The Unconscious Need for Punishment:

 Expression or Evasion of the Sense of Guilt?*


Donald L. Carveth, Ph.D.



In the end we come to see that we are dealing with what may be called a 'moral' factor ... which is finding its satisfaction in the illness and refuses to give up the punishment of suffering. ...  But as far as the patient is concerned this sense of guilt is dumb; it does not tell him he is guilty; he does not feel guilty, he feels ill (Freud, 1923, pp. 49-50).


At the center of the sickness of the psyche is a sickness of the spirit.  Contemporary psychoanalysis will have eventually to reckon with this Kierkegaardian point of view (Barrett, 1958, p. 170).





In Civilization and Its Discontents and other writings, Freud equates the unconscious need for punishment expressed in various patterns of self-torment and self-sabotage with the unconscious sense of guilt. But there are cogent clinical and theoretical grounds for distinguishing between genuine guilt and the unconscious need for punishment that serves as a guilt-substitute the function of which is precisely to ward off an unbearable sense of guilt.  Whereas guilt embodies the depressive anxiety and the capacity for concern for the other that characterize the depressive position and that motivate the desire to make reparation, the unconscious need for punishment reflects the narcissistic and sado-masochistic dynamics associated with the paranoid-schizoid position. The discontent Freud links with civilization is not a manifestation of guilt but of the self-torment resulting from its evasion.  The enlarged capacity to experience and bear guilt that is a mark of civilization reflects the healing, not the deepening, of our cultural malaise.





In the final section (VII) of Civilization and Its Discontents, Freud (1930) states that the primary intention of this work is "to represent the sense of guilt as the most important problem in the development of civilization and to show that the price we pay for our advance in civilization is a loss of happiness through the heightening of the sense of guilt" (p.134).  It is the thesis of this paper that our discontent in civilization arises not through the heightening of the sense of guilt, but rather through the heightening of the unconscious need for punishment that defends against the sense of guilt.  An advance in civilization through a heightening of the capacity to confront and bear guilt leads to a decrease, not an increase, in discontent.  Genuine guilt, understood as depressive anxiety or concern, represents not the problem of civilization but its solution.  It is the path toward genuine happiness and peace.

According to Freud (1930), "men are not gentle creatures who want to be loved, and who at the most can defend themselves if they are attacked," but are, on the contrary, "creatures among whose instinctual endowments is to be reckoned a powerful share of aggressiveness. ... Homo homini lupus. Who, in the face of all his experience of life and of history, will have the courage to dispute this assertion?" (p.111). It follows that if the Hobbesian "war of each against all" in which life is, of necessity, "nasty, brutish and short" is to give way to civilized order, such "cruel aggressiveness," this "primary mutual hostility of human beings" (p.112), must in some way or another be inhibited. Freud offers us three options by which this may be achieved: repression, suppression and sublimation. Since most of us do not possess the strength of character for conscious suppression and self-mastery without self-deception, and lack the talent for much sublimation, the majority will be forced to fall back on repression, with the disguised return of the repressed that this inevitably entails. A major manifestation of the disguised return of our repressed aggressiveness is in the operations of the sadistic superego that retroflects id aggression away from the object world and against the ego. This results in diverse forms of self-punishment, the "moral masochism" Freud (1916) described in "the criminal from a sense of guilt," "those wrecked by success," and other self-sabotaging and self-tormenting character-types.

Freud (1930) was quite aware that the unconscious need for punishment resulting from such retroflected aggression often operates in people's lives without any accompanying conscious sense of guilt. But even where, as in some cases of obsessional neurosis, "the sense of guilt makes itself noisily heard in consciousness," on exploration it often turns out that the ostensible sins of omission or commission with which it is consciously linked bear only the remotest connection to the true, unconscious sources of the guilt feeling—the true crimes, if you will, whether these be acts or merely wishes.  In The Ego and the Id, Freud (1923) writes:

In certain forms of obsessional neurosis the sense of guilt is over-noisy but cannot justify itself to the ego. Consequently the patient's ego rebels against the imputation of guilt and seeks the physician's support in repudiating it. It would be folly to acquiesce in this, for to do so would have no effect. Analysis eventually shows that the super-ego is being influenced by processes that have remained unknown to the ego. It is possible to discover the repressed impulses which are really at the bottom of the sense of guilt. Thus in this case the super-ego knew more than the ego about the unconscious id (p.51).


While one may agree with Carroll (1985) regarding the limitations of psychology in approaching the domain of the spirit and the fallacy of Freud’s reduction of religion per se to the categories of neurosis (pp. 225-6; see also Carveth, 1998; Forster & Carveth, 1999), his notion that, according to Freud and his followers, “you should do away with your guilt” (p.225) is as oversimplified as Freud’s view of religion. While there is little doubt that “some psychotherapists feel that soothing the patient’s superego is part of their job,” attempting to “de-guilt” the patient by suggesting “he or she has nothing to feel guilty about” (Carveth & Hantman, 2003, p. 470), this is a distortion of Freud’s own stance and practice and that of those analysts who, following Freud, far from repudiating the patient’s guilt seek to help him or her discover the unconscious impulses and phantasies that may justify it. In Freudian and Kleinian “soul-study,” far from soothing the psyche or the self and dismissing guilt as unfounded, the analyst seeks to make conscious its unconscious, real or imagined, grounds.





In Civilization, Freud (1930) writes, "in most other cases and forms of neurosis it [the sense of guilt] remains completely unconscious, without on that account producing any less important effects." "Our patients do not believe us," he writes, "when we attribute an 'unconscious sense of guilt' to them. In order to make ourselves at all intelligible to them, we tell them of an unconscious need for punishment, in which the sense of guilt finds expression" (p.135).

The fact that his patients do not believe him when he attributes an unconscious sense of guilt to them does not trouble Freud. He gets around their objection by equating the unconscious sense of guilt with the unconscious need for punishment. The self-damaging or self-tormenting behaviors are observable and although at first patients may be unconscious of the role they themselves are playing in bringing such suffering on themselves, they can often come to recognize their own unconscious agency in their misfortune when it is pointed out to them. Since Freud assumes self-punishing behavior is driven by and a manifestation of guilt, and since conscious guilt is absent, he postulates the existence of unconscious guilt, equating this with the unconscious need for punishment.

Just as the sense of guilt or fear of the superego may not be conscious in the moral masochist, so "it is very conceivable," Freud (1930) writes, "that the sense of guilt produced by civilization is not perceived as such either, and remains to a large extent unconscious, or appears as a sort of malaise, a dissatisfaction, for which people seek other motivations" (p. 135-6). Here we are introduced to the important concept of the guilt-substitute. Just as the unconscious operations of the punitive superego that Freud equated with unconscious guilt may find expression in the patterns of self-punishment seen in manifold forms of masochism, so it may appear in various forms of malaise, dissatisfactions, discontents and mysterious neurotic afflictions, many of which appear to have little or nothing to do with issues of guilt, crime and punishment, but which may nevertheless be the work of the unconscious punitive superego.

Here I would include such conditions as Erikson's (1956) "identity diffusion" and the states of fragmentation and depletion of the self that Kohut (1971, 1978) described in the "self disorders" of the so-called "Tragic Man" that he claimed has replaced the "Guilty Man" of the Freudian era, as well as the “postmodern” hystero-paranoid syndromes discussed by Showalter (1997) and the present author (Carveth & Hantman, 2003).  Although these postmodern, narcissistic conditions of fragmentation, emptiness, boredom and irritability are nowadays widely conceptualized in terms of defect, deficit, failures of mentalization, etc., resulting from parental or “selfobject” failure, the fact is they are experienced as tormenting by those who suffer from them and, like more obvious forms of self-punishment, I believe they function as substitutes for, and defenses against, guilt.  In other words, the tragedy of "Tragic Man" has less to do with deficits in psychic structure per se, than with the latent ongoing self-annihilation, the manifest traces of which appear as defects in the ego or the structure of the self.

As we have seen, Freud equates such unconscious activities of the punitive superego with the unconscious sense of guilt. But as Freud (1924) himself recognizes in The Economic Problem of Masochism, there is a problem with this association:


Patients do not easily believe us when we tell them about the unconscious sense of guilt. They know well enough by what torments--the pangs of conscience--a conscious sense of guilt, a consciousness of guilt, expresses itself, and they therefore cannot admit that they could harbour exactly analogous impulses in themselves without being in the least aware of them. We may, I think, to some extent meet their objection if we give up the term "unconscious sense of guilt," which is in any case psychologically incorrect, and speak instead of a "need for punishment," which covers the observed state of affairs just as aptly (p.166).


In the same essay, writing of the "negative therapeutic reaction," Freud (1924) places the adjective "unconscious" in inverted commas in referring to "patients to whom ... we are obliged to ascribe an 'unconscious' sense of guilt" (p.166). He does so because he recognizes as problematic the notion that a feeling or affect, as distinct from its associated ideation, could be unconscious. Only a few years later, in Civilization, Freud (1930) is struggling with the same issue. He associates the unconscious sense of guilt with fear of the superego and refers to this as an "unconscious anxiety" and continues "or, if we want to have a clearer psychological conscience, since anxiety is in the first instance simply a feeling, of possibilities of anxiety" (p.135). Strachey feels compelled to add a footnote here stating that "Feelings cannot properly be described as 'unconscious'" (p.135).

Without entering into the metapsychological complexities of the issue of whether affects or feelings or only their ideational correlates can be unconscious, it suffices for our purposes to indicate that Freud himself was uneasy with respect to his own notion of unconscious guilt.  But this uneasiness had to do with the question as to whether an affect could properly be said to be unconscious, not with the equation of the unconscious need for punishment with unconscious guilt. In the present essay I am less concerned with the question of whether the sense of guilt properly speaking (i.e., concern and the drive toward reparation) may be unconscious, than with the misleading equation of guilt, conscious or unconscious, with the need for punishment.[2] Although Freud suggests we could "give up the term 'unconscious sense of guilt,' which is in any case psychologically incorrect, and speak instead of a 'need for punishment'" (p.166), he in fact does not do so.  Instead, he continues to use these terms and concepts interchangeably with the unfortunate consequence that the role of self-punishment in the evasion of guilt, rather than as an expression of it, has been obscured.[3]

In attempting to de-link the concept of the unconscious need for punishment from the concept of guilt I am in no way seeking to cast doubt upon Freud's important discovery of the role of the former in psychopathology. I am merely seeking to draw attention to the fact that equating the need for punishment with guilt obscures the former's primary defensive function: the evasion of guilt. Freud (1924) writes:



  The satisfaction of this unconscious sense of guilt is perhaps the most powerful bastion in the subject's (usually composite) gain from illness--in the sum of forces which struggle against his recovery and refuse to surrender his state of illness. The suffering entailed by neuroses is precisely the factor that makes them valuable to the masochistic trend. It is instructive, too, to find, contrary to all theory and expectation, that a neurosis which has defied every therapeutic effort may vanish if the subject becomes involved in the misery of an unhappy marriage, or loses all his money, or develops a dangerous organic disease. In such instances one form of suffering has been replaced by another; and we see that all that mattered was that it should be possible to maintain a certain amount of suffering (p.166).



            There is little doubt that the unconscious need for punishment and the unconscious operations of the punitive superego occupy a central place in psychopathology.  But there are good clinical and theoretical grounds, some of which were even pointed out by Freud and Strachey, for regarding as quite justified the skepticism of the patients to whom Freud (1924) refers who "do not easily believe us when we tell them about the unconscious sense of guilt" because "They know only too well by what torments--the pangs of conscience--a conscious sense of guilt, a consciousness of guilt, expresses itself ..." (p.166).  Where an unconscious need for punishment exists the unconscious superego clearly regards the subject as culpable and, hence, as deserving of punishment.  But to refer to this unconscious superego judgment and the self-punitive activity that results from it as "unconscious guilt" obscures the fact that its function is to foreclose the experience of genuine guilt as concern for and the drive to repair the damage done to the other.[4]

To view the unconscious superego activity resulting in self-punishment as guilt is to blur the crucial difference between the subject's self-torment and what Winnicott (1963) called "the capacity for concern" for the object.  Unconscious self-punitive activity is narcissistic.  Authentic guilt moves beyond narcissism toward object love.  It only leads to theoretical confusion when we employ the same term to refer to such different realities as the narcissistic, paranoid-schizoid phenomena of self-torment on the one hand, and the object-oriented, depressive position phenomena of guilt and concern on the other.

Part of our difficulty here arises from an ambiguity contained in the single word "guilt" which can refer both to the ontological state of being or being judged to be guilty and the psychological or experiential state of feeling guilty.  Someone who does not feel guilty may be judged by his own or another's superego to be guilty; sometimes someone judged to be guilty also feels guilty; occasionally someone who feels guilty turns out not to be guilty (though, more often, while not guilty of the charges of which he accuses himself, he turns out to be quite guilty of other crimes that are the real but hidden source of the guilt-feeling).  Hence, when we encounter the term "unconscious guilt" we cannot, apart from context and often not even then, determine whether what is being referred to is a state of being guilty of which the subject is unaware, or a state of feeling guilty of which the subject is unconscious.

Like Freud and Strachey, I find the notion of unconscious feeling problematic.  But the unconscious superego often judges someone to be guilty even though they do not consciously feel any guilt.  I have no problem with the notion of being guilty but unconscious of this fact and unconscious of the fact that the superego considers one so: the superego finds one guilty but one does not feel guilty.  My problem (and Freud's and Strachey's) is with the notion of a person unconsciously feeling guilty.  Without feeling guilty a subject's unconscious superego frequently judges him or her to be guilty.  Frequently, instead of coming to feel guilty (whether such guilt is justified or not is another matter), the subject often unconsciously seeks punishment.  Such self-punishment, I submit, serves as a defense against the process of becoming conscious of one's guilt and feeling guilty.  The feeling of guilt that might accompany the state of being or being judged to be guilty is absent because this guilt-feeling is unbearable.  Consequently, its development is short-circuited through mechanisms of self-torment, the pain of which is somehow preferable to unbearable guilt-feeling.[5]

I propose, then, that we reserve the term guilt for pangs of conscience that lead to reparation, as distinct from pangs that substitute for reparation, and stop confusing it with the unconscious patterns of self-torment and self-sabotage that Freud described.  Although they do not challenge Freud's association of such behaviors with unconscious guilt, Reisenberg-Malcolm (1980) and, more recently, Safan-Gerard (1998) have shown how unconscious self-punishment, which the latter refers to as "expiation," serves to evade and defend against the experience of guilt and concern.  Freud's equation of the unconscious need for punishment with unconscious guilt has obscured the defensive function of unconscious self-torment and its role in the chronic evasion of the mental suffering, depressive anxiety, guilt and remorse that must be confronted and contained in working-through the depressive position.  Frequently, when the unconscious superego judges us guilty, we evade feeling guilty by going directly to self-punishment. Unfortunately, evading guilt-feeling in this way precludes the rational evaluation of such guilt that would enable us to decide whether to accept and make reparation for it, or reject it as irrational and ungrounded.

Ury (1998) has recently drawn attention to the contradiction in Freudian theory between its developmental affirmation of superego formation as a sign of maturity and its clinical recognition of the role of the superego in psychopathology. She writes:


There is a tendency in psychoanalytic literature to view the nature of guilt in two contradictory ways. The first is often found in the theoretically derived developmental premise of the tripartite structural model of intrapsychic differentiation, which states that unconscious guilt emerges from an internalized superego, which presupposes a structured and mature ego. An assumption follows that the "capacity for guilt" is a higher and more adaptive form of mental functioning: it is healthy, civilized, and mature, and equated with notions of repair and concern. It is also often interchanged with the concept of conscience. The second view of guilt is to be found in clinical formulations of pathology where the destructiveness of guilt in psychic functioning is highlighted, especially in relation to the sadism of the superego. Despite the observation that guilt is usually, if not always, associated with destructive pathology, the developmental framework that positions guilt as a mature affect is left intact. This contradiction begins with Freud, who suggested that guilt is not only the height of civilization, but also a deep-seated, intractable form of aggression ... (p.51).

            Ury proposes to resolve this contradiction by distinguishing between guilt, as a superego function observed in pathological states of self-torment, and conscience as an ego function involving thought and anticipation of the consequences of our actions for others and ourselves. To my mind, there are two main problems with this proffered solution. First, it requires us to abandon our everyday association of guilt with normal and even healthy experiences of the voice and pangs of conscience--i.e., for what Grinberg (1964) calls depressive as distinct from persecutory guilt. Second, in excluding the operations of mature conscience from the experience of guilt and shrinking the latter to include only the operations of the archaic, persecutory superego--i.e., the pathological states of self-torment that Grinberg calls persecutory guilt--Ury, like Grinberg, confirms Freud's association of such pathological self-punishment with guilt when, in my view, it frequently functions as a defense against unbearable guilt, i.e., as a guilt-substitute. I do not wish to surrender to the widespread inclination in our “culture of narcissism” to derogate guilt as pathology rather than recognize it as an essential component of maturity and mental and spiritual health. 

The trouble with Grinberg’s (1964) strategy of referring to two kinds of guilt, persecutory and depressive, is that it obscures the role of the former as a defense against the latter and, by associating guilt with self-persecution, concurs with its pathologization. Why is guilt at times so unbearable that it must be short-circuited through processes of unconscious self-punishment? I think the answer is that the subject, caught up in paranoid-schizoid splitting or polarization, feels it cannot admit of any wrongdoing or badness without being revealed as a poisonously all-bad object. In other words, at the root of the problem is a difficulty in self and object constancy, in holding both bad and good simultaneously, in being able to acknowledge the badness without forgetting the goodness and so achieving ambivalence. In the pre-ambivalent, paranoid-schizoid position, to admit of any imperfection is to reveal oneself as hopelessly defective.[6]  

In the paranoid-schizoid position, the archaic, sadistic superego reigns. Whatever the surface effectiveness of the defensive denial, displacement or projection of blame, the archaic superego demands its pound of flesh in the form of the unconscious need for punishment (for parricidal, matricidal and fratricidal impulses) that Freud equated with unconscious guilt but that, in reality, is a consequence of guilt evasion. Far from colluding with such evasion, psychoanalysis works against it, both by making unconscious guilt conscious, and by reawakening conscience through analysis of the self-tormenting unconscious superego activities by means of which guilt is evaded. Because the need for punishment substitutes for and defends against genuine guilt, learning in analysis how to face and bear one's guilt (i.e., working though the depressive position) is the road to freedom from the grip of the unconscious need for punishment and for the soothing of the pain arising from self-torment. For, however effective on the surface, such soothing (whether derived from substances, “selfobjects,” or other sources) cannot eradicate the savage god, the unconscious superego and its punitive operations. By blocking the development of mature guilt and reparation, such soothing strategies set up a vicious cycle: the superego is forced into the regressive channel of self-torment resulting only in an increased need for soothing.






Although it is open to question whether the Judeo-Christian discourse of sin and redemption in and of itself made guilt more acknowledgeable and bearable by rendering it universal, at least that tradition drew attention to the inherent destructiveness of human nature in its central doctrine of man's fallenness, brokenness and sinfulness.  It is a measure of his personal estrangement from the tradition of his father and his people that Freud (1933) could attribute resistance to recognition of the destructive element of human nature to religion. He writes (Lecture 32 of the New Introductory Lectures on Psychoanalysis):

We should probably have met with little resistance if we had wanted to ascribe an instinct with such an aim to animals. But to include it in the human constitution appears sacrilegious; it contradicts too many religious presumptions and social conventions. No, man must be naturally good or at least good-natured. If he occasionally shows himself brutal, violent or cruel, these are only passing disturbances of his emotional life, for the most part provoked, or perhaps only consequences of the inexpedient social regulations which he has hitherto imposed on himself (pp.103-4).

            Here Freud appears to have confused the Bible with Das Kapital or Rousseau's Social Contract and to have forgotten the central Biblical doctrine of the fall of man and the Judeo-Christian understanding of human nature as corrupted and of the essence of evil as man's sinful self-centeredness.  Freud mistakenly blames religion for idealizing human nature because he does not wish to blame the secular, anti-religious, Enlightenment tradition with which he identifies, but that is the real source of the cultural denial, not of human destructiveness per se (which it often diagnosed brilliantly, especially in its socioeconomic forms), but of its deepest roots.  One need not accept Freud’s false naturalization and ultimate reductive biologization of such destructiveness to reject the environmental determinism that views it exclusively as the outcome of particular social arrangements.  There is no need to deny the “surplus” destructiveness engendered by social conditions, nor to devalue social critique, in order to recognize as a viable alternative to both biologism and sociologism the psychoanalytic existentialism that views our “basic” destructiveness as imposed neither by biology nor society but as a manifestation of human narcissism (Carveth, 1996).






As early as 1950, Erikson asserted that "The patient of today suffers most under the problem of what he should believe in and who he should--or, indeed, might--be or become; while the patient of early psychoanalysis suffered most under inhibitions which prevented him from being what and who he thought he knew he was" (Erikson, as quoted by Karen, 1992, p. 60). Thirteen years later, Marcuse (1963) argued that this shift entailed the growing obsolescence of the Freudian concept of man. He claimed that while Freud's theory had validly described the relatively structured personality of the modern era, it had grown increasingly irrelevant in a postmodern culture in which: "the 'individual' as the embodiment of ego, id and superego has become obsolescent in the social reality" (p.44) and in which a new personality type had emerged whose ego-identity is diffuse and shifting due to its lack of inner support from internalized values and ideals. In this situation, he observed, "the mediation between the self and the other gives way to immediate identification" while "the ego shrinks to such an extent that it seems no longer capable of sustaining itself, as a self, in distinction from id and superego" (p. 47).

By the mid-seventies, Kohut (1977) was arguing that classical psychoanalysis, in its focus upon conflict, repression and neurosis, amounted to a psychology of "Guilty Man" (pp. 132-33). It accurately reflected the society in which it arose, a society with sufficient normative integration to sustain a family life with enough coherence and stability to permit its offspring to develop the relatively well-structured or cohesive self and the internalized superego, ego-ideal and ego-identity that serve as a kind of gyroscope for the "inner-directed" (Riesman et al, 1950) personality of Freudian theory. However, like Marcuse, Kohut saw the emergence in our postmodern era of an unstructured and "other-directed" personality whose ego-identity or "self" is fragmentary, diffuse and shifting, a "Tragic Man" who, in the relative absence of firmly internalized values and ideals, suffers from a sense of inner emptiness and disconnection that lead him to chronically and hungrily seek a self in the mirror constituted by significant others toward whose expectations and responses he develops a radar-like sensitivity.

Recently, Finlay de Monchy (1997) wrote as follows about what she calls (p. 9) "the post-modern, psychotic, non-subject":


Post-modernism refers to a body of theory about the subject and discourse, including the works of such thinkers as Foucault, Deleuze and Guattari, Blanchot, Lyotard, Baudrillard etc. But post-modernism also refers to a set of cultural practices and discursive productions now commonly labeled "post-modern" and in the literature often likened to the psychical presentation of the psychotic. Post-modernism has been described in terms of culturally generalized psychosis. A few cultural examples: The emptying of the star of any personal identity is analogous to the depersonalization of the psychotic. We know nothing of the personal lives of the members of ABBA, a very different approach to the star than the pages of gossip about personal lives that fill the traditional "Hollywood" popular press. This depersonalization is perhaps best exemplified in the title of the rock group, "Talking Heads." No identities, no subjective substance, just a thing out of which flows discourse, the new post-modern non-subject. Rather than considering the subject to be some sort of totalizable unity, the post-modern creature is a kind of universalized "bad object" in the Kleinian sense of the term. The lead singer of the "Sex Pistols" calls himself "Johnny Rotten." Rock groups, even in their performances, go to great measure to eradicate the subject as the source and amplification of sound. "Kraftwerk"--factory work--presents a concert in which the members of the group do not appear on stage, merely tend the computers from behind the scene. This production is likeable to the "Disembodied eye" of the psychotic subject that has no continuity in time and place and exists nowhere; the habit of jet-setting, so well practiced by post-modern theorists as well, being yet another manifestation of an individual who exists nowhere, unpositioned as a subject in time and space. Subjects don't exist, rather personae are manufactured for a temporary experience and then remade for another set of desires and experiences, a bit like being made up cosmetically (pp. 5-6).


            Despite widespread agreement among social scientists and psychoanalysts regarding such psychosocial shifts, I believe they have more to do with the manifest forms and expressions of character and psychopathology than with their latent dynamics.  It is often claimed that in today's "culture of narcissism" (Lasch, 1979) analysands seldom present with the defined psychoneurotic symptoms characteristic of the patients of Freud's day.  But if we look more closely we are usually able to find a myriad of hysterical and obsessional manifestations in our contemporaries and, in any case, the patients Freud regarded as oedipal-level neurotics look highly narcissistic, borderline and even psychotic by current diagnostic standards.  Nevertheless, it seems to be the case that in contemporary society manifest symptomatology has shifted toward the subjective states of emptiness, detachment, painful social isolation and unrelatedness, fragmentation anxiety, narcissistic self-obsession and the various newer forms of hysteria (chronic fatigue syndrome, fibromyalgia syndrome, multiple chemical sensitivity, environmental illness, dissociative identity disorder, the “alternative health” obsession with internal pollutants and parasitic infestation, etc.) some of which Showalter (1997) calls Hystories and that I (Carveth & Hantman, 2003) consider variants of an hystero-paranoid syndrome.  But despite such manifest changes, on the latent level psychodynamic processes with which we have long been familiar, especially the unconscious operations of the persecutory superego, still obtain in these conditions.  Although they arise from guilt-evasion, in much contemporary post-Freudian and post-Kleinian (i.e., guilt-evading) psychoanalysis, which even now is not the only kind of psychoanalysis, they are more likely to be conceptualized in terms of deficit than in those of conflict and defence.

This is not the place to discuss the complex technical issues involved in the clinical handling of these problems.  Suffice it to say that persons suffering from a persecutory superego are all too ready to hear its confrontation and interpretation as accusation or attack and to flee from or, alternately, submit to and even be gratified in being, as they imagine, attacked in this way.  But the fact that the sadistic superego can turn interpretations of the sadistic superego to its own purposes does not mean that the sadistic superego does not exist or need, eventually, to be interpreted.  It merely means that it must be approached tactfully, skillfully and strategically.  It is here that respect for patients' resistances is most important.  Such patients have good reasons for evading guilt by resorting automatically to self-torment: fixed in the paranoid-schizoid position as they are, any admission of fault appears to confront them with a traumatic and unbearable sense of badness, inadequacy and shame. 






While it is true that no one can feel guilt about the damage one has done or wished to do to others without simultaneously feeling ashamed of the fact that one is the sort of person who has done or wished to do such damage, the reverse does not follow. It is possible to experience shame without guilt—that is, to be so self-obsessed that one loses sight of the object altogether except as a mirror or audience or resource for the self. In this sense, while it may be incorrect to say that guilt is a more mature emotion than shame—in that mature people continue to experience both—it is certainly true that the person who can experience guilt is more mature than the person who can experience only shame. In such a mature person, despite shame for the self, concern for the object (i.e., guilt) is maintained.  On the more primitive level of Klein's paranoid-schizoid position one may experience predominantly shame—one can be suffused with shame without having to cease one's self-obsession long enough to feel any concern for the object.  If, as the old saying has it, the superego is soluble in alcohol, then in narcissism it may be liquidated altogether.  On the other hand, one may mature to the point of becoming ashamed of one's narcissism and incapacity to experience guilt. This is perhaps a turning point initiating an advance to a level of object-relating, Klein's depressive position, at which "the capacity for concern" (Winnicott, 1963) is finally achieved.

This is not to say that guilt may not seem at times to be a defence against shame, as Fairbairn (1952) recognized in describing "the moral defense" in which the unloved child attempts to escape traumatic helplessness through the illusion of control afforded by blaming the self for the parental failure to love.  In order to escape intolerable shame in the face of one's unmet needs and one's helpless dependence on others who cannot be controlled, one resorts to an illusion of guilt which at least moves the trauma into the field of one's own (defensive) omnipotence.  There is no doubt that this mechanism exists and is important in psychopathology. But, far from constituting an argument for reducing guilt to an underlying shame, it merely points to a spurious or false guilt. For such "guilt" that exists to escape a painful state of shame is entirely narcissistic: it reflects no genuine sense of concern for the object; its function is purely defensive. The very idea that the phenomenon of human guilt could be reduced to such pseudo-guilt and in this way made subordinate to shame is itself a symptom of a widespread desperation to somehow find a way to sidestep the real guilt that is an inevitable part of mature object-relations. The motive for this wish to reduce guilt to shame is simply the wish to continue to forget the superego, to continue to live in a culture of narcissism, with a "psychology of the self" that evades concern for the object and rejects as a "health and maturity morality" (Kohut, 1979, p.12) the developmental demand that we move beyond issues of shame and the self and take up the cross of object-relating and the inevitable struggles with guilt that such relating entails.

Towards the end of his perceptive essay on shame, Karen (1992) writes of a patient whom he says wants to know "the real me" but is afraid to find out. She is afraid to face "the shameful fact that she is a shrew to her husband and children ... [and suffers from] the desperate fear that she will be found in the wrong" (p.69). In other words, overwhelming shame, itself a symptom of pathological narcissism, prevents her from being able to acknowledge her guilt. Karen goes on to explain the narcissistic basis of this flight from or inability to experience guilt: "To stop running and experience the shame is to give herself a chance to recognize that being in the wrong for acting like a shrew does not mean that her husband isn't also in the wrong in his way, nor does it make her into a poisonously deformed and unlovable thing" (p.69). In other words, it requires a certain level of self-esteem and a certain cohesion or constancy of the self to be able to acknowledge guilt without feeling totally annihilated—that is, without regressing from the depressive position of guilt and concern to the paranoid-schizoid sense of the self as all-bad.

According to Karen, "Putting shame into words with a trusted companion enables one to step outside it—it no longer seems to permeate one's entire being—and allows some self-forgiveness to emerge" (p.70). This, he suggests, is what used to be accomplished, to some degree, in the days when the Judeo-Christian discourse of sin and redemption was still culturally available and when, every Sabbath, in the company of an entire congregation, one confessed one's sins and accepted forgiveness for them and acceptance back into the community of the faithful.  "In medieval Christendom," he writes, "the belief that all people were sinners ... used this sense of universal defect to bind the community ... and ... to drain off some shame that might otherwise have become individual and narcissistic. From our distant perspective in a diametrically different world, we can only imagine how comforting it might have been to know that one was not alone in one's flaws and vulnerabilities, to feel assured of one's place despite everything, to be confident that all were equal in God's eyes" (p.70).

However appealing such sociohistorical speculation may be, especially for those who wish to re-valorize a culturally marginalized Judeo-Christian discourse of sin and redemption, it must be pointed out that such beliefs and practices, however useful they may have been to those already struggling to work through depressive anxieties, seem to have been of limited use in helping those caught up in schizoid and paranoid processes to overcome the sadomasochism, scapegoating, magical thinking and other individual and collective forms of destructiveness that have been far from absent in ages and communities of faith.  Carroll (1985) describes the typical man of the Middle Ages as “violent, immoderate, perceiving the world in childlike extremes, his mind inhabited by a phantasmagoria of devils and angels, ghosts and holy relics” (p. 97) and states that “The disposition of medieval man was that of the delinquent.  It was violent and impulsive, without capacity for restraint or moderation.  Tempestuous uninhibited passion was never far from the surface.  Everything was in extreme, of the cruelest punishment one moment, mercy the next, barbarism giving way to copious weeping” (p. 102).  In calling for the reculturing of guilt and the provision of cultural forms for its expression, we must keep in mind this disjunction between the level of cultural discourse on the one hand and that of the actual experience and psychic functioning of individuals and groups on the other.

Without denying the role of broader social processes in shaping the familial and childhood experiences that structure character and disposition, it seems to me that guilt is and always has been hard to bear and guilt-evasion is and always has been prevalent.  Aside from the sociocultural factors that shape and channel guilt, the forms taken by guilt-evasion vary as different cultures and historical periods offer differing versions of what Shorter (1992) has called the "legitimate symptom-pool" (p.x).  If one culture legitimates hysterical paralyses and compulsions, another recognizes multiple personality disorder and environmental illness.  The guilty subject who cannot bear feeling guilty will evade guilt-feeling one way or another.

"A person will spend his whole life writhing in the clutches of the superficial, psychological symptoms of guilt unless he learns to speak its true language" (Carroll, 1985, p.15).  One of such superficial psychological symptoms is the manifestly absent subjectivity of the so-called "postmodern psychotic non-subject."  The challenge facing this self-evading subject is that of recognizing it is a subject after all and, like all culpable subjects, of facing and bearing its guilt, integrating as a part of the tragic dimension of human existence the reality of our primordial ambivalence, and accepting as an aspect of "common human unhappiness" the need to shoulder the burden of responsibility to make reparation and to change that genuinely facing our guilt entails.  Facing and bearing guilt opens the path toward restoration of a sense of inner goodness through reparative processes mediating identification with resurrected, surviving, good and forgiving internal objects.  If advance in civilization entails an increased capacity to confront and bear guilt, then a first step may be to learn to speak its true language, not least by ceasing to confuse it with the self-torment that represents its evasion.




* This is a substantially revised version of a paper originally presented at an International symposium on Freud’s Civilization and Its Discontents, McMaster University, Hamilton, Ontario, September 1999, and that subsequently appeared in Psychoanalytic Studies 3, 1 (March 2001): 9-21.

[2] I think true guilt or concern may be temporarily repressed, but if its repression is at all extended it rather quickly ceases to be guilt at all—that is, its repression represents the beginning of a regression in which the superego resorts to self-punishment as a substitute for and a defense against guilt.

[3] Although in principle it should be possible to punish oneself and simultaneously make reparation to the other, there seems to be a tendency, as described in Freud’s (1914) U-tube theory, to invest either in narcissism or object love, to attend to the wounded other or allow him to bleed while self-flagellating instead of bandaging. Perhaps this is another instance of the question as to whether paranoid-schizoid and depressive dynamics can operate simultaneously or necessarily oscillate.

[4] When Freud refers to the "torments--the pangs of conscience" by which "a conscious sense of guilt, a consciousness of guilt, expresses itself" (p. 166), one wonders whether such pangs represent authentic guilt or the self-torment that so often defends against it.  Experiencing the pangs of conscience either initiates reparative efforts that tend to reduce self-torment by restoring some positive self-esteem, or leads to chronic, conscious or unconscious, self-punishment. In my view, bearing guilt does not mean suffering perpetual pangs, but acknowledging and seeking insofar as possible to repair the damage done, in this way restoring one’s good objects and one’s self-esteem, yet living in the awareness of the badness that inevitably accompanies one’s goodness.


 [5] Painful pangs of conscience can either lead to constructive reparative activity or substitute for it.  In the latter case, self-torment replaces guilt.  I am defining guilt in terms of its consequences.  By its fruits you shall know it.  If it results in reparation toward the object, it is guilt.  If it results in self-torment it is not. 


[6] This inability to be bad while at the same time being good—not needing to be all-good as the only alternative to being all-bad—interferes with the subject’s enjoyment of the pleasures of playful transgression and leads to a flattening, an impoverishment, in the domain of sexual and other forms of play and creativity.  As Stoller (1974) and Kernberg (1991) have both emphasized, “Sadomasochism, an ingredient of infantile sexuality, is an essential part of normal sexual functioning and love relations, and of the very nature of sexual excitement”—as are “bisexual identifications, the desire to transgress oedipal prohibitions and the secretiveness of the primal scene, and to violate the boundaries of a teasing and withholding object” (Kernberg, 1991, p.333).




Barrett, Wm. (1958).  Irrational Man: A Study in Existential Philosophy.  Garden City, New York:   Doubleday, 1962.

Carroll, J. (1985.  Guilt: The Grey Eminence Behind Character, History and Culture. London: Routledge & Kegan Paul.

Carveth, D. (1996).  Psychoanalytic conceptions of the passions. In Freud and the Passions, ed. J. O'Neill.  University Park, Pennsylvania:

    The Pennsylvania State University Press, 1996, ch. 2., pp. 25-51.

-----. (1998).  Freud's flawed philosophy of religion: a reply to Rempel. Canadian Journal of Psychoanalysis/Revue Canadienne de

    Psychanalyse 6, 1 (Spring 1998): 141-7. 

-----. & J. Hantman Carveth (2003).  Fugitives from guilt: postmodern de-moralization and the new hysteriasAmerican Imago 60, 4: 445-479.

Erikson, E. (1950). Childhood and Society.  New York: Norton.

-----. (1956).  Identity and identity diffusion. Journal of the American Psychoanalytic Association 4: 56-121.

Fairbairn, W.R.D. (1952). An Object-Relations Theory of the Personality. London: Routledge.

Forster, S. & D. Carveth (1999).  Christianity: a Kleinian perspective. Canadian Journal of Psychoanalysis/Revue Canadienne de

    Psychanalyse 7, 2: 187-218.

Finlay de Monchy, M. (1997). Post-modernizing psychoanalysis / psychoanalyzing post- modernity--modernity before emancipation--

    re-ontologizing the subject in discourse.   Paper presented at the Conference on After Postmodernism, University of Chicago,

    November 14-16, 1997.  A shorter version of this text appeared in Free Associations 16 (1989). Available online at:

Freud, S. (1924).  On narcissism: an introduction.  S.E., 14: 69-

-----. (1916). Some character-types met with in psycho-analytic work. S.E., 14: 311-333.

-----. (1923). The Ego and the Id. S.E., 19: 3-66.

-----. (1924). The economic problem of masochism. S.E., 19: 157-173.

-----. (1930). Civilization and Its Discontents. S.E., 21: 59-145.

-----. (1933). New Introductory Lectures on Psychoanalysis. S.E., 22: 1-182.

Grinberg, L. (1964). Two kinds of guilt: their relations with normal and pathological aspects of   mourning. Int. J. Psycho-Anal., 45:366-371.

Karen, R. (1992). Shame: The rediscovery of what some psychologists regard as the primary    cause of emotional distress. The Atlantic 269,

    2 (February1992): 40-70.

Kernberg, O. (1991).  Sadomasochism, sexual excitement, and perversion.  J.A.P.A.,   39: 333-362.

Kohut, H. (1971). The Analysis of the Self.  New York: Int. Univ. Press.

-----. (1978). The Restoration of the Self.  New York: Int. Univ. Press.

-----. (1979). The two analyses of Mr. Z. Int. J. Psycho-Anal., 60:3-27.

Lasch, C. (1979). The Culture of Narcissism: American Life in an Age of Diminishing Expectations. New York: Warner.

Marcuse, H. (1963). The obsolescence of the Freudian concept of man. In Five Lectures: Psychoanalysis, Politics, and Utopia.

    Trans. J.J. Shapiro and S.M. Weber. Boston: Beacon, 1970.

Riesenberg-Malcolm, R. 1980. Expiation as a defense. International Journal of Psychoanalytic Psychotherapy 8; reprinted in Revista de la

    Asociacion Psicoanalitica de Buenos Aires 3 (1981); and, as "Self-punishment as defence," in On Bearing Unbearable States of Mind.

    Ed. P. Roth. London & New York:  Routledge, 1999, chapter 5, pp. 93-112.

Reisman, D., Glazer, N. & R. Denney (1961). The Lonely Crowd: A Study in the Changing American Character. New Haven: Yale

    University Press.

Safan-Gerard, D. (1998). Bearable and unbearable guilt: a Kleinian perspective. Psychoanal. Quart., 67: 351-378.

Shorter, E. (1992).  From Paralysis to Fatigue: A History of Psychosomatic Illness in the   Modern Era.  New York: The Free Press.

Stoller, R.J. (1974).  Hostility and mystery in perversionInt. J. Psychoanal., 55: 426-434.

Ury, C. (1998). The Nietzschean monster: reconsidering guilt in developmental theory. Canadian Journal of Psychoanalysis/Revue

    Canadienne de Psychanalyse 6, 1 (Spring, 1998): 51-74.

Winnicott, D.W. (1963). The development of the capacity for concern. In Maturational Processes and the Facilitating Environment.

    London: Hogarth Press, 1965, pp. 73-82.