By Jonas Knatz and Nuala Coamhánach

Ruth Leys is Professor Emerita of the Humanities at Johns Hopkins University. Her work has focused on aspects of the history and theory of psychoanalysis; the history of psychiatry; the history of the neurosciences; trauma theory; the mind-body problem; and the history of approaches to the emotions.

Jonas Knatz & Nuala Caomhánach: In 2000, you published Trauma: A Genealogy, the first book in what you consider a trilogy of works on certain crucial aspects of the history of the human sciences. You continued with the publication of From Guilt to Shame: Auschwitz and After in 2007 and concluded with The Ascent of Affect: Genealogy and Critique in 2017. In Trauma you argue that the concept of trauma was, from its origins in late 19th-century psychology, inherently unstable and aporetic, because it was informed by two apparently mutually exclusive paradigms: a “mimetic” understanding of trauma and its “anti-mimetic” counterpart. Could you quickly define what you mean by these two paradigms and illustrate how they featured in Sigmund Freud’s understanding of trauma?

Ruth Leys: The word trauma derives from the Greek for “wound” and originally referred to a surgical wound. It was only in the late 19th century that the concept of trauma started to refer to a psychic process. This development was associated with the rise of hypnosis as an instrument to probe the lives of hysterics in the hands of the great French psychiatrist Jean-Martin Charcot, and more theoretically as a constitutive feature of the traumatic situation.

The theoretical idea was that, in the moment of trauma, the victim is suddenly precipitated into a state of suggestive imitation or identification with the scene or perpetrator. This produces a state of hypnotic dissociation that protects the victim from the horror of the situation by driving her into a kind of mimetic loss of consciousness, but also prevents her from remembering the traumatic event afterwards. The trauma victim was thus compared to a post-hypnotic subject who cannot subsequently remember what she did when, in a state of unconsciousness, she imitatively and blindly followed a hypnotic command.

On this “mimetic” hypothesis, trauma was thus understood as a situation of (physical or moral) violence that hypnotically immerses the victim in the scene so deeply that it precludes the kind of spectatorial distance necessary for knowledge of what has happened. The mimetic theory had the virtue of explaining the commonly-reported loss of memory in victims of trauma, while also providing a sympathetic account of why the victim was so disabled in this way.

But the mimetic theory also produced, as a troubling consequence of its assumptions, doubts as to the veracity of the subject’s testimony about the event. This is because her dissociated state at the time of trauma was thought to preclude her ability to tell the truth about what had happened. Indeed, because the trauma victim was imagined as thrown into a state of terrorized imitation of, or unconscious identification with, the perpetrator, the mimetic theorist could not help worrying about the problem of confabulation. Even more problematically, since the victim was imagined as imitatively incorporating the perpetrator’s violence, the mimetic theory assumed a certain complicity on the part of the victim who was seen to share the hostility directed against herself. At the time, the tendency of the subject to mimetically identify with the perpetrator was called “identification with the aggressor.” Today it is known as the “Stockholm Syndrome.”

Ruth Leys, Trauma: A Genealogy, University of Chicago Press (2000)

For these and related reasons, an alternative “anti-mimetic” theory was posited, according to which a certain kind of imitation or identification remained basic to the traumatic experience, but the imitation or identification was theorized differently. Instead of imagining that the victim of trauma was so immersed in the scene of violence as to lose all sense of herself, she was conceptualized as remaining aloof from the event. Although she imitatively succumbed to the situation, she remained a spectator of the scene, which she could therefore see and represent to herself. In principle she could therefore testify to what had happened to her, even if the therapeutic process of helping the patient recover and narrate the memory of the trauma as means to a cure might be long and arduous.

The result of the anti-mimetic theory was a tendency to relegate the problem of imitative-hypnotic identification to a secondary position in order to establish a strict dichotomy between the autonomous subject and the external event or perpetrator.  The anti-mimetic theory was compatible with and often gave way to the idea that trauma is a purely external event or act of violence that befalls a passive subject. The anti-mimetic theory had the advantage not only of envisaging the victim of terror as capable in principle of conveying the truth of what had occurred to her but as in no way mimetically collusive with the assaultive violence, even as the absence of complication as regards her testimony shored up the idea of the unproblematic actuality of the traumatic event. The anti-mimetic theory has also lent itself to various positivist interpretations of trauma epitomized by the turn to neurobiological explanations that are widely accepted today.

Sigmund Freud, c.1921. By Max Halberstadt. Courtesy WikiCommons.

(Sigmund Freud (1856-1939), WikiCommons)

In my book on trauma I emphasized the fact that there has always been an oscillation between the mimetic and anti-mimetic theories. The result is that from its inception the concept of trauma has been fundamentally unstable, balancing uneasily, or veering uncontrollably, between the mimetic and anti-mimetic poles. Both the mimetic and the anti-mimetic accounts of trauma have advantages and disadvantages, which is why neither can be completely embraced or completely abandoned. So tortuous has been the relationship between the two approaches that one frequently finds both represented in the very same text.

This is especially true of Freud, whose complex attitude toward the problem of hypnosis and suggestion pervades his treatments of the topic. One of the aims of my book was to bring out how deep Freud’s engagement with hypnosis-suggestion was, and how “aporetic” his texts were when he theorized trauma. He, too, oscillated between the mimetic and anti-mimetic poles of analysis, sometimes treating the victim of trauma in mimetic terms as so immersed in the trauma as to be unable in principle to remember it, so that his “talking cure” with its goal of getting the patient to narrate the past was doomed to failure, and sometimes treating the victim in anti-mimetic terms.

JK & NC: In 1980, the third edition of the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorder (DSM-III) adopted, under the pressure of psychiatrists, social workers, activists and Vietnam war veterans, a definition of Post-traumatic stress disorder (PTSD) to account for traumatic symptoms that appeared long after the traumatizing event. This definition, as Allan Young writes, had been “glued together by the practices, technologies, and narratives with which it is diagnosed, studied, treated, and represented and by the various interests, institutions, and moral arguments that mobilized these efforts and resources.” (p.5). Yet, this lack of definitional cohesion does not seem to have stopped trauma from becoming the “signum of our times” (p.194), as Wulf Kansteiner critically remarks in his review of literature on trauma in the 20th century. How did the definition of PTSD change the concept of trauma and how do you explain the interest that PTSD generated?

RL: Allan Young is quite right that PTSD was stitched together in 1980 in ways that left the diagnosis open to constant revisions in subsequent editions of the DSM. PTSD changed the concept of trauma in the sense that it firmly oriented research in an anti-mimetic direction. It posited the trauma as an event “outside ordinary human experience” that assaulted the autonomous subject entirely from the outside, and although DSM-III purported to be atheoretical, it encouraged scientists to rethink PTSD in neurobiological term on the grounds that at bottom mental disorders have organic causes, not psychological ones.

In my book on trauma I focus especially on the contributions of the psychiatrist Bessel van der Kolk because his views are characteristic in this regard. Van der Kolk claimed that although the trauma patient suffers from traumatic forgetting, nevertheless the imprint of the event is preserved in a hidden memory with timeless accuracy; this traumatic memory is not available to conscious recall but causes the long-term effects of PTSD. In particular, van der Kolk suggested that the traumatic event is encoded in the brain in a different way from ordinary memory. Traumatic memory, on this model, is less like “declarative” or “narrative” memory involving the subject’s ability to consciously remember and narrate events of the past, and more like “implicit” or “nondeclarative” memory, involving bodily responses that lie outside conscious linguistic-semantic-representation and recall. When traumatic memories belatedly return to haunt the patient they do so in the form of repetitive nightmares, flashbacks and other reenactments that are held to be completely true to the event.

Why has PTSD—or trauma more generally—caught the imagination of our times? There are several reasons for this: the rise of interest in trauma is “over-determined,” as Freud might put it. There is the fact of the Vietnam War, which left angry American veterans back home stranded in a society that had turned against the war and denied the returning soldiers recognition and support. Then there was the belated recognition of the Holocaust, whose traumatic effects had been recognized prior to 1980 in the post-war diagnosis of the “survivor syndrome,” but an accounting of whose psychiatric consequences had not yet been integrated with the literature on shell-shock and the war neuroses in WW1 and WW2 until the efforts of DSM-III brought them together for the first time. These developments within psychiatry gave salience to the idea of trauma.

There was also the influence of the women’s movement and the rise of interest in child sexual abuse which coincided with these events. Ian Hacking has written brilliantly on aspects of the genealogy of the concept of child abuse. The post-WW2 resurgence of interest in dissociation and multiple personality was also part of this story too.

Yet another reason why trauma has captured the imagination of so many people, especially in academia, is the fascination with the idea of what can’t be said or narrated or represented in trauma. In scientific circles this takes the form of emphasizing the difficulty the victim has in telling the story of the traumatic event and thereby coming into conscious possession of her past.

In more literary and cultural circles, interest in the non-narratable aspect of trauma has taken the post-modernist, deconstructive form of the idea of trauma as an unrepresentable event resulting from a general breakdown of all representation. It is striking in this regard that the post-modernist literary critic, Cathy Caruth, has acknowledged the convergence of her ideas about the failure of representation in trauma and the neurobiological work of recent scientists, such as Bessel van der Kolk. She approvingly cites his findings in her interpretation of trauma, and van der Kolk returns the compliment by citing her work. She interprets the flashback symptoms and repetitive nightmares that are held to be characteristic of PTSD indicate the existence of a kind of haunting of the victim by past traumas that are “unclaimed” (as in the title of her book, Unclaimed Experience: Trauma, Narrative, and History, 1996). As I have tried to show in Trauma, both offer accounts of trauma that lend themselves to facile speculation about the complete breakdown of meaning and representation in our post-Holocaust times. I am not a fan of such claims, and the aim of my chapters on van der Kolk and Caruth in my book on trauma was to dismantle their arguments.

JK & NC: In your book you argue that van der Kolk’s studies are “haunted by the same problem of mimetic suggestibility that the theory is designed to forestall.” This critical remark betrays your skeptical conclusion: that, despite the technological developments and increases in funding of neurobiological research, “the debates over trauma are fated to end in an impasse.” Could you explain your skepticism towards neurobiological trauma research and why it attracts not only considerable public attention but also high levels of funding?

RL: A key feature of van der Kolk’s approach is the idea that because the trauma victim is not able to process the traumatic experience in a normal way, the event leaves a “reality imprint” in the brain that, in its insistent literality, testifies to the existence of a timeless traumatic truth uncontaminated by subjective meaning or unconscious symbolic elaborations. His theory is anti-mimetic in the sense that it seeks to avoid the issue of psychic imitation and identification with the perpetrator or traumatic scene that had informed and troubled earlier psychoanalytic interpretations of trauma– even as, in a return of the mimetic problematic, psychiatrists have been obliged to acknowledge the tendency of PTSD patients to be highly hypnotizable in ways the “atheroetical” DSM has been helpless to explain.

Van der Kolk’s aim has been to identify the neurohormonal-brain bases of such untold memories and to develop methods of treatment designed to alleviate the patient’s haunting symptoms. The theory of trauma as literally and timelessly encapsulated in a special memory system has gained plausibility by the attempt to frame it in neurobiological terms, given the authority wielded by the biological sciences in contemporary American psychiatry. While van der Kolk has been remarkably effective in this regard, Allan Young among others has offered powerful critiques of the scientific evidence on offer, suggesting that the experiments supposedly proving the biological theory have failed to produce the predicted outcomes and have necessitated attempts to “save the phenomena” in various, mutually incompatible ways. Nevertheless, the anti-mimetic approach to trauma as exemplified by van der Kolk’s work has continued to garner widespread support. It’s as if the entire medical-psychiatric culture wants to find biological solutions to psychological problems, and van der Kolk’s and related attempts to prove the neurobiological basis of psychic trauma fit the bill.

JK & NC: In the second book of your trilogy, From Guilt to Shame, you trace how the discourse of guilt in the assessment of Holocaust survivors has slowly been replaced by a discourse of shame. You demonstrate that the notion of survivor guilt had been central to postwar psychoanalytic works, such Bruno Bettelheim’s famous studies on concentration camps and William Nederland’s and Henry Kristal’s work, which fought for the acknowledgement of traumas in Holocaust survivors and their right for reparation against the Federal Republic of Germany. Yet the psychoanalytic notion of survivor guilt came under sustained criticism for allegedly making the victims complicit in the Holocaust. As in your book on trauma, you argue that this dispute over survivor guilt concerns the tension between mimetic and anti-mimetic approaches to trauma. What role did this mimetic understanding play in early theories about survival in the camps and why did shame become the central emotion for anti-mimetic theories?

RL: After finishing my Trauma book, I turned my attention to a new topic that intrigued me. When in 1980 Post-Traumatic Stress Disorder was first introduced in DSM-III, survivor guilt was listed as one of the diagnostic criteria of the condition. But in the subsequent revision of DSM-III in 1987 (DSM-IIIR), survivor guilt was dropped from the list of criteria and demoted to merely an associated feature of the condition.

I thought I knew why. It was because the official diagnosis of PTSD occurred at the moment American psychiatry abandoned a somewhat watered-down psychoanalytic template for understanding and organizing mental diseases and reverted instead to a descriptive-classificatory approach associated in the past with the work of the influential German psychiatrist Emil Kraepelin. The aim of the classificatory approach was to provide an inventory of mental disorders grouped according to distinct, differentiating diagnostic criteria and associated non-criterial features and symptoms. Survivor guilt had had a long history of association with trauma and stress as a term for the victim’s feeling of guilt, whether for surviving when others had died, as in the case of the Holocaust survivors, or for feeling somehow responsible for the violence or abuse victims had experienced at the hands of others, as in cases of rape and other acts of abuse. The feeling of guilt had been explained in psychoanalytic terms as the outcome of the victim’s imitative identification with the aggressor, understood as an unconscious process of psychic assimilation of the perpetrator’s violence- violence that, in its incorporated form, was held to be directed against the innocent victim herself and felt as a self-reproach produced by the lacerating superego or conscience.

But in the wake of the 1980 revision of DSM-III and the abandonment of the previous psychoanalytic framework, survivor guilt no longer had an obvious rationale, and it was therefore not surprising that in the next revision or PTSD, in DSM-IIIR (1987), it was relegated to merely an “associated feature” of the disorder. The decision to downgrade survivor guilt as a diagnostic criterion for PTSD may be seen as exemplifying the anti-mimetic turn in American psychiatry, because by downgrading survivor guilt appeals to identification with the aggressor that had explained it were likewise demoted, enabling instead the kinds of neurobiological theories advocated by van der Kolk and others.

What emotion would take its place? It soon became obvious to me that the affect that had begun to take the place of survivor guilt in discussions of trauma– but not just in discussions of trauma– was the affect of shame. At roughly the same moment, post-PTSD psychiatrists decided that shame was a more important affect than guilt in trauma; the influential post-modernist literary critic Eve Kosofsky Sedgwick rejected guilt as the defining affect of subject formation and turned instead to the role of shame in the formation of queer identity; and the philosopher Giorgio Agamben likewise rejected notions of survivor guilt in favor of shame to discuss the nature of testimony in the concentration camp experience.

The wager of my book From Guilt to Shame: Auschwitz and After then became to show what these apparently disparate authors had in common. My answer was twofold: that that these authors’ shift from guilt to shame reflected the general move from mimesis to anti-mimesis that had already occurred in the psychiatric sciences with the arrival of DSM-III and PTSD; and that the shift also reflected a preference for a non-intentionalist rather than an intentionalist approach to the affects.

Ruth Leys, From Guilt to Shame: Auschwitz and After, Princeton University Press, (2007).

In the first place, as I’ve already observed, in the pre-PTSD diagnosis writings of psychiatrists on the survivor syndrome, the concentration camp survivor’s sense of guilt for having survived when others had died was explained in mimetic terms as the outcome of an unconscious imitation of, or identification with, the aggressor. Since in camp conditions of abject powerlessness, the prisoner’s incorporated feelings of aggression could not be directed against the powerful perpetrator, the violence was turned back against the victim, who experienced it in the form of a lacerating feeling of guilt and remorse. But today’s shame theory conforms to the anti-mimetic pole of trauma theory because it displaces attention from the guilty subject’s unconscious yielding to the enemy to the shamed subject’s anti-mimetic awareness of being seen. As a result, shame theory downplays the mimetic interpersonal dynamic central to the formulation of guilt in order to portray shame as an expression of the consciousness of the self when an individual becomes aware of the disapproving gaze of another person. As I put this in my book, “shame enacts the shift from the mimetic to the antimimetic by emphasizing the realm of the specular.”

Second, the shift from the mimetic to the anti-mimetic was also a shift from an intentionalist to a non-intentionalist interpretation of the affects. It is generally agreed that guilt concerns your actions, which is to say, what you do– or what you wish or fantasize you have done, because according to Freud the unconscious does not distinguish between the intention and the deed, between the virtual and the actual. Equating intention with the deed, psychoanalysis maintains the link to intention and action that has been held to be intrinsic to the notion of guilt. Shame, however, is held to concern not your actions but rather who you are, that is your deficiencies as a person as these are revealed to the shaming gaze of the other. This a shift of focus from actions to the self makes the question of personal feeling and identity of paramount importance.

Moreover, as part of this shift of focus, many theorists define the affects, including shame, in non-intentionalist, materialist terms. According to them, shame and the other emotions are the result of evolved and inherited states of the brain and body that are independent of intentions and meanings because, in the spirit of Silvan Tomkins and Paul Ekman on whose ideas Sedgwick and other new shame theorists lean (see the second part of this interview for further clarification), they are “triggered” by what might be called emotional objects– but these objects are nothing more than stimuli or tripwires for an inbuilt behavioral response. Even though Giorgio Agamben has no interest in current psychological theories of the kind adopted by Tomkins, Ekman, Sedgwick and others, I try to show that he, too, offers a materialist, non-intentionalist account of shame. The overall import of these developments is an account of shame that makes questions of agency, intention, and belief beside the point and privileges instead issues of personal identity and personal difference.


This is the first installment of a two-part interview with Ruth Leys about her work on the history of the human sciences. Continue reading here.

Jonas Knatz is a PhD Student in New York University’s History Department. He works on 20th-century European intellectual history

Nuala Caomhánach is a PhD Student in New York University’s History Department. She works on 19th and 20th century History of Science and Environmental History.

Featured Image: Jean Fautrier, Les Massacrés, c. 1945.