Book Search:  


Google full text of our books:


From Guilt to Shame:
Auschwitz and After
Ruth Leys

Book Description | Endorsements | Table of Contents

COPYRIGHT NOTICE: Published by Princeton University Press and copyrighted, © 2007, by Princeton University Press. All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means (including photocopying, recording, or information storage and retrieval) without permission in writing from the publisher, except for reading and browsing via the World Wide Web. Users are not permitted to mount this file on any network servers. Follow links for Class Use and other Permissions. For more information, send e-mail to

This file is also available in Adobe Acrobat PDF format


From Guilt to Shame

WHAT is the logic of torture? In an article on prisoner abuse at Abu Ghraib prison in Iraq, Mark Danner has shown that the methods used to soften up and interrogate detainees by American military personnel can be traced back to techniques developed by the CIA in the 1960s. The best known manual of such procedures, the CIA’s Counterintelligence Interrogation of Resistance Sources, produced in 1963 at the height of the Cold War, states that the purpose of all coercive techniques of interrogation is “to induce regression.” The result of external pressures of sufficient intensity is the loss of those defenses “most recently acquired by civilized man . . . Relatively small degrees of homeostatic derangement, fatigue, pain, sleep loss, or anxiety may impair these functions.”1 The programmatic manipulation and control of the environment, including the use of blindfolds or hooding, sleep and food deprivation, exposure to intense heat and cold, sensory deprivation, and similar methods, are meant to disorient the prisoner and break down resistance. “Once this disruption is achieved,” a later version of the manual observes, the subject’s resistance is “seriously impaired.” He experiences a “kind of psychological shock” as a result of which he is far more open to suggestion and far likelier to comply with what is asked of him than before. Frequently the subject will experience a “feeling of guilt.” If the interrogator can intensify those guilt feelings, it will “increase the subject’s anxiety and his urge to cooperate as a means of escape.”2 Viewed in this light, Danner remarks, the garish scenes of humiliation documented in the photographs and depositions from Abu Ghraib “begin to be comprehensible; they are in fact staged operas of fabricated shame, intended to ‘intensify’ the prisoner’s ‘guilt feelings, increase his anxiety, and his urge to cooperate’” (“LT,” 72).

The terms of Danner’s analysis imply that there has existed a single logic of torture extending uninterruptedly from the 1960s to the occupation of Iraq, according to which there is no important distinction to be drawn between the emotions of guilt and shame. Yet if we focus on the details of the manuals, reports, and protocols to which he has so usefully drawn our attention, differences become apparent. The CIA’s approach to interrogation in 1963 was largely based on a watered-down Freudianism that emphasized the psychic relationship between prisoner and interrogator, especially the tendency of the latter to assume the role of a parental figure with whom the prisoner might unconsciously identify in an ambivalent and guilty manner. The 1963 manual says that its procedures aim not only to “exploit the resistant source’s internal conflicts and induce him to wrestle with himself” but also to bring a superior outside force to bear upon his resistance. In other words, by virtue of his role as the sole supplier of satisfaction and punishment the interrogator seeks to assume the stature and importance of a paternal figure in the prisoner’s feelings and thoughts. Although there may be “intense hatred” for the interrogator, it is not unusual for the subject also to develop “warm feelings” toward him. Such ambivalence is the basis for the suspect’s guilt reactions, and if the interrogator nourishes those feelings of guilt, they may prove strong enough to influence the prisoner’s behavior. “Guilt makes compliance more likely.”3 The ultimate goal of the physical abuse and other manipulative techniques described in the manual is thus the production of a docile, compliant, and guilt-wracked prisoner so regressively bonded with his interrogator as to be willing to confess. We might define this 1963 logic of torture as an identificatory logic of guilt. According to the manual, hypnosis, suggestion, and narcosis may serve the same purpose, since they too are capable of inducing a regressive identification with the interrogator. Even the efficacity of pain is understood to depend on the prisoner’s guilty attitudes. One telling detail in the manual is the advice that if audio and video recording devices are to be used, the subject should not be conscious that he is being recorded—as if the psychological dynamic between prisoner and interrogator conducive to successful interrogation can only develop when the captive is unaware of being seen or overheard by someone other than the interrogator.

Contrast this with the implicit logic of torture at Abu Ghraib forty years later. All the methods that have been described in the current scandal are designed to publicly humiliate and shame the prisoner. An American military pamphlet instructing troops on Iraqi sensitivities warns against shaming or humiliating a man in public, since shaming will cause him and his family to be anti-Coalition. According to the pamphlet, the most important qualifier for all shame is for “a third party to witness the act.” It cautions that if an American must do something likely to cause an Iraqi shame, he should “remove the person from the view of others.” Acts such as placing hoods over a detainee’s head, placing a detainee on the ground, or putting a foot on him should be avoided because they cause Arabs shame. Likewise, the pamphlet says, Iraqis consider a variety of things to be unclean: “‘Feet or soles of feet. Using the bathroom around others. Unlike Marines, who are used to open-air toilets, Arab men will not shower/use the bathroom together. Bodily fluids’” (“LT,” 72).

As Danner observes, these precepts are emphatically reversed at Abu Ghraib. It is precisely because such methods induce shame that they have been exploited there and at other American interrogation sites, where detainees have been kept hooded and bound, made to crawl and grovel on the floor, forced to put shoes in their mouths, and worse. “And in all of this, as the Red Cross report noted, the public nature of the humiliation is absolutely critical: thus the parading of naked bodies, the forced masturbation in front of female soldiers, the confrontation of one naked prisoner with one or more others, the forcing together of naked prisoners in ‘human pyramids’” (“LT,” 72). The torture carried out at Abu Ghraib almost seems to demand what was counterindicated in 1963, the open use of the camera, because shame depends on the subject’s consciousness of exposure. As Danner again notes: “And all of this was made to take place in full view not only of foreigners, men and women, but also of that ultimate third party: the ubiquitous digital camera with its inescapable flash, there to let the detainee know that the humiliation would not stop when the act itself did but would be preserved into the future in a way that the detainee would not be able to control” (“LT,” 72). As a “shame multiplier” (“LT,” 72) the camera epitomizes the logic of torture at Abu Ghraib, which can be defined as a spectatorial logic of shame.4

In one sense, the resort to shaming techniques may represent a specific adaptation to the Arab context. But it is also true that the shift from a logic of torture based on guilt to a logic of torture based on shame reflects a more general shift that has taken place in the course of the last forty years from a discourse of guilt to a discourse of shame. It is not just a question of assuming, as anthropologists used to do, that the Iraqis belong to a more primitive “shame culture” than our own Western “guilt culture.” Today, shame (and shamelessness) has displaced guilt as a dominant emotional reference in the West as well. A major purpose of my book is to examine and evaluate that displacement. In his recent study of humiliation and associated emotions, such as shame and embarrassment, William Ian Miller has noted the recent depreciation of guilt and resurgence of interest in shame among the self-help and related disciplines, but disputes the idea that a major paradigm shift has really occurred. He claims that “the makeover makes shame look not at all unlike guilt.”5 I disagree. For all the interest of his study, Miller fails to see what is original and important about shame theory today, and misconstrues the stakes involved in the upsurge of books and articles that take shame as their primary point of reference. I argue instead that the change from a culture of guilt to a culture of shame in Western thinking about the emotions is highly significant and has important consequences.

My story begins with the centrality of guilt to post–World War II assessments of survivors of the concentration camps. The terms used in the CIA’s 1963 training manual for the interrogation of resistant detainees bear an uncomfortably close proximity to those used by victims and researchers alike in the same postwar period to describe the psychodynamics of the tortured and shocked survivors of the Holocaust.6 Giorgio Agamben has recently observed in this regard that the survivor’s feeling of guilt is a locus classicus of the literature on the camps.7 “That many (including me) experienced ‘shame,’ that is, a feeling of guilt during the imprisonment and afterward, is an ascertained fact confirmed by numerous testimonies. It is absurd, but it is a fact,” Primo Levi observes in his last (and most troubled) book about his time in Auschwitz.8 Similarly the psychoanalyst Bruno Bettelheim, who was imprisoned in Dachau and Buchenwald in 1938–39, writes: “One cannot survive the concentration camp without feeling guilty that one was so incredibly lucky when millions perished, many of them in front of one’s eyes . . . In the camps one was forced, day after day, for years, to watch the destruction of others, feeling—against one’s better judgment—that one should have intervened, feeling guilty for not having done so, and, most of all, feeling guilty for having also felt glad that it was not oneself who perished.”9 Or in the words of Elie Wiesel, cited by Agamben: “‘I live, therefore I am guilty. I am here because a friend, an acquaintance, an unknown person died in my place’” (RA, 89). And in a statement also cited by Agamben, Ella Lingens asks: “‘Does not each of us who has returned go around with a guilt feeling, feelings which our executioners rarely feel—‘I live, because others died in my place?’” (RA, 89).

In attempts during the 1960s to explain the phenomenon of survivor guilt, American psychoanalysts such as William Niederland, Henry Krystal, Bettelheim, and others borrowed from the work of Freud, Sandor Ferenczi, and Anna Freud in theorizing that the guilt feelings associated with survival were the result of an unconscious imitation of, or identification with, the aggressor. They argued that the humiliated prisoner, in the moment of shock, regressively defends against the persecutor’s violence by unconsciously yielding to, or imitatively incorporating, the violent other. And since under camp conditions of abject powerlessness the incorporated aggression cannot be projected onto the aggressor, the violence is turned back against the victim, who experiences it in the form of a self-lacerating conscience. In short, from the start the notion of survivor guilt was closely connected to the theme of imitative identification and to the idea of the victim’s defensive, unconscious bond of collusion with the situation of terror. In his highly influential discussion of the “gray zone,” where “the two camps of masters and servants both diverge and converge” (DS, 42), Primo Levi—no admirer of psychoanalysis—likewise explored the question of the unconscious identification, “mimesis,” or imitation of the aggressor that complicitously binds the victim to the violence directed against himself (DS, 48).

But it is precisely because of the taint of collusion associated with the notion of survivor guilt that almost from the start objections have been raised against it. To take one influential example, Terrence Des Pres, in his widely admired book The Survivor (1976), repudiated both the notion of identification with the aggressor and that of survivor guilt, emphasizing instead the role played in the victim’s survival by social bonding, mutual care, and outright resistance. According to Des Pres, if imitation of the SS took place in the camps, as Bettelheim and others claimed, the imitation was not an unconscious, collusive identification with the enemy but merely a strategic mimicry undertaken consciously by political prisoners in order to obtain positions of power and to assist other victims in the struggle for life.10 A similar repudiation of the notion of unconscious imitation and survivor guilt marks the more recent literature on trauma. In 1980, when the diagnosis of Posttraumatic Stress Disorder (PTSD) was introduced into the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders (DSMIII), survivor guilt feelings were regarded as a characteristic symptom of the disorder and were included in the list of diagnostic criteria.11 But in the revised edition of the manual of 1987 (DSM-IIIR), the American Psychiatric Association after considerable controversy downgraded survivor guilt to the status of an “associated” and noncriterial feature of the condition.12 As we shall see, now that survivor guilt has disappeared from the official list of criteria for PTSD, shame has come to take its place as the emotion that for many investigators most defines the condition of posttraumatic stress.

In a similar movement, literary critic Lawrence Langer, known for his analyses of Holocaust video testimony, has rejected the notion of survivor guilt, not only because he thinks it deflects blame from the real culprits onto the victims themselves, but more generally because it belongs to what he regards as a normalizing, therapeutic, redemptive approach to the misery of the Holocaust that estranges us from the ultimately incomprehensible and unredeemable reality of the camps. This leads him to call for a post-Holocaust revision of ethics that would go beyond the dilemmas and contradictions posed by the unheroic “choiceless choices” that ruled the victims of the Nazis.13 Although it is not clear what Langer thinks an alternative, post-Holocaust ethics might look like, he hints at one direction to follow when he suggests that “shame” might be a better word than the troubling concept of “guilt” for the anguish experienced by the survivor. Embracing a distinction between guilt and shame that Primo Levi himself does not observe, Langer states that Levi did not enjoy using a word like “guilt” when raising the problematic topic of collaboration. In the end, he claims, Levi preferred to speak of “shame” as the “primary legacy of the moral swamp into which German coercion had sunk its prey.”14 Agamben makes a similar gesture when he criticizes the notion of survivor guilt and rejects as “puerile” Levi’s self-reproaches for minor wrongs committed by him during his time in the camps. Agamben suggests that the reader’s supposed unease with Levi’s writings on this topic can only be a reflection of the survivor’s embarrassment at being unable to master shame (RA, 88). Agamben’s larger claim, which goes far beyond Langer, is that the concentration camps were nothing less than an “absolute situation” that revealed shame to be “truly something like the hidden structure of all subjectivity and consciousness” (RA, 128).

Although Langer and Agamben represent different approaches to the Holocaust and stand for different ideas about shame, the general privilege they accord to shame over guilt can be situated in the context of a broad shift that has recently occurred in the medical and psychiatric sciences, literary criticism, and even philosophy away from the “moral” concept of guilt in favor of the ethically different or “freer” concept of shame. Today’s “vogue of shame,” to use Christopher Lasch’s phrase, is manifest not only in the work of the American Psychiatric Association and books by Langer and Agamben.15 It is also apparent in such widely disparate texts as Bernard Williams’s Shame and Necessity (1993), a study of ancient Greek tragedy and thought; Eve Kosofsky Sedgwick and Adam Frank’s Shame and Its Sisters (1995), an anthology of texts by the psychologist Silvan Tomkins (1911–89), who from the 1960s through the 1980s advocated a nonpsychoanalytic affect theory centered on shame; Sedgwick’s more recent collection of essays on a variety of topics, including shame, Touching Feeling: Affect, Pedagogy, Performativity (2003); Joseph Adamson and Hilary Clark’s literary-critical anthology, Scenes of Shame: Psychoanalysis, Shame, and Writing (1999), which presents a series of studies of the role of shame in nineteenth- and twentieth-century literature, making use of contemporary shame theory; Jacqueline Rose’s On Not Being Able to Sleep: Psychoanalysis and the Modern World (2003), which deals with the role of shame in South Africa’s Truth and Reconciliation Commission, as well as in the cult of celebrity; Martha C. Nussbaum’s Hiding from Humanity: Disgust, Shame, and the Law (2003), a critique of the recent legal use of shaming sanctions; Elspeth Probyn’s Blush: Faces of Shame (2005), a cultural studies approach to the psychology and politics of identity; and three somewhat earlier books, Shame: The Power of Caring (1980) by Gershen Kaufman, which was the first text to introduce Tomkins’s affect theory into psychotherapy, The Mask of Shame (1981) by psychoanalyst Leon Wurmser, and The Many Faces of Shame (1987) by Donald L. Nathanson, Tomkins’ best-known follower in the psychotherapeutic domain. Each of these works—and there are many others—posits a clear differentiation between guilt and shame in order to make use of shame theory for various philosophical, postpsychoanalytic, postmodernist, and political projects and critiques. It is a measure of how much has changed that the author of a recent biography of Bruno Bettelheim, one of the architects in the United States of the postwar concept of survivor guilt, treats Bettelheim’s deeply covered-over feelings of shame, not guilt, as the key to the self-doubt and sense of fraudulence that haunted him throughout his life and career.16

The original intuition informing the present book was that the current tendency to privilege shame over guilt could at least partly be understood in terms of the perennial conflict between the “mimetic” and “antimimetic” tendencies internal to trauma theory, as those terms are defined and tracked historically in my Trauma: A Genealogy (2000).17 In that book I argued that from the moment of its invention in the late-nineteenth century, the concept of trauma has been fundamentally unstable, balancing uneasily, or veering uncontrollably, between two antithetical poles or theories. The first, or mimetic theory, holds that trauma, or the experience of the traumatized subject, can be understood as involving a kind of hypnotic imitation of or regressive identification with the original traumatogenic person, scene, or event, with the result that the subject is fated to act it out or in other ways imitate it. Trauma is understood as an experience of violence that immerses the victim in the scene so profoundly that it precludes the kind of specular distance necessary for cognitive knowledge of what has happened. The mimetic theory explains the tendency of traumatized people to compulsively repeat their violent experiences in nightmares or repetitive forms of acting out by comparing the traumatic repetition to hypnotic imitation. Trauma is therefore interpreted as an experience of hypnotic imitation and identification that disables the victim’s perceptual and cognitive apparatus to such an extent that the experience never becomes part of the ordinary memory system. This means that the amnesia held to be typical of psychical shock is explained as a kind of posthypnotic forgetting.

An aspect of the mimetic theory that should be stressed—and indeed is featured in the CIA’s discussion of the resistant source’s identification with the interrogator—is that mimesis or unconscious imitation leads to doubts about the veracity of the subject’s testimony, since the identificatory process is thought to take place outside of, or dissociated from, ordinary awareness. Because the victim or detainee is imagined as thrust into a state of suggestive-hypnotic imitation, the mimetic theory cannot help worrying about confabulation, or the problem of testimonial authenticity. Finally, since the mimetic theory posits a moment of terrorized identification with the aggressor, prisoners are imagined as incorporating and therefore complicitously sharing the hostility directed toward themselves. The concept of survivor guilt finds its explanation in the mimetic theory by assuming that the identification is always ambivalent because structured by hate and love and hence is inherently rivalrous and guilty.

The second, or antimimetic theory, also tends to make imitative identification basic to the traumatic experience, but it understands imitation differently. The mimetic notion that victims of trauma are completely caught up or blindly immersed in the scene of shock is repudiated in favor of the opposite idea that the subject remains aloof from the traumatic experience, in the sense that he remains a spectator of the scene, which he can therefore see and represent to himself. The result is a tendency to relegate the problem of mimesis to a secondary position in order to establish a strict dichotomy between the autonomous subject and the external event. The antimimetic theory is compatible with, and often gives way to, the idea that trauma is a purely external event that befalls a fully constituted if passive subject. Whatever damage there may be to the victim’s psychical integrity, there is in principle no problem about his eventually recovering from the trauma, though the process of bringing this about may be long and arduous. And in contrast to the mimetic theory’s assumption of an unconscious identification with the aggressor, the antimimetic theory depicts violence as simply an assault from without. This has the advantage of portraying the victim of terror as in no way mimetically collusive with the violence directed against him, even as the absence of hypnotic complication as regards the reliability of his testimony shores up the notion of the unproblematic actuality of the traumatic event.

Des Pres’s claim that if the victims of the camps did imitate the SS, they did so only for strategic purposes and always maintained an inner resistance to, or spectatorial distance from, the scene in question conforms to the antimimetic model of trauma. The antimimetic theory also lends itself to various positivistic interpretations of trauma epitomized by the neurobiological theories that have won widespread acceptance today. The American Psychiatric Association’s decision in 1987 to remove survivor guilt from the criteria of PTSD may therefore be seen as exemplifying the antimimetic tendency of contemporary American psychiatry to suppress any reference to the mimetic dimension and to enforce instead a strict dichotomy between the autonomous subject and the external trauma. The framers of the definition of PTSD aimed at precisely such a dichotomy for forensic reasons: the division justified the claims of the anti-Vietnam war movement that veterans were suffering from combat-related psychiatric disorders against skeptics who doubted the need for a new diagnostic category. The PTSD committee of DSM-III tried to guarantee that strict dichotomy between subject and event by carefully defining the “stressor criterion.”18

Is it possible that today’s shame theory reflects a similar antimimetic tendency? In my book on trauma, I argued that from the end of the nineteenth century to the present there has been a continual oscillation between mimetic and antimimetic theories, indeed that the interpenetration of one by the other, or alternatively the collapse of one into the other, has been recurrent and unstoppable. Put slightly differently, my claim was that the concept of trauma has been structured historically in such a way as simultaneously to invite resolution in favor of one pole or the other of the mimetic-antimimetic oscillation and to resist and ultimately defeat all such attempts at resolution. The wager of the present study is that recent efforts to displace the concept of survivor guilt by that of shame may be understood as yet another manifestation of the fluctuation or tension between the mimetic and the antimimetic paradigms that has structured the genealogy of trauma from the start.

In this book I therefore plan to show that the concept of survivor guilt is inseparable from the notion of the subject’s unconscious identification with the other. Conversely, I seek to demonstrate that, in spite of the apparent diversity of approaches proffered by today’s theorists, shame theory conforms to the antimimetic pole of trauma theory because it displaces attention from the guilty subject’s unconscious yielding to the enemy to the shamed subject’s antimimetic consciousness of being seen. As a result, shame theory downplays the mimeticimmersive, interpersonal dynamic central to the formulation of guilt in order to depict shame as an experience of consciousness of the self when the individual becomes aware of being exposed to the diminishing or disapproving gaze of another. In other words, shame enacts a shift from the mimetic to the antimimetic by emphasizing the realm of the specular.

This, however, is not to say that the tension or oscillation between mimesis and antimimesis at work in the guilt-shame debate can be resolved into a simple opposition between a guilt concept governed exclusively and unproblematically by mimetic assumptions and a shame concept governed solely by antimimetic presuppositions. It has been my argument all along that the oscillation between mimetic and antimimetic tendencies in trauma theory can never be fully resolved. Accordingly, we would expect those same tensions to surface within the theorizing of guilt itself, just as we would expect them to manifest themselves in the conceptualization of shame. Thus I shall show that the psychoanalytic explanation of survivor guilt is marked by aporias and inconsistencies that can best be understood as a legacy of unresolved and unresolvable tensions within the theorization of imitation defined simultaneously in mimetic and antimimetic terms. And we shall see that contemporary shame theory, which emphasizes the antimimetic pole of the mimetic-antimimetic oscillation, struggles to maintain a coherent anti-mimetic position. Nevertheless, I claim that the tension between mimesis and antimimesis that is internal to the conceptualization of both survivor guilt and shame does play out over time in the form of a general shift from a conceptualization of survivor guilt, understood in mimetic terms as involving the subject’s unconscious identification with the other, to a conceptualization of shame that transforms passionate identifications into identity and in so doing posits a rigid dichotomy and specular distance between the autonomous subject and the external other. Accordingly, in this book I shall generally be less interested in tracking the inevitable mimetic-antimimetic tensions that arise within the discourses of guilt and shame than in charting the general shift from guilt to shame.

Moreover, I aim to demonstrate that whereas in the past the theorization of survivor guilt remained within an intentionalist or cognitivist paradigm of the emotions, current shame theory shares the positivist ambitions of the medical sciences by theorizing shame in antiintentionalist (or anticognitivist) terms. By common agreement, guilt concerns your actions, that is, what you do—or what you wish or fantasize you have done, since according to Freud the unconscious does not distinguish between the intention and the deed, the virtual and the actual. Equating intention with the deed, psychoanalysis maintains the link to intention and action that is held to be intrinsic to the notion of guilt.19 Shame, however, is held to concern not your actions but who you are, that is, your deficiencies and inadequacies as a person as these are revealed to the shaming gaze of the other, a shift of focus from actions to the self that makes the question of personal identity of paramount importance. (It makes no difference to my argument in this regard whether the self is considered a unified, fixed essence or identity or whether, after Freud, Lacan, or deconstruction, the self is regarded in antiessentialist terms as fragmentary, destabilized, and unfixed.)

Accompanying such a shift of focus to the self, and as an alternative to an intentionalist account of the emotions, many of today’s theorists define the affects, including shame, in materialist terms. According to them, shame is the result of inherited, neurophysiological responses of the body that are held to be independent of our intentions and wishes. As I showed in Trauma, these same materialist (or literalist) assumptions govern modern trauma theory as well, so that shame theory and trauma theory here overlap. In the works on shame I shall be discussing, the turn to materialism is partly a function of a general displacement of psychoanalysis by postpsychoanalytic and/or biological-evolutionary approaches to the study of human behavior. That displacement has been going on now for more than twenty years and amounts to a major conceptual and methodological paradigm shift. But even in the work of Agamben, for whom the theory of evolution is irrelevant, we find a similar materialist, antiintentionalist approach to shame. The general result is an account of shame that makes questions of agency, intention, and meaning beside the point and privileges instead issues of personal identity and difference.

The success of current shame theory, as I see it, can be explained by its ability to support and reinforce a self-declared postmodernist and posthistoricist commitment to replacing disputes or disagreements about intentions and meaning with an emphasis on who one is, or differences in personal experience. This development is as much a historical phenomenon as it is a theoretical one, and accordingly I conceive of my enterprise as a contribution to our understanding of that history. My book is not intended as a comprehensive study of the vicissitudes of guilt and shame in Western thought.20 Rather, it is presented as a contribution to the understanding of the changing fortunes of survivor guilt and shame from the post–World War II period to the present. It is an effort to take a step back from the routine, almost somnambulistic way in which notions of shame have recently come to dominate discussions of trauma, violence, and the self in order to examine, in the mode of what might be called a “genealogy of the present,” the steps by which the shift from guilt to shame has come about.

Some further remarks are in order. Since four of the five chapters of my book chart responses to survivors of trauma, including centrally survivors of the Holocaust, there is a clear sense in which my book can be understood as a contribution to the field of Holocaust and trauma studies. However, my goal is broader than that. By aiming to grasp the significance of the replacement of postwar notions of guilt by those of shame, my narrative as it unfolds simultaneously loosens the connection to trauma as such—since the most influential writings on shame that I discuss in chapter 4 have little to do with questions of trauma—and expands the scope of the argument. In particular, my objective is to show that with few exceptions all the recent shame theorists I discuss, whether they are interested in trauma or not, are alike bound to a set of linked commitments—to antiintentionalism, materialism, and the primacy of personal identity or difference—that decisively alters the terms of the analysis. For what I have come to see is that the conflict between the mimetic and the antimimetic with which I began this project is part of a larger set of oppositions—between intentionalism or cognitivism versus antiintentionalism or anticognitivism, between antimaterialism versus materialism, and between identification versus identity—at work in general questions of interpretation today. What is new is my claim that there is a particular logic at work in the shift from guilt to shame, a logic according to which if you think that the emotions, including shame, are to be understood in nonintentionalist terms, then you are also committed to the idea that they are to be defined in material terms, indeed that they are a matter of personal differences such that what is important is not what you have done, or imagined you have done, but who you are. It seems to me that this logic, pervasive in emotion theory today, is unsound— empirically unsound, because as I shall try to show, the experimental evidence does not support a coherent antiintentionalist position, and theoretically unsound, because it means giving up disagreement about intention and meaning in favor of an interest in simply what an individual person experiences or feels, that is, in favor of questions of personal identity.

My book is not intended as an exhaustive study of psychiatric or institutional responses to survivors of the Holocaust, or as a detailed examination of German reparation law, or as a comprehensive history of psychotherapeutic approaches to the survivor, or as a thorough examination of laboratory experiments that touch on the question of shame, although it engages with aspects of all these topics. It is rather a work of intellectual history in which I focus in a systematic fashion on the shift from guilt to shame that has taken place in the United States in the post–World War II period in an attempt to evaluate the stakes of that change, a change amounting to a major paradigm change in concepts of affect, self, and personal agency.

My study is divided into five chapters. In chapter 1, “Survivor Guilt,” I trace the post-Holocaust development of the concept of survivor guilt. I pay special attention to the contributions of those psychoanalysts who, encouraged by West Germany’s new reparation laws, which recognized emotional disability as a basis for a compensation claim, tried to help victims of the Nazis make claims for restitution by establishing a “survivor syndrome” diagnosis that linked the victim’s characteristic symptoms of persistent depression and guilt-ridden anxiety to his or her disastrous experiences during the war. In the course of my analysis I note that many of the figures I discuss, including Primo Levi, do not make a clear distinction between guilt and shame, treating the latter as a variant of the former—the present tendency to treat guilt and shame in binary terms is a recent development (I myself do not believe that these emotions are necessarily mutually exclusive). I also discuss some of the difficulties those same analysts experienced in theorizing survivor guilt coherently in mimetic-identificatory terms. I suggest that revisionist modifications in the Freudian approach to the traumatized subject introduced by Robert Jay Lifton, a psychiatrist recognized for his 1968 study of survivors of Hiroshima, disarticulated the concept of survivor guilt from that of identification with the aggressor in such a way that the connection between guilt and aggression was dissolved or at least attenuated and more traditional notions of individual responsibility and consciousness began to take over. I show that Lifton’s ideas about guilt were taken up by critics who were opposed to the whole idea of a “survivor syndrome,” with the result that his ideas were soon put to uses that were fundamentally hostile to the psychoanalytic enterprise.

Chapter 2, “Dismantling Survivor Guilt,” centers on critiques of the concept of survivor guilt by Terrence Des Pres and others in the political context of the postwar controversy launched in the 1960s by Hannah Arendt and others over the question of Jewish “complicity” in the Holocaust. In his book The Survivor (1976), Des Pres claimed that the notion of survivor guilt ended up blaming camp prisoners because it implied they were collusive with perpetrator violence. Instead, Des Pres proposed a sociobiological definition of survivors as ethical and caring persons who, thanks to their biological endowment, had emerged from the camps with their integrity and minds intact. His critique belongs to that general movement in the human sciences in America that during the 1970s and 1980s displaced psychoanalysis from its previous position of importance. In particular, Des Pres crystallized a tendency in the wake of the Eichmann trial to question the authority of the Freudians to give a just portrait of the survivor. At the center of Des Pres’s attack on the notion of survivor guilt was his account of survivors as prisoners who, if they were obliged to imitate the enemy for self-protection, did so not in the mode of an unconscious identification with the aggressor but in the mode of a conscious mimicry that concealed the victims’ true wishes and feelings. Des Pres’s recasting of imitation in antimimetic terms as deliberate simulation permitted a reevaluation of survivors not as neurotic or ill but as capable of resisting power by performatively disguising their true intentions. At the same time, by reinterpreting survival in terms of a biological “talent for life,” Des Pres treated the human being’s capacity for intentions as a function of his or her biological-corporeal endowment—in other words, he interpreted survival in materialist terms. Des Pres made use of Erving Goffman’s dramaturgical ideas to support his arguments. Similar approaches to imitation governed sociological and psychoanalytic critiques of the notion of survivor guilt in the 1970s and 1980s. Shame is not yet the dominant motif of the works I examine in this chapter, but the account of survival and the self that they offer helped set the stage for shame theory’s rise to influence.

In chapter 3, “Image and Trauma,” I turn to the American Psychiatric Association’s decision in 1987 to drop survivor guilt as one of the diagnostic criteria for PTSD. The third, 1980 edition of the American Psychiatric Association’s Diagnostic and Statistical Manual (DSM-III), which officially introduced the diagnosis of Posttraumatic Stress Disorder (PTSD), represented a revolution in the approach of American and hence worldwide psychiatry. Psychoanalytic norms that for thirty years or more had dominated the field were abandoned in favor of a more positivist and ostensibly atheoretical description and classification of mental disorders. The introduction of PTSD stimulated a large number of research projects designed to further clarify and operationalize the disorder. In this chapter I explore the importance of the concept of the traumatic “image” to the ongoing process of reformulation. In particular, I argue that the reconceptualization of PTSD in the 1980s around the traumatic image, defined as an externally caused mental content or “icon” uncontaminated by any mimetic, fictive, or fantasmatic dimension, made the notion of survivor guilt, which depended for its rationale on a now-discredited Freudian theory of identification with the aggressor, an incoherent element in the theory of posttraumatic stress, so that its elimination from those criteria made sense. This doesn’t mean that the notion of survivor guilt completely disappeared from the ordinary or daily language of trauma, only that within psychiatry it now lacked any obvious theoretical justification. I therefore link the demise of survivor guilt in trauma theory to the coalescence of the question of traumatic violence around the concept of image. I shall focus on the work of several trauma theorists, including Mardi J. Horowitz, whose use of stress-inducing films to operationalize and objectivize trauma contributed to the formulation of posttraumatic stress. The chapter ends with a brief discussion of the ways in which shame has come to take the place of survivor guilt in recent discussions of PTSD.

In chapter 4, “Shame Now,” I examine a variety of recent texts on shame in order to exhibit and critically assess the fundamental logic of shame theory today. In the course of my discussion I shall take into account a wide range of psychological, biological, and literary-critical works. I shall pay special attention to the post-modernist, postpsychoanalytic writings of literary critic Eve Kosofsky Sedgwick, in my view the most brilliant and articulate of recent shame theorists. Sedgwick is indebted to the work of psychologist Silvan Tomkins, who proposed an “affect program theory” of the emotions that defined the affects, including shame, in biological, antiintentionalist terms. Although Sedgwick presents the intentionalist or cognitivist theory of the emotions as the entrenched position she wants to challenge, I see things rather differently. In fact, the opposite seems to me true: versions of Tomkins’s “affect program theory”—especially those associated with the work of Tomkins’s followers Paul Ekman and Carroll E. Izard—have dominated modern psychological, if not philosophical, work on the emotions for many years now, and Sedgwick’s recent commitment to Tomkins’s work suggests that the influence of such theory has now spread to the humanities as well.

In this chapter the larger implications of my argument begin to emerge. Since neither Tomkins, Ekman, or Izard formulate their work on the affects in terms of trauma theory, the theme of trauma, which up to this point has marked my discussion of survivor guilt, recedes, at least for the moment (it returns in chapter 5). At the same time, my discussion broadens to include a discussion and critique of the general emotion theory associated with the work of Silvan Tomkins, and of the experiments performed by Ekman and others to support the new materialist approach to shame. This chapter, arguably the heart of my book, contains a detailed, critical (though by no means exhaustive) assessment of the arguments and evidence adduced in support of the antiintentionalist position in emotion theory. My aim is not only to say what I think is wrong or incoherent about the Tomkins-Ekman-Izard approach to the affects but to analyze the identitarian consequences of that approach as those consequences are especially made evident in the work of Sedgwick and her followers. The overall purpose of the chapter is thus to critique the antiintentionalist, materialist paradigm that governs modern shame theory and that has displaced the intentionalist paradigm that had previously informed the concept of survivor guilt.

In chapter 5, “The Shame of Auschwitz,” I return to the theme of the Holocaust in order to analyze Giorgio Agamben’s influential effort to dislodge the notion of survivor guilt from its position of importance in the literature of the Nazi camps in favor of a highly personal, but in the end representative, notion of shame. My goal is to demonstrate the similarities between Sedgwick’s materialist and antiintentionalist approach to shame and Agamben’s ideas. In other words, I aim to show that, in spite of an apparently very different intellectual agenda, Agamben’s hostility to the idea of survivor guilt and his concomitant valorization of shame are based on similar ideas about the absence of intention and meaning in shame that we find developed by Sedgwick and many recent affect theorists. This entails a brief discussion of Agamben’s ideas about testimony, in the course of which I shall also draw attention to the similarities between his materialist ideas about language and those of trauma theorists Shoshana Felman and Cathy Caruth.

One last point, which brings me back to where I began. Many Americans, including myself, would not hesitate to declare that they experience intense shame for the prisoner abuse scandal at Abu Ghraib. Nothing that is said critically about contemporary shame theory in the pages that follow is meant to criticize the view that shame can be an appropriate response to such situations.

Return to Book Description

File created: 8/7/2007

Questions and comments to:
Princeton University Press

New Book E-mails
New In Print
PUP Blog
Princeton APPS
Sample Chapters
Princeton Legacy Library
Exam/Desk Copy
Recent Awards
Princeton Shorts
Freshman Reading
PUP Europe
About Us
Contact Us
PUP Home

Bookmark and Share