International Forum of Psychoanalysis

ISSN: 0803-706X (Print) 1651-2324 (Online) Journal homepage: https://www.tandfonline.com/loi/spsy20

Identification with the aggressor and the ‘normal

traumas’: clinical implications

Jay Frankel

To cite this article: Jay Frankel (2004) Identification with the aggressor and the ‘normal
traumas’: clinical implications, International Forum of Psychoanalysis, 13:1-2, 78-83, DOI:
10.1080/08037060410027101

To link to this article: https://doi.org/10.1080/08037060410027101

Published online: 05 Aug 2006.

Submit your article to this journal

Article views: 220

View related articles

Citing articles: 6 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=spsy20
Int Forum Psychoanal 13:(78±83), 2004

Identification with the Aggressor and the ‘Normal Traumas’: Clinical

Implications
Jay Frankel, New York

Frankel J. Identification with the aggressor and the ‘normal traumas’: clinical implications. Int Forum
Psychoanal 2004;13:78–83. Stockholm. ISSN 0803-706X.
Ferenczi’s (1933) surprisingly unknown concept of identification with the aggressor – an abuse
victim’s ‘eliminating’ her own subjectivity and ‘becoming’ precisely what an attacker needs her to
be – has radical implications for our understanding of analytic technique. Its very frequent
occurrence also forces us to broaden our understanding of what constitutes trauma. Ferenczi saw
the experience of ‘traumatic aloneness’ or ‘emotional abandonment’ as the key element of trauma,
since this is what enforces the traumatic responses of dissociation and identification with the
aggressor. Identification with the aggressor operates in the analytic relationship in both patient and
analyst. This has various consequences, including the structuring of the relationship through
unconscious collusions – mutually coordinated, defensive identifications designed to help both
participants feel secure. This view of the analytic relationship has clinical implications in at least
four areas: the understanding of the patient’s free associations, which may reflect the patient’s
compliance with the analyst’s wishes rather than the contents of the patient’s own unconscious;
the need for some kind of mutuality of analysis; the traumatizing potential of the analyst’s
authority; and the tendency of some patients to take blame and responsibility reflexively, as a way
of protecting the analyst.
Key words: analytic relationship, compliance, collusion, aloneness, free association, mutuality
Jay Frankel, 14 East 4th Street, #402, New York, NY 10012, USA. E-mail: [email protected]

Seventy years after he introduced it, Ferenczi’s (1) frequent occurrence of identification with the
concept of identification with the aggressor re- aggressor – which Ferenczi discovered to be a
mains surprisingly unknown and misunderstood. response to trauma – forces us to broaden our
Even the name of the concept is usually associated understanding of what constitutes trauma and to
with Anna Freud (2), who, several years later, gave consider the idea of ‘normal traumas’: events that
the term a very different and much more limited are ubiquitous, yet in important ways traumatic.
meaning (see 3). More importantly, Ferenczi’s idea Ferenczi used the term ‘identification with the
of identification with the aggressor has radical aggressor’ to refer to a victim’s ‘eliminating’ her
implications for our understanding of the analytic own subjectivity and ‘becoming’ precisely what an
relationship and for analytic technique – even attacker needs her to be, as a way to survive. His
today – that may contribute to the repression of this concept is part of a trauma theory that is surpris-
concept by the analytic community. ingly contemporary and sophisticated. A little
The concept of identification with the aggressor background about certain aspects of this theory is
contains the seeds of a new and very different idea necessary in order to more fully grasp the concept
about what the analytic relationship has to be, as I and its implications for the understanding of the
will describe. Even today, it undermines some of analytic relationship and for analytic technique.
the very tenets of what has become standard
analytic technique: that free association is a
dependable route to understanding someone’s Ferenczi’s Trauma Theory and Emotional
unconscious; that the patient can be understood Abandonment
in isolation, with the analyst exerting minimal In his paper, ‘Confusion of tongues’ (1), Ferenczi
influence on the process; that insight is a reliable said that the aspect of child abuse that is truly
road to cure; even that standard analytic technique unbearable for the child and causes lasting damage
is essentially benign. What is more, the extremely is being emotionally abandoned at a moment of

 2004 Taylor & Francis. ISSN 0803-706X DOI: 10.1080/08037060410027101

Int Forum Psychoanal 13, 2004 Identification with the aggressor: clinical implications 79
great need (p. 160). This is a quantum leap in the and it works in tandem with dissociation (1, 3, 10).
theory of trauma that I believe still has not been In the moment of trauma, the victim dissociates his
fully appreciated. own perceptions and feelings – he empties his
He developed this idea by exploring the role of mind, both in order to get away from the
the ‘other’ parent – the one who colludes with the unbearable experience and also so he can replace
abuser by denying that abuse happened, or that it his own perceptions, thoughts, even feelings with
was abuse, or that the child was its victim. Ferenczi those the attacker needs him to have. The victim
believed that children can cope with almost any- identifies with – virtually becomes – the attacker’s
thing as long as there is someone with whom to image of him as a way to survive. At the same time,
share one’s fear and pain. But when even the other the victim’s closely attuned identification guides
parent turns their back and insists on a ‘reality’ that his dissociation: it tells him precisely which of his
contradicts the child’s own perceptions – insists perceptions, thoughts, and feelings must be ‘elimi-
that she not see, not know, not speak – then the nated’ in order to be safe with the attacker.
child is left unbearably alone with her traumatic Identifying with an aggressor requires acute
experience. sensitivity toward him. The victim must know
If a child knows something her parents tells her what the aggressor intends, exactly who the
is not true, her choice is between knowing what she aggressor wants her to be, and how the aggressor
knows, which puts a desperately needed relation- will respond. When trauma becomes expectable,
ship at risk, or else trying to preserve the relation- this sensitivity may develop into a chronic, anxious
ship by blotting out what she knows. There is no hyperalertness and also an extraordinary inter-
contest: children need the relationship (see 4). personal attunement toward others (5, p. 203; 1, p.
They cannot let themselves know something if 165).
knowing it will leave them feeling profoundly, As suddenly as these capacities are gained, the
intolerably alone. A child is likely to ‘accept’ victim’s contact with his own emotional life is lost,
reality as presented by her parents. The resulting dissociated. Genuine emotions are too dangerous –
dissociation of her own perceptions, along with they can distract from the urgent task upon which
associated thoughts and feelings, is the splitting of life may depend. Instead, the victim takes on the
the personality that constitutes the lasting damage. emotions, as well as the perceptions and thoughts,
I believe that Ferenczi thus places ‘traumatic which the aggressor expects of him. These ten-
aloneness’ (5, p. 193) or ‘emotional abandonment’ dencies can also persist, resulting in an emotional
(6, p. 138; 1, pp. 163–64; 5, pp. 115, 164, 202; also life that feels false, mistrust of one’s own percep-
see 7) as the key element in trauma – this is what tions, and lack of conviction in one’s beliefs.
enforces dissociation (7). Automatic compliance with the wishes of the
It is also important, as background, to name aggressor may lead to a chronic tendency to
some other events Ferenczi thought were trau- comply.
matic. These include: not being loved (8); parental Yet identification with an aggressor is never
cruelty and eroticism that are disguised (9, p. 121) complete – some piece of the victim resists total
– in other words, children respond to what parents compliance and total loss of self. And so there is
feel rather than to how they behave; ‘tenderness’ in likely to be some subtle rebellion and resistance.
excess of what a child wants (9, p. 121; 1, p. 164; 5, Ferenczi felt the most damaging aspect of
pp. 115–16, 175); parents’ narcissistic overidenti- identification with the aggressor is that the victim
fication with their children, in the form of requiring feels he himself is the bad one and responsible for
precocious achievements (1, p. 165); and the his own abuse. The victim may take on this
‘terrorism of suffering’: parents overwhelming badness to exonerate the attacker in his own mind
their children with displays of pain (1, pp. 165– and continue to feel the attacker is a good internal
66). object (4) or to placate the attacker and to stave off
actual emotional abandonment (3). There may also
simply be a sense of shame inherent in feeling
Coping with Trauma weak and victimized (7, 11). Whatever the reason,
Identification with the aggressor is part of how the long-term result is often chronic confusion
people cope with trauma, according to Ferenczi; about responsibility – feeling both culpable, as part
80 J. Frankel Int Forum Psychoanal 13, 2004

of one’s identification, and innocent, based on patients is that patients may detect the underlying
one’s own perceptions (1, p. 162). falseness, tension, and resentment and feel further
emotionally abandoned. They may even feel they
must collude with the analyst’s mask of goodness.
How Identification with the Aggressor Plays This replays the coercive, traumatic family situa-
out in the Therapeutic Relationship tion that Ferenczi (1) called ‘parental hypocrisy.’
Ferenczi (1, 5) regularly observed identification Collusions between patient and analyst may
with the aggressor in his patients. They refused to result from the mutual, coordinated operation of
confront him with their anger and criticism even identification with the aggressor (12, 13). Since
when he encouraged them to do so. Instead, they both people, on some level, see the other as a
were compliant. Ferenczi threat, each person tends to comply with the
other’s implicit demands. Yet each person also
came to the conclusion that the patients have an grasps that the other is threatened by oneself, and
exceedingly refined sensitivity for the wishes, tenden- therefore that one has some power over the other
cies, whims, sympathies and antipathies of their analyst,
even if the analyst is completely unaware of this person. So a tenuous, unconscious ‘deal’ is struck
sensitivity. Instead of contradicting the analyst or that the most vulnerable parts of each person – the
accusing him of errors and blindness, the patients deepest anxiety and shame – will never be touched
identify themselves with him; only in rare moments of directly by either party. Collusions are essentially
an hysteroid excitement . . . can they pluck up enough coordinated dissociations and identifications with
courage to make a protest; normally they do not allow each other. One implication of this idea of
themselves to criticize us, such a criticism does not even collusion is that what is manifest in the relationship
become conscious in them (1, pp. 157–58).
is often a diversion from the most important
My own observation is that identification with the interpersonal dynamics. What is most real is likely
aggressor occurs often even in patients who have to be invisible.
not been grossly traumatized as children. If true,
this suggests that more subtle forms of trauma,
such as those discussed by Ferenczi – especially Therapeutic Implications
emotional abandonment or other forms of trauma How does Ferenczi’s concept of identification with
of which emotional abandonment is a part – are the aggressor force us to rethink our clinical work?
extremely frequent among patients who have not I will address four areas: free association, mu-
been abused and are indeed experienced in certain tuality, the analyst’s authority, and taking blame
ways as genuinely traumatic – as unbearable. and responsibility.
Ferenczi’s Clinical Diary (5) makes clear that he
also observed in himself a tendency to identify
with his patients as aggressors. This phenomenon, I Free association
think, is very widespread among analysts, many of Free association is patients’ fundamental assign-
whom have grown up having to cope with the ment during analysis. But in light of what has been
demands of narcissistic or needy parents and who said, the patient’s associations, rather than being
may have been exposed to the ‘terrorism of the route to enlightenment about the patient’s
suffering’ by these parents. Analysts’ identification unconscious, are often – to a greater extent than we
with patients as aggressors can manifest in many would expect – decoys whose purpose is to
ways: compulsive compassion; empathy that does perpetuate the mutual illusions upon which collu-
not feel deeply rooted; warmth or involvement that sions are based and hide the most vulnerable parts
feels superficial; complying with patients’ de- of both patient and analyst. Why are these decoys
mands against one’s own better judgement; diffi- effective? Because the patient may be speaking
culty negotiating fees; not knowing what you feel partly as the object the analyst needs him to be,
or not thinking clearly in a patient’s presence; rather than from his own subjectivity, so the
feeling obligated to be extra ‘good’; or else analyst likes what he is hearing. It may sometimes
tiredness, boredom, or resentment as a resistance be the analyst’s mind, more than the patient’s – the
to identifying and ‘losing’ one’s self. The irony of object in the analyst’s mind – that can best be
these strained attempts to comply and be with understood through the patient’s associations. The
Int Forum Psychoanal 13, 2004 Identification with the aggressor: clinical implications 81
patient, perhaps mostly unconsciously, senses what pathological compliance with an intimidating
the analyst is looking for and gives it to him; she patient, or else a retaliation against such a patient.
may even make herself believe that what she is
disclosing to her analyst is sincere and deeply felt.
This sacrifice of one’s own healthy skepticism, of The Dangers of the Analyst’s Authority
independent perception and thinking, amounts to a Another implication of the high degree of fre-
maneuver to disarm and neutralize the analyst. quency of identification with the aggressor among
Whether this takes the form of subtly flattering the patients is that analysts’ perceived power is often
analyst, not noticing his faults, or not confronting subtly traumatizing to patients. I believe that, even
him – or more likely all of these – the result is to today, most analysts still do not fully grasp the
blind the analyst by boosting his self-esteem at the effects of their power in the analytic relationship.
expense of one’s self-awareness and a sense of While many of us have given up a stance of
ownership of one’s own experience. Hiding one’s authoritative certainty, we nevertheless share with
own vulnerable parts from the analyst requires the patients our observations of them; we make
patient to disconnect from these aspects of himself. interpretations; and these moves are inevitably
Another function of free associations is implied heard by patients as conveying some degree of
in Ferenczi’s observation that there is always some authority. Gentleness and humility toward patients
resistance to fully giving up one’s self and may counteract their feeling intimidated and are
identifying with an aggressor. Indeed, patients’ therefore to be recommended. Yet even an analyst
associations often contain disguised criticisms of who adopts this unoffensive approach can be
the analyst (cf. 14). Ferenczi commented on the experienced as oppressive if the patient feels
‘hidden hatred and contempt that can be so required to confirm the analyst’s benign view of
cunningly disguised in our patients’ associations’ himself.
(1, p. 158). And if an analyst needs to be powerful or feels
hostile, these feelings will infiltrate any stance.
Through his experiments with his ‘relaxation
Mutuality technique’ of maternal kindness and indulgence,
Since the concepts of identification with the Ferenczi (6, 9) discovered that patients will often
aggressor and collusion suggest that what patients react primarily to how an analyst feels – even
do and say cannot be understood apart from their unconsciously – and not to how he behaves.
interaction with their analysts, the next implication Ferenczi (1, 5) came to believe that honesty and
is that some kind of mutuality of analysis, openness – the routes through which the patient
involving analyst as well as patient, is necessary. can discover what the analyst feels – are more
Ferenczi’s (5) formal, alternating format of mutual important than kindness in helping patients feel
analysis is not the only way to get at how the safe.
analyst structures the analytic interaction. But How can we foster a necessary atmosphere of
there has to be some kind of forum, both for the safety? First, if our emotional reality will override
open expression of the analyst’s subjectivity and any of our procedures, we need to try to know what
for a way to get at its unconscious aspects. And we need from our patients. Secondly, we should try
patients must be part of that exploration, both in to be genuinely emotionally present, since even
order to add an outside perspective to the analyst’s subtle indications of our emotional absence may be
understanding of himself and because the patient somewhat traumatic for some patients. The ana-
also needs to know something about what is going lyst’s real interest and connection to the patient is
on in the analyst’s mind. perhaps the most important element in healing.
But there are dangers to mutuality in analysis, as Thirdly, if honesty is more important even than
Ferenczi well understood. Patients may be quite kindness or interest, it is better to acknowledge
upset about what the analyst feels or does not feel lapses in either of these than to pretend to feel them
(5, p. 28). Patients also may use mutuality – we must own up to our mistakes. And honesty
sadistically, lording it over an analyst who makes can help restore our emotional presence with the
himself vulnerable. And sometimes an analyst’s patient.
decision to disclose reflects his identification and Finally, despite our good intentions, knowing
82 J. Frankel Int Forum Psychoanal 13, 2004

our own requirements, being emotionally present, attuned to the patient, the patient copes with her
and even being honest will not always happen. We consequent sense of aloneness by taking the blame
may not even know when we are not. So we should and feeling something is wrong with her.
help our patients talk about these dimensions of our
behavior: to encourage them to give voice to their
(perhaps barely expressed) observations and criti- Conclusion: Healing
cisms of us. This dovetails with what I recom- Where a lot of the ideas in this paper lead is that
mended about mutuality. Benjamin Wolstein what patients are after and sometimes express in a
(personal communication) used to remark that the disguised way through symptoms – what people
best supervisor for an analyst on any given case are after, and what a sense of well-being requires –
was his patient. We need to listen to the super- is a feeling of being deeply acceptable to others, or
vision they constantly offer us. at least to one other person, in a genuine way, as
oneself. Jessica Benjamin’s (16) idea of mutual
recognition suggests that the person from whom
Taking Blame and Responsibility we seek acceptance must be someone who also has
Ferenczi said that the most damaging element of value to us. Such recognition can begin to dissolve
identification with the aggressor was victims’ debilitating feelings of shame and unacceptability.
taking on the responsibility for the attack – feeling Analytic recognition is based on the complicated
that they are the bad ones. The resulting long-term work of unmasking the sundry ways the patient
distortion in one’s sense of responsibility plays out identifies with the analyst that deflect the analyst
in the clinical situation. Some patients may feel from discovering the patient’s innermost places.
they should be able to cope with anything and are This kind of hard-won recognition may be the core
weak or deficient if they cannot. Some patients will of the healing that analysis can provide.
be almost eager to ask ‘What’s wrong with me?’,
‘What did I do to provoke the attack?’, ‘How did I
contribute to this problem?’ Yet the genuine References
insights into oneself that can result from these 1. Ferenczi S. Confusion of tongues between adults and the child
questions may have little healing effect, perhaps (1933). In: Balint M, ed. Final contributions to the problems and
because the sense of responsibility is part of an methods of psycho-analysis. London: Hogarth; 1955. pp. 156–
67.
identification with the aggressor and is undermined 2. Freud A. The ego and the mechanisms of defense, revised
by a simultaneous awareness of innocence and a edition. New York: International Universities Press; 1936.
3. Frankel J. Exploring Ferenczi’s concept of identification with the
lingering sense of injustice. aggressor: its role in trauma, everyday life, and the therapeutic
The pursuit of insight can be a way to exonerate relationship. Psychoanal Dial 2002;12:101–39.
others, including the analyst. For instance, one 4. Fairbairn WRD. The repression and the return of bad objects
(1943). In: Psychoanalytic studies of the personality. London:
patient, thinking about why the previous session Routledge & Kegan Paul; 1952. pp. 59–81.
felt bad, went through a long list of his own 5. Ferenczi S. The Clinical Diary of Sandor Ferenczi (1932).
Dupont J, ed. Balint M, Jackson NZ, trans. Cambridge (MA):
neurotic dynamics as possible explanations before Harvard; 1988.
tentatively advancing the idea that perhaps some- 6. Ferenczi S. Child-analysis in the analysis of adults (1931). In:
thing had been ‘off’ with the analyst the previous Balint M, ed. Final contributions to the problems and methods of
psycho-analysis. London: Hogarth; 1955. pp. 126–42.
day. And, of course, it had. Some patients can whip 7. Frankel J. Identification and ‘traumatic aloneness’: reply to
up self-accusations or dismiss themselves as ‘too commentaries by Berman and Bonomi. Psychoanal Dial 2002;12:
159–70.
dramatic,’ ‘too sensitive,’ ‘too dependent,’ or self- 8. Ferenczi S. The unwelcome child and his death instinct (1929).
indulgent. In: Balint M, ed. Final contributions to the problems and methods
The basic tasks of free associating and exploring of psycho-analysis. London: Hogarth; 1955. pp. 102–107.
9. Ferenczi S. The principle of relaxation and neocatharsis (1930).
inner fantasy can become substitutes when the In: Balint M, ed. Final contributions to the problems and methods
patient feels he really needs some response from of psycho-analysis. London: Hogarth; 1955. pp. 108–25.
the analyst (see 15): the patient puts herself on the 10. Frankel J. Ferenczi’s trauma theory. Am Journal Psychoanal
1998;58:41–61.
hotseat in place of the analyst. Some patients may 11. Jones JM. Affects as process: an inquiry into the centrality of
be ‘good patients,’ essentially working in isolation affect in psychological life. Hillsdale (NJ): Analytic Press; 1995.
12. Frankel J. Collusion and intimacy in the analytic relationship. In:
and avoiding any demand on the analyst that might Aron L, Harris A, eds. The legacy of Sandor Ferenczi. Hillsdale,
prove disappointing. Often, when the analyst is not NJ: Analytic Press; 1993. pp. 227–47.
Int Forum Psychoanal 13, 2004 Identification with the aggressor: clinical implications 83
13. Frankel J. Collusion and intimacy. Forms and functions of 14. Hoffman IZ. Ritual and spontaneity in the psychoanalytic
identification in the analytic relationship. Presented at the 23rd process. Hillsdale, NJ: Analytic Press; 1998.
Annual Conference of the Israel Association of Psychotherapy, 15. Balint M. The basic fault: therapeutic aspects of regression.
‘Sandor Ferenczi: The ‘mother’ of modern psychoanalysis and London: Tavistock; 1968.
psychotherapy’. Tel Aviv: Israel; 8 May 1999. 16. Benjamin J. The bonds of love. New York: Pantheon; 1988.

Summaries in German and Spanish

Frankel J. Identifizierung mit dem Aggressor und die Frankel J. Identificación con el agresor y los” traumas
“normalen Traumata”: Klinische Implikationen normales”: las implicaciones Clı́nicas.
Ferenczis überraschend wenig bekanntes Konzept der Identi- El concepto de Identificación con el agresor de Ferenczi
fizierung mit dem Aggressor hat grundlegende Konsequenzen (1933) tan sorprendentemente desconocido- una victima del
für unser Verständnis der analytischen Technik – gerade bei abuso “eliminando” su propia subjetividad y “convirtiéndose”
Opfern eines Missbrauchs, bei denen die eigene Subjektivität precisamente en lo que un asaltante necesita que sea- tiene
“verschwindet”, und die genau so “werden” wie sie der Täter implicaciones importantes para nuestra comprensión de la
haben will. Das häufige Auftreten dieser psychischen Dyna- técnica analı́tica. Su frecuente ocurrencia nos obliga a ampliar
mik legt uns nahe, unser Verständnis davon zu erweitern, la comprensión sobre lo que constituye el trauma. Ferenczi
woraus ein Trauma entsteht. Ferenczi betrachtete als Schlüs- vio la experiencia de “soledad traumática” (traumatic alone-
selelemente des Traumas das Erleben von “traumatischer ness) o “abandono emocional” (emocional abandonment)
Einsamkeit” und “emotionaler Verlassenheit”. Diese bewir- como el elemento clave del trauma, como lo que obliga a las
ken die charakteristischen traumatischen Reaktionen wie respuestas traumáticas de disociación e identificación con el
Dissoziation und Identifizierung mit dem Aggressor. Die agresor. La identificación con el agresor opera en la relación
Identifizierung mit dem Aggressor ist in der analytischen analı́tica entre paciente y analista. Tiene varias consecuencias
Beziehung in beiden Beteiligten wirksam, im Patienten und incluyendo la estructuración de la relación a través de la
im Analytiker. Eine von mehreren Konsequenzen ist die confabulación mutua coordinada, identificaciones defensivas
Strukturierung der Beziehung durch unbewusste Kollusionen. designadas para ayudar a ambos participantes a sentirse
Diese gegenseitigen der Abwehr dienenden Identifizierungen seguros. Este punto de vista de la relación analı́tica tiene
helfen den beiden Beteiligten sich sicher zu fühlen. Dieses implicaciones clı́nicas en al menos cuatro áreas: la compren-
Verständnis der analytischen Beziehung wirkt sich in sión de las asociaciones libres del paciente, las cuales pueden
wenigstens vier Bereichen aus: 1. Die freien Assoziationen reflejar la complacencia del paciente con los deseos del
des Patienten drücken öfters weniger das Unbewusste des analista más que los contenidos del propio inconsciente del
Patienten aus sondern das Eingehen des Patienten auf die paciente; la necesidad de algn tipo de mutualidad de análisis;
Wünsche des Analytikers. 2. Die Notwendigkeit einer el potencial traumatizante de la autoridad del analista; y la
gewissen Gegenseitigkeit in der Analyse. 3. Das traumatisier- tendencia de algunos pacientes de sentirse culpables y
ende Potential der Autorität des Analytikers. 4. Die Tendenz reflexivamente responsables, como una forma de proteger al
von gewissen Patienten, Schuld und Verantwortung auf sich analista.
selbst zu beziehen, um auf diese Weise den Analytiker zu
schützen.