The Treatment of Traumatic Memories:
Synthesis, Realization, & Integration

Onno van der Hart, Ph.D., Kathy Steele, R.N., M.N., C.S.
Suzette Boon, Ph.D. & Paul Brown, M.D.

Originally published in Dissociation, 1993, 6(2/3), 162-180. Posted at with permission of the first author (OvdH) and Editor (R. Kluft). Note that this online version may have minor differences from the published version.
Acknowledgement: The authors thank Barbara Friedman, M.A., M.F.C.C., for her helpful comments on an earlier draft of this article.
Onno van der Hart, Ph.D., is a professor in the Department of Clinical Psychology, Utrecht University, The Netherlands, and a psychologist at the Regional Institute for Ambulatory Mental Health Care Amsterdam South/New West, Amsterdam, The Netherlands. Kathy Steele, R.N., M.N., C.S., is a psychotherapist with the Metropolitan Psychotherapy Associates, Atlanta, Georgia, and is Clinical Administrator of the Post-traumatic Stress and Dissociative Disorders Program at Parkwood Hospital, Atlanta, Georgia. Suzette Boon, Ph.D., is a research fellow in the Department of Psychiatry, Free University, Amsterdam, The Netherlands, and a clinical psychologist at the Regional Institute for Ambulatory Mental Health Care Amsterdam South/New West, Amsterdam, The Netherlands. Paul Brown, M.D., is a psychiatrist at the Northpark Hospital Centre, Bundoora, Melbourne, Victoria, Australia.
Correspondence: Onno van der Hart, Department of Clinical Psychology, Utrecht University, Heidelberg 1, 3584 CS Utrecht, The Netherlands. Phone: +31(30)253-1785; fax +31(30)253-4718; Email:


This paper is based on Pierre Janet's dissociation theory and his concept of the non-realization of a traumatic event. A model of treatment that integrates Janet's dissociation-integration theory with contemporary trauma-based models of therapy is delineated. The nature of traumatic memories is described, and a stage-oriented model for their treatment in patients with multiple personality disorder (MPD) is presented. Ideally a discrete phase in the overall treatment of MPD, this phase can itself be subdivided into the following stages: (1) preparation; (2) synthesis; and (3) realization/integration. Although a number of treatment recommendations are offered, the emphasis here is more on clarifying concepts than on the description of techniques.

Multiple Personality Disorder (MPD) usually results from severe and repeated traumatization in childhood (Putnam, Guroff, Silberman, Barban, & Post, 1986; Ross, Norton, & Wozney, 1989; Schultz, Braun, & Kluft, 1989). The core pathogenic problem, i.e., the presence and disrupted impact of dissociated traumatic memories, is often independent of the form or complications of the MPD. Experienced therapists concur that this disorder cannot be completely resolved until these traumatic memories have been successfully processed. Modern clinicians working with MPD have developed effective treatment strategies to achieve this (e.g.. Fine, 1991; Putnam, 1989,1992; Ross, 1989; Steele & Colrain, 1990), but in our opinion the conceptual language in which these strategies are couched is inadequate. In two related papers, some of the authors drew attention to the uncritical use of the term "abreaction," and they argued in favor of the concept of dissociation over repression in describing and explaining the origins and maintenance of traumatic memories (Van der Hart, & Brown, 1992; Van der Hart, Steele, & Brown, manuscript in preparation). Recently, Peterson also underlined the need to substitute more appropriate concepts for the term "abreaction" (Peterson, 1993). Subsequently, in their treatment manual on processing traumatic memory, Sachs and Peterson (1994) only retained the term "spontaneous abreaction" as a synonym for flashback, and otherwise used the term "memory processing" when referring to treatment of traumatic memory.

The purpose of this paper is to develop a phase-oriented treatment model based on Janet's concepts of dissociation and the "non-realization" of traumatic memories. Within this model a number of existing treatment techniques are comprehensively integrated. Since other publicationso in particular Steele and Colrain (1990) o already provide excellent practical guidelines, the present emphasis is on conceptual aspects of the treatment model.

The authors first describe MPD, in Janet's terms, as a "disorder of non-realization." They then explore the nature of traumatic memories, describing their various forms, and the dissociative reactions which they may evoke. Next, they describe the treatment of traumatic memories as a distinct core phase within the overall treatment of MPD. This phase can be clinically and theoretically divided into the following stages: (1) preparation stage; (2) synthesis stage; and (3) realization/integration stage. Finally, they discuss "synthesis" of traumatic memories versus expression of emotions; the substitution of traumatic memories with positive mental images; and finally "realization" as a social act, as a continuous mental operation, and as the beginning of integration.

MPD as a Disorder of Non-Realization

Janet (1904, 1919/25) observed that traumatized individuals appear to have had the evolution of their lives arrested.

They are "attached" to an obstacle which they cannot go beyond. The happening we describe as traumatic has brought about a situation to which the individual ought to react.

Adaptation is required, and adaptation is achieved by modifying the outer world and by modifying oneself. Now, what characterizes these "attached" patients is that they have not succeeded in liquidating the difficult situation (Janet, 1919/25, p. 660).

Janet (1935, 1945) added that these traumatized individuals have not "realized" the traumatic event. Thus, realization requires putting the event into words, relating it as a narrative, and reconciling the experience within the personality, thereby restoring continuity to the individual's personal history. Non-realization of the trauma can exist to varying degrees. Many traumatized people have a subliminal awareness of their traumatization, but cannot bear to put it into words. They tend to evade all references to the event. If they are confronted with it, they become highly anxious, a phenomenon which Janet (1904) called a phobia for the [traumatic] memory, and which Van der Kolk (1988) regarded as an inability to tolerate the feelings associated with the trauma. Their anxiety is, in fact, an act of separating themselves from the traumatic memory, a flight from the act of realization (Janet, 1945). Complete non-realization is found in those patients who

do not make any allusion anymore to the verbal formula which expresses the suppressed event, nor to the anxiety which stops them: for them, the event ... seems never to have occurred. (Janet, 1905, p. 184)

Janet called this the hysterical form of non-realization, which was most clearly demonstrated in MPD patients (Janet, 1910). In a comparable contemporary vein, Ross (1989, p. 59) characterized MPD as occurring when a little girl who is abused imagines it happens to somebody else. As an adaptation to continuing abuse, she creates alter personalities who alone suffer the abuse. Non-realization is complete in this child in the sense that for her, the abuse seems not to exist at all. For her traumatized alter personalities, the experience of the traumatic events continues to exist as dissociated "traumatic memories" and, as such, these too are also not realized.

A central focus in the treatment of traumatized patients is the processing of their traumatic memories. This enables them to overcome their phobia for, and avoidance of these traumatic memories, to reverse the dissociation of these memories, to realize the distressing experiences, and to integrate them into the whole of their personality. The total process of therapy, moving as it does inexorably towards integration may, thus, also be described in terms of increasing realization.

The Nature of Traumatic Memories

One of the central features of traumatized patients, such as those suffering from MPD, is that their traumatic memories continue to exist as highly distressing dissociated psychophysiological states. These traumatic memories are states in which the traumatic events are partially or completely, re-experienced. In the clinical literature on PTSD they are regarded as forms of intrusive recall (e.g., Blank, 1985), while in the MPD literature, they are often referred to as "spontaneous abreactions." The latter expression was the subject of the authors' recent critique (Van der Hart & Brown, 1992), as it is based on the Freudian hydraulic model of pent-up emotions.

However denoted, traumatic memories are part of experiential recall, and must be distinguished from ordinary or narrative memory (Van der Kolk &; Van der Hart, 1991). Traumatic memory is characterized by a sense of timelessness and immutability, has no social function, and is reactivated by trigger stimuli. It also follows the principle of restitutio ad integrum, the tendency of all elements of the trauma to be reactivated, once one element is called into play (Janet, 1907,1928).

The term "traumatic memories" should also be introduced in the clinical literature on MPD to describe those states which are partial or complete re-experiences of the trauma. Because of its lack of specificity, Jung's (1921/2) alternative expression "complex" is much less adequate.

When only in the dispositional or dormant state, traumatic memories are not usually a hindrance. When reactivated, they then exert their harmful influence. Like "parasites in the mind," they overgrow consciousness (Charcot, 1887; Janet, 1894). Traumatic memories never exist as wholly isolated states. They are always "part of some personality" (McDougall, 1926, p. 543). In describing acutely traumatized combat soldiers, Myers (1940) thus titled their traumatized state the "emotional personality." He observed that this state went on to alternate with the "apparently normal personality." He spoke about the "apparently normal," because the traumatic memory could still manifest itself indirectly, for instance through intrusive somatic dissociative phenomena such as motor paralyses.

In MPD, this doubling of the personality is more complex. During actual trauma itself, through the process of dissociation, a number of alter personalities sequentially experience different aspects of the trauma. The authors prefer to call this process linear dissociation (as proposed by B. Friedman, personal communication, (November, 1993). Alter personalities may also contain a range of different (non-successive) dimensions of the traumatic event, in which case the authors speak of simultaneous dissociation. Janet (1907) distinguished the visual, auditory and kinesthetic dimensions, among others. Braun (1988a&b) systematized these dimensions in his BASK model. BASK is an acronym for behavior, affect, sensation, and knowledge (in which we would also include the cognitions of the individual at the time of the trauma). For many MPD patients, traumatic memories can be manifested both in a linear and a simultaneous manner.

It is now widely recognized among clinicians and researchers in the field of post-traumatic stress, as it was by Janet before them, that traumatic memory is reactivated under particular conditions, specifically in situations which are reminiscent of the original traumatic situation (Courtois, 1988; Gelinas, 1983; Steele & Colrain, 1990; Solomon, Garb, Bleich & Grupper, 1987; Van der Hart & Friedman, 1992; Van der Kolk, 1994). These situations are called triggers or reactivating stimuli (Van der Hart, Boon, Friedman & Mierop, 1992). For practical purposes, the following types of reactivating stimuli may be distinguished: (1) sensory data; (2) time-related stimuli; (3) daily life events; (4) events during the therapeutic session; (5) emotional states; (6) physiological states; (7) stimuli recalling intimidations by perpetrators; and (8) current trauma (Van der Hart & Friedman, 1992; Van der Hart et al., 1992).

Blank (1985) distinguished four types of reactivated traumatic memories in Vietnam combat veterans with PTSD: (1) vivid dreams and nightmares of the traumatic event; (2) dreams from which the dreamer awakens, and then has difficulty shaking off the sense of the recollected traumata, possibly even manifesting motor behavior related to the dream content; (3) conscious flashbacks, in which the subject vividly experiences images of traumatic events; and (4) unconscious flashbacks, wherein the survivor has a sudden, discrete experience that leads to actions, and where the manifest psychic content is only indirectly related to the trauma. Furthermore, in unconscious flashbacks the survivor does not have a conscious awareness of reliving traumatic events, either at the time of the flashback or later. Amnesia can follow both conscious and unconscious reminiscences.

In MPD the re-experiencing of a trauma by alter personalities may, in some cases, involve complete loss of contact with present reality as well as amnesia for both the occurrence and its contents. The authors therefore suggest the following distinction between two forms of reactivated traumatic memories: (1) partial re-experiences, and (2) complete re-experiences. The former concern the reactivation and re-experience of the trauma, while the executive personality and its habitual consciousness also remains, to a certain degree, in touch with the current situation. By contrast, during a complete re-experience, the trauma becomes the total current reality for the dominant personality.

Partial re-experiences are reflexive, incomplete, and fragmentary, with much of the content occurring outside the awareness of the executive personality (Steele & Colrain, 1990; Loewenstein, 1991). Any dimension of the trauma can be reactivated and "relived." During a flashback, the trauma survivor is usually still relatively aware of his or her current surroundings and situation. There is a kind of doubling of consciousness, i.e., of reliving the past (to which the person also begins to respond) while at the same time partially staying in the present. The dissonance occurring between these parallel experiences can induce survivors to feel "crazy."

An important distinction is made in the literature between conscious flashbacks and unconscious flashbacks (Blank, 1985). During a conscious flashback, there is usually a strong visual component of the traumatic memory, implying that there is a cognitive dimension to the re-experienced trauma. Trauma survivors are also partially aware of their present circumstances, and usually remember the contents of the flashback afterwards. This combined awareness of both the trauma and the present situation can give rise to a confused sense of time. Although both the trauma and the present situation are experienced in the present, the current circumstances may be experienced in a more depersonalized manner.

It can also occur that the executive personality is more aware of other sensory dimensions of the trauma than the visual one. Janet (1898) described in detail examples pertaining only to hearing certain sounds such as screaming, to certain physical sensations such as pain, and to emotions such as fear and anger. These elements were dissociated from the rest of the traumatic memory, and may be re-experienced at a subconscious level by other alter personalities. The executive personality may then be unable to comprehend what s/he is (re) experiencing.

As previously mentioned, Blank (1985) distinguished the so-called unconscious flashback. In this the trauma survivor has a sudden, discrete experience leading to actions, where the manifest psychic content is only indirectly related to the original trauma. The survivor does not have conscious awareness of reliving the trauma, either at the time of the flashback or later (Blank, 1985, p. 297). The authors regard this phenomenon as being due to the dominance of the behavioral dimension of the traumatic memory, with other (e.g., cognitive) dimensions dissociated from the executive personality.

Complete or near complete re-experiences were described in the classical literature as hysterical attacks, or as somnambulistic crises (Briquet, 1859; Charcot, 1887; Janet, 1889, 1907). Recently, Loewenstein (1991, 1993) spoke of major dissociative reenactments. During these, the executive personality is completely disoriented, and experiences him or herself back in the original traumatic situation without any grounding in present reality. When the behavioral dimension is also involved, one usually speaks of a behavioral re-enactment (e.g., Van der Kolk, 1987b). Patients may be completely oblivious to the current situation and circumstances, such as the presence of persons in the same room. It is also possible that these persons are perceived, but only in the traumatic context. For example, when a rape is re-experienced during a therapy session, the therapist may be perceived as the rapist. In complete re-experiences, the patient has amnesia afterwards, not only for the contents of the experience but often for the fact that anything unusual occurred at all. Instead, the subject experiences a puzzling loss of time or "amnesia for amnesia" (Kluft, 1987).

The partners of some MPD patients report observing them having had severe nightmares, during which a trauma was apparently re-experienced. In such cases, alters gain executive control and completely relive the traumatic experience, sometimes with an intense behavioral re-enactments.

Dissociative Reactions to Reactivated Traumatic Memories

Re-experiencing a trauma can also be a traumatic experience in its own right. Just as in the original acute trauma, further dissociative responses may occur. Depersonalization, for example, not only can take place during or directly after traffic accidents (Noyes, Hoenck, & Kupperman, 1977), but also during or after a reactivated traumatic memory. The authors distinguish two broad categories of dissociative responses to reactivated traumatic memories:

1. Brief Reactive Dissociative disorder. Sometimes the dissociative reaction is so intense, that it can be compared to acute stress disorder, the diagnostic category proposed for inclusion in DSM-IV (Spiegel & Cardena, 1991). It occurs during or directly after a distressing event, and is characterized by one or more of the following intense and often persisting symptoms: (1) stupor; (2) derealization; (3) depersonalization; (4) perceptual distortions; (5) subjective sense of numbing; (6) amnesia following the distressing event; (7) sudden experience of terror, or other signs of physiological arousal.

2. Switching to Alters. For MPD patients, the distressing event has usually been experienced by several alter personalities. During a reactivated traumatic memory the personalities can alternate again. In response to this reactivation, it is also possible for switches to occur either to alters who know only of the trauma, or to those who are completely ignorant of it.

This switching may occur very rapidly. Many agitated alters, especially child alters, alternate rapidly with one another without being able to reduce the attendant terror. Since the primary psychological function of dissociation is to avoid re-experiencing traumatic elements, it would appear that rapid switching defeats the original defensive purpose. It instead generates intense dysphoria and dysfunction, and promotes further dissociation in reaction to the dysphoric rapid switching process itself.

In summary, traumatic memories are sensorimotor states that may be partially or completely re-experienced by the individual. In addition they may be conscious or unconscious. They may remain temporarily in a quiescent state, but are often reactivated by stimuli that resemble some aspect of the trauma, either literally or symbolically. The evocation of one dimension of dissociated memory tends to reactivate all dissociated components, and reactivated traumatic memories may themselves re-traumatize the patient.

A Phase Model for the Treatment of Traumatic Memories

The introduction of phase-oriented treatment models for traumatically-induced disorders, dissociative and non-dissociative, is widespread, and the treatment of traumatic memories per se is usually seen as a distinct phase. In Janet's original treatment approach to post-traumatic stress, three phases could be distinguished (Van der Hart, Brown & Van der Kolk, 1989): (1) containment, stabilization, and symptom reduction; (2) modification of traumatic memories; and (3) personality integration and rehabilitation. Janet's stage model is very similar to modern models of treatment for post-traumatic stress disorder (PTSD) and dissociative disorders such as MPD (Boon & Van der Hart, 1991; Braun, 1986; Brende, 1984; Brown & Fromm, 1986; Herman, 1992; Horowitz, 1986; Parson, 1984; Kluft, 1987; Putnam, 1989; Ross, 1989; Sachs, Frischholz & Wood, 1988; Sakheim, Hess & Chivas, 1988). For example, Sachs et al. (1988) identified five phases in the treatment of MPD: (1) making and sharing the diagnosis; (2) identifying the various personality states, and understanding their purpose and function; (3) sharing with the therapist, and with each personality state, the specific traumata associated with those personality states; (4) integrating the various personality states into a single functional whole; and (5) learning new coping skills that enable functioning of the unified personality, and prevent future personality splitting.

In this article, the prime focus is on Janet's stage two, the modification of traumatic memories, which corresponds to phase three in the MPD treatment model of Sachs et al. (1983). Their definition of this phase as the sharing of specific traumata with the therapist and other personality states is well chosen. Successful treatment of traumatic memories involves relinquishing related dissociations. This means that alters both involved in the traumatic experience and independent of it must share with each other their respective post-traumatic dissociations. The authors call this communication the synthesis of the traumatic memory. In the older literature, the term "re-synthesis" was used (cf. Brown, 1920/1; Myers, 1920/1). However, this term incorrectly implied that the traumatic memory had not been dissociated. This is in contrast to the authors' view that dissociation is an immediate defensive reaction to trauma.

The purpose of synthesis is to alleviate the dissociation by helping alters to share their respective traumatic experiences of the trauma both with each other, and with the therapist. This orientation is, at the very least, semantically different from those approaches which encourage abreaction and catharsis of emotions. The authors consider "abreaction" to be an incorrect explanatory principle, and an inappropriate treatment technique (cf. Van der Hart & Brown, 1992).

However, the present approach does not stand for blanket discouragement of the expression of emotions. On the contrary. When the patient needs to express the cry which she suppressed during the actual abuse, she is told that she can now cry safely, and she is encouraged to do so. This expression of emotion leads to realization and completion. Abreaction or uncontrolled catharsis of overwhelming traumatic affects leads to states of hyperarousal and, at times, to complete psychological decompensation. The hydraulic view of affect ("It is in there, and needs to get out!") contrasts with the Janetian notion that the affect, as well as the other elements of the trauma, needs to be realized and integrated or "metabolized."

Therapeutic synthesis of traumatic memories is a purposeful mental action, based on a conscious decision either by the whole personality, or by the alters involved in the traumatic experience. For this endeavor to succeed the patient, or the alters involved, needs the complete support of the therapist, not only through his or her sharing in the traumatic experience (as emphasized by Sachs et al., 1988), but also through encouragement and guidance

When synthesis of a trauma occurs, the memory ceases to be re-experienced in a traumatic way. The trauma becomes a "neutral" historical fact. In order to achieve this, the patient, or the group of alters concerned, undergoes a process of "realization" of the personal disaster which has occurred. The patient substitutes and integrates into his or her life history or autobiography, a narrative memory of the event in place of the experiential traumatic reminiscences (Janet, 1919/25), i.e., into the whole of the personality. This integrative mental work is both a time and an energy-consuming reorganization of the personality.

The treatment phase of traumatic memories can be divided into the following stages, each of which will be described below: (1) preparation stage, (2) synthesis stage, and (3) realization/integration stage.

Stage 1: Preparation

The treatment of traumatic memories can be a stormy process and so must be planned carefully. The preliminary groundwork follows the general guidelines for establishing and maintaining a secure therapeutic setting, with particular attention paid to boundaries, consistency, and predictability (Kluft, 1991,1993; Courtois, 1991; Fine, 1991,1993; Putnam, 1992; Steele, 1989; Steele & Colrain, 1990). In addition, in working with victims of childhood trauma and with dissociative disorders, the following strategies should be employed: (1) identification of alters and their respective conflicts; (2) development of a reasonable degree of internal communication and cooperation among alters; (3)containment of traumatic intrusion phenomena; (4) development of ego strength in the host personality, as well as in the various alters; and (5) development of self-care capacities, i.e., capacities to tolerate affects, to self-soothe, to tolerate aloneness, and to regulate self-loathing (McCann & Pearlman, 1990); and (6) the introduction and use of hypnosis. The latter is for the purpose of enabling the patient to develop control over dissociation, for containment and affect modulation, for creation of a safe psychological space, internal communication, symptom relief and/or substitution, and for management of other dissociative crises.

There are two important caveats regarding the application of hypnosis: (1) It must be implemented within the context of well-conducted psychotherapy and therefore should be used only by therapists with specialized training; and (2) particularly in the later phase of treatment, care should be taken to prevent the defensive use of trance to avoid painful issues. During the earlier synthesis phase, however, hypnosis is particularly useful.

Once treatment of traumatic memories has been initiated, the stage is set to prepare for the synthesis and realization/integration of traumatic memories. The preparation phase addresses: (1) safety factors; (2) issues of controlling or containing reactivated traumatic memories; (3) exploration of traumatic memories; (4) correcting cognitive errors and distortions; and (5) careful explanation and planning of synthesis sessions (Steele & Colrain, 1990).

Issues of Safety

Steele and Colrain (1990) presented an excellent overview of the preparations required to ensure safety both during and between synthesis sessions. Preparation pertains to: (1) intrapsychic safety; (2) interpersonal safety, both of the patient and the therapist; and (3) a safe therapeutic environment. The latter is best provided in a minority of cases by a short term hospital stay. Additional safety issues which should also receive careful consideration include: (1) the degree of cooperation and communication among alters; (2) the general level of hyperarousal and hypersensitivity to affective stimulation, as well as the level of affective maturity; (3) the potential for suicide and self-harm; (4) the general level of coping of the individual and the alters involved in maintaining function in the world (i.e., will these alters participate in the integrative memory work, and what effect will that have on functioning?); (5) the range of support systems and resources available; and (6) the level of motivation to work through traumatic memories.

Controlling or containing reactivated traumatic memories

Traumatic memories do not usually emerge in a planned or controlled manner. They are often triggered by a reactivating stimulus. When the timing is not right for their synthesis, either in the overall course of treatment, or because the particular therapeutic session is too far advanced, these re-experiencing phenomena should be controlled or contained using any of the following techniques either alone or, still better, in combination: (1) identifying the reactivating stimulus and, when indicated, relating it to the original trauma; (2) removing the reactivating stimulus; (3) neutralizing the reactivating stimulus (involves teaching the patient to regard the trigger as belonging to two different contexts, the present neutral one and the past traumatic one, and helping the patient realize that the object itself holds no emotional charge); (4) promoting and emphasizing the safe present; (5) suggesting temporary amnesia for, or creating more mental distance from, the traumatic memory [for example, hypnotic interventions such as the movie screen technique, the use of adult observers, and the use of affective regulation techniques as described by Fine (1991) and Kluft (1989)]; (6) cue-conditioning to help the patient return to the safe present (the patient can be trained to respond to a certain cue in returning to the present if he or she should get lost whilst reliving a traumatic memory); (7) screening traumatized alter personalities from the outside world; (8) introducing a co- or deputy therapist; and (9) medication and hospitalization as crisis intervention (Van der Hart & Friedman, 1992; Van der Hart et al., 1992).

Exploration of the Traumatic Memory

Either parallel to, or following safety preparation, the therapist begins to collect information about the content of the traumatic memories.

1) Charting the Context of the Trauma. The authors first recommend defining the historical context, including the personal history (development, school history, training and occupation, supportive figures, pets, objects and other attachment objects, etc.), and the family circumstances (social, cultural and economic factors); domestic factors (e.g., who lived in the home, who shared bedrooms, family rules and discipline) (Steele & Colrain, 1990).

2) Completing the Story of the Trauma. Although most alters involved cannot, by definition, relate a narrative of the unassimilated trauma, there are often some alters who have a cognitive awareness without any associated affect, and who will be able to relate what happened. Often this is an observing alter who watched happened "from a distance." In completing the story, it is important to collect information from the whole sequence, including directly before the trauma, during the trauma itself, and directly thereafter. This not only provides a context for the trauma, but also ensures temporal continuity.

When an observing alter is asked to relate what happened, care is taken to inquire if the traumatized alters can be protected from hearing the story. In this way, untimely reactivation of the whole traumatic memory can be prevented.

3) Identifying the Alters and Their Role in the Trauma. It is preferable to know prior to the actual synthesis which alters are involved, and what are their functions in the traumatic experience (Sachs, Braun, & Shepp, 1988). Ignorance of which alters are involved, and which should participate, may allow some alters to withdraw and refrain from sharing their experiences with the others. Dissociated aspects of the traumatic memory will then continue to exist, and can eventually be reactivated. Braun's (1988a&b) BASK Model may serve as a checklist to determine whether all relevant dimensions of the trauma are covered by the alters known to the therapist.

The purpose of this inquiry is therefore not acquisition of details for their own sake, but rather to become aware of all of the dissociated elements, or "pathogenic kernels" (Van der Hart & Op den Velde, 1991), which must be synthesized in order to neutralize the traumatic memory. Often the personal meaning in these "pathogenic kernels" is more important than the objective factual details. Personal or subjective meaning of the trauma is most often experienced in the form of existential crises. Pathogenic kernels frequently develop in response to, and contain in an undiluted and unassimilated form, intense existential crises that are related to the trauma. Such crises are partially related to "stuck points" created by discrepancies between prior belief schemes and new schemata generated by the traumatic experience (Resick & Schnicke, 1992). These existential issues will be discussed further in a later section of this paper, as well as specific approaches to the synthesis and integration of pathogenic kernels.

Apart from knowing which alters were involved in the trauma, and which should therefore participate in the synthesis session, the therapist should discover which of the other alters can remain in the session as well, and which of them should be protected from it, e.g., by encouraging withdrawal behind an inner, dissociative "wall." It is also important to ascertain whether there are alters who still actively object to the memory work. If so, these should be engaged in negotiation to prevent internal punishment for premature revelation of trauma-related "secrets."

Correcting Cognitive Distortions

This phase of therapy also enables the patient to recognize and confront any cognitive distortions or errors that create dysfunction and misperception of the self and others (Fine, 1988, 1990, 1991, 1993; Fish-Murray, Kolb, & Van der Kolk, 1987; Janoff-Bulman, 1992; McCann & Pearlman, 1990; Orzek, 1985; Ross, 1989; Ross & Gahan, 1988). Cognitive schemas that were developed in environments of chronic abuse and neglect are inevitably negative, with little associated flexibility or adaptive personality functioning. They remain fixed along with the traumatic memories. The latter are reinforced by intense post-traumatic affects. The negative cognitions, in their turn, reinforce these negative affects. During synthesis the cognitions also become available for therapeutic exploration and change.

The trauma is encoded within specific cognitive schema, and continues to be interpreted within this context by the patient until the therapist assists in the reconstruction of new assumptions, beliefs, and expectations about the self and others. Cognitive distortions frequently pose obstacles to initial memory work, and contacting the alters involved before the actual synthesis helps the therapist to motivate them for this intense and painful work. Resistances often emanate from deeply embedded beliefs about the trauma, e.g., "It lasted forever"; about the perpetrator, e.g., "He'll know if I ever tell"; and about the self, e.g., "It was my fault," or "It didn't happen to me."

Thus, in a case of sadistic sexual abuse by the father, a female patient's "father-alter" had first to acknowledge that this was the part other personality which, by identifying with the abuser, had helped her to survive the trauma. At the time, her defensive dissociation helped her to believe that the intolerable abuse did not occur to her but to somebody else. Another alter, in the same person, had to accept in treatment that she was not really crazy. She believed that expressing her anger directly against her father would have made him abuse her even more.

Assisting the patient, and alters, to change inappropriate and distorted cognitions is an approach which may be required through out the whole course of therapy (Fine, 1988, 1990,1991; Ross & Gahan, 1988; McCann & Permian, 1990). Fine has extensively delineated the necessity of cognitive work throughout the course of MPD treatment, and has utilized Beck's (1979) cognitive therapy approach (Fine, 1990, 1991, 1993). Many cognitive errors and distortions may be accessed prior to synthesis, but a number emerge in the course of eradicating the post-traumatic dissociation. These distortions must be identified and processed following synthesis. Others even emerge after the synthesis, when integration is taking place.

It must be realized that such cognitive distortions have created a gestalt of schema that comprise the individual's entire worldview, a view that has been held for a lifetime. The survivor's defenses are based on these cognitions, and usually have been incorporated into the entire character structure itself. Change on such a scale of magnitude progresses slowly, yet such change is necessary if the survivor is to heal. New, adaptive cognitions must be reinforced over and over in the course of therapy in order to completely replace old, negative injunctions (Steele & Colrain, 1990).

Explaining and Planning the Synthesis Session

During the preparation phase, it is necessary to give a clear explanation of the various steps and procedures to be followed, and to provide a cognitive framework for the assimilation of the traumatic memories.

A global explanation will have been given earlier in therapy. Now it is augmented in a detailed manner. It must be repeated a number of times, initially with the host personality (or the alter who is most active in therapy), followed the observing alters who are able to provide information about the trauma, and then the other alters who are involved in the traumatic memories. Sometimes the observing alter will be the one to inform these latter alters.

Prior to synthesis, the possibility of using artwork or other non-verbal therapeutic modalities is explored. For example, severe abuse which is dominated by intense physical pain destroys language and impedes verbal communication (Scarry, 1985). Thus, initial use of non-verbal modalities in severely traumatized individuals may facilitate verbal communication. Symbolic images may also provide a framework and a "container," perhaps for the first time, for the traumatic experiences of all the alters.

The rationale provided for this treatment approach is that reactivated traumatic memories continue to exert their intrusive and disturbing effects because of their dissociative nature. They have yet to become shared and integrated parts of the alters involved, as well as of the whole of the personality. To achieve this, the alters involved with the trauma share their respective part-experiences with each other, unifying them into one whole, and sharing this eventually with the remaining alters. The therapist emphasizes that the success of the procedure is dependent on the completeness of the synthesis. This implies that all dissociated "pathogenic kernels" must be synthesized. To share the trauma with each other, the alters need to experience it intentionally, while also staying in the safe present, and keeping contact with the therapist.

In many instances, all of the alters involved may accomplish the relief of the dissociation, and the synthesis of the traumatic memory, in one session. This is called parallel synthesis. With greater severity and duration of the trauma, resolution may take a number sessions, which is called serial synthesis. Whether synthesis is accomplished in one or in several sessions, it is crucial for the patient to pursue his or her own optimal pace. If two or more sessions are needed, great care should be taken that the traumatic memories are well contained between the sessions.

Stage 2: Synthesis

Synthesis of a traumatic memory involves its controlled reactivation in a collaborative effort by the therapist and the patient. The involved alters re-experience and share their respective parts of the trauma with each other, and with other alters. In this way the existing dissociations are relieved.

A century ago Delboeuf formulated this reactivation as follows: Using hypnosis the therapist "puts the subject back into the state in which his trouble manifested itself, and combats the same trouble with spoken words, but in a state of rebirth" (1889, p. 71; cf. Macmillan, 1991, p. 82). The "therapist's words" help the patient to stay in touch with the present, and with the therapist, while the patient is, and the alters involved are, returning to the traumatic experience. All dissociated aspects of this experience must be re-experienced. The memory must eventually be shared with every part of the personality. That the presence of the therapist is an essential factor in this process was aptly formulated by Jung (1921/2):

The support and understanding of the therapist raise the level of the patient's consciousness and enable him to consciously bring the autonomous traumatic memory once more under the control of his will. (p. 15).

While (re) experiencing the trauma, the patient enters an altered state of consciousness, and loses the sense of chronological time. There is no sense of past or present relative to the trauma, and so the trauma seems timeless. Synthesis must counteract this and add a sense of continuity and finitude (Steele & Colrain, 1990). Using details collected during the preparation stage, the therapist helps the patient to evolve from the period directly before the trauma, through the trauma itself, to the period directly after it. The patient thereby experiences a beginning and an end to the trauma (Putnam, 1989).

In addition to ensuring the chronological context, it is critical to help the patient recover the subjective experience of the trauma, which is often held in the various alters. This subjective experience includes a number of dimensions: the various meanings given to the trauma (e.g., "this can't happen," "this has never happened to anyone else," "this is too terrible to ever tell," etc.); internalized injunctions that are either implied or verbalized (e.g., "if you tell, I'll kill you," "you know you like this," etc.); the various BASK dimensions including cognitions, and existential crises (Steele & Colrain, 1990).

Existential Crises

Yalom (1980) saw an existential crisis as "a conflict that flows from the individual's confrontation with the givens of existence" (p. 8). The crisis occurs at the moment when the issues shift from the abstract to the experiential dimension, most likely when some form of realization occurs. They become so real and intolerable that disintegrative anxiety occurs (Steele & Colrain, 1990). Lifton (1980) aptly referred to the moment of existential crisis in the traumatic memory as the "ultimate horror." Dissociation of the personality often occurs at this point, and particular alters may be the embodiment of different aspects of the crisis. There are six important aspects relating to: (1) death or psychological annihilation, (2) isolation, (3) meaninglessness, (4) freedom and responsibility, (5) identity, and (6) intolerable pain.

Janoff-Bulman (1985, 1992) extensively discussed the destructive impact of trauma on basic ontological assumptions (i.e., belief in a meaningful and benevolent world, personal invulnerability, and positive self-worth). These assumptions are shattered by trauma, and, in the case of early and prolonged child abuse, never develop. In both cases, the result is a propensity to existential crises. The moment of the crisis is unbearable, and becomes fixed in the traumatic memory without a temporal or meaningful context. It is accompanied by disintegrative anxiety and terror that is diffuse, intense, and non-verbal, i.e., the classic state of hyper-arousal in response to overwhelming trauma (Van der Kolk, 1987a, 1988, 1994; Krystal,1988).

For each trauma there may be one or a series of existential crises. Since these moments are so laden with affect, it is crucial to recognize them in the synthesis session, so they may be processed during realization.

Parallel and Serial Synthesis

Two fundamental approaches to the synthesis of traumatic memories in MPD patients can be distinguished: serial and parallel synthesis. The first often involves a number of sessions, each returning to the original sequence of events. The second is concentrated in one session with the therapist and patient striving towards maximum involvement of all alters concerned.

Serial Synthesis

The rule in classic serial synthesis of traumatic memories is to follow the patient, and to stay where the patient is at (Putnam, 1989,1992). Clinicians pursuing this approach risk not knowing which alters are involved, and which should be excluded from participating at any given point in the actual synthesis work.

The essence of serial synthesis is to bring the patient, after the necessary preparations, back before the traumatic event. Then, while constantly remaining oriented to the therapist and to the present, the patient is chronologically guided through the experience (i.e., by asking, "And after that, what happened next?") Using the BASK Model (Braun, 1988a&b), for example, the therapist attempts to elicit various dissociated traumatic responses, and to include missing aspects requiring synthesis.

Sometimes the whole process must be repeated a number of times, so that all alters involved in the original traumatic event can make their respective contribution. Should synthesis not be completed, reactivation of the same traumatic memory can occur following the session. Putnam spoke of this in regard to the resulting contradictions and the problems of credibility which ensue:

The patient may relate very different versions of the same event and express widely divergent emotions. In MPD patients, different alters will often embody these contradictory perceptions. In some cases, specific events or specific versions of an event will have clear fantasy components; in many other cases, the therapist and patient will never be sure what actually occurred, and what was fantasy, in these cases. I do not know of any way to sort out truth from fantasy in these cases. (Putnam, 1989, p.245)

First, it is important to be aware that different alters may indeed have experienced widely different aspects of the trauma. An alter identifying with the abusive perpetrator might relate a quite different version from an alter experiencing the pain and fear during abuse; another alter who is completely dissociated from all pain and fear might even experience sexual arousal.

Secondly, it has frequently been observed that some alters may have fantasy experiences during a traumatic event, which should also be synthesized. Charcot (1887) described a classic case of adult traumatization in which a fantasy played a major part in the post-traumatic symptomatology. His patient Lelog was in a traffic accident with a horse-drawn wagon, after which his legs were paralyzed. Although he fell against the ground, and was unconscious, there were no neurological signs indicating a somatic cause of the paralysis. Instead it was discovered that as he fell. Just before losing consciousness, he saw the wheels of the car approaching him, and strongly believed that he would be overrun. This fantasy was dissociated and gave rise to his paralysis.

Hypnotic Techniques

To ensure the safe and effective course of synthesis, the therapist may draw upon specific hypnotic techniques. Some of these, such as creating a safe imaginary place for alters not yet ready to participate in the actual synthesis, usually have been implemented during earlier therapy phases or during the preparation stage. Other hypnotic techniques may be used to ensure gradual synthesis over a number of sessions. One such technique involves the use of an imaginary video screen on which elements of the trauma can be displayed. The patient may then relate associated thoughts, emotions and sensations. Another approach involves placing the incomplete traumatic memory in "safe keeping" (e.g., in an imaginary vault or similar location). More extensive descriptions of these and other relevant hypnotic techniques can be found in Brown and Fromm (1986), Hammond (1990), Kluft (1988, 1989), Peterson (1991; 1993), Putnam (1989), Ross (1989), Spiegel (1981), Sachs and Peterson (1994), Steele and Colrain (1990), and Van der Hart, Boon, and Van Everdingen (1990).

Parallel Synthesis

Parallel synthesis is a much more structured approach. The development of the idea of parallel synthesis, as well as the several techniques involved in its execution, like many other ideas and techniques in the field of MPD, may be attributed to several different sources. The authors have attempted to attribute credit as accurately as possible, while acknowledging that on occasion it is difficult, if not impossible, to determine a primary source in a field with such a strong orientation to workshops and oral tradition. Kluft, Comstock and Sachs, for example, taught variations of parallel synthesis in workshops and conference presentations during the 1980's, and Judith Peterson did the same during the early 1990's (Peterson, 1991). The variation described below was traced to Stephen Ray, as personally related to the authors by Barbara Friedman (personal communication, September 1990) and Pamela Reagor (personal communication, November 1990). In the literature, Kluft (1988,1989), Fine (1991,1993), and Sachs and Peterson (1994) have made significant contributions to this concept and associated techniques.

Parallel synthesis is based on a detailed simultaneous inventory of the traumatic event. Making this inventory involves seeking answers to questions such as what actually happened, which alters were involved in the trauma, and in which ways, e.g., on the dimensions of the BASK Model. It also involves determining with the patient, or, preferably, with an alter who has an overview, which alters should attend the synthesis, and which alters should temporarily be protected from such knowledge and participation.

The actual synthesis starts by repeating the rationale and the actual steps involved. The therapist emphasizes that the success of the technique depends on the completeness with which it is executed. At the beginning of the synthesis session, the patient is guided into a state of intense concentration (i.e., hypnosis). Those alters who are required to synthesize their respective parts of the traumatic memory are brought forward, and others permitted to attend are also invited. By contrast, those who should withdraw are instructed to go their safe places behind the pre-established protective barrier.

A written account of the traumatic event has been divided into 10 segments. The therapist counts from 1 to 10, for example, beginning at "1 "with the onset of the trauma, and ending with "10" after the traumatic experience is over. Each segment that the therapist subsequently mentions is preceded by its appropriate number. This counting gives a clear structure to the session, reminding the patient of the stage reached in the process, and how much more is still to come. During the synthesis, all alters involved are encouraged to share their respective experiences, so that they coalesce into a single whole, and relieve existing dissociations.

After the last count, the therapist offers additional appropriate support to all alters involved, and asks what percent age of the total traumatic event has been brought together. Often a significant percentage, particularly pertaining to the worst pain and fear, is left behind, and must be more explicitly included in the next round. Within a given session, the whole procedure is executed no more than three times. This is often more than adequate.

It is useful to intersperse this "titrated" (re) experiencing with episodes of hypnotic "resting," in which the patient is induced into a very deeply relaxed state. This employs hypnotic time distortion, with the suggestion that the patient may rest as long as is necessary before the next synthesis module during the session. Suggestions include: "seconds seem like minutes, and minutes like hours," "in a short time you may feel relaxed, refreshed and renewed as though you had rested for a long time."

Where there is little outstanding dissociated material, say up to five percent, this may in some cases be processed safely by the patient on his or her own. However, reassurances by patients that they can do this last bit alone, must be tempered by the therapist's clinical judgment. In one female patient the last five percent "at the end" was responsible for the activation of serious suicidal tendencies. It pertained to an earlier suicide attempt made in response to a rape by her father at the age of nine. She had subsequently believed she was pregnant, and failed in an attempt to induce an abortion. Thus it is important to be sure that the remaining dissociated elements do not contain material that will be overwhelming, and that they do not link to further traumas that are not yet ready to be synthesized.

Parallel synthesis may involve many variations. Fine emphasized the importance of first working with "like-clusters" of personalities so that affect and knowledge is shared among groups of alters grouped according to affective or cognitive themes. In this way, the affect is processed and modulated prior to being shared with other alters within other clusters (Fine, 1991,1993). Fine advocated fractionated synthesis in order to avoid re-traumatization to the patient. Comstock (1986) suggested various approaches to managing the treatment of traumatic memories through modulating affect and content. As presented already in his workshops during the early 1980's, Kluft (1988,1989) developed specific approaches to fractionate traumatic material, in which it is suggested to the patient and alters that they need not re-experience their part in the trauma to the full extent, but for instance to a degree of 4 on a scale of 1 to 5. They are further instructed that they "need only experience that which is necessary to know, to understand, and to heal." Another variation is the instruction that, if necessary, one specific alter or one dimension of the traumatic experience, may be left out during the first attempt and may be included later. This transforms parallel synthesis into a highly structured serial synthesis. It is also the case when subsequent sessions are systematically focused on the synthesis of successive episodes of a complex traumatic event.

While the parallel approach initially involves active participation by the therapist, utilization of detailed information regarding the traumata, and sharing the experiences of the respective alters involved, it also lends itself to relatively autonomous work by the patient. Some patients become so accustomed to its application, that they can do the necessary work with certain traumatic memories without the therapist being informed about all of the relevant details. At first counting is often experienced as very helpful, but after a while some patients indicate that they do not need it anymore. This implies that they are not utilizing dissociation as much as before, and that they are now better able to tolerate and modulate affects related to the traumatic memory.

An important issue is the degree of involvement of the host personality (where one exists) in the synthesis of the traumatic memory. When this has been present during the trauma, the answer is of course in the affirmative. In other cases it depends on the host's ego strength o or "psychological tension" in Janet's terms (cf. Van der Hart& Friedman, 1989) o and motivation, among other factors. In general, it is important to tailor synthesis to the patient's needs.

Termination of the Synthesis Session

Towards the end of the synthesis session, while the patient is still in an altered state of consciousness, the therapist ensures that all alters involved are supported, and suggests a positive sense of accomplishment and unity. After having shared their terrible ordeals, they can now share their warmth and wellbeing, not only with each other, but also with all other alters as well. The therapist may use additional healing imagery, such as projecting the image of a white, healing light, which embraces them, and brings them all the more closely together (Kluft, personal communication January 1987). The authors believe that such experiences provide a powerful model for future integration and unification. Finally, the patient is brought back to a state of full awareness and orientation to the present time and place (Putnam, 1992).


Good preparation for a synthesis session requires knowing how (and, in some cases, with whom) the patient will return home after the session, and what type of support will be available, if needed (Putnam, 1992). It is sometimes useful to have the patient contact the therapist at a pre-deter-mined time at the end of the day, or at the beginning of the next day, and to make available a follow-up session at short notice.

Stage 3: Realization and Integration

With successful synthesis the traumatic memory ceases operating at a sensorimotor level. McDougall (1920/1) formulated this as follows:

As soon as the dissociation is overcome, though the same train of thoughts may recur, its power to produce distress is greatly weakened by the patient's accompanying awareness of his present surroundings and his knowledge that the experience belongs to the past. The process of readjustment of his emotional attitude can then begin, or, in other words, he makes progress in 'autognosis' (1920/1, p. 27; 1926,p.457).

What McDougall called the "process of readjustment" requires both acceptance of the memory as real, and a conciliation between this narrative memory and the patient's personality. The traumatic memory has now been transformed from an intrusive re-experiencing, to a trauma-related narrative within the overall stream of consciousness. At this point, therapy shifts to mitigating the impact of the trauma on the self and on relationships with others. Herman stated that the core experiences of this stage are "empowerment and reconnection" (1992, p. 197).

Janet (1935, 1945) called this conciliation "realization" of the event and its consequences. Realization is the formulation of a belief about what happened (the trauma), when it happened (in the past), and to whom it happened (to self). The trauma becomes personalized, relegated to the past, and takes on symbolic rather than sensorimotor properties. In order to resolve existential crises, realization requires the development of new cognitive schema, such as accepting the event and resigning oneself to its occurrence, reframing one's attitudes to the event, the perpetrator, and oneself, and reorganizing one's overall behavior (Janet, 1945). Yet, it is more than cognitive process; realization is an experience of one's history as one's own. It provides a transition from the non-verbal dissociated realm of traumatic intrusion into secondary mental process in which words can construct meaning and form, thereby facilitating the transformation of traumatic memory into narrative memory. Integration or assimilation involves multi-dimensional change in the life of the survivor. Realization then becomes lived experience. As one survivor said, "I don't have to live my old story anymore now that I know it; I'm creating a new one every day. I'm no longer stuck with one script!"

Realization evolves over time as more of the events and their meanings are digested and resonate throughout the mind and as the various dissociated layers are uncovered and integrated. The trauma thus becomes gradually assimilated or integrated into the patient's personality; it is incorporated into the patient's narrative memory or autobiography (Janet, 1919/25). The patient should not only be able to narrate the story of the abuse without re-experiencing it, but:

must know how to associate the happening with the other events of his life, how to give it its place in that life-history which each one of us is perpetually building up and which for each of us is an essential element of our personality. A situation has not been satisfactorily liquidated, has not been fully assimilated, until we have achieved, not merely an outward reaction through our movements, but also an inward reaction through the words we address to ourselves, through the organization of the recital of the event to others and to ourselves, and through the putting of this recital in its place as one of the chapters in our personal history. (Janet, 1919/25, p.662)

Clinical experience teaches us that this realization can be a very difficult o if not initially impossible ochallenge for some patients. This is sometimes due to the fact that surviving dissociated elements of the traumatic memory have been reactivated. But even when synthesis of the traumatic memory has been completely successful, patients may initially feel unable to reconcile the narrative memory with their self-representation. The realization for example that their mother always hated them, that their father sadistically abused them, that they themselves came to perform horrible acts, is overwhelming and defies categorization or attribution of meaning.

Integration and realization involve confronting enormous loss. Grief and mourning are a crucial transition between realization and integration. The patient must learn to grieve deeply the lost childhood that can never be recovered; the loneliness and the pain that have been and must continue to be endured; the lost time, money, education, jobs, relationships, and energy spent avoiding or struggling with the aftermath of the trauma; and finally, the awful truth that must be carried throughout the rest of his or her life. Yet grief, with all its pain and anguish, is ultimately healing. It enables the survivor to relinquish unrealistic expectations, to acknowledge what "is," and therefore, to move fully into the present with new clarity and purpose o with sadness, but also with newfound wisdom.

The first task of this stage is for the therapist to listen empathically to the narration of the story, repeatedly, if need be. Telling the story is an important component of realization and integration; the narrative makes the story more real and personal, as well as shifting the trauma from an autistic re-experiencing, to a relational sharing and linking in with the therapeutic alliance. The survivor is no longer alone in the trauma. Words begin to allow the patient to form new semantic structures that will assist him or her to assimilate the trauma over time. Ager and Jensen (1990) stated that the telling becomes a "testimony," a ritual that is healing and restorative, and gives the story not only a personal, but an interpersonal and social dimension, so that it can be put in its broader context.

Cognitive work should also continue during this stage (Ross, 1989). The patient develops more insight into the origins of cognitive distortions, while the affective and behavioral realms follow suit. The capacity for self-care becomes established, and with it, a new-found confidence and self-esteem. Personal psychological needs are acknowledged, and the survivor begins to take an active role in creating an environment in which trust, healthy dependence, autonomy, power, esteem, and intimacy can flourish (McCann & Pearlman, 1990). In short, the patient learns (or relearns) to love, to work, and to play.


Sometimes realization of a traumatic event requires making the original trauma more acceptable (Janet, 1935). This can be done in two principal ways: either by changing the content of the traumatic memory itself, or by altering aspects of the already transformed narrative recall.

The first approach is called the substitution technique, and was originally developed by Janet (1889, 1898; cf. Van der Hart, Brown, & Van der Kolk, 1989; Van der Hart, Brown, & Turco, 1990). In it he suggested a completely different course to the original traumatic event. Thus, with his patient Marie, for whom the unanticipated first menstruation had been traumatic, he used hypnosis to regress her before it and gave her appropriate preparatory instruction. He then suggested that she would now "remember" her first menstruation as having occurred in a normal and painless way.

A number of therapist have recently reintroduced this technique (Lamb, 1985; Miller, 1986). With regard to traumatic events involving perpetrators (e.g., childhood sexual assault), clinicians are mostly reluctant to employ substitution, since it might be perceived by the patient as an invalidation of the crime, something which has been done so often by the perpetrators and by others who knew about the abuse. However, minor variations can be employed with survivors of child abuse. After synthesis, for example, the patient is encouraged to express anger, and to do symbolically what he or she had wanted to do against the perpetrator (s) during or after the original traumatic event. Watkins (1949), who applied this technique in the treatment of traumatized combat veterans, stated that it may help the patient to bring the original frustrating situation to a more satisfactory conclusion.

However, for true realization to occur, substitution must occur in parallel with retention of the actual traumatic memory. What actually happened and what one wished to happen must exist simultaneously, so that substitution does not lead to denial.

The second approach consists of changing narrative and semantics aspects of the traumatic event. These often involve reframing the patient's own interpretations (e.g., when the patient blames herself for the abuse she suffered as a child from her father), exploring different meanings, and correcting cognitive distortions, all of which have been discussed above.

Whichever the approach, the goal is always to "liquidate the happening" (Janet, 1919/25, p. 681), i.e., to effect closure. In an era when the veracity of memories of abuse is being questioned, and the possibility has been raised of "memories" being suggested by the therapist, it is prudent to consider carefully in advance the implications of memory substitution. This technique offers the potential for resolution and relief. In the highly hypnotizable and suggestible patient, it can also lead to confusion, and should thus be used with caution.

Resolution of Existential Crises

Resolution of existential crises identified in the synthesis stage, particularly those associated with a sense of meaninglessness, is of prime importance in the process of realization and integration. Chronically reenacted existential crises often dominate the present reality of the survivor. These crises are resolved once they, and the associated memories are integrated into the personality structure. The patient then accepts that the moment of crisis has past, and chronic responses can be transformed into more adaptive cognitions, affects, and behaviors. First, the patient must accept that she or he has, indeed, survived. The crisis of death has been continually re-experienced in the present as suicidality, self-destructive tendencies, internal alters who kill or threaten to kill other alters, dead or dying alters, and a sense of foreshortened future. One patient described the synthesis and realization of the existential crisis of death as follows:

I couldn't remember for a very long time; I couldn't bear to remember that I was so close to death so many times. When I finally got to the point of being able to remember, something became crystal clear: I wasn't dead and I wasn't dying anymore. I realized the reason why I was always so suicidal was because somehow I had to be close to death all the time. Somehow I got stuck in the very moment I thought I was going to die. Terror is a very sticky thing, you know, it sticks you like glue right to the thing that happens, right to the part you are most afraid of. Once I got unstuck, I didn't have to be dying anymore. Now I can focus on living, and I have a lot of catching up to do.

Psychological annihilation is experienced and associated with the onslaught of overwhelming and primitive states of hyper-arousal which have been termed "affective storms." In those who do not have the capacity to soothe themselves or to modulate or tolerate affect, the latter can occur with frightening regularity. Severe trauma is often accompanied by a terrifying loss of all self boundaries: "My body seemed like it was flying apart, and then my insides shattered into a thousand million pieces, and there was no more 'me' anymore." Although dissociation during trauma is usually considered a natural defense, the process itself is quite often experienced as intensely dysphoric and terrifying: "I melted into the wall, and although it hurt and I knew this was the end, I spread myself so thin I was invisible." "I shrank myself until it was like a black hole in space, and it sucked me in there and I was the black hole." "I became no-one and went to nowhere into a black nothing."

Once the process of dissociation is more under the survivor's central control, these dysphoric and annihilatory experiences gradually cease. The work on the central self, begun early in treatment, and continued throughout, should by now have contributed significantly to the patient's capacity to tolerate affects without being overwhelmed by them. There is a growing realization that feelings need not be life-threatening events. Through synthesis, dissociative re-enactment of annihilation in the trauma ceases. The survivor comes to understand that there are experiences that none of us can tolerate. Having tolerated these in the past by dissociating, nonetheless, she or he gradually learns how to achieve distance from them in the present without splitting or disintegration of the personality.

The experience of intolerable physical and emotional pain is the basis of further existential crises which impede healing. Repetitive intrusion of overwhelming affect and sensation related to the abuse convinces the patient that she or he cannot tolerate living. The fear is not one of dying, but rather of living with such intolerable pain. This process underlies much of the suicidal ideation of survivors, since death can be seen as leading to relief from pain.

Severe pain is thus a devastating and fragmenting experience (Scarry, 1985). It destroys the self, eradicates the internal locus of control, and dismantles one's world view. Emotional self-care functions do not develop adequately in many childhood abuse victims. As they are unable to tolerate or regulate affect or the hyperaroused state generated by pain, they therefore continue to be overwhelmed by it. Tolerance for pain varies, and survivors with low thresholds readily enter states of overwhelming hyper-arousal. Once dissociated affects and sensations are integrated, significant relief is obtained, and alters who contained the pain begin to heal.

Grief is an important part of the emotional pain which must be worked through. With the passage of time, episodes of grief gradually decrease in intensity and duration. Survivors come to understand and accept that loss is an inevitable part of trauma, and that it is ultimately a lifelong task to assimilate the ebb and flow of re-experienced grief with equanimity.

Assigning Meaning and Processing Information

Once the survivor has accepted that she or he will neither die nor be annihilated, and has successfully resolved the pain, the difficult task of reconstructing meaning continues. A sense of meaninglessness accompanies the obsessive "approach-avoidance" process involved in attempting to realize what has happened: "This is real" versus "I made it up"; "This is real" versus "It didn't happen to me. "As realization proceeds, the survivor is helped to make sense out of what has been senseless. Herman (1992) wrote eloquently about the struggle to assign meaning to the trauma:

[The survivor] stands mute before the emptiness of evil, feeling the insufficiency of any known systems of explanation. Survivors of atrocity of every age and every culture come to a point in their testimony where all questions are reduced to one, spoken more in bewilderment than in outrage: Why? The answer is beyond human understanding. (p.178)

Assigning meaning is a basic need underpinning the expression of all cognitions and affects. Intrapsychic meaning provides the frame for organizing self concepts, and ultimately all cognitive schema (Frankl, 1963; Janoff-Bulman, 1992; Ulman & Brothers, 1988; Yalom, 1980). Trauma not only shatters existing meaning, but also impedes provision of new meaning to the trauma itself. The extent to which cognitive transformation is required depends upon the severity and chronicity of the trauma. Single trauma mainly requires a focus on integration of the traumatic memories, while prolonged and repeated trauma may require transformation of the entire existing framework of thought. Thus, for those individuals who experience chronic trauma in childhood, early meaning structures and cognitive schema are faulty, and completely new ones must be constructed. This is an arduous and time-consuming process, but the trauma cannot be integrated without such reconstruction.

The processing of traumatic memories has been described in terms of relief of dissociation, realization, synthesis and integration. In Janet's words, it is a process aimed at "a conciliation between two terms, between on the one hand the narrative of the trauma, and on the other the personality of the subject" (Janet, 1935, p. 357). Integration of traumatic memories into existing frameworks of thought, as for example in single trauma, was called "transformation of the first term" by Janet, and "assimilation" by Piaget and his followers (Janoff-Bulman, 1992; Kleber & Brom, 1992. The much more radical reconstruction of existing frameworks of thought required in conditions characterized by chronic trauma, such as in MPD, was described as "transformation of the second term" by Janet, and "accommodation" by Piaget.

Such frameworks of thought were first called "cognitive schema" by Sir Henry Head in 1920. The concept was elaborated by Bartlett (1932), and taken up by those researching traumatic neuroses in the 1970's (e.g., Horowitz, 1976/1986). Mandler (1979) stated that "a schema is formed on the basis of past experience with objects, scenes, or events and consists of a set of (usually unconscious) expectations about what things look like and/or the order in which they occur" (p. 263). Consistent with Janet's two forms of conciliation, Hollon and Garber (1988) stated that schema-discrepant information such as trauma may be assimilated into existing schema ("That never happened," "I caused it to happen") or schema may be altered to accommodate the new information ("There is no safe place in the world." "This happened because people chose to hurt me; it wasn't my fault.").

Schema formed following trauma tend to be narrow and fixed, and tend to "determine to what extent new information is absorbed and processed" (Van der Kolk & Van der Hart, 1991, p. 440). Frequently the survivor misinterprets the meaning of current experiences as being dangerous and hostile (the meaning given to the interaction in the trauma) . Survivors often speak of the helplessness and hopelessness they feel as they perceive a "random and chaotic" world, where nothing is predictable or safe. In addition, many struggle with issues of spirituality, rejecting their faith, and questioning "Why would God allow this to happen?" The concept of God inevitably acquires a negative connotation as well as a negative transference: that of the sadistic abuser, or of the passive non-offending parent who knew but did not protect.

Thus, without ontological, psychological, or spiritual meaning structures that offer safety, hope, control, trust, and esteem, the survivor lives in a seemingly random, chaotic, and cruel inner and outer world. This continuing traumatic experience sets up vicious cycles that perpetuate the dissociative and related defenses, and which ultimately prevent the individual from resolving the trauma and from moving toward health. During the realization/integration stage of treatment, new meaning structures are therefore developed, and an internal locus of control is gradually restored. The world then gradually becomes a less hostile and formidable place.

Once new meaning, cognitions and expectations have been established, the possibility of improving relationships with others becomes more plausible; in the words of Lifton (1988), "The survivor seeks vitality both in immediate relationships and ultimate meaning, the one impossible without the other" (p. 26).

Survivors with severe PTSD and dissociative disorders depend on dissociation and denial to maintain the pre-traumatic identity and intact related basic assumptions: "That never happened" (temporal dissociation, or amnesia), or "It happened to that other little girl" (parallel dissociation, and the development of alters). The victim identity, denied as it is of selfhood, is intolerable, and the survivor goes to great lengths to develop any other identity: "Nothing really happened; I just made it up because I am crazy... because I am bad... because I am evil." Byway of contrast, some survivors tenaciously cling to the victim identity, since the trauma overshadows or even destroys any other: they simply become what happened to them.

Survivors are also unable to develop a post-trauma identity because of fragmentation and lack of realization. They have a peculiar and frightening sense that they are not 'whole " or "real" (cf. Putnam, 1990). Those with MPD have many identities conflicting with one another. The vicissitudes of multiple identities in one body have been amply documented in both the professional and the popular literature. Survivors express the terror of experiencing a complete lack of identity or sense of self: "I don't know who I am until somebody asks me to do something"; "Maybe, after all these parts integrate I won't be anybody; maybe there isn't a 'real me' in here; maybe I'm just empty and nothing." For the dissociative patient, realization of the self must coincide with realization of the trauma.

As the intrapsychic structure of survivors is stabilized and reinforced throughout treatment, they will begin to feel more confident of selfhood. Many express feelings that they deserve to exist or even to live for the first time. A core sense of shame dissolves, as the therapist provides a positive identification for the patient in the context of the therapeutic alliance. Survivors are urged to conclude that they have a right to a selfhood, and can begin building on that foundation.

Realization as a Social Act

The existential crisis of isolation is experienced at the moment when the patient realizes that she or he was completely alone in experiencing the trauma, and that there were often no consistent supportive figures in their childhood. This isolation is re-enacted in the immense struggles to find and maintain connection with the self and with others. It is frequently associated with failures and narcissistic wounding; the constant sense of abandonment and terrifying aloneness; the utter certainty and shame associated with the feeling that one is ultimately unworthy to be loved (e.g., "If you really knew me, you would be disgusted"), and finally, with the keeping of the secrets, not only from others, but from oneself (Schultz, 1990). Chronic dissociation and depersonalization leave the individual unable to feel sufficiently connected in the present. Disrupted internal object relations preclude mutually satisfying and interdependent external relationships.

Realization is not only an intrapsychic process, it is also a social act. Janet (1935) described realization as a way of unifying the total life experience, of linking the past with the present and future. With regard to traumatic memories, this requires an ability to tell the story, not only to one's self, but also to others. Furthermore, testimony must not only be given, it must also be heard.

This leads to consideration of social processes that actually promote non-realization and continued isolation of the traumatized individual: collective denial. Realization is usually associated with social expectations. Thus, in the case of loss of a family member, those around us, as well as society at large, expect us to realize this loss and to take it into account in our social actions and in our personal belief systems (Janet, 1937). We are even provided with personal, social, and religious rituals to assist in realization and integration. However, in the case of childhood abuse, society, including the mental health field, has been extremely reluctant to acknowledge (realize) either the occurrence or the impact of abuse (Goodwin, 1985).

There is no social expectation or demand for adults to realize their highly traumatized childhoods. In fact, there is often enormous pressure from perpetrators and families to continue the pattern of dissociation and denial that characterizes collective non-realization (cf. Herman, 1992). The mental health community merely expects survivors to become asymptomatic, and if any acknowledgement is given at all, it is in the form of: "That's all in the past; forget about it and get over it." Goodwin (1985) formulated this social demand for non-realization of abuse as follows:

We observe, in interactions with patients with MPD and abused children and their families, a shared negative hallucination. (...) The MPD patient and the physician cling to the series of false symptoms and false diagnoses in proportion to their mutual need to blot out the reality of multiplicity, and to blot out the unbearable experiences of real pain that triggered it (p. 14).

Realization as a social act may thus be hindered, less by the individual's reluctance to acknowledge what happened, than by society's inability to accept the truth about abuse. The survivor remains isolated in a dissociative "no man's land" between truth and denial. Hopefully, as society becomes increasingly aware of abuse and its dramatic negative personal and social consequences, this will change. Until that time, the therapeutic relationship will be the primary medium through which the survivor can explore the terrifying abyss of isolation and engage in the social act of realization. Perhaps for the first time, the survivor can re-experience the trauma in the presence of an empathic listener, break the isolation and emerge from the crisis, no longer alone or forgotten.

The secure boundaries of the therapeutic relationship, and the honesty, clarity and consistency of the therapist allow the patient accept the impossibility of the hoped for "perfect" relationship, while sensing that the therapist understands the anguish of this perceived loss. The survivor learns to negotiate a balance between closeness and distance, to constructively utilize time spent alone, to remain in touch with others even when alone, and to understand the impact of the boundary violations experienced in the trauma. The therapist aims to enhance the survivor's ability to discern and maintain awareness of interpersonal and intrapsychic boundaries, and the capacity to view self and others as integrated, the "good" with the "bad."

While the therapeutic relationship is a primary healing force, the therapist must also assist the patient to develop additional support systems. Partners and other supportive family members may be engaged in family therapy with the survivor (Sachs, Frischholz, & Wood, 1988). This area is often much neglected in the treatment of survivors. Intensive individual therapy can often disrupt the delicate balance of domestic relationships, and this must be attended to, if the survivor is not to sustain further loss. In many cases, this can be prevented by early intervention. Partner support groups are a growing phenomenon as therapists become more aware of the survivor's current family needs.

The survivor must also find ways of rejoining the community at large. These may include a survivors support or therapy group. Survivors often begin by expressing their hopes and dreams for the future. These may include studies, a job change, or volunteer work. The therapist should encourage even the smallest desire of the survivor to be a part of the "world."

Adult survivors frequently perceive themselves as having no choice, no control over, or capacity to direct the course of their lives. They passively await the next disaster, because they are convinced they can do nothing about it. For some survivors the very process of making a choice can lead to hyperarousal, and they become anxious and unable to decide (Van der Kolk & Greenberg, 1987).

The roots of this loss of control are in childhood (Summit, 1983). Children are, by definition, powerless, dependent, and vulnerable. By virtue of developmental limitations, they are unable to take adequate responsibility for themselves or others. Yet, the abused child is often "parentified," becoming the caretaker and/or the surrogate "spouse" of a parent. The child develops an egocentric set of beliefs that she or he is responsible for the abuse. At the same time, the abused child exists in an environment of inconsistency, arbitrary rules, and unpredictable violence. The child lives in the throes of continual double binds in which responses may be punished at one time, and rewarded at another (Bateson, 1972; Spiegel, 1986; Fine, 1990, 1991, 1993). She or he is unable to control or stop the abuse, and usually, other adults also make no attempt to stop it. The child learns that any action results in the same outcome as no action: nothing is effective. This is the genesis of learned helplessness (Seligman, 1975; Garber & Seligman, 1980).

In one case, a woman had not only sustained childhood abuse and severe neglect, but had then married a vicious sadist. Double binds of forced choice were so great that she became catatonic at the thought of having to make even the smallest decisions. In tracing the roots of this extreme response, it was discovered that her husband had repeatedly forced her to choose which of their children he would torture, or which method of torture she would "prefer." He persistently told her that he would kill her if she tried to leave, as well as telling her it was her fault that he "had" to hurt her and the children. This patient was so fragmented and her ability to make choices so severely incapacitated that she could not make the simplest decisions, and found elaborate ways to have others make choices for her. This made her extremely vulnerable to exploitation, and to further revictimization.

In the course of long and arduous treatment, the patient one day triumphantly reported that she had been able to shop for groceries without assistance, the first "choice" she had been able to make in well over a decade. Although this is an extreme example, it highlights the survivor's difficulties in learning to be responsible and to successfully make choices.

In addition to an impaired capacity to choose and act, the survivor feels a core sense of guilt, shame, and responsibility for the abuse. In the case above, the patient emphatically believed that she had been responsible for both what had happened to her as a child, and for hurting her children. All three children had MPD, and both she and the children had well documented medical evidence of the torture. This led to the belief that she must be punished for her "sins," and that she had no right to recover or to have a good life so long as her children continue to suffer. It was only after a number of years of therapy that she could even begin to acknowledge and realize her previous helplessness and lack of choice, and could move gradually toward empowerment in the present.


This paper, and related literature on trauma, presents the treatment of traumatic memories as a separate step in the phase-oriented therapy of MPD. In actual practice, it is important to realize that this specific phase often alternates several times with other treatment phases. Traumatic memories frequently present in a layered fashion. A number of fusions of alter personalities o usually seen at a later phase in therapy o may have occurred before all traumatic memories have been processed adequately.

In this regard, the realization and integration of past traumata should also be seen as an ongoing mental process. Traumatic events, together with other significant aspects of the patient's past, are shared with an increasing number of alter personalities, whose fusions demand further reorganization of the patient's "autobiography" and personality. Only some time after the last fusion may this reorganization be considered completed, albeit for the time being. The more patients succeed in linking their past to the present and future, the more they are able to live fully independent lives. Yet, for all of us, integration continues as a lifelong process.

The psychology of realization and integration of chronic trauma still remains in its infancy. Much more comprehensive work has been done regarding the processing of single traumatic events or series of events that occurred in adulthood. Much of this work is not directly applicable to MPD patients, since their defenses have had years to solidify and become thoroughly incorporated into the personality structure. This is particularly true of the dissociative defense, which often becomes the preferred method of coping with even the smallest of inconveniences. Many MPD patients were never able to develop what are considered to be basic adaptive assumptions about themselves and others. Their cognitive schema and meaning structures are distorted and entrenched, and there are often no pre-traumatic structures on which to begin to build a new foundation.

This lends a complexity to the therapeutic work that is inadequately addressed in the literature on adult trauma. The realization and integration of a lifetime, or at the very least, of an entire childhood of trauma is a monumental task, requiring all the skills of the therapist, the energy and motivation of the patient, and the strongest therapeutic rapport. The work is slow, grueling, and, at times, tediously repetitious, again and again contrasting the old worldview with the new.

There is still much to learn about this process, and much in this paper will need investigation and elaboration. This has been a pioneering effort to clarify current conceptions about traumatic memories, and to integrate the work ofPierre Janet with contemporary trauma-based models of therapy. It provides a sound conceptual framework for the resolution of traumatic memories in adult survivors of chronic childhood abuse.


Ager, I., &Jensen, S.B. (1990). Testimony as ritual and evidence in psychotherapy for political refugees. Journal of Traumatic Stress, 3, 115-130.

Bartlett, F.C. (1932). Remembering. Cambridge: Cambridge University Press.

Bateson, G. (1972). Steps to an ecology of mind. New York: Ballantine Books.

Beck, A.T. (1979). Cognitive therapy and the emotional disorders. New York: International Universities Press.

Blank, A.S. (1985). The unconscious flashback to the war in Viet Nam veterans: Clinical mystery, legal defense, and community problem. In S.M. Sonnenberg & J.A. Talbot (Eds.), The trauma of war: Stress and recovery in Viet Nam veterans (pp. 293-308). Washington, DC: American Psychiatric Press.

Bliss, E. (1986). Multiple personality, allied disorders and hypnosis. Cambridge: Cambridge University Press.

Boon, S., & Van der Hart, 0. (1991). De behandeling van de multiple persoonlijkheidsstoornis [Treatment of multiple personality disorder]. In 0. van der Hart (Ed.), Trauma, dissociatie en hypnose [Trauma, dissociation and hypnosis] (pp. 159-187). Amsterdam/Lisse: Swets & Zeitlinger.

Braun, B.G. (1986). Issues in the psychotherapy of multiple personality disorder. In B.G. Braun (Ed.), Treatment of multiple personality disorder , (pp. 3-28). Washington, DC: American Psychiatric Press.

Braun, B.G. (1988a). The BASK model of dissociation: Part I. DISSOCIATION, 1(1), 4-23.

Braun, B.G. (1988b). The BASK model of dissociation: Part II: Clinical applications. DISSOCIATION, 1(2), 16-23.

Brende, J.O. (1984). An educational-therapeutic group for drugs and alcohol abusing combat veterans. Journal of Contemporary Psychotherapy, 14, 122-136.

Briquet, P. (1859). Traite clinique et therapeutique de l'hysterie (2 vols.). Paris: J.-B. Bailliere & Fils.

Brown, W. (1920/1921). The revival of emotional memories and its therapeutic value (1). British Journal of Medical Psychology, 1, 16-19.

Brown, D.P., & Fromm, E. (1986). Hypnotherapy and hypnoanalysis. Hillsdale, NJ: L. Eribaum Associates.

Charcot, J.M. (1887). Leçons sur les maladies du systéme nerveux faites a la Salpêtriére, Tome III. Paris: Progres Medical and A. Delahaye & E. Lecrosnie.

Comstock, C. (1986). The therapeutic utilization of abreactive experiences in the treatment of multiple personality disorder. In B.C. Braun (Ed.), Dissociative disorders 1986: Proceedings of the Third Annual Conference on Multiple Personality and Dissociative States. Chicago: Rush.

Courtois, C. A. (1988). Healing the incest wound: Adult survivors in therapy. New York: W.W. Norton & Co.

Courtois, C.A. (1991). Theory, sequencing, and strategy in treating adult survivors. New Directions for Mental Health Services, 5-7 (Fall), 47-60.

Delboeuf, J.R.L. (1889). Le magnétisme animal: A propos d'une visite a líécole de Nancy. Paris: Félix Alcan.

Fine, C.G. (1988). Thoughts on the cognitive perceptual substrates of multiple personality disorder. DISSOCIATION, 1(4), 5-10.

Fine, C.G. (1990). The cognitive sequelae of incest. In R.P. Kluft (Ed.), Incest related syndromes of adult psychopathology (pp. 161-182). Washington, DC: American Psychiatric Press.

Fine, C.G. (1991). Treatment stabilization and crisis prevention: Pacing the therapy of the multiple personality disorder patient. Psychiatric Clinics of North America, 14(3), 661-675.

Fine, C.G. (1993). A tactical integrationalist perspective on the treatment of multiple personality disorder. In R.P. Kluft & C.G. Fine (Eds.), Clinical perspectives on multiple personality disorder (pp. 135-153). Washington, DC: American Psychiatric Press.

Fish-Murray, C.C., Koby, E.V., & Van der Kolk, B.A. (1987).Evolving ideas: The effect of abuse on children's thought. In B.A. Van der Kolk (Ed.), Psychological trauma (pp.89-110). Washington, DC: American Psychiatric Press.

Frankl, V. (1963). Man's search for meaning. New York: Bantam Books.

Fromm, E. (1965). Hypnoanalysis: Theory and two case excerpts. Psychotherapy: Theory, Research and Practice, 2 127-133.

Garber,J., & Seligman, M. (Eds.) (1980). Human helplessness: Theory and application. New York: Academic Press.

Gelinas, D. (1983). The persisting negative effects of incest. Psychiatry, 46, 312-332.

Goodwin, J. (1985). Credibility problems in multiple personality disorder patients and abused children. In R. P. Kluft (Ed.), Childhood antecedents of multiple personality (pp. 2-19). Washington, DC: American Psychiatric Press.

Hammond, C.D. (1990) (Ed.). The handbook of hypnotic suggestion and metaphors. Washington, DC: W. W. Norton & Co.

Herman, J. L. (1992). Trauma and recovery. New York: Basic Books.

Hollon, S. D., & Garber, J. (1988). Cognitive therapy. In L.Y. Abramson (Ed.). Social cognition and clinical psychology: A synthesis, pp. 204-253. New York: Guilford Press.

Horowitz, MJ. (1986). Stress response syndromes, 2nd ed., Northvale, NJ:J.Aronson. (Original publication: 1976).

Horowitz, MJ. (1976). Stress response syndromes. New York: Jason Aronson.

Janet, P. (1889). L'Automatisme psychologique. Paris: Felix Alcan. Reprint: Societe Pierre Janet, Paris, 1973.

Janet, P. (1894). Histoire D'une idée fixe. Revue Philosophique, 37, 121-163.

Janet, P. (1898). Névroses et idées fixe. Paris: Félix Alcan.

Janet, P. (1903). Les obsessions et la psychasthénie (Vol. 1). Paris: Félix Alcan.

Janet, P. (1904). L'Amnésia et la dissociation des souvenirs par l'émotion. Journal de Psychologie, 1, 417-453. Also in P.Janet, (1911). L 'Etat mental des hysteriques, 2nd. ed. (pp. 506-544). Paris: Felix Alcan. Reprint: Lafitte Reprints, Marseille, 1983.

Janet, P. (1907). The major symptoms of hysteria. London & New York: Macmillan.

Janet, P. (1910). Une Félida artificielle. Revue Philosophique, 69, 329-357; 483-529.

Janet, P. (1911). L'Etat mental des hystériques (Reprint: Lafitte Reprints, Marseilles, 1983 ed.). Paris: Felix Alcan.

Janet, P. (1919/25). Les médications psychologiques. Paris: Félix Alcan. English edition Psychological healing (2 vols ) New York Macmillan, 1925. Reprint Arno Press, New York, 1976.

Janet, P. (1928). L'Evolution de la mémoire et de la notion du tempts. Paris: A Chahine.

Janet, P. (1929). L'Evolution de la personnalité. (New Société Pierre Janet 1984 ed.). Paris: Chahine.

Janet, P. (1935). Réalisation et interprétation. Annales Médico-Psychologiques, 93, 329-366.

Janet, P. (1937). Les troubles de la personnalité sociale. Annales Médico-Psychologiques, 95, 421-468.

Janet, P. (1945). La coyance délirante. Schweizerische Zeitschrift für Psychologie, 4, 173-187.

Janoff-Bulman, R (1985): The aftermath of victimization: Rebuilding shattered assumptions. In C. R. Figley (Ed ), Trauma and its wake, pp 15-35. New York: Brunner/Mazel.

Janoff-Bulman, R. (1992). Shattered assumptions: Towards a new psychology of trauma. New York: Free Press.

Jung, C. G. (1921/1922.) The question of the therapeutic value of abreaction. British Journal of Medical Psychology, 2, 13-22.

Kleber, R.J. , & Brom, D., with Defares, P. B. (1992). Coping with trauma: Theory, prevention and treatment. Amsterdam: Lisse Swets & Zeitlinger.

Kluft, R.P. (1987). An update on multiple personality disorder. Hospital and Community Psychiatry, 38, (4), 363-373.

Kluft, R. P. (1988). On treating the older patient with multiple personality disorder: "Race against time " or "Make haste slowly? " American Journal of Clinical Hypnosis, 30 (4), 257-266.

Kluft, R.P. (1989). Playing for time: Temporizing techniques in the treatment of Multiple Personality Disorder. American Journal of Clinical Hypnosis, 32, 90-98.

Kluft, R.P. (1991). Multiple personality disorder. In A Tasman & S Goldfinger (Eds ), American Psychiatric Press Annual Review of Psychiatry, 10, 161-188.

Kluft, R. P. (1993). Basic principles in conducting the psychotherapy of multiple personality disorder. In R. P. Kluft & C. G. Fine (Eds. ), Clinical perspectives on multiple personality disorder (pp 19 50). Washington, DC: American Psychiatric Press.

Krystal, H. (1988). Integration and self-healing: Affect, alexithymia, and trauma. Hillsdale, NJ: The Analytic Press.

Lamb, C. S. (1985). Hypnotically-induced deconditioning: Reconstruction of memories in the treatment of phobias. American Journal of Clinical Hypnosis, 28, 56-62.

Lifton, R.J. (1980.) The concept of the survivor. In J. E Dimsdale (Ed ), Survivors, victims, and perpetrators: Essays on the Nazi holocaust (pp 113-126. New York: Hemisphere Publishers.

Lifton, R. J. (1988.) Understanding the traumatized self: Imagery, symbolization, and transformation. In J.P. Wilson, Z. Harel & B. Kahana (Eds ), Human adaptation to extreme stress From the Holocaust to Vietnam (pp 7-31). New York: Plenum Press.

Loewenstein, R.J. (1993). Posttraumatic and dissociative aspects of transference and countertransference in the treatment of multiple personality disorder. In R. P. Kluft & C. G. Fine (Eds ). Clinical perspectives on multiple personality disorder, pp 51-85. Washington, DC: American Psychiatric Press.

Loewenstein, R.J. (1991) An office mental status examination for complex chronic dissociative symptoms and multiple personality disorder. In R.J. Loewenstein (Ed ). Psychiatric Clinics of North America Multiple Personality Disorder, 14(3), 567-604.

Macmillan, M. (1991). Freud evaluated: The completed arc. Amsterdam/New York North-Holland.

Mandler, J.M. (1979). Categorical and schematic organization of memory In C .R. Puff (Ed ), Memory organization and structure. New York Academic Press.

McCann, I. L. , & Pearlman, L. A. (1990). Psychological trauma and the adult survivor: Theory, therapy, and transformation. New York: Brunner/Mazel Publishers.

McDougall, W. (1920/1) The revival of emotional memories and its therapeutic value (III). British Journal of Medical Psychology, 1, 23-29.

McDougall, W. (1926). An outline of abnormal psychology. London: Methuen.

Miller, A. (1986). Brief reconstructive hypnotherapy for anxiety reactions: Three case reports. American Journal of Clinical Hypnosis, 28, 138-146.

Myers, C. S. (1920/1). The revival of emotional memories and its therapeutic value (II). British Journal of Medical Psychology, 1, 20-22.

Myers, C.S. (1940). Shell shock in France 1914-1.8 Cambridge: Cambridge University Press.

Noyes, R. , Hoenck, P. R. , & Kupperman, B.A. (1977). Depersonalization in accident victims and psychiatric patients. Acta Psychiatrica Scandinavica, 66, 451-458.

Orzek, A. M. (1985). The child's cognitive processing of sexual abuse. Child and Adolescent Psychotherapy, 2, 110-114.

Parson, E. R. (1984). The reparation of self: Clinical and theoretical dimensions in the treatment of Vietnam combat veterans. Journal of Contemporary Psychotherapy, 14, 4-56.

Peterson, ,J.A. (1991, November). Hypnotic techniques recommended to assist in associating the dissociation: Abreaction. Paper presented at the 8th International Conference on Multiple Personality/Dissociative States: Chicago, Illinois.

Putnam, F. W. (1989). Diagnosis and treatment of multiple personality disorder New York: Guilford Press.

Putnam, F.W. (1990) Disturbances of self in victims of childhood sexual abuse In R. P. Kluft (Ed ), Incest-related syndromes of adult psychopathology (pp 113-131). Washington, DC: American Psychiatric Press.

Putnam, F.W. (1992). Using hypnosis for therapeutic abreactions. Psychiatric Medicine, 10 (1), 51-65.

Putnam, F.W., Guroff.JJ., Silberman, E.K., & Post, R.M. (1986). The clinical phenomenology of multiple personality disorder. Journal of Clinical Psychiatry, 47, 285-293.

Resick,P.A.,& Schnicke,M.K. (1992). Cognitive processing therapy for sexual assault victims. Journal of Consulting and ClinicalPsychology, 60 (5), 748-756.

Ross, C.A. (1989). Multiple personality disorder: Diagnosis, clinical features and treatment. New York: John Wiley & Sons.

Ross, C.A., Norton, G.R., & Wozney, K. (1989). Multiple personality disorder: An analysis of 236 cases. Canadian Journal of Psychiatry, 34, 413-418.

Ross, C.A. &: Gahan, P. (1988). Cognitive analysis of multiple personality disorder. American Journal of Psychotherapy, 42, 229-239.

Sachs, R.G., Braun, E.G., &Shepp, E. (1988). Techniques for planned abreactions with MPD patients. In B.G. Braun (Ed.), Dissociative Disorders 1988: Proceedings of the Fifth International Conference on Multiple Personality/Dissociative States (p. 85). Chicago: Rush.

Sachs, R.G., Frischholz, E., & Wood, J.I. (1988). Marital and family therapy in the treatment of multiple personality disorder. Journal of Marital and Family Therapy, 14, 249-259.

Sachs, R. G., & Peterson.J. A. (1994). Processing memories retrieved by trauma victims and survivors: A primer for therapists. Tyler, Texas: Family Violence and Sexual Assault Institute.

Sakheim, D.K., Hess, E.P., & Chivas, A. (1988). General principles for short-term inpatient work with multiple personality-disorder patients. Psychotherapy, 25, 117-124.

Scarry, E. (1985). The body in pain: The making and unmaking of the world. New York: Oxford University Press.

Schultz, R. (1990). Secrets of adolescence: Incest and developmental issues. In R.P. Kluft (ed.), Incest-related syndromes of adult psychopathology (pp. 133-159). Washington, DC: American Psychiatric Press.

Schultz, R., Braun, B.G., & Kluft, R.P. (1989). Multiple personality disorder: Phenomenology of selected variables in comparison to major depression. DISSOCIATION, 2(1), 45-51.

Seligman.M. (197'5). Helplessness: On depression, development and death. San Francisco: Freeman Press.

Solomon, Z., Garb, R., Bleich.A, & Grupper, D. (1987). Reactivation of combat-related post-traumatic stress disorder. American Journal of Psychiatry, 144, 51-55.

Spiegel, D. (1986). Dissociation, double binds, and post-traumatic stress in multiple personality disorder. In B.G. Braun (Ed.), Treatment of multiple personality disorder (pp. 61-77). Washington, DC: American Psychiatric Press.

Spiegel, D. (1981). Vietnam grief work using hypnosis. American Journal of Clinical Hypnosis, 24 289-294.

Spiegel, D., & Cardeña, E. (1991). Disintegrated experience: The dissociative disorders revisited. Journal of Abnormal Psychology, 100, (3), 366-378.

Steele, K. (1989). A model for abreaction with MPD and other dissociative disorder. DISSOCIATION, 2(3), 151-159.

Steele, K., & Colrain, J. (1990). Abreactive work with sexual abuse survivors: Concepts and techniques. In M.A. Hunter (Ed.), The sexually abused male. Volume 2, Applications of treatment strategies (pp. 1-55). Lexington, MA: Lexington Books.

Summit, R.C. (1983). The child sexual abuse accommodation syndrome. Child Abuse and Neglect, 7 177-193.

Terr, L.C. (1991). Childhood traumas: An outline and overview. American Journal of Psychiatry, 148, 10-20.

Ulman, R.B., & Brothers, D. (1988). The shattered self: A psychoanalytic study of trauma. Hillsdale, NJ: The Analytic Press.

Van der Hart, 0., Boon, S., Friedman, B., & Mierop, V. (1992). De reactivering van traumatische herinneringen [The reactivation of traumatic memories]. Directieve Therapie, 12(1), 12-55.

Van der Hart, 0., Boon, S., & Van Everdingen, G.B. (1990). Writing assignments and hypnosis in the treatment of traumatic memories. In M.L. Fass & D. Brown (Eds.), Creative mastery in hypnosis and hypnoanalysis: A Festschrift for Erika Fromm (pp. 231-254). Hillsdale, NJ: L. Erlbaum Associates.

Van der Hart, 0., Brown, P., & Van der Kolk, B.A. (1989). Pierre Janet's treatment of post-traumatic stress. Journal of Traumatic Stress, 2 (4), 379-395.

Van der Hart, 0., Brown, P., & Turco, R.N. (1990). Hypnotherapy for traumatic grief: Janetian and modern approaches integrated. American Journal of Clinical Hypnosis, 32(4), 263-271.

Van der Hart, 0., & Brown, P. (1992). Abreaction re-evaluated. DISSOCIATION, 5(4), 127-138.

Van der Hart, 0., Steele, K., & Brown, P. (manuscript in preparation) . Traumatic memories: Repression and dissociation re-evaluated.

Van der Hart, 0., & Friedman, B. (1989). A reader's guide to Pierre Janet on dissociation: A neglected heritage. DISSOCIATION, 2(1), 3-16.

Van der Hart, 0., & Friedman, B. (1992). Trauma, dissociation and triggers: Their role in treatment and emergency psychiatry. In J.B. van Luyn et al. (Eds.), Emergency psychiatry today (pp.137-142). Amsterdam: Elsevier.

Van der Hart, 0., & Op den Velde, W. (1991). Traumatische herinneringen [Traumatic memories]. In 0. van der Hart (Ed.), Trauma, dissociatie en hypnose [Trauma, dissociation, and hypnosis] (pp. 71-90). Amsterdam/Lisse: Swets & Zeitlinger.

Van der Kolk, B.A. (1987a). The psychobiology of the trauma response: Hyperarousal, constriction, and addiction to traumatic reexposure. In B.A. van der Kolk (Ed.), Psychological trauma (pp. 63-88). Washington, DC: American Psychiatric Press.

Van der Kolk, B.A. (Ed.) (1987b). Psychological trauma. Washington, DC: American Psychiatric Press.

Van der Kolk, B.A. (1988). The biological response to psychic trauma. In F.M. Ochberg (Ed.), Post-traumatic therapy and victims of violence (pp. 25-38). New York: Brunner/Mazel.

Van der Kolk, B.A. (in press, 1994). The body keeps the score: The evolving psychobiology of post-traumatic stress. Harvard Review of Psychiatry.

Van der Kolk, B.A., & Greenberg, M. (1987). The psychobiology of the trauma response: Hyperarousal, constriction, and addiction to traumatic exposure. In B.A. van der Kolk (Ed.), Psychological trauma (pp. 63-87). Washington, DC: American Psychiatric Press.

Van der Kolk, B.A., & Van der Hart, 0. (1989). Pierre Janet and the breakdown of adaptation in psychological trauma. American Joumal of Psychiatry, 146, 1530-1540.

Van der Kolk, B.A., & Van der Hart, 0. (1991). The intrusive past: The flexibility of memory and the engraving of trauma. American Imago, 48(4), 425-545.

Van Putten, T., & Emory, W.H. (1973). Traumatic neurosis in Vietnam returnees: A forgotten diagnosis? Archives of General Psychiatry, 29, 695-698.

Watkins,J.G. (1949). Hypnotherapy of war neuroses. New York: The Ronald Press Company.

Yalom, I. (1980). Existential psychotherapy. New York: Basic Books.