After decades of relative neglect, due mostly to the concurrent neglect of the effects of real-life traumatic experiences in psychopathology and psychotherapy, there has been an upsurge of interest in the dissociative disorders (DDs). The introduction of this diagnostic category in the third edition of the Diagnostic and statistical manual of mental disorders (DSM), has been instrumental, since 1980, in calling attention to disturbances in the integrative functions of memory, consciousness, and identity. This renewed interest notwithstanding, the nosology of the DDs remains a problematic issue (Dell, 2001). Arguments supporting the clear differentiation of the DDs from other disorders—e.g., personality disorders of the dramatic cluster (especially borderline personality disorder, BPD), conversion disorders, chronic posttraumatic stress disorders, and trauma-related psychotic states (as described by Kingdon and Turkington, 2002)—are not compelling. The prototypic disorder in the category, dissociative identity disorder (DID, formerly known as multiple personality disorder), is the subject of particular controversy. While many psychotherapists, especially in North America, diagnose it with an impressive frequency, others, mainly in Europe, doubt the real existence or the prevalence of DID, and regard many published clinical cases of DID as artifacts. Moreover, current DSM-IV nosology, while it focuses on the presence of alternate personalities (alters) for the diagnosis of DID, does not provide clear guidance with regard to the phenomenology of alters. The possibility of clinical syndromes in which ego states are only partially dissociated (and therefore cannot be considered full-blown alters) is acknowledged by DSM-IV in the subcategory of ‘dissociative disorders not otherwise specified’ (DDNOS). This possibility adds to the difficulty of identifying patients suffering from DID insofar as it is not always easy to decide when an ego state is fully rather than only partially dissociated. As a result, whereas most papers on the psychotherapy of the DDs are concerned with DID, reliable information on the prevalence of this disorder, on the boundaries between it and related DDs, and on the treatment of DDs different from DID, is scarce.
The unsatisfactory status of the nosology and the epidemiology of the DDs, and the paucity of controlled outcome studies of their treatment, requires a particularly careful consideration of what we know about their etiology in order to understand the logic and the potentialities of their psychotherapy.
The DDs can be regarded as very complex and long-term types of posttraumatic stress disorder, beginning acutely during childhood and becoming chronic throughout adolescence and adulthood. As trauma has repeatedly impinged during crucial years of development, many different dimensions of the experience of self are affected (Mollon, 1996). Childhood sexual abuse, often perpetrated by family members, is particularly prominent among the traumatic memories of patients suffering from DDs (Allen, 2001). Other types of traumas (e.g., neglect, physical violence, severe humiliations) also play a part in the etiology of the DDs.
Trauma elicits dissociation, which is a discontinuity of experience (consciousness) and memory. Allen (2001) suggests there are broadly two components to this—detachment from the overwhelming experience, and compartmentalization of the experience. Dissociation may serve initially as an adaptive function of making fear and psychological pain accompanying trauma more tolerable. Over time, however, dissociation distorts personality development and the ongoing integration of memories, self-perception, and perception of emotions in other people. It also acts as a fragmented representation of the original traumatic experiences through the perpetuation of hyperarousal in response to stimuli reminiscent of trauma, as well as through its other manifestations in numbing, intrusive flashbacks, and nightmares. Dis-integrated perceptions of self and the environment may give way to depersonalization and derealization. To these reverberating echoes of the traumatic experience, patients may react with panic and the deepest discouragement. Thus the flashbacks create secondary trauma. Greater discontinuities of memory (amnesia) and even of identity (fugue) may grow out of the dissociative defenses against the original traumas. Reciprocally dissociated ego states (alters) may form, giving rise to the experience of internal conflict and sometimes resulting in different ego states and identities taking executive control at different times. Understandably, when personality development takes place in such a traumatic-dissociative climate, the ability to regulate emotions and to control aggressive impulses (both toward oneself and toward others) is usually underdeveloped. Also undermined is the development of those metacognitive abilities that allow for critical reflection on one's own or other people's states of mind (‘theory of mind’ and mentalization: Fonagy et al., 2002). Interpersonal relationships, as a consequence, become often stormy or otherwise very difficult.
It may be noted that this description of the DDs may widely apply also to BPD: high rates of childhood trauma are reported by BPD patients, dissociation of mental states is among the clinical features required for diagnosing BPD, and the comorbidity between DID and BPD is very frequent. Some psychoanalysts hold that different defense mechanisms (dissociation and splitting, respectively) are responsible for the different features of DID and BPD. Other clinicians, however (Ross, 1989; Blizard, 2001) deny any substantial difference between DID and BPD. It is therefore debatable whether the conceptualization of psychotherapy interventions in cases of DID and of BPD should differ significantly or not.
However important the role of traumatic experiences in the genesis of the DDs, it should be noted that other risk factors concur in their development, the understanding of which is important in treatment conceptualization. These other risk factors explain the existence of cases, although comparatively rare, of DDs where no history of childhood trauma can be reconstructed, and of cases of severe childhood trauma associated with other disorders (anxiety disorders and mood disorders). Temperamental traits (e.g., susceptibility to hypnosis, linked to the genetic make-up of the individual: Bliss, 1986) may increase the tendency to react to trauma with dissociation. Unbearable loneliness may also contribute to the development of imaginary friends and alternative ‘realities’. These factors may allow for the possibility that a DD develops as the consequence only of subtle relational traumas in the absence of obvious child maltreatment. Particular types of early attachment relationship (attachment disorganization: see below) seem to exert an adverse influence, since the earliest phases of life, on the integrative functions of memory, consciousness, and identity (Main and Morgan, 1996; Liotti, 1999; Lyons-Ruth and Jacobvitz, 1999; Schore, 2001): they may therefore constitute risk factors in the development of disorders implying dissociation.
It should also be emphasized that, very often, the perpetrator of the abuses reported by the wide majority of dissociative patients is a caregiver. Therefore, most dissociative patients had to face, as children, a series of relational dilemmas whose core is that, in order to maintain attachment to the caregiver, the abuse must be denied, forgotten, and dissociated (Freyd, 1996), while at the same time, in order to protect themselves from abuse, attachment wishes must be disavowed (Blizard, 2001). These relational dilemmas and the related defensive strategies (usually taking place within previously disorganized attachment relationships), rather than traumas as such, set the stage for the development of dissociated ego states.
From this consensus view of the etiology and pathogenesis of the DDs, follows the conceptualization of the psychotherapy. The goals to be pursued are:
the stabilization and reduction of symptoms of anxiety, depression, or impulse dyscontrol;
the processing of traumatic experience;
the integration of memory, consciousness, and identity;
the development of the capacity of trusting interpersonal relationships and of the relational skills necessary to both self-protection and secure attachment.
Although the restoration of the integrative functions of memory, consciousness, and identity requires the processing of trauma, the emphasis is not, as in some earlier recommendations, on abreaction (i.e., on the supposedly cathartic ‘reliving’ of the original painful experience within the therapeutic dialog). Rather, it requires attention to developmental, relational, and self-regulation issues.
The importance, for the efficacy of psychotherapy in the DDs, of achieving integration (of traumatic memories, ego states, or alters), is asserted by two follow-up studies (Ellason and Ross, 1997; Coons and Bowman, 2001). However, methodological limits of these studies mean they can only be regarded as preliminary findings.
Despite the paucity of methodologically satisfactory outcome studies, three areas of research have provided reliable findings that are of great relevance in the psychotherapy of the DDs. The first is concerned with the strong links between early childhood trauma (sexual abuse and notably incest, physical and emotional abuse, especially intrafamilial) and dissociation in adolescence and adulthood (Ross et al., 1989; Coons, 1994; Simeon et al., 2001; Pasquini et al., 2002). The second has to do with the reliability of traumatic childhood memories, as they may be retrieved during adult psychotherapy. The third deals with the developmental consequences of attachment disorganization in increasing the risk for abnormal dissociative reactions to trauma.
Extensive and diverse research findings show that autobiographical memory is an active process of reconstruction, liable to influence by suggestion and the incorporation of information from various sources, rather than an accurate and stable registration/retrieval of events (Loftus, 1993; Mollon, 2002a). Moreover, trauma can differentially affect explicit and implicit memory. The nonverbal (nondeclarative) aspects of the experience, such as the fear response, may be encoded as implicit memory, while the verbal aspects (i.e., the comments of the perpetrator, the victim's inner dialog during the experience, the narrative of the experience) may not be encoded. According to recent neuropsychological findings, the fact that the declarative or narrative aspects of the traumatic experience may not be encoded as explicit memory is due to disruption of the functions of the hippocampus in the brain. This hippocampal dysfunction, in turn, is linked to the high levels of stress neurohormones produced during serious traumas. At the same time, the perceptual, physiological, emotional, and motor aspects of the traumatic experience, and particularly the fear response, are recorded in other neural maps involving the amygdala (which, unlike the hippocampus, is not affected by the stress neurohormones). Therefore, the traumatic experience cannot be retrieved in any narrative form during verbal interchanges concerning the patient's memories. While the body memory of the trauma persists in the form of phobias and psychosomatic disturbances, the traumatic experience may never become part of autobiographical memory (Allen et al., 1999; Allen, 2001; Mollon, 2002a).
Because of the complex interplay between declarative and nondeclarative memory, traumatic memories verbally reported by patients in psychotherapy may combine confabulation and accurate recall (Mollon, 2002a). The psychotherapist should therefore be alert to the dangers of suggesting (either explicitly or inadvertently) that the patient may have been abused as a child if this has not already been reported, and should be tolerant of uncertainty regarding the accuracy of traumatic memories emerging during the treatment of dissociative patients (see below, particularly ‘Middle phase: processing trauma and beginning integration’, for details on what such ‘tolerance of uncertainty’ may mean in clinical practice).
Infants are said to be disorganized in their attachments when they show a mixture of approach and avoidance behaviors toward the caregiver during a standard sequence of brief episodes of separation from and reunion to the caregiver, known as Strange Situation (Main and Hesse, 1990; Lyons-Ruth and Jacobvitz, 1999). Disorganized attachment may also show as disoriented attitudes of the infant toward the caregiver (e.g., trance-like states during the interactions with the caregiver in the Strange Situation).
Disorganized attachment patterns develop in infants as the consequence of caregiving parental behaviors that are frightening to the child, either because they are violent or because they express fear and/or dissociative experience in the parent (Schuengel et al., 1999). These frightened/frightening parental attitudes, in turn, are linked to unresolved losses and/or traumas (Main and Hesse, 1990; Lyons-Ruth and Jacobvitz, 1999).
The relevance of such studies for an understanding of dissociation is suggested by certain similarities between the experiences and behavior of dissociative patients, the behavior of infants showing disorganized attachment, and the mental states of these infants’ parents (Liotti, 1992, 1999; Main and Morgan, 1996). Furthermore, the early, implicit representations of self-with-others (or internal working model, IWM) stemming from disorganized attachment are very likely to be multiple, incoherent, and dissociated. This is at striking variance with the IWM of the organized patterns of early attachment (secure, insecure-avoidant, and insecure-resistant), that are single and coherent (Main and Hesse, 1990; Liotti, 1999). The IWM of disorganized attachment conveys dramatic emotions that quickly shift from rage to fear to hopelessness, closely mimicking the mutable dramatic interpersonal emotions so easily observed in dissociative and borderline patients. Moreover, the implicit memory structures composing this type of IWM convey information that facilitates the later construction of dissociated representations of self and others according to the three stereotypes of the ‘drama triangle’: persecutor, rescuer, and victim (Liotti, 1999). These three stereotypes correspond to the basic structure of the most common dissociated ego states, or alternate personalities, which have been observed in DID (protective alters, persecutor alters, and victim alters that often rehearse the traumatic memories).
Controlled studies, both longitudinal (Ogawa et al., 1997; Carlson, 1998) and correlational (Liotti et al., 2000; Pasquini et al., 2002), support the hypothesis that early disorganized attachment is linked, throughout development, to propensities toward dissociation, DDs and BPD. These studies not only contribute to clarifying the etiology of the DDs; they also suggest how the knowledge of attachment disorganization helps in dealing with the complex dramatic type of therapeutic relationship these patients tend to establish (Liotti, 1995; Liotti and Intreccialagli, 1998; Fonagy, 1999; Blizard, 2001; Steele et al., 2001).
Over the past 20 years, clinicians treating dissociative patients and other adult survivors of childhood abuse have achieved, according to informed reviewers (Courtois, 1997; Chu and Bowman, 2000), wide consensus as to the treatment of these disorders. This consensus model is based on the idea that the psychotherapy should be phase oriented, with attention to the therapeutic relationship, belief systems, and the structure and experience of self taking precedence over the exploration of trauma (see, e.g., Putnam, 1989; Herman, 1992; Davies and Frawley, 1994; Courtois, 1997; Fine, 1999; Kluft, 1999; Blizard, 2001; Gold et al., 2001; Steele et al., 2001). The preliminary phase of the treatment, according to the consensus model, is devoted to alliance building and safety. The intermediate one is focused on processing traumatic memories. The late phase aims at personality integration and relational rehabilitation (i.e., further integration of dissociated mental functions and development of self-care and relational skills). This sequence is not strictly linear; although phase oriented, the treatment alternates as a spiral between the themes of the three stages. Knowledge of disorganized attachment, as it will argued below (‘Preliminary phase: alliance building and safety operations’ and especially ‘Middle phase: processing trauma and beginning integration’), may usefully guide the psychotherapist in deciding the individualized timing and manner of such alternations.
It should be noticed that, while the consensus model provides a schematic conceptualization of the therapeutic process, some (e.g., Mollon, 2002a,b) are less impressed with the idea of a clearly sequential approach in practice, and emphasize extreme caution regarding therapeutic goals.
Dissociative patients are prone to expect dramatic shifts in the attitudes of the caregivers, reflecting their experience of being abused or neglected by the primary caretakers. In more traditional psychodynamic terms, these expectations have been linked to the simultaneous operations of sadistic and masochistic defenses (Blizard, 2001). In terms of attachment theory, these expectations, reflecting the patients’ early experiences, make them unconsciously prone, whenever their attachment system becomes active, to shift or switch between reciprocally dissociated implicit mental structures (IWMs) for organizing experience and behavior. Liotti (1999) has suggested that these switches can be captured by the model of the drama triangle, where three basic narrative templates for representing self-with-others alternate with each other. The first template has, as a theme, the helplessness of self and/or others (theme of the victim). The second narrative template is defined by the theme of the persecutor, and the third by the theme of the rescuer. Whenever this drama triangle regulates the inner narrative and the emotional experience, the integrative functions of the mind (metacognitive capacity, theory of mind, self-reflection) are seriously hindered.
The therapeutic relationship is particularly apt to facilitate the activation of the attachment system among the various motivational systems of which human beings are endowed (e.g., the competitive system, the cooperative system, the sexual system, and the exploratory system: Gilbert, 1989; Lichtenberg, 1989). This means that during psychotherapy it is very likely that the patients will come to represent self and the psychotherapist according to the drama triangle, with consequent hindrance to metacognitive ability (metacognition is the capacity to monitor one's own mental states, and to reflect critically upon them). Patients (and their therapists) are often confronted during psychotherapy with a relational dilemma, in which it seems impossible to achieve both self-protection and protective closeness (Blizard, 2001). Because of the hindrance to metacognitive abilities, it is difficult to deal with this dilemma by critically reflecting upon it. In these situations, the patient may oscillate between abnormal dependence on the therapist (construed as the rescuer) and equally abnormal independence (e.g., when the therapist is perceived as a potential persecutor: Steele et al., 2001).
Thus, the too early, too frequent, or too intense activation of the attachment system could create a situation that exceeds the capacity of the patient to regulate interpersonal emotions and reflect on interpersonal experience. It is therefore vital that the therapeutic relationship is so structured as to titrate the activation of the patients’ attachment system. In order to avoid the risk of too strong an activation of the attachment system, the treatment guidelines of the International Society for the Study of Dissociation (2000) suggest as optimal the frequency of two or three sessions each week; more than three sessions should be considered only after having carefully evaluated the risk of fostering excessive dependence. For the same reason, at the beginning of treatment, dealing too closely on traumatic memories should be avoided; the painful experience of retrieving such memories would bring with itself the powerful activation of the attachment system (see Bowlby, 1982, for an account of how the activation of the attachment system may relate to the experience of either physical or emotional pain). Rather, the therapist should strive to facilitate the activation of the patients’ cooperative system, that corresponds to the building of a therapeutic alliance in which both patient and therapist explicitly perceive themselves as sharing a common goal.
In order to build up the therapeutic alliance at the beginning of psychotherapy, some clinicians ask the patient what his or her goals for the treatment are, and accept them explicitly as a preliminary aim (provided, of course, that they are reasonable and ethically acceptable). Patients who may often have experienced being powerless at the hand of an abusing caretaker, or being unheeded by a neglecting caregiver, are thus empowered within the therapeutic relationship (Courtois, 1997). Mollon (2002a,b) emphasizes the importance of seeking an internal consensus among dissociative parts concerning whether to proceed or not with psychotherapy; to go ahead only on the basis of the wishes of one dissociative part is to risk an internal ‘civil war’.
In this phase of treatment, when the patient may ask for relief from anxiety symptoms or depression, the therapist may reply with standard cognitive-behavioral techniques for anxiety and mood disorders (Kennerley, 1996), and/or with the prescription of serotonergic or mood-stabilizer drugs. Thus, while aiming at symptom reduction and stabilization, it may become clear to the patient that the therapist not only listens empathically to him/her, but actively and efficiently cares for his/her well-being. In so doing, it is important to avoid any violation of therapeutic boundaries (e.g., through protective overinvolvement, or collusion with the patients’ fantasies of having met a loving rescuer from their sufferings). Boundary violation, always a great danger in psychotherapy, is particularly harmful to dissociative patients, as it subtly repeats and confirms structurally similar violations in the relationship with abusive parents. The therapist should never attempt through physical contact (of any variety) to meet the dissociative patient's often desperate quest for affection and comfort. Similarly, the therapist should also avoid acting upon the often powerful countertransferential wishes to offer reparation for what may have been the patient's extreme childhood experiences of pain and betrayal. While offering professional protection and understanding, they should be wary of the risk of overprotecting their patients (see also below, ‘Difficult situations and their solution’).
While boundary violation should never be allowed, sporadic, prudent boundary crossing for therapeutic purposes may be beneficial to dissociative patients (Dalenberg, 2000).
Lisa, suffering from a DDNOS that could be described as an incomplete multiple personality (she had partially dissociated ‘parts of herself’, rather than fully dissociated ‘alters’), came to a session with a large bandage on her upper arm and forearm. She reported having cut herself with a lancet she had previously stolen from her husband (a surgeon). Her comments and her tone of voice on reporting the episode expressed only contempt toward the ‘cowardly’ and ‘bitchy’ part of herself (‘That little bitch, what a coward… unable to stand even this little pain… and she had deserved quite a bigger one…’). While listening to Lisa's cruel report of how she had cut herself, the therapist—who was supervised by G.L.—was aware of the extremely brutal and sadistic sexual abuse which she, when only 4, had suffered from her father (Lisa's older brother had witnessed the abuse and reported it to the therapist). He felt moved at the idea of the little victim Lisa had been, now again victimized by herself. He decided not to conceal his feelings from the patient, and let a tear run through his face. On the next session, Lisa commented that the therapist's tears had been a topic on which she had reflected. Her insight may be summarized as follows: ‘Maybe I needed that somebody else could cry over my pain, to become able to cry over it myself. Nobody ever cried or was moved when I suffered as a child’.
Another instance of rather courageous therapeutic boundary crossing is reported by Brenner (1996, p. 791), when he describes his reaction to an assaultive and suicidal patient, suffering from DID, who revealed during a session that she was hiding a blade. While realizing that she could have easily cut him or herself, the therapist offered his outstretched hand asking for the blade, which the patient, after a menacing look, carefully handed over to him. The patient was much relieved by this interaction and ‘the incident became a nodal point in the treatment’ (Brenner, 1996, p. 791).*
Brenner's clinical vignette illustrates also the second main theme of the preliminary phase of psychotherapy: together with the building of alliance, it is of vital importance that dissociative patients experience, quite explicitly and from the beginning of treatment, that the therapist regards their safety as a primary goal. Whenever the patient, verbally or behaviorally, raises an issue concerning any type of self-harm (e.g., grossly abnormal eating behavior, self-mutilation, promiscuous sexuality without prevention of infection, threats of suicide), the therapist should immediately and primarily focus attention on it, suspending any other type of therapeutic work. In addition to exploring the meaning and function of the behavior, safety contracting, along the lines so profitably suggested by Linehan (1993) for the cognitive-behavioral therapy of borderline patients, may be instrumental in conveying, without emotional overinvolvement, that the therapist wants the patient alive and well. This creates the condition for a type of corrective relational experience that is much needed by dissociative patients. As their previous attachment relationships may have been with caregivers who were abusive in the extreme, as well as neglecting—and who were themselves extraordinarily vulnerable, being traumatized and dissociative—these patients usually hold, consciously or not, the belief that nobody cares for their life and well-being. The experience of a secure attachment to a therapist who, within the boundaries of a cooperative therapeutic relationship, explicitly and coherently values the patient's life, safety, and well-being, challenges this pathogenic belief. In the mind of the patient, this ongoing experience of cooperation and secure attachment presents an alternative to the IWM of previous disorganized attachments.
The interpersonal meanings associated with disorganized attachment (drama triangle) are thus mobilized within the therapeutic relationship in a context of relative safety, enabling a test of the validity of the new interpersonal information as against the old IWM (cf. Weiss, 1993). Patients can act on the assumption that they are evil, guilty, and not deserving of care (self in the ‘persecutor’ role of the drama triangle); that they are hopelessly deemed to suffering and no type of care could ever heal them (the ‘victim’ role); that they have finally found in the now idealized therapist an omnipotent rescuer who will heal them without any effort on their part (therapist in the ‘rescuer’ role); or that the therapist is only concealing his real, evil intentions under the mask of a faked protective attitude only to exploit, disillusion, and perhaps abuse them later on (therapist in the ‘persecutor’ role of the drama triangle). Within a group of sessions or even within the same session, the patient's attitudes may shift between these roles, in such a quick manner as to deserve the label ‘kaleidoscopic’. In DID, these shifts may correspond to one or more alternate personalities entering the stage of therapeutic dialog. In DDNOS and in other DDs, they may correspond to ego states that are less ‘autonomous’ and less reciprocally dissociated. Also, previously avoided images or narratives of abuse may begin to surface. This opens up the second phase of the treatment, as in the following clinical illustration.
Tina had been in psychotherapy for about a year when she had the first, and only, occurrence of generalized dissociative amnesia (acute and total loss of autobiographic memory). She woke up one morning having forgotten her name, her history, and the identities of her husband and her child. In a panic, speaking with a childlike voice that was unfamiliar to her frightened husband, she stated that she did not recognize anybody in the house, nor the house as her home, and was unable to remember anything of her past.
At the beginning of treatment, she had been severely depressed. Her recurrent depressive episodes were further complicated by dissociative experiences, in the form of prolonged trance-like states (that had been mistaken, by her GP, for the apathy of depression). Her clinical history revealed, since adolescence, a mild and atypical anorexia nervosa, orgasmic dysfunction, and somatoform symptoms suggestive of childhood traumatic experiences. The daughter of an alcoholic man, she had reported severe emotional abuse from her mother during the first phase of treatment, but not from her father.
Tina felt much helped by her psychotherapist (G.L.) during the first months of treatment. Soon thereafter, she begun to report feelings of sexual arousal during the sessions, and became explicitly seductive toward the therapist. He politely refused her advances, stating that the only relevant thing in their relationship was the pursuing, through joint reflections on her experiences, of the therapeutic goals they had agreed upon during the very first session. After a brief phase of resentment for the refusal, Tina seemed relieved, as if she now felt safer in the relationship. At the same time, she started having dreams concerning apparent memories of incest with her father when she was about 7. She was bewildered by the veracity of these dream memories, and begun to wonder with the therapist whether or not they could reflect actual, until then forgotten events of her late childhood. The episode of dissociative amnesia took place in this context.
The therapist responded to the crisis by accepting the necessity of crossing the boundaries of psychotherapy. He responded to the request of Tina's frightened husband, to visit her at home as soon as possible (she was obviously unable to go anywhere on her own, and she refused to be accompanied by him, whom she now saw as a stranger). When the therapist arrived at Tina's home, a few hours from the beginning of the generalized amnesia, she did not recognize him, and asked ‘Who are you?’. He replied: ‘I am a doctor. Could I help you?’ She then expressed her extreme fear at finding herself in that unknown place, and at having ‘totally lost memory’. The therapist empathically said that he, too, would have felt frightened if he had lost his memory and awoke in a place he was unable to remember, thereby reframing Tina's way of constructing her experience (while ‘being in an unknown place together with strangers’ and ‘having lost memory’ were separate issues to her, the therapist indirectly suggested that, as she had lost her memory, she may not remember her home and her own family). Tina looked at him perplexedly for perhaps a minute, than she cried ‘Now I remember! You are my psychotherapist! I have told you that I have been abused by my father!’
The subsequent phase of the treatment was devoted to exploring both her emerging memories of incest, their veracity, and the meaning that remembering them had to her. This exploration was guided by the principles outlined in the following section.
* The reader should notice that the two therapists in these clinical vignettes, although taking a risk and crossing the boundaries of the therapeutic relationship, did not violate them (e.g., by touching the patient or by expressing any affect that should be illegitimate to express within a therapeutic dialog).
The processing of traumatic memories and of their meanings is the core of the treatment of DDs. In DID, this exploration of painful childhood memories is intertwined with the need to deal with alternate personalities that show up in the therapeutic dialog.
Some psychotherapists advocate a careful exploration of the alternate identity system once the first alter has spontaneously established direct contact with the therapist and the patient has been stabilized and strengthened (e.g., Kluft, 1996; Fine, 1999). They thus ask (most of them nowadays avoiding hypnosis during this exploration) if there are other identities willing to share their issues and concerns besides the one that had already established contact with the therapist. Other therapists fear that this clinical choice may encourage confabulation and iatrogenic expansion of the number of alters, and prefer to dialog only with those that may spontaneously enter the stage of psychotherapy. There is, however, wide consensus that, if an ego state does not present him/herself as a separate identity with a different name, therapists should be wary not to reify it (e.g., asking for his/her name or speaking as if they believed in the existence of different persons sharing the same body with the patient's host—i.e., primary— personality). Whenever possible without invalidating the patient's experience of switching to an alternate and separate identity, the therapist should address dissociated mental structures and behaviors (such as Tina's speaking with the voice of a child) as ‘parts’ or ‘states’ of the patient's self. That there is a unitary self for each human body may be an illusion, as has been argued by authoritative philosophers and psychologists (see, e.g., Dennett, 1991), but then it is a necessary and universal illusion. Equally, the idea that disorganized parts are separate ‘individuals’ is an illusion—a pretence that has structured the personality—as all are part of the overall holonomic mind (Mollon, 1996, 2002a,b).
Even wider is the consensus on the need, in order to achieve the integration of hitherto dissociated mental structures, to deal carefully in this phase of the treatment with the patients’ traumatic memories (whether they be of frank abuse or of more subtle relational traumas in the early attachments). Courtois (1997) has described three main scenarios concerning the status of traumatic memories in dissociative patients. In the first, traumatic memories are accessible to the patient from the beginning of treatment, but not divulged because of shame, guilt, and family loyalty. In the second scenario, they are not known to the patient at the beginning of treatment. In the third scenario, they are not known with certainty but are suspected by the patient. Moreover, the position is complicated by processes of dissociation, which mean that what may be known in one state of mind is not known in another.
In the first phase of treatment, even if strongly suspecting the existence of traumas, therapists must tolerate the patients’ not knowing or not disclosing, and avoid any pressure toward remembering or disclosure. In the second phase, therapists not only empathically listen to the patients’ spontaneous report of traumatic memories, but should also actively inquire on them. Hypnosis should be avoided in this exploration of past traumatic experiences, because of the risk of creating false memories. The exploration should not aim at mere abreaction, but at meaning: the therapist should empathically inquire on the meaning that both the remembered experiences and the experience of remembering has to the patient (Mollon, 2002a,b).
Sally, an intelligent but very troubled woman of 19, was hospitalized after becoming very disturbed following a sexual assault. She appeared at times disoriented to time and place, and displayed signs of extreme fear. In the calm ambiance of psychotherapy she began to settle. However, she would still sometimes express anxieties that the therapist (P.M.) would attack her physically. Gradually she disclosed more about an internal image that had begun to haunt her. It was of a little girl alone in a room. At times she spoke of hatred and fear of this little child. On other occasions she would deny that there was such a child in her mind. In lucid moments she would speak of her realization that the image was of her child self—the child who had been extensively abused by the ‘uncle’ she had been sent to live with after her mother became ill when Sally was age 3. She described a process whereby she had believed she could omnipotently repudiate her abused child self and create a new version of herself; she would ‘pretend’ that the bad experiences had not happened—but then found that she was confused about what was pretence and what was real. This process was spontaneously enacted within the therapy, in that she would speak of her uncle's terrifying behavior towards her, but then a moment later would state that none of what she had just said was true, that there was no little girl, and that there was nothing wrong with her. She would say she had just pretended there was a little girl—but might then express confusion about whether she was pretending to pretend. Repudiating what she had just said became a recurrent pattern in the therapy. There were periods when she would appear extremely childlike, but these episodes were associated with considerable anxiety. Overall, Sally seemed to become more relaxed and trusting as she discovered that the therapist remained calm, interested, and inquiring, while not reaching for premature conclusions about the content and meaning of her memories and fantasies.
The above vignette illustrates how, during trauma work, considerable uncertainty must sometimes be tolerated both by patient and therapist. When patients ask for the therapist's assurance that their traumatic memories are totally real, honesty in the therapist's reply is particularly important. The therapist should acknowledge that there is no method to distinguish with absolute certainty accurate memories from inaccurate ones, and help the patient in accepting this fact. Empathy for the anguish with which patients reflect on their uncertainty about the real occurrence of traumas is essential. Therapists should be clear and explicit that they consider the painful meaning of the uncertainly surfacing memories absolutely real and dramatically important in the patients’ life, even if they cannot confirm or disconfirm the reality of their content.
Therapeutic techniques such as ‘eye movement desensitization and reprocessing’ (EMDR: Shapiro, 2001), which have been promisingly used in the treatment of posttraumatic stress disorder (see Chapter 13 for a description of the technique), have also been advocated for working with the traumatic childhood memories of adult dissociative patients. This method, involving bilateral stimulation of the two hemispheres, through eye movements or auditory stimulation alternating in either ear, can be viewed as a method of accelerated processing of emotional information (Mollon, 2001b). It has been argued that incorporating EMDR in the trauma work with dissociative patients may provide a protective format for the processing of otherwise overwhelmingly painful memories, by reducing the risk of negative transferences during trauma work (Twombly, 2000). Extreme caution, however, should be exercised in the use of EMDR with patients who have suffered extensive trauma in childhood (Mollon, 2002a), because of the danger of ‘opening the floodgates’ to unmanageable levels of dissociated memory and affect.
The therapeutic work on memories of abuse is a phase, often unavoidably long, of mourning and resolution of the traumas. It is also a phase of psychotherapy in which integration begins. The exploration of traumatic experiences allows for the joint understanding, by patient and therapist, of the coping reactions that have led to dissociative experiences (e.g., numbing, trance-like states, amnesia, depersonalization) and dissociated ego states, which thus begin to be integrated in the patients’ explicit self-knowledge. The function of alternate personalities, if present, becomes intelligible when matched with the experience of abuse and with the dramatic status of attachment relationships. For instance, protective personalities may have had the function of coping with aspects of reality that exceeded the coping capacity of the traumatized host personality. Persecutor personalities may have been created in order to express both overwhelming rage and guilt (linked to the belief of being responsible for the abuse or of having deserved it). Victim personalities may had the function of preserving both the memory of the abuse and the associated meaning of being totally helpless at the hand of the perpetrator.
As a primary attachment figure may have been the perpetrator of abuse or may have been neglecting, or may have been perceived as exceedingly fragile and unable to protect the patient from the abuse, the therapist's close attention to the dynamics of attachment is usually rewarding. In this phase, the patient may begin to share with the therapist and understand the simultaneous presence of utterly incompatible and dramatically strong feelings toward, for instance, a parent who was at times frightening, abusive, and deeply emotionally ill, while at other times offered them at least some protection and comfort (otherwise they, as children, would not have survived). The therapist's awareness of the dynamics of disorganized attachment assists in expecting and understanding the patients’ dramatically shifting transferences. This understanding is invaluable in protecting therapists from untoward countertransferential reactions and from misunderstanding of the meaning of a patient's shifts, say, from gratitude and hope to expressed hopelessness, fear, suspicion, or even hatred toward the therapist—and it must be recognized that at certain points in the vicissitudes of the transference the therapist will be perceived as being as bad as the original abuser. The therapist is offering empathy and support to a deeply suffering patient who is mourning over very painful memories. This interpersonal situation—a therapist offering empathic understanding to a deeply suffering patient—unavoidably activates the patient's attachment system within the therapeutic relationship (the attachment system is activated when one's suffering is met by a person perceived as ‘stronger and wiser than the self’: Bowlby, 1979, p. 129). As the patient's IWM of attachment is disorganized, it is likely that he/she will be prone to construe the therapist's role, alternately, as that of the rescuer, the perpetrator of abuses, and even the victim of the patient's alleged evilness.
When changes of ego state in the transference are understood as shifting aspects of the relational dilemmas of attachment disorganization, therapeutic exploration of their meaning begins to center on a unitary meaning. Patients, while beginning to experience a secure attachment to the therapist (an important emotional corrective experience), may understand that a unitary motive—the wish to be understood and of having their suffering soothed by another person—is at the base of their manifold shifts from idealizing to devaluing or attacking self and others. This corrective emotional experience and these reflections foster integration at the level of the patient's basic meaning structures of self-with-other.
In cases of DID, reciprocally dissociated protective, persecuting, and victim ‘personalities’ alternate during the sessions. Attachment theory, in these cases, offers to the therapist a way to conceive the basic, unitary psychic structure of self-with-other from which the alters are created out of reciprocally incompatible and disowned memories, expectations, beliefs, affects, and wishes. Bearing in mind this structure (however, it is conceived within different theoretical frameworks) it becomes easier to establish moments of dialog with the patient in which the attitudes of two or more alters are considered together as different ways to deal, in the same traumatic interaction with the caregiver, with the needs for both attachment and self-protection. Two or more ‘alters’, so to speak, ‘sit together with the therapist’ and become able to consider their common origin.
In this therapeutic interaction lies the integrative power of psychotherapy in the DDS, common to different types of theoretical approaches. To it, some therapists add ‘fusion rituals’, often utilizing hypnosis, aimed at further facilitating the blending of alters in a unitary sense of self (Kluft, 1993). Others instruct patients to bring the alters together in their mind, in a sort of imagined group meeting, in order to develop ‘group thinking’ and ‘group feeling’ as a preliminary to a unitary sense of self (Fine, 1999). Outcome studies evaluating the specific advantages of these techniques are needed.
There is, however, an integrative flow inherent in the psychotherapeutic process. This follows from the point that, although the shifting mental and behavioral states are dissociated within the patient's mind, they are not dissociated within the therapist's mind. Thus, integration takes place first within the therapist, who hears about and observes many different aspects of the patient. The therapist can reflect upon (mentalize) these multiple experiences, behaviors, narratives, and affects—and, indeed, multiple transferences—and gradually communicate the emerging meanings and perspectives to the patient.
Over the course of treatment, a sense of self less encumbered by intrusions of traumatic memories and dissociative experiences is developed. Many patients, in the late phase of treatment, express the wish for a more thorough mastery of what they now understand as a tendency to dissociate in response to specific situations, e.g., attachment-related feelings of anger or anxiety. Cognitive techniques of journal-keeping, through which patients may more carefully assess the external contingencies, the emotions and the thoughts related to the tendency to dissociate, may be useful at this juncture. Patients may also benefit from repeating mentally simple verbal formulas, such as ‘I am here now’, as an instrument for keeping attention on ongoing experiences and thus countering the tendency to dissociate (Kennerley, 1996).
Mario had achieved many insights on the childhood, traumatic origins of his shifting ego states (besides having been brutally beaten by his father, he had been the victim of extreme neglect by both his parents). He had also achieved a good capacity for metacognitive monitoring of his tendency to enter into dissociative mental states. In an advanced phase of his treatment, he had been instructed by his therapist (G.M.) to register in a journal information concerning where he was, what he was feeling and what he was thinking every time he noticed in himself the tendency to dissociate. Here is a page of his journal:
Context in which the dissociative experience tend to emerge: ‘I am traveling by train, and I am alone in the compartment’.
Feelings: ‘Loneliness, anxiety, wish to have Anna (wife) here.’
Thoughts: ‘That curtain… the door for another world… I do not want to enter it… I'd rather stay here… maybe have a coffee…’
On that occasion, then, Mario had been able to resist the temptation to absorb his attention in the swinging rhythmical movements of the curtain, and to enter thereby in the trance-like state that was so easily accessible to him whenever he felt afraid, distressed, or lonely. It was in this state that most of his shifts between different ego states took place, as he had learned during psychotherapy. Keeping the journal was instrumental in reminding him of his decision—one that he had painfully reached during a long therapeutic work—to give up his dissociative defenses. The journal also made it increasingly clear that most of his dissociative tendencies emerged concurrently with his attachment wishes (e.g., in situations of loneliness, mental pain, threatened losses).
Cognitive-behavioral techniques of self-control and self-regulation may be used in the late phase of psychotherapy also for problems (e.g., sexual dysfunctions, abnormal eating patterns, addictions, and compulsions) that were not amenable to therapeutic influence before the resolution of core traumatic issues. The relational message implicit in the use of such techniques is that the therapist considers the patient as both potentially able and wholly entitled to take care of themselves. Before using these techniques, trauma-related pathogenic beliefs of not deserving care or of utter helplessness should therefore have been corrected. Patients should also have been able to modulate the emotions of guilt, shame, and abnormal dependency (abnormal anxiety at separations) accompanying such beliefs. The completion of such therapeutic accomplishments could be the task of the late phase.
Another task of the late phase is to clarify the difference between self-care and compulsive self-sufficiency. Issues of expected separation from the therapist may set the stage for completing the therapeutic work on this distinction. The prospect of ending the therapy may facilitate comments on normal emotional reactions to separation, on the difference between separation and loss, and on how to aspire, in other relationships, to the standards of secure attachment and mutual cooperation now experienced in the therapeutic relationship. The dramatic past relational experiences of dissociative patients may induce them to search for support to these reflections by testing the therapist's availability after having agreed upon ending treatment. Follow-up or booster sessions (e.g., once every 2–3 months), scheduled for two or more years, may be instrumental in passing this test successfully. Patients should, as the outcome of treatment, become able to rebalance their old relationship and to select/build up new ones according to the normal needs for relational safety, mutuality, and respect.
Most therapists agree that the more severe forms of DD require at least 3–5 years of intensive individual psychotherapy to reach these goals (International Society for the Study of Dissociation, 2000).
Dissociative patients are prone to harm themselves in various ways, and many of them are at risk of suicide. Some of them may also severely harm other people. These harming tendencies may be particularly difficult to deal with in the usual outpatient setting when they are menaced by persecutor alters emerging in the therapeutic dialogs. When hospitalization needs be considered to cope with these risks, it should be planned so that inpatient treatment aims at achieving specific goals of the psychotherapy. An instance is the planned processing of traumatic material (to clarify the meaning of the aggressive tendencies and regulate them), resorting to the protective hospital setting during such a planned exploration.
Particularly during trauma work, dissociative patients may feel utterly destabilized and disoriented by the surfacing memories and their meaning. Even when the risk of harming themselves or others in such moments of destabilization is not so serious as to require hospitalization, particular interventions should be considered that may assist in soothing the patients and reorienting them. Some therapists find hypnosis useful in this respect, as a context in which patients may more easily accept useful suggestions, e.g., to terminate spontaneous flashbacks and reorient themselves to present reality, or to momentarily ‘put to sleep’ a particularly troublesome alter.
When shame or family loyalties or overwhelming emotions prohibit trauma work in the individual setting, participation in group therapy involving other survivors of childhood abuse may be of great value (Buchele, 1993)—although some authors caution that this may provoke vicarious traumatization and provide suggestive stimuli for confabulated memories (Mollon, 2002a,b). While pharmacotherapy is not a primary treatment for the DDs, it may help in managing the destabilizing features of trauma work, or in dealing with comorbid mood or anxiety disorders. Group interventions and pharmacotherapy should preferably be the responsibility of a different clinician, with whom the individual psychotherapist keeps constant cooperative dialog. In this way, having a relationship with two different but communicating therapeutic attachment figures, the activation of the patients’ attachment system toward the individual psychotherapist becomes usually less intense and more easily manageable (because of the concurrent lessening of the dissociating influences of the disorganized IWM).
Another instance in which the individual psychotherapists may usefully cooperate with another clinician is when a family therapist is consulted (e.g., because of risk of repeated abuse in the patients’ new family, or because of sexual problems in the patient's conjugal couple: Porter et al., 1993). The cooperation between two different therapists in the treatment of particularly difficult cases of dissociative pathology is so potentially useful that even the simultaneity of two individual psychotherapies has been advocated (Wine and Carter, 1999).
Therapists of patients reporting memories of childhood intrafamilial abuse should also be aware of two major legal problems that can complicate the treatment: the risk of being accused by family members of having induced false memories and the possible request by the patient to be supported in suing an abusive family member. Every care should be taken in order to avoid the risk of inducing false memories, or of sanctioning any memory as certainly factual rather than reconstructive and therefore potentially fallible. As to actively assisting patients in taking legal actions against family members, this possibility is appropriate only when the adult patient's present safety is threatened, i.e., in the face of ongoing abuse. In all other cases, the patient should be aware that therapy's task is concerned with meaning and the exploration of memory, feeling, thought, intention, wish, or behavior pattern, but not with assistance in (nor dissuasion from) claiming legal acknowledgment of any past injustice the patient may have suffered.
Finally, it cannot be emphasized enough that psychotherapeutic work with dissociative patients, while frequently rewarding, can be extremely difficult and hazardous. Perhaps more than with any other patients, considerable clinical skill and experience are required—to understand and manage such aspects as the complexity of the presented material, the shifting self states, the enactments in the consulting room, the multiple transferences, the extent of anxiety and overwhelming affect held within the dissociative structure, and the profound ambivalence that the patient will feel about allowing the therapist access to the secrets of their internal world. Although the hope may be that the patient can surrender the dissociative mode of being and achieve greater integration, the reality in some cases may be that instead of improving through therapy he or she deteriorates. Instead of gratitude for hard therapeutic work, the patient feels hatred for the therapist who has undone the dissociative defenses against unbearable annihilatory pain—and now it may be the therapist that has to be both clung to and annihilated. This is the danger of malignant regression. It is not exclusive to DDs, but is a hazard in all cases where a patient has experienced severe emotional deprivation and interpersonal trauma in childhood; the combination of rage, envy, and intense need may mean that the therapist becomes ultimately the patient's victim. In a state of manic triumph, the ‘bad’ or vulnerable self is projectively located in the therapist—and there it is condemned and persecuted.
A young female clinical psychologist, working in relative isolation from colleagues with more psychotherapeutic experience, found herself working with a woman with DID. She felt fortunate to have such an interesting case, finding the patient's shifting self states fascinating, and soon the therapy sessions were becoming more and more frequent and longer in duration—these and other boundary violations being rationalized as adaptation to what she viewed as the patient's obvious need for ‘special’ conditions. Child alter states began to emerge and the psychologist allowed the patient, in these states, to sit on her knee cuddling her for long periods. She felt shock and deep compassion at the narratives of severe childhood abuse—and felt she must work extra hard for such a damaged and deserving patient. However, the psychologist became increasingly alarmed when angry and very demanding states were presented. As she attempted to withdraw and limit her ‘therapeutic’ involvement, the patient became increasingly agitated, alternating between threatening and pathetic pleading modes. The psychologist experienced rage, fear, and bewilderment. She felt guilty about her own hatred towards the patient, resulting in attempts to compensate by trying even harder to meet the patient's ‘needs’—thus alternating between being overly gratifying and rejecting. Steadily the psychologist became more and more exhausted—and her judgment increasingly impaired. She was herself suffering traumatic stress, as a result both of hearing terrible narratives of childhood abuse, and also through the enactments in their interaction. Her feelings of shame at the unorthodox position she found herself in, with many deviations from the normal boundaries of therapy, meant that she did not feel able to seek supervisory consultation. The patient assumed greater and greater power over the psychologist, even succeeding in getting the latter to agree to social meetings. Four years after the therapy began, the patient made an official complaint about the psychologist, alleging that that she had encouraged the development of a multiple personality and made her worse as a result of malpractice.
On the other hand, with appropriate caution, modesty of therapeutic aims, and continual attention to pacing of the work, in such a way that the patient can feel some degree of control over the process while also experiencing the reassurance of a secure frame, the psychotherapist may enjoy the privilege and grace of witnessing moments of true healing.
Individual outpatient psychotherapy is the treatment of choice for DDs. The most commonly cited treatment orientation is psychodynamic-eclectic, focused on relational themes, and often incorporating cognitive therapy techniques and/or hypnosis (for soothing and containment, not for abreaction of traumatic memories). Pharmacotherapy, group, and family interventions may be of great but ancillary value with respect to the individual therapy. It is essential to establish the therapeutic alliance and to fortify the patient before working on traumatic memories. Trauma work, in turn, is necessary to achieve integration. Although conclusive evidence is still lacking, integration seems both a possibility and a necessity for the successful treatment of these disorders.