss the male analyst's com-■ ments with remarks like 'you could be right' or 'I hadn't thought of that', implying that these were the crazy comments of a psychoperson, hardly, relevant to a down-to-earth businessman like himself. When his rivalry with the analyst was interpreted he began to recall childhood feelings of misery and utter betrayal around the time his mother was pregnant with his brother, and episodes in which he was humiliated by his father who tried to force him out of his taciturn sulks at meal times. He began to realise how stuck he was in a battle with his father which he felt he could never win, and a secret intimacy with his mother which no girlfriend could ever match, and even that his bachelorhood (which worried her dreadfully) was a reproach to her for her 'faithlessness' in producing two more boys when surely he was good enough? In talking about this he began to cry, and fully expected the analyst to ridicule or attack him. After this session he began to be much more assertive with his father, found a new girlfriend who lived abroad, and, by protesting his need for holidays that did not conform to analytic breaks, became much less overtly compliant with analysis.
From an interpersonal perspective, this case illustrates the importance of parental handling and containment of their child's oedipal
64   Tlieoiry
feelings: a mother needs to be able to enjoy her son's infatuation with her while, at the same time, not being seductive; the father mast be able to accept this infatuation without feeling threatened by it, and be prepared to offer himself as a model, while accepting his son's wish to attack and belittle him: 'a boy needs his father both to protect him from the danger that his mother represents for him, and to protect her from the danger he represents for her' (Horroclcs 1994). The case might also be seen as representing the father's failure to provide Peter with a valid initiation into manhood, and so help him overcome his fear of being merged with his mother and., by extension, with his girlfriends (Ely 198S). Peter's 'castration anxiety' embodied this fear, as did his sense of psychic mutilation.
The KEeiMam perspective ©ra the oenläpras complex
For Klein, the seeds of the oedipus complex are to be found in the earliest stages of infancy (Britton et at. 1989). Separation from the mother symbolises the existence of father, and the child's own aggressive phantasies that are aroused by the separation are then projected into the father. This explains the ubiquitous phantasy of the punitive and prohibitive father of the oedipus complex, who seems far removed from the normal reality of contemporary paternal attitudes. For Klein, negotiating the oedipal stage is closely linked with the maturation associated with the movement from paranoid-schizoid to the depressive position. In the latter, good and bad have to be kept separate until the child is able to accept the guilt and sadness associated with the realisation that they are directed at one and the same person. Similarly, the - resolution of the oedipus complex requires the child to accept his or her temporary exclusion from the parental couple, to allow mother and father to come together, and to tolerate the phantasy of "good intercourse' or 'primal scene'. If this can be achieved the child 'loses* the mother, but gains the capacity to think - creative thought requires the bringing together of ideas mn iicw combinations - and to lead his or her own life, having internalised the parental couple. He or she also 'gains' the father as a real object, and not just as an alternative mother. The latter position is encapsulated in Klein's famous equation, 'breast = penis', which can be used to account for the way in which some women are continuously disappointed by men in whom they are seeking the elusive breast, and also some homosexual men's unconscious
Origins of the interna! world    65
decision to renounce the unreliable breast in favour of the ever-available penis.
ILaiCSin
Some feminists have seized on Freud's views on female sexuality, as expressed in the oedipus complex, as the ultimate example of patriarchy and male chauvinism. But here again we encounter a paradox. One of the strengths of psychoanalysis is its emphasis on bodily experience and the biological fundamentals of feeding, elimination, sexuality, physical illness and death ('the ego is first and foremost a bodily ego' (Freud 1923)). On the other hand, if taken literally, Freud's idea of females as castrated males, only to achieve fulfilment through producing a 'penis-baby', is absurd, inaccurate and insulting. The importance of Lacan (Lacan 1977; Bowie 199!) was his notion that Freud's ideas are best seen as metaphors rather than scientific facts, and if couched in the terms of linguistics and anthropology constitute a devastatingly perceptive account of the psychological structures of a patriarchal society. To describe is not necessarily to condone.
Lacan sees a crisis in development around the age of two when the child begins to acquire self-a ware nes s and language. The primitive pre-oedipal unity of mother and child is shattered by the advent of the 'no(m) du pere' - the name of the father (as opposed to the 'maternal' 'Christian' name), and also the 'no', the prohibition placed like the archangel's sword at the gates of paradise by the jealous father. Language (a 'mother-tongue', but policed by patriarchy) is the reality to which the child must accommodate his primeval experience of the world; sensory impressions coalesce into objects, named and classified by the power of society, and the most potent symbol of these is the phallus, the father's penis. It is not so much, as Freud implied, that because females lack power because they lack a penis, but, rather, because they lack power they believe they must 'submit to the rule of the phallus.
Just as a child's experience at the oedipal stage-is shaped and potentially alienated by language, so his self-perception begins to -estrange him from his original self, a moment which is epitomised in what Lacan calls the 'mirror-stage' (Lacan 1966). If a coloured dot is crayoned on a child's forehead and he is shown himself in the mirror he will, up to the age of about two, try to touch the dot in the mirror, but after two years will touch his own forehead, suggesting
66
thai self-awareness arises at around that age. Unlike Freed and Klein who see the origins of narcissism in the pre-oedipai phase, Lacan contrasts the child's inchoate self-experience with the perfection of the image in the glass. The 'mirror stage' can also be understood in oedipal terms of the T (eye) seeing 'me', an objectiácation of the self that depends on the 'third term' (i.e. the father), a new vantage point outside the mother-infant couple, creating a triangle with the possibilities of feelings of exclusion and envy, but also freedom of írnovenient and abstract thought.
Lacao's conception of the mirror stage is thus very different from Winnícotťs emphasis on mirroring. For Lacan the mirror epitomises the insertion of the father into the child's blissful pre-oedipal life: alienation and seif-estrangement start here. For Winnicott, mirroring is a quintessentially maternal function, helping to build up the infant's sense of self and contributing to healthy narcissism.
FeißHiiBsirira ands <u>edIijTOs
The early women analysts - Klein, Horney, Brunswick - responded to Freud's patriarchal bias not so much by challenging him directly but by moving away from the role of the father to a focus on the early mother-child relationship (Chodorow 1978; Sayers 1992). In place of 'penis envy' in women (and oedipal boys who never grow up) arose the spectre of the'all-powerful, all-giving maternal breast, which males respond to by denigration and distancing, and into which females get sucked and sometimes depressively stuck. For Chasseguet-Smirgel (1985) the essence of the oedipal stage is the discovery and working through of the 'double difference' - the difference betvt/een the sexes and the difference between the generations. She sees perversity - at an individual and social level - as based on a denial of this difference, the oedipal boy building his perverse and false potency out of a 'faecal penis' composed of hatred and fear. Similarly 'penis envy' is based on a woman's devaluation of the F0vVi,r and generativity of her own body, deriving from a failure of identification with her mother.
Chodorow (1978), Mitchell (1989) and Benjamin (1990) have brought a feminist perspective to bear on contemporary psychoanalytic thought. The latter emphasises Freud's notion (borrowed from Fleiss) of the inherent bisexuality of the human psyche. Benjamin visualises an eariy oedipal stage before gender identity is
Origins of the internaS world    67
fixed, at which activity and receptiveness, exploration and passivity, inner space and outer space, assertion and submission are available to both sexes and emerge in children's play as much in little boys as in little girls. She sees the role of the father not just as representing prohibition and power, but as vital at this stage in offering excitement for the child and recuperative space for the mother. She proposes a non-oedipal (i.e. non-conrlictual) identificatory love (Benjamin Í995) in which boys and girls alike 'contain' the attributes of the 'other' parent. In the conventional family this means for boys the security and innerness of the mother, for girls the difference and exploratory possibilities of the father. Like Temperley (1993), she is critical of the overemphasis in the Freudian-Lacanian account of power and domination, stressing instead the need to recognise and value playfulness and 'inner space' in a child who, in Winnícotťs terms, is not just 'alone in the presence of the mother' but can also tolerate the parents' separate relationship without anxiety. As Benjamin (1990: 163) puts it:
In the oedipal experience of losing the inner continuity with the mother and encountering instead the idealised, acutely desirable object outside, the image of the female as the dangerous regressive siren is born. The counterpart of this image is the wholy idealised, masterful subject who can withstand and conquer her ... As we give greater value to the pre-oedipal world, to a- more flexible acceptance of difference, we can see that difference is only truly established when it exists in tension with likeness, when we are able to recognise the other in ourselves.
The reality of the moderm fetHEaEly: aSjserat amd alrasive fathers
Benjamin's perspective contrasts a real relationship with an actual father with the distortions that occur in the all too common single parent families of Western societies. 'Masculinity' becomes idealised and invested with dominance and unwarranted power, or is projected into a 'phallic mother', who is unattainably desirable and terrifying, creating what Lasch (1979) has called a 'culture of narcissism' in which the individual has to rely on self-love as the only means of survival in an objectless world.
Several authors (e.g. Young 1994) have pointed out that Oedipus was himself an abused child, left to die by his father Laius, who feared the oracle's prediction that he would be killed by his own son.
68   Theory
This perspective on the myth opens out an Interpersonal, inter-generational viewpoint that fits with the realities of the modern family. The abusive father, or, more often, step-father, with a mother who fails to protect ör even turns a blind eye, represents for the little girl the nightmare version of the oedipal myth in its modern form. Both abused and abuser are trying to escape from this cold oedipal world of separation to a pre-oedipal state of fusion and obliteration of difference. What is 'traumatic5 and psychologically damaging in trauma is the piercing of the normal barrier between oedipal phantasy and reality (Garland 1991): that which the child so desperately wants in phantasy becomes overwhelming in reality. The regressive transferences seen in patients who have been abused reflect both the search for a secure and responsive mother who will protect, and the need for a father who can allow closeness while still respecting separateness.
Example: oedipal themes in the transference of an abused patient
Eila, a 50-year-old divorced teacher with a grown-up daughter, who suffered from depression in the context of borderline symptoms, had been systematically humiliated and mentally, physically and sexually abused by her father throughout her early childhood, until her parents' separation when she was 11. Her mother was herself Intimidated, exhausted and neglectful. Ella's earliest memory was of being hurled across the room at the age of three when she had asked her father to play with her, not long after he had returned from the war. She was hypervigilant In her sessions, sensitive to the slightest lapse of concentration from the analyst. She was the antithesis of Winnlcotťs (1971) notion of being 'alone In the presence of another', finding it almost Impossible to relax enough in analysis to explore her own Inner world without worrying about what her analyst would think of her or want her to be saying. lust as her father had abused his paternal role by his invasiveness while her mother was unavailable, so the analyst's countertransference swung between feeling detached and sleepy, and ui!comfortably Intruded upon.
On one occasion, just before a break, Ella brought a set of pocket chess into the session, inviting the analyst to play with her. He was torn and confused between the wish not to make her feel rejected, while resisting vigorously this invitation to a mutual 'acting in5 (see Chapter 9). After a while, when Ella evidently had felt humiliated by his refusal to join in, he interpreted this along
Origins of the internal world    69
the lines of her terror of intimacy, heightened by the impending break, and her imperative need to control at all times. He Indicated that he saw the wish to play as positive, but suggesting that the idea of a board game was a reflection of her despair at being unable to 'play' spontaneously on the couch. The 'good' oedipal parent, which Ella felt so cut off from, is simply 'there' (Balint 1993), neither present nor absent, not too close nor too far away: like Goldilocks' three bears, neither too hot nor too cold, too hard nor too soft, but a 'transformational object' (q.v. pp. 74—5) grounded in reality, able to be incorporated into phantasy.
Summary
The oedipal period is seen as arising out of the two-person phase which preceded it. The main issues of the pre-oedipal period are: (a) the forming of affectional bonds with a nurturing parent and hence the achievement of a secure base from which exploration of the inner and outer world can occur; (b) learning to tolerate separation and absence and to respond with healthy protest, rather than split-off hatred and envy, when appropriate; (c) learning to think and to feel, via the sensitivity of the nurturing environment which turns a preconception into a conception, and modulates affect so that it becomes manageable; and (d) the emergence of a stable sense of self and others, able to survive separation and angry attacks.
In the oedipal phase the themes of intimacy and separation, of similarity and difference, are projected onto the three-dimensional screen of two parents and child. Children learn how to be close enough to their parents so as to feel special and loveable, but not so close as to be engulfed; how to respect limits and boundaries without feeling unbearably excluded; how to tolerate envy without being overwhelmed by it, or using it to destroy.
Compared with its emphasis on the early years of life, psychoanalysis has tended to have less to say about the later phases of the life cycle. Space allows us to touch only lightly on these important topics. However, epigenesis continues to apply. Developmental pathways are never fixed, and later good experiences (including psychoanalytic treatment) can compensate for earlier environmental failure, leading an individual away from the vicious circle of neurosis towards the benign cycles of psychic health.
í'O1   Theory
ADOLESCENCE
After the dbramas of the oedipal phase and before their continuation in adolescence, there follows aperiod of comparative quiet 'Latency1 provides respite, a period in which psychosexua! development and emotional maturation continue in a much more inuted vein, and in which the acquisition of cognitive and motor skills, and the capacity to go beyond the family into a world of peer relationships, are the predominant developmental tasks.
Just as the ways in which attachment and loss have been handled and felt in the pre-oedipal phase will affect the oedipal stage, so too a child's oedipal experiences will equip him for good or ill during the turbulence of adolescence. Each phase is both a continuation of the past and offers opportunities for new beginnings. The adolescent faces the twin tasks of separation from his family of origin, and preparing himself for the intimacies of his family of generation -putting the developmental trajectory of the first two decades into reverse by moving from loss to bonding. Two central issues here are the body (Lauifer and Laufer 1984) and identity (Erikson 1968). The adolescent no longer relies on the parent to regulate and modulate his bodily affective states, but has to undertake this task for himself. The anorexic who does not know when she is hungry, or when she has had enough, is struggling with this issue. At the same time the adolescent has to learn how to entrust the other with his anger and sexuality, and not to feel that they will be destructive or rejected. He has to know who he is so as to make choices and begin to create a "world of his own. In place of parents come ideas, systems, role models, fashions, aspirations. Their purpose is to contain and define the self whose lineaments are beginning to solidify. A negative identity built around protest and a preoccupation with what one is not, or a conferred srn based on compliance with parental aspirations may equally conceal an inner sense of emptiness and lack of connection. Psychoanalysis with adolescents (cf. p. 200) is difficult, since it represents their worst fears of being odd or abnormal, but also c?r o-ffer a 'moratorium' or 'Speilraum' (Erikson 1968) within which adaptive regression can occur.
Example: oedipal inhibitions
David was a pleasant 19-year-old presenting with panic attacks which had temporarily crippled him, making him unable to leave the house alone or to study effectively for his exams. His movements were tentative, as though he did not quite inhabit his
Origins of the internal world   71
adult body; his hair and voice slightly soft and child-like. A middle child, he had a successful older brother and a much adored younger sister. The family was supportive, but tense. David's father, a builder, suffered from severe asthma which had kept him off work for long periods. The family was in financial difficulties and there was pressure on David to leave school and earn his living. David was very close to his mother, sympathising with her worries about his father's health, but he resented her pushing and domineering manner, and he envied her more straightforward relationship with his sister, feelings he expressed by sulky withdrawal. During his sessions David was polite but wary, communicating a sense of helplessness and a passive wish that the analyst should instruct him how to live without his fears. He felt guilty about following his own interests, rather than helping in the home where he felt he was needed. At times he felt his life was empty and meaningless. He had a girlfriend - of whom his parents did not entirely approve - with whom he slept, but did not make love. Interpretations focused on the possible connection between his presenting anxiety and the oedipal fears of breaking the bond with his mother if he made a sexual relationship with his girlfriend, and of triumphing over his damaged father if he was successful in his exams and achieved sexual potency, linking this with his cautious and deferential attitude towards the analyst. David was at first outraged at this 'ridiculous' suggestion, insisting that he and his father were the best of friends, but he then admitted that he did resent the way he felt that his father favoured his eider brother. This open conflict with the analyst seemed to shift things. By the end of David's time-limited treatment he was feeling better, had been to two school dances, had made love enjoyably with his girlfriend, and his exams had gone reasonably well.
ADULTHOOD
Psychological development does not come to an end once physical growth is complete. It takes a lifetime to learn to love and to work. The two motors of psychological development, biological change and the vicissitudes of attachment and separation, continue throughout the life cycle. Erikson (1968) schematically saw the task of young adulthood in terms of relatedness versus self-absorption; the task of the middle years as generativity versus stagnation; and the task of old age as integrity versus despair.
'(2.   Theory
A somewhat arbitrary, and no doubt culturally biased, list of the developmental 'tasks' of adult life would include: the acquisition of skills and the capacity to submit to tutelage but to challenge it when necessary; the capacity to love a partner 'in sickness and in health'; tolerance of the transition from romantic to conjugal love in a long-term relationship; capacity to love and to hate without fear (Skynner 1976); ability to hold the balance between immersion and detachment as a parent; ability to achieve stability and security while still being ready to explore new territory (Bowlby 1988); being able to come to terms with loss and to grieve appropriately; acceptance of the inevitability of one's limitations and eventual death (Jaques 1965); acceptance of loneliness (Nemiroff and Colarusso 1990); being able to detach onself from one's children, occupation at retirement, and, in the end, life itself (Porter 1991); retaining appropriate optimism and not being overwhelmed by despair.
From a pssfchoanalytic perspective an individual brings the history of his early relationships to these 'tasks'. Being a parent reawakens oedipal phantasies, so that through our children we look back upon our own childhood, and are made aware of the limits of our abilities and lifespan. Awareness of the reality of death in middle life forces a reworking of the paranoid-schizoid/depressive interface as an individual moves towards acceptance or violent repudiation of his limitations (Jaques 1965). Psychoanalysis examines the interplay between the three basic polarities of (a) connection and separation, (b) destruction and reparation and (c) self-love and other-love, which are played out on the broader canvas of adulthood. The way someone responds to 'slings and arrows of outrageous fortune' is shaped by the mental structures that derive from childhood. While it is not possible to survey the whole field, we shall touch briefly on three psycho-analytically central topics: mourning, marriage and maturation.
McWUMEIlg
The idea thai suppressed mourning leads to psychological difficulties and, conversely, that facing loss and expressing grief are curative, is central to much psychoanalytic work. Following Abraham's suggestion, Freud (1917) drew attention to the parallels between depression and normal grief. He saw that in depression the sufferer is struggling not just with himself, but, often unaware, with a relationship: 'the shadow of the object falls on the ego'. The work of mourning involves the paradox that in acknowledging what is lost,
Origins of the internal world   73
the bereaved person is at the same time reclaiming it. Based on 'her own mourning for her son who died in a climbing accident, Klein (1940) realised that what was at stake in loss was the integrity of the whole inner world. She therefore linked loss with internalisation. The bereaved:
'through the work of mourning is reinstating all his loved internal objects which he feels he has lost'. . .. Every advance in the process of mourning results in a deepening of the individual's relation to his inner objects, in the happiness of regaining them when they were felt to be lost.
(Klein 1940: 356)
Many of the tasks and transitions of adult life involve loss. Klein is sensitive to the feeling that with one loss 'all' feels lost, and to the continuities between the handling of loss in childhood and adult responses to separation. Where loss can be mourned (including by healthy protest) then the lost object is 'reinstated' internally. This leads to an enrichment of the inner world which can balance the sadness about what has gone; where it cannot, depletion and depression may follow. Much of the work of analysis centres around this regaining and reinstatement of lost objects, especially as the analysis itself becomes an object that, between sessions and at breaks, is constantly lost and found, so that finally, at successful termination, the patient will have internalised a therapeutic function, even though the relationship with the therapist has come to an end.
Marriage
The psychoanalytic literature specifically devoted to marriage is surprisingly scant (but see Dicks 1967; Chilow 1985; Ruszczynski 1993). Perhaps' this is because the majority of people seeking psychoanalysis are so obviously suffering from the problems of 'relationships', or the lack of them, that much of the psychoanalytic project is a treatise on relationships. Falling in love, and the search for the physical, emotional, intellectual and moral intimacy of marriage, requires a retracing of the history of the inner world and its mutual alignment with the partner - an inner process that parallels the way in which lovers exchange their 'external' life stories and incorporate one another into their respective families.
Marriage is therefore the most potent and fertile source of 'transference' outside the psychoanalytic relationship. The partner
74    ii heory
is a 'transformational object' (Bollas 1987), a vehicle for projection, a receptacle for unwanted aspects of the self, a source of delight and terror, bringing one into touch with one's deepest desires and disappointments. The potential for transformation or destruction suggests that health is not an escape from transference, but integral to it. Transference is, in Slavin and Kreigman's (1992) terms, a uretranscribing of the relational environment', a 'probe' based on past affective experience embodied in 'semantic memory', which, when marriage (or therapy) goes well, leads to the mutual emergence of new patterns of maturity.
Materstvom
Implicit in the developmental perspective of psychoanalysis is the idea of maturation. For Freud, this meant becoming more reality-oriented, as he expressed in his Zuider Zee metaphor of the reclamation of the unconscious by the conscious, his view of neurosis as a 'turning away from reality', and his famous aphorism 'where id was there ego shall be'. Implicit in Freud's work is also the notion of the coherence of an inner world illuminated by self-knowledge, and the acceptance of the different aspects of oneself. Mahler et al. (1975) combine both a Jungian and Freudian perspective in their concept of 'separation-individuation'. Some psychoanalytic writers emphasise the outgrowing of childish preoccupations, stressing a painful but heroic autonomy as the goal of treatment and as an ideal state of emotional health. This viewpoint has been criticised by authors such as Bowlby and Fairbairn who see dependency as integral to the human condition and autonomy as the false goal of a consumedst society intent on producing 'normosis' (Bollas 1987), not health. Fairbairn (1952) described maturation as a movement from immature to mature dependence, and Kohut (1977) insisted on the need for the persistence of 'self-objects' - i.e. 'narcissistic' and 'special' relationships with spouse, children and parents as well as ideas, places, pets, mementoes, etc. - throughout life. These authors see development as a dynamic system, an equilibrium balancing past and present, and maturation a! and regressive tendencies. Klein saw a shifting balance between paranoid and depressive positions continuing throughout life. This perspective explains the way in which an overwhelming trauma can bring out 'primitive' responses in the most mature of individuals (Garland 1991), and how psychoanalysis, like life itself, is always an 'unfinished journey'.
Origins of the interna! world   75
CONCLUSION
The developmental perspective of psychoanalysis provides a rich source of clinical metaphor for understanding the affective interactional matrix between patient and analyst. Analysts constantly search for the 'child' in the adult patient and try to reconstruct from the transference and countertransference the developmental situation in which the patient is ensnared. But the search for the 'inner child' is not just a metaphor: it connects the hermeneutic project of psychoanalysis - the construction of meaning through a life narrative stretching hack to infancy - with the empirical world of developmental science. The mind has a developmental history no less than the body, and the evidence linking coherent or incoherent narrative styles in adulthood with patterns of secure or insecure attachment in infancy and childhood (Bruner 1990; Fonagy 1991; Holmes 1993) provides striking support for this view (see p. 255).
Chapter 4_______________
Mechanisms of defeno
The ego makes use of various procedures for fulfilling its task, which, to put it in general terms, is to avoid danger, anxiety and uinpleasure. We call these procedures 'mechanisms of defence".
(Freud 1937: 235)
We saw in Chapter 2 how models of the mind may be divided into those that are predominantly intrapsychic, interpersonal or mixed. Similarly, the concept of defence, and even the individual defence mechanisms themselves, may also be viewed from an intrapsychic, interpersonal/relational or mixed point of view. Some defences refer primarily to internal life (for example, repression), others to interactional or interpersonal phenomena (for example, projective identification, splitting), and yet others to both, such as denial.'
TME CONCEPT OF ©EFENCE
Classical psychoanalysis views defences primarily from an intrapsychic perspective, placing conflict at the heart of psychic life. First, conflict occurs between wishes and external reality which produces inner tension and anxiety. Second, conflict develops between the different agencies of the mind. Adaptation is made possible by defences. These are psychological configurations operating outside the realm of consciousness which minimise conflict, reduce tension, maintain intrapsychic equilibrium, regulate self-esteem and play a central role in dealing with anxiety whether it arises from internal or external sources.
Mechanisms of defence   77
Repjressnora
Repression, the pushing back of unacceptable wishes from consciousness, is the classical primary mechanism of defence. Repression ensures that wishes which are incompatible with reality, superego demands, or other impulses, remain unconscious or disguised. The inherent tendency of repressed wishes and impulses to return to consciousness - the 'return of the repressed' - means that tension and anxiety remain, and that an array of further defences are mobilised to alleviate the resulting conflict, reduce tension and stabilise the personality. But all this occurs at the cost of distorting internal reality.
Example: a happy abandonment?
A borderline patient told her analyst that she was pleased to have' a break from treatment when he told her that he had to cancel a session. Her mood changed from one of generalised anxiety and depression to overt happiness, accompanied by curt comments. Her apparent happiness at the cancelled session was clearly a defence against feelings of abandonment. On the missing day she came, expecting her session, and, on rinding no one present, she became acutely anxious, feeling that her existence was meaningless. She was admitted to hospital, complaining that the analyst had discarded her because he preferred to see his family. The cancelled session had reawakened previously repressed feelings of abandonment by her mother, who had sent her away to stay with her aunt when her younger brother was born.
Further clinical work led to an understanding of her repressed wish never to be left, the consequent rage towards her mother and its displacement onto the analyst, and the denial of the emotional impact of the cancellation of the session.
In contrast to this classical picture, relational models see defence mechanisms as a protective shield within which the authentic self is held: defences form part of the attempt to facilitate the development of a 'true' (Winnicott 1965) or 'nuclear' (Kohut 1984) self in the face of a defective relational environment. Alvarez (1992), in keeping with Freud, has taken this point further and considers some uses of defence as developmentally necessary. The boasting of the little boy becomes a powerful force in overcoming inferiority and attaining manhood; omnipotent and paranoid defences, rather than avoiding inherent de struct! vene ss or innate division and conflict, are
7S   Theory
desperate attempts to overcome and recover from states of terror and despair resulting from environmental failure.
Bowlby refrained defences in interpersonal terms, basing his view on attachment theory (Hamilton 1985; Holmes 1993). Secure attachment provides a positive primary defence while secondary or pathological defences retain closeness to rejecting or unreliable attachment figures. In 'avoidant attachment', both neediness and aggression are split off and the individual has no conscious knowledge of the need to be near the attachment figure, appearing aloof and distant; in 'ambivalent attachment', omnipotence and denial of autonomy lead to clinging and uncontrolled demands (cf. p. 256).
A markedly different view of defence is put forward by experimental and social psychologists who conceptualise defences as 'coping mechanisms' (Lazarus et at. 1974), primarily aimed at dealing with problems in the external world. In contrast to the unconscious nature of defence mechanisms, coping mechanisms are. supposedly (a) conscious and (b) mobilised to deal with external rather than internal threat. This distinction has also been questioned (Murphy 1962; Haan 1963). First, a number of everyday coping activities occur automatically, much like a reflex, while a refusal to listen to something or a denial of particular feelings may be conscious. Second, changes in the external world may evoke unacceptable affects which are then dealt with through the mobilisation of the mechanisms of defence. Third, the perceived danger of an external threat requires internal assessment, which is itself dependent on unconscious antecedents, and so there can be no clear distinction between internal and external conflict, with the two interacting in a complex way (Bond 1992).
Coping strategies can be taught and further developed into cognitive-behavioural strategies, which can be operationally defined for research purposes. Horowitz et aL (1990) have attempted to conceptualise mechanisms of defence within cognitive psychology; they see defences as the outcome of cognitive control processes which sequence ideas and join meanings together. Thus, defence and coping mechanisms are related phenomena, being both adaptive and potential sources of pathology.
Mechanisms of defence   79
Amma Fremd
The adaptive aspect of defence was elaborated by Anna Freud (1936). She showed how phantasy and intellectual activity can be used defensively and how defences can be directed against external situations, superego demands, cultural requirements, etc. She contrasted dynamic defences with permanent or character defence phenomena such as those described by Wilhelm Reich (1928, 1933) as 'Charakterpanzerung' (character-armour). At the same time Hartmann (1939), the founder of ego psychology, placed special emphasis on 'conflict-free spheres' of the ego which were not solely associated with defence and conflict, and this led to greater emphasis on normal aspects of the personality.
Kleäm
By contrast, Klein (1946) emphasises a new array of defences, including splitting of the object (a different use of the term from Freud), projective identification, omnipotent control over objects, idealisation and devaluation (see p. 219). Her followers have developed these ideas further, and now consider defences not so much as transient psychological processes brought into play when necessary, but psychological configurations that coalesce to form a rigid and inflexible system. These defence systems of the personality have been variously known as narcissistic organisations (Rosenfeid 1964), defence organisations (O'Shaughnessy 1981) and pathological organisations (Steiner 1982). They are associated with powerful, controlling internal objects (q.v. pp. 222-7). Meitzer (1968) described a patient dominated by a 'foxy part' of the self that continually persuaded him of the attractions of grandiose and destructive aspects of relationships. Rosenfeid (1971) identified an internal 'mafia gang' that demanded emotional 'protection money' from the good parts of the personality, which were then forced to collude with the 'gang's' idealisation of destructiveness and devaluation of love and truth. Sohn (1985) writes of the omnipotent self created by identification with an external object, the wealc and needy parts having been discarded, to form an arrogant 'identificate', which then takes over the whole personality (cf. p. 220).
This 'structuralisation' of defence systems has been applied to such complex areas as social systems and groups (Bion 1961; Pines 1985) which may themselves be under the sway of destructive interactions,
30   Theory
or driven by the equivalent of 'foxy* parts and. 'mafia gangs'. Soch understanding forms a major aspect of psycfaoanalyticaUy informed intervention into organisations (Jaques 1955, Trist and Eamforth 1951, Menzies-Lyth 1988, Hinshelwood 1987, 1993, 1994b).
Implicit in all the formulations is the fact that defence may become maladaptive in certain circumstances and lead to the formation of symptoms in the individual. This is especially likely if primitive methods of defence are reawakened through regression or remain active through developmental arrests at early 'fixation points'. Such ideas stem from Freud (1894, 1896, 1926) and there is now empirical evidence to suggest a relationship between psychological adjustment and maturity. Vaillant (1971, 1977) has shown that there is a continuum of defences from normal or mature through to pathological, and that the use of more mature defences is correlated with successful life adjustments in work, relationships and medical history.
The differentiation of defences into (a) psychotic/immature or primitive, (b) neurotic and (c) mature, links specific aspects of childhood psychological functioning to emotional difficulties in adulthood. However., the use of primitive mechanisms is not in itself pathological: as mentioned, they emerge in psychologically healthy individuals exposed to extreme stress (Garland 1991); it is their persistent use that is maladaptive.
Sľdimnusiry
In summary, the main features of defences are as follows:
° They may be normal and adaptive as well as pathological.
*   They are a function of the ego. ci They are usually unconscious.
*   They are dynamic and everchanging but may coalesce into rigid, fixed systems in pathological states and in character formation.
° Different defences are associated with different psychological states, e.g. repression in hysteria, isolation and undoing in obsessional neurosis.
° They are associated with levels of development, with some defences being seen as primitive and others as mature.
We shall now consider some of the individual defences in more detail, starting with the psychotic/immature or primitive mechan-řsms, and then move on through some of the neurotic defences to those of maturity. A list of defences is given in Table 4.1.
Table 4,1 Mechanisms of Defence
Condensation                Humour
Externalisation Identification with aggressor
Regression
Repression
Following Klein, contemporary psychoanalysts use the term 'splitting' to refer to a division of an object into 'good' and 'bad'. A child,
frustrating mother whom he hates and the good, idealised mother whom he loves. By mentally keeping the good and bad mother strictly separate, the ambivalent conflict between loving and hating a mother who is, in reality, one and the same person, and a mixture of good and bad, can be avoided.
Example: the envied siblings
A depressed patient who was one of four children continually
complained that her mother had neglected her and favoured her
siblings. She repeated stories of her mother's favouritism and unceasing concern for her sisters and brothers, from whom she was now estranged. In between sessions she began to feel that the analyst was kinder and more helpful to his other patients. She envied them and started to wait outside his consulting room, asking them whether he saw them more often than he saw her. In the sessions themselves she felt better and couldn't understand why outside the sessions she accused the analyst of hating her and
íe willingly gave to others.
32   Theory
As a result of the patient's ambivalence, her internal representation of the analyst had become split. Outside the sessions she could only see him as someone who was cruel and neglectful. Within the sessions she experienced him as ideal, caring and thoughtful. Thus the 'good' analyst was protected from her envious attacks.
Klein also recognised that, since internal and external objects are intrinsically related to the ego, a split, in the ego may also occur. This was in keeping 'with Freud's original use of the term 'splitting'. He referred to a splitting of the ego in fetishism, allowing a quasi-psychotic simultaneous holding of contradictory ideas (Freud 1927). The split coincided with the contradiction between a wishful fantasy and a reality, rather than between object representations -'the instinct is allowed to retain its satisfaction and proper respect is shown to reality' (Freud 1940, our italics). This descriptive use of the term is compatible with Bleuler's (1924) account of the loosening of associations in schizophrenia. However, splitting is now viewed, especially by lOeinian analysts, as a primary phenomenon of mental life in infancy potentially leading to the development of borderline and psychotic disorders later in life (Sternberg 1975). In these cases splitting is extreme and leads to a distortion of perception, a diminution in the capacity to think, and fragmentation of objects.
As mentioned in Chapter 3, splitting can be benign as well as pathological. The ordering of the internal world relies on splitting. Its success is, therefore, a precondition of later integration and the basis of the faculty of judgement. Splitting is also used to pay attention, to suspend one's emotional distress in order to come to a decision, to make moral choices, or form an intellectual judgement (Segal 1973). In this aspect it has similarities to Freud's (1909) concept of negation (see p. 189). The widespread tendency to split the world into good and bad, right and wrong, black and white, or heaven and hell persists throughout life and profoundly affects our attiiudes not only to individuals but also to social institutions and political, religious and other organisations.
ProjectMMS, iéemtfficfflt!»B smú pirojeeítiYe Memtifficatikm
Á recent enthusiast compared the significance of Klein's 'discovery' of projective identification in psychoanalysis with the discovery of
Mechanisms of defence   83
gravity or the mechanism of evolution (Young 1994). Projective identification is undoubtedly an important but complex subject, partly because of its inherent difficulty, partly perhaps because its name is misleading, and partly because, as one of the fundamentals of Kleinian psychoanalysis, it provokes political controversy disproportionate to its clinical role and relevance.
The notion of projection is relatively straightforward, and has entered the vernacular of 'folk psychology' (Bruner 1990). The depressed young man lying on a beach who stated 'everyone on this beach looks utterly miserable' was clearly attributing to others his own affective state. We commonly attribute our more difficult and unacceptable feelings to others - for example, blaming those that are close to us for our own shortcomings. Externalisation, the outward limb of projection, allows us to disown responsibility and to feel an illusory sense of mastery over our impulses. If our unwanted Impulses and feelings are rejected, like a boomerang, resulting in a feeling of being under constant attack, the projection has gone full circle and leads to anxiety or, if extreme, paranoid delusions.
Identification, similarly, is relatively straightforward, referring to the process by which self-representations are built up and modified during development, as distinct from the conscious copying of imitation. The little boy who shuffles around in his father's shoes is simply imitating, but as his internal image of himself is influenced and later transformed into a personality characteristic, identification has occurred, especially if he eventually 'steps into his father's shoes' and takes oyer the family business. Piaget's (1954) concepts of 'assimilation' and 'accommodation' are similar although referring to the development of cognitive ability rather than self-representation. Piaget" suggests the young infant has internal 'schemas' of only actions and perceptions, but later the child represents one thing by another through the use of words and symbols. New experiences are 'assimilated' into existing schemas, and may be distorted by them, much as the external world is introjected and modified by unconscious phantasy within a psychoanalytic model. Schemas are modified, extended, and combined to meet new situations through 'accommodation'. Similarly, self-representations are modified and built-on through new identifications.
As Klein (1946) originally conceived it, projective identification combines these two notions in a highly specific way. She described projective identification as a phantasy in which bad parts of the infantile self are split off from the rest of the self and projected into
84   Theory
the mother or her breast. As a result, the infant feels that his mother has 'become' the bad parts of himself. Of particular importance is, first, that the projection is 'into' rather than 'onto' the object -prototy pic ally the mother or the analyst, and, second, that what is projected is not so much a feeling or an attitude, but the self, or part of it. Klein imagined thai in the paranoid-schizoid position the infant might project 'bad' sadistic parts of himself into the mother's body in order to control and injure her from within. If these are then reintrojected - 'introjective identification' - the individual contains a 'bad' identificate, a potential source of low self-esteem or self-hatred. In contrast, 'good' parts of the self may also be projected and reintrojected, increasing self-esteem and enhancing good object relations if not carried to excess.
In this original formulation, projective identification was defensive, intrapsychic, and solipsistic, a mental transaction involving the self and a perception, but not the participation, of the other. How then does projective identification differ, if at all, from projection? Klein, herself, was clear about this. Projection is the mental mechanism underpinning the process and projective identification is the specific phantasy expressing it. Spillius (1988) suggests that projective identification adds depth to Freud's original concept of projection by emphasising the fact that a phantasy of projection is only possible if accompanied by a projection of parts of the self. She comments that British authors rarely consider the distinction between projection and projective identification to be of particular importance. In contrast, many American writers have devoted a great deal of discussion to the topic (Malin and Grotstein 1966, Langs 1978, Ogdesii 1979), often distinguishing projection and projective identification by whether or not the recipient of the projections is emotionally affected or not by the phantasy. In projection the target of the projections may be blissfully unaware of his role - as no doubt were the holiday makers on the beach in the example above. The paranoid person projects malevolent intentions onto politicians, pop stars, Freemasons, etc., with whom he never comes in contact, or indeed onto inanimate objects. This distinction has arisen from developments of Klein's original idea of projective identification emphasising its communicative aspect.
The communicative aspect of projective identification, as opposed to its defensive nature, means that it can be used to describe three distinct processes. First, if projective identification is seen as an interactive phenomenon, then the recipient of the projection may be
Mechanisms of defence   85
induced to feel or act in ways that originate with the projector. This accounts for the realisation by Heimann (1950), Grinberg (1962) and Racker (1968) that countertransferential feelings evoked in the analyst can reflect aspects of the patient's inner world. Ogden (1979) argues that identification occurs within both projector and recipient, while Grotstein (1981)'and Kernberg (1987) both feel that the term should be confined to identification within the projector. These ideas, often in a diluted form, have become widely accepted in psychodynaraic circles: if the analyst is feeling bored or irritable or sad, these feelings may, via projective identification, originate with the patient. Here the 'identification' is occurring within the target of the projective identification rather than, as Klein first saw it, within the projector (i.e. a 'misperception').
Second, by extension, projective identification becomes a mutual process in which projector and recipient interact with one another at an unconscious level. The analyst who is unaware of the feelings induced in him by projective identification may enact them by, for example, being rejecting or sluggish in the session - feelings which may, in turn, be identified by the patient. Spillius (1994) suggests the term 'evocatory projective identification' to describe this pressure put onto the analyst to conform to the patient's phantasy. Sandler (1976a, 1976b) and Sandler and Sandler (1978) see this aspect of projective identification in terms of 'actualisation' and 'role responsiveness' in which the analyst has to be flexible enough to respond slightly to the role in which he is cast by the patient's projective identification, and must also remain sufficiently centred in himself to observe and interpret this process as it happens.
The ramifications of the communicative aspects of projective identification are so great that it can eventually cover almost all that happens in the analytic situation. However, it is ciinicaiiy mistaken to assume that everything that the analyst experiences is a result of what the patient is 'putting into' him. It is important to distinguish between 'patient-derived countertransference' and 'analyst-derived countertransference' (seep. Ill), however difficult in practice this may be (Money-Kyrle 1956). The former is based on projective identification, while the latter most definitely is not.
The third extension of projective identification derives from Bion (1962, 1963). Klein saw projective identification primarily in negative terms: the projection of sadistic feelings as part of the paranoid-schizoid position. Bion realised that there was also a 'positive' form of projective identification underlying empathy, and the processes
86   Theory
by which the mother contains projected painful and hostile feelings, 'detoxifies' them, and returns them to the infant in a more benign form at a phase-appropriate moment. However, some of Bion's clinical uses of the idea of projective identification have been controversial. He advocated speaking to psychotic patients in concrete ways; thus he might say 'You are pushing your fear of murdering me into my insides'. While this kind of interpretation may occasionally be successful, in inexperienced hands it can be at best incomprehensible, at worst dangerous, and as a standard technique has been much criticised (Sandler 1987). It is rarely used now.
Many authors, including Bion (1955), have stressed the importance of projective identification as a method of control of the object and of unmanageable feelings. In this aspect of projective identi-ficatiomu whole aspects of the ego are split off and projected into another person, animal or inanimate object, who then represents and becomes identified with the split-off parts; attempts are then made to control these split-off parts of the self by asserting control over the other person (Sandler 1987).
Example: the controlling teacher
A man in his late twenties was working as a probationary teacher and was put in charge of a class of adolescent boys; he soon began to resent any boy who was disobedient, rude or otherwise badly behaved so much so that be became frightened before each lesson lest he would be unable to control the class. Indeed, the class was becoming increasingly badly behaved. He began to hate these pupils and feared that he might himself lose control and hit one or other of the boys. During analysis it emerged that throughout his own adolescence he had been a good boy, obedient, hardworking and never causing trouble either at school or at home. He had split off all those destructive and rebellious aspects of himself which, as a result of bis upbringing, he considered to be evil, dangerous and unacceptable. By means of projective identification he was now projecting all these unacceptable aspects of himself into his adolescent pupils. They had become identified with his own split-off rebellious parts. He hated these unwanted aspects of himself and desperately but unsuccessfully tried to control them by trying to control and punish his pupils. He suddenly understood this during a session and then reported that as a boy he used to collect and piay with toy animals, such as tigers and crocodiles, making them attack and devour other animals and human beings. These
Mechanisms of defence   87
toy animals had at that time also become identified with the projected aggressive parts of himself. By projecting all these parts into them he had, however, lost any inner sense of being strong enough to stand up for himself and to fight other boys and, later, adults who might insult or oppose him.
Good aspects of the self can also be projected into others. Projective identification can thus leave an individual feeling deprived of essential aspects of his own personality. A central task of analysis is to help the patient recover these lost aspects of the self (see p. 178).
Example: the worthless woman and her worthy friend
A young woman who had always tended to regard herself as unattractive, and unlovable became .dependent on and envious of a slightly older woman friend whom she regarded as very competent, beautiful and successful in all her relationships with men. During analysis it became clear that she had projected ail her own positive qualities of which she felt unworthy, and hence unable to contain inside herself, into this other woman, who thus became identified with all her own assets. This made her feel even less attractive and less worthy of being loved so that she felt impoverished inside herself. She made constant demands on her friend, as only in her presence and with her backing did she feel able to do anything worthwhile. When her friend married and moved away she became depressed, feeling her friend had taken away with her everything good that she no longer contained within herself.
Projective identification is therefore important because it tackles the lifeblood of psychoanalysis, the interplay of phantasy in intimate relationships. It is both defensive and communicative and the response of the analyst may be the primary factor in determining which aspect is uppermost (Joseph 1987). It is 'difficult' because the concept originates in a rather casual definition given by Klein (1952) and retains a title that does not really capture post-Kleinian extensions. Although 'communicative projection', or 'projective interaction' may be preferable expressions, Spillius (1988) suggests that projective identification should be retained as a general term within which various subtypes can be differentiated. The many motives behind the process - to control the object, to acquire its attributes, to evacuate a bad quality, to protect a good quality, to avoid separation, to communicate - may be useful starting points to identify subtypes.
38   Theory
NEU&OTIC MECHANISMS
E^ipiressnoini;, demuaH ajsuá disavowal
Repression has already been mentioned as the primary mechariism of defence during development which enables the child to maintain a balance between internal wishes and the constraints of the external world without excessive psychic pain. However, later in life, when ased excessively, it leads to marked dissociation of whole areas of emotional life from consciousness and according to classical theory leads to the formation of symptoms such as the 'belle indifference' associated with hysteria.   .
Example: the deadly memory
Á clear example of repression was observed in a patient suffering from depression. During analysis, he burst into tears when talking about the death of his mother when he was a child- He became aware that he had 'forgotten' that the cause of her death was an overdose of tablets. Only after having faced the fact that she had killed herself was he able to work through his long delayed grief reaction and recover from his depression.
En contrast to repression, which aims to remove an aspect of internal reality from consciousness, denial or disavowal (Freud 1940) deals with external reality and enables an individual to repudiate or to control affectively his response to a specific aspect of the outside world. Denial involves splitting in which there is cognitive acceptance of a painful event while the associated painful emotions are repudiated. The protective aspect of denial is illustrated by Greer et aľ's (1979) finding that women with breast cancer who showed denial (or defiance) when told the diagnosis, had a significantly higher survival rate after mastectomy than women who reacted with hopelessness or depression.
Reacťiím foiTMaiiitöEii? MemMfficatJkMs w5& fte aggressor
If an individual adopts a psychological attitude that is diametrically opposed to his conscious wish or desire, it is known as a reaction formation. Reaction formations often appear during latency and act as a bridge to more mature defences such as sublimation. Reaction formations may be highly specific - for example, showing excessive deference to some person one hates, or caring for others when one
Mechanisms of defence   89
wishes to be cared for oneself- or more generalised, in which case they form part of a character trait. Conscientiousness, shame and self-distrust may be examples of this when they are associated with obsessional personality and obsessional neurosis. As with all defensive pathologies, reaction formations alter the structure of the ego in a permanent way, so that the defence is used although the danger is no longer present.
Although Freud (1920) had alluded to 'identification with the aggressor', and Ferenczi (1932) had used the term to describe the behaviour of a child towards an adult in which there was a total submission to the adult's aggression and a resulting internalisation of profound feelings of guilt, it was Anna Freud (1936) who described the mechanism in detail and related it to the early formation of the superego. Identification with the aggressor has links with both reaction formation, in that there is a reversal of affects, and identification.
Example: the unassuming lawyer
A quietly spoken and timid patient was frequently threatened, humiliated and beaten during childhood by his father. Prior to ■ being beaten he would run away to his room while his increasingly angry father chased him. After the chase the little boy would suddenly go quiet and bend over and his father would then beat him, while he remained completely silent and entered a dissociated state. At this moment, the boy had dis-identified with his self-representation and identified with his father (the aggressor) who was going to beat his naughty bad body. In adult life, he himself continued the identification with the abusive father by taking illegal drugs and cutting himself, thereby allowing both the abuser and the abused to continue living out their interaction through his raind and body.
This process is particularly important in the intergenerational transmission of child abuse, in which the abused of one generation becomes the abuser in the next.
Isolation aed iimiloimg
These mechanisms are part of the psychoanalytic conceptualisation of obsessional disorders. Freud first described isolation as a distinguishing feature separating hysterical conversion and obsessional
neurosis. He suggested that if the individual did not 'convert* painful
90   Theory
affects through repression into bodily symptoms, then the affect was neutralized by isolation. The affect was rendered unconscious while the idea, stripped of feeling, remained conscious.
This contrasts with repression in which the idea rather than the affect is banished, and also with dissociation, in which the affect and idea both remain conscious but the connection between them is rendered meaningless. In isolation a traumatic memory may be easily retrieved but is denuded of any feeling - patients who have recently taken an overdose or cut themselves often talk about the circumstances leading up to the event with considerable cairn and may induce a false sense of ease in the assessing doctor. Fenichel (1946) suggested that the sexual and tender components of male sexuality are often isolated', leaving men more able to enjoy sensual feelings towards those whom they do not love, although his point could also be seen as an example of splitting.
Undoing is often referred to as 'doing and undoing' or 'magical undoing' and is well known to anyone who has treated patients with ao obsessive-compulsive disorder. Undoing enables the individual to reverse hostile wishes which he believes he has already perpetrated in the "doing7. In the obsessive-compulsive disorders this is often seen in the behaviour of the sufferer who may have to do things in a specific order only to have to repeat the whole ritual if one component is missed. The attempt to undo has a magical quality and is not just an attempt to make up for some error. It aims to reverse time, attack the reality of the original hostile thought or wish, and recreate the past as though such intentions had never existed.
HetereaUňsaStikwm samdl ůncoirporatíiom
Internalisation is a ubiquitous psychological process that is not necessarily defensive but can form part of adult defensive manoeuvres such as projective identification (see p. 83) and identification with the aggressor (see p. 83). Internalisation is a supra-ordinate term which subsumes introjection, incorporation and identification (see p. 83) and refers to all those processes by which the individual builds up his inner representational world by taking in and modifying the external world. Introjection is synonymous with internalisation early in development and applies to the formation of memories of perceptions."
When object relationships are well established, introjection does not consist so much of 'taking in' and laying down memories but
Mechanisms of defence   91
more of modifying their status in relation to the self - a student, after adolescent battles with his parents, who returns home from his first term at university and expresses surprise at how much his parents have changed, may have identified with his parental introject, or, more correctly, with that part of the introject which, in his rows, he was trying to differentiate himself from, and become more like them. This process is also common when young people become parents for the first time and finally identify with introjected parental attitudes in their approach to their child. Introjection, therefore, has an important role to play in the formation of the superego.
'Identification with the lost object' (Klein 1955) (cf. p. 72-3) is part of the normal mourning process which, along with introjection, may serve to delay the loss of the object. It is also a common feature in abnormal grief reactions:
Example: an internal object
A woman mourning the loss of her granddaughter appeared to have resolved her grief. However, some months later she developed intractable abdominal pain for which no physical cause was found. Her granddaughter had died from intestinal volvulus and the patient's symptoms were a result of an identification with the symptoms of the dead child. She was helped to recognise this during bereavement counselling and this led to the disappearance of her symptoms.
Incorporation is the psychological correlate of eating and refers to the 'swallowing whole' of an identificate (q.v. pp. 219-20) without modification or assimilation, and can be part of a psychotic form of internalisation.
Example: psychotic identification
A 45-year-old borderline patient with severe bulimic symptoms repeatedly cut herself and attempted suicide on a number of occasions. She exercised excessively - biking at least 10 miles a day and swimming a kilometre in the evenings. Her fantasy when exercising, cutting and inducing vomiting was that all the 'evil things' were being expurgated, and it was especially relieving to her if the vomit had streaks of blood in it. indeed, she believed that the only way to find absolute calm and tranquillity, and to be released from everything inside her, was if she stuck a knife into her body to see all the red evil things gush out of her inside. Her
3S   Theory
continually prayed that it was not true. This had allowed her to feel less wicked. Her mother's view of her had become a non-assimilated incorporated object. It had never fully resulted in a change in her self-representation which was protected by prayer. Only when she accepted these aspects of herself did her self-destroctive acts diminish.
A ssroilar process may occur in globus hystericus in which some 'thing' is experienced as being lodged in the throat.
EistellectaaSĚssitSoini antust ratioimaMsaitioEH ■
InteHectualisation and rationalisation are common in politics, business and medicine. They bridge the gap between immature mechanisms and those of maturity and often persist into adult life without leading to any overt problems. InteHectualisation covers a range of subdefences, including thinking instead of experiencing, and paying undue attention to the abstract in order to avoid intimacy. It may be associated with isolation, magical doing and undoing as well as rationalisation. The adolescent, fearful of his developing sexuality, may talk intellectually about premarital sex or earnestly discuss the sexual behaviour of young people. Rationalisation similarly offers logical and believable explanations for irrational behaviours that have been prompted by unconscious wishes.
MATURE MECHANISMS
Sublimation and humour were honoured by Freud as mature defence mechanisms for their capacity to allow partial expression of underlying wishes and desires in a socially acceptable way, while simultaneously enriching society. In sublimation, wishes are channelled rather than dammed or diverted: aggressive urges may find expression in gashes and sport; feelings are acknowledged, modified and directed towards significant goals. Similarly, narcissistic needs may be fulfilled by becoming a successful stage actor.
Example: the paint pot and the palate
Á 50-year-old engineer and amateur artist entered analysis suffering from severe depression after he had been made redundant by a firm to which he had devoted the best years of his life. He had never known his father, who had been killed in the war, and had grown up as an exceptionally controlled and 'good' boy who had
Mechanisms of defence   93
devoted himself to his widowed mother. When she remarried when he was 7 years old he submitted without murmur to his stepfather's harsh discipline. He prided himself on never losing his temper. He had married a subservient wife. On one occasion near the start of his depression, while he was decorating the house, she supplied him with the wrong sort of paint. To his horror he found himself simply upending the entire gallon tin on the floor and walking out of the house, leaving her to clear up the mess. As the analysis progressed he began to get in touch with deeply suppressed feelings of longing for closeness and intimacy, and anger that no one realised what he really wanted. Later, to his surprise he returned to painting, but instead of the rather meticulous line drawings he had done in the past, he started to smear paint thickly with the palate knife directly onto the canvas. Thus his aggression and need for emotional expression, which earlier had been a rageful manifestation of depression, were now sublimated into creativity.
Freud saw sublimation as the vehicle by which a society's basest and deepest desires as well as aspirations and ambitions gain expression, through carnival, drama, music, poetry and religious and political aspirations. Humour allows us to share emotion, often aggressive, without discomfort, to regress without embarrassment, to play games with freedom, to laugh with impunity and relax with pleasure; it includes rather than rejects and may at times allow terrible tragedy to become bearable.
Vaillant and Drake (1985) and Vaillant et al. (1986) have confirmed the adaptive value of mature defences. Vaillanťs longitudinal studies of men, followed from graduation over a forty-year period, show that those who use them are consistently happier and more successful and stable occupational ly, and in their family lives, than those using less mature defences. Perry and Cooper (1989) similarly found that immature mechanisms of defence are associated with psychological symptoms, personal distress and poor social functioning. Such findings, along with the general usefulness of defence in understanding psychological function, led Sartorius et Í2/.099Q) to suggest that descriptive (i.e. non-psychoanalytic) psychiatry's efforts to link clinical symptoms, causal factors, pathogenic models and prognostic types had generally failed. He argues that the return of 'allegedly out-dated Meyerian reaction patterns and Freudian defence mechanisms is warranted'. This comment emphasises the
94   Theory
continuing importance of psychodynamic contributions to psychiatric diagnosis, and Vailiant (1992) also argues cogently for an 'axis VI' (Defence Mechanisms) to be added to future multi-axial classi-ficatory systems. Reliable measurement of many of the defence mechanisms is now possible (Bond et al. 1983; Vailiant et al. 1986) and study of their change during psychoanalytic sreatment may allow a better understanding of the interventions needed to help overcome s-ieurosis and to achieve psychological health.
It cannot be disputed that controlling the phenomenon of transference presents the psychoanalyst with the greatest difficulties.
(Freud 1912b: 108)
The hallmark of psychoanalysis is the use of transference and countertransference as a guide to understanding the inner world. But what precisely is transference, and how does it relate to the different models of the unconscious? Here again we come up against the 'elasticity' of psychoanalytic concepts discussed in the previous chapters. We have seen how variations between psychoanalytic models of the mind, and contrasting theories of development, may affect the analytic approach to the patient. These theoretical differences inevitably also impinge on the concepts of transference and countertransference since both are inseparable from the framework within which they are discussed. To put it schematically, interpersonal approaches are likely to describe transference in terms of a two-person interaction with contributions from both analyst and patient; ego psychology considers transference in terms of the expression of instinctual wishes; and a Kleinian perspective in terms of representation of unconscious phantasy. As has already been mentioned in Chapter 2, some of these are differences of substance; others no more than variations of language. However, no approach is pure and the interpretation of unconscious phantasy is no more the prerogative of a Kleinian position than consideration of instincts is of ego psychology or a contemporary Freudian approach. There is general agreement about the existence of transference as a phenomenon, but a real debate exists concerning its content ~ i.e. what exactly is 'transferred'. There are also important differences of opinion about the centrality or otherwise of transference interpretation as the only