Thomas J. Scheff
(Aug. 4, 1997)


This essay formulates a theory of violence, showing how it is generated at the individual and group levels. I propose that killing rage is generated in individuals by insults or rejection which result in shameful feelings that are not acknowledged. Collective violence has a parallel dynamic: when persons feel they do not belong anywhere, but do not acknowledge this feeling to themselves or others, they may form a group which rejects the group or groups rejecting them. The culture of such groups generates techniques of neutralization which encourage hatred and mayhem. First I explore the development of rage sequences between individuals, tying them to the state of social bonds and the emotion which signals that state, shame. Then I outline the social and cultural conditions that give rise to collective hatred and rage. Finally, I review what is known about the opposite of rage, seeking forgiveness through apology. In the last section, I discuss possible steps toward forgiveness and reconciliation.


Rage is an important component of killing and maiming on a large scale. There are indications in primitive epics of warfare that warriors worked themselves into paroxysms of rage. It is known historically that in early times Irish warriors fought naked, with nothing but a sword, running and screaming at a high pitch, often casting terror into their enemies. Here is an example from a 12th century Irish epic, The Tain (cited in Cahill 1995):


[Cuchulainn then] went into the middle of them and beyond, and mowed down great ramparts of his enemies' corpses, circling completely around the armies three times, attacking them in hatred. They fell sole to sole and neck to headless neck, so dense was that destruction. He circled them three times more in the same way, and left a bed of them six deep in a great circuit, the soles of three to the necks of three in a circle round the camp.... Any count or estimate of the rabble who fell there is unknown, and unknowable. Only the names of the chiefs have been counted.... In this great Carnage on Murtheimne Plain Cuchulainn slew one hundred and thirty kings, as well as an uncountable horde of dogs and horses, women and boys and children and rabble of all kinds. Not one man in three escaped without his thighbone or his head or his eye being smashed, or without some blemish for the rest of his life. And when the battle was over Cuchulainn left without a scratch or a stain on himself, his helper or either of his horses.

This passage tries to convey intense fury by exaggeration, since it is unlikely that a single individual, no matter how powerful, could have enough stamina to cause such wholesale destruction.

I propose that certain emotions in sequence, and the social and cultural settings which generate these emotions, are key causes of mass rage and violence. But most social science writing on violent conflict assumes a "realist" or materialist perspective, that the real causes of human conduct always involve physical, rather than social and psychological reality. But eliminating emotional and relational elements as causes of violence is a gross error. Of course I am not arguing that material conditions are unimportant, only that violence is caused by a combination of physical and social/psychological elements. I will describe rage and violence first at the level of individuals, then at the collective level, showing how hatred and rage are products of unacknowledged emotions, which are in turn generated by alienation and by cultural scripts for demonizing purported enemies.

An immediate problem in making this argument persuasive is the difficulty of describing in words the experience of rage and other compelling emotions. When my readers are sitting the comfort of their study, feeling more or less safe and secure, it will take some effort to help them visualize the intensity of "war fever," or of the feelings that lead to massacres on a vast scale. The intensity and primitiveness of fury beggars verbal description. How is one to convey intense feelings with mere words? Once again, I resort to archaic literature where this difficulty was dealt with by florid exaggeration, so that the words could point the reader toward the intensity of the actual feelings. These words, I take it, are not meant to describe outer reality, they are far too gross, but instead, to convey inner, experiential reality, the objective correlative, as T.S. Eliot called it, of a fit of rage.


This is another example from the Tain (Cahill 1995) describing the outward appearance of a warrior in a fit of rage:

The first warp-spasm seized Cuchulainn, and made him into a monstrous thing, hideous and shapeless, unheard of. His shanks and his joints, every knuckle and angle and organ from head to foot, shook like a tree in the flood or a reed in the stream. His body made a furious twist inside his skin, so that his feet and shins and knees switched to the rear and his heels and calves switched to the front. The balled sinews of his calves switched to the front of his shins, each big knot the size of a warrior's bunched fist. On his head the temple-sinews stretched to the nape of his neck, each mighty, immense, measureless knob as big as the head of a month-old child. His face and features became a red bowl: he sucked one eye so deep into his head that a wild crane couldn't probe it onto his cheek out of the depths of his skull; the other eye fell out along his cheek. His mouth weirdly distorted: his cheek peeled back from his jaws until the gullet appeared, his lungs and liver flapped in his mouth and throat, his lower jaw struck the upper a lion-killing blow, and fiery flakes large as a ram's fleece reached his mouth from his throat. His heart boomed loud in his breast like the baying of a watch-dog at its feed or the sound of a lion among bears. Malignant mists and spurts of fire...flickered red in the vaporous clouds that rose boiling above his head, so fierce was his fury. The hair of his head twisted like the tangle of a red thornbush stuck in a gap; if a royal apple tree with all its kingly fruit were shaken above him, scarce an apple would reach the ground but each would be spiked on a bristle of his hair as it stood up on his scalp with rage. The hero-halo rose out of his brow, long and broad as a warrior's whetstone, long as a snout, and he went mad rattling his shields, urging on his charioteer and harassing the hosts. Then, tall and thick, steady and strong, high as the mast of a noble ship, rose up from the dead centre of his skull a straight spout of black blood darkly and magically smoking like the smoke from a royal hostel when a king is coming to be cared for at the close of a winter day.

The extraordinary intensity of enraged actions, as described in the first quotation, and of the experience of rage, as suggested in the second, leads to the belief that rage is a virtually irresistible force and that it is an elemental component of human nature. This essay will contradict both of these beliefs, first that it is an elemental, and secondly, that it is irresistible.

The ability of primitive warriors to work themselves into a state of rage suggests that it is something that can be constructed, rather than an elemental. How did these warriors do it? We will probably never know the answer to that question. But studies of actual discourse suggest a sequence of events that inevitably occur prior to the outbreak of violent rage. At the group level, it would seem that alienation and certain cultural beliefs militate toward states of hatred and rage, and violent behavior. I will begin with individuals.

I suggest that rage is a composite affect, a sequence of two elemental emotions, shame and anger. This idea has been advanced by other authors, notably Heinz Kohut (1971), and Helen Lewis (1971). Kohut proposed that violent anger of the kind he called "narcissistic rage" was a compound of shame and anger. Helen Lewis suggested that shame and anger have a deep affinity, and that it might be possible to find indications of unacknowledged shame occurring just prior to any intense hostility. This sequence has been demonstrated to occur in marital quarrels by Retzinger (1991), and in Hitler's writings and speeches (Scheff 1994), exactly as Lewis proposed. With all sixteen of the episodes of escalation of verbal violence in her data, Retzinger was able to demonstrate that prior to each episode, there had been first an insult by one party, indications of unacknowledged shame in the other party, and finally intense hostility in that party. This sequence can be seen as the motor of violence, since it connects the intense emotions of shame and anger to overt aggression.

Although there has been little research focused explicitly on pure, unalloyed anger, there are indications from the studies of discourse by Lewis (1971), Retzinger (1991) and my own work, (such as Scheff 1990) that pure anger is rare and unlikely to lead to violence or even social disruption. On the contrary, anger by itself is usually brief and instructive. A person who is frustrated and unashamed of her anger is mobilized to tell what is going on, and to do what is needed, without making a huge scene. In my own personal case, I can testify that most of my experiences of anger have involved shame/anger, either in the form of humiliated fury, or in a more passive form, what Labov and Fanshel (1977) call "helpless anger." Both of these variants are long lasting and extremely unpleasant, especially for me. Shame-induced anger was unpleasant while happening, and even more unpleasant when it was over, since I inevitably felt foolish and out of control.

But in the very few episodes of what seems to have been, in retrospect, pure anger, the experience was entirely different. I did not raise my voice in any of them, nor did I put any one down or any other kind of excess. I simply told my view of what was going on directly, rapidly and with no need of calculation or planning. I was overcome with what I call "machine gun mouth." Every one who was present to one of these communications suddenly became quite respectful. As for me, I did not feel out of control, even though my speech was completely spontaneous; on the contrary , I was wondering why I had not said my say before. It would seem that anger without shame has only a signal function, to alert self and others that one is frustrated.

When anger has its source in feelings of rejection or inadequacy, and when the latter feelings are not acknowledged, a continuous spiral of shame/anger may result, which is experienced as hatred and rage. Rather than expressing and discharging one's shame, it is masked by rage and aggression. One can be angry one is ashamed, and ashamed that one is angry, and so on, working up to a loop of unlimited duration and intensity. This loop is the emotional basis of lengthy episodes or even life-long hatred and rage.


Social and cultural conditions for the development of intergroup animosity

In a companion paper to this one, I describe how a certain type of alienation, which I characterize as bimodal, generates violence at the collective level (Scheff 1997). I will give only a brief synopsis here. In my terminology, bimodal alienation between groups occurs when there is "isolation" between them, but "engulfment" within them. That is, the members of group A are distant from the members of group B, and vice versa. But at the same time, the members of each group are suffocatingly close, to the point that they give up important parts of themselves, in order to securely belong to the group.

The initial motor in this theory is the need to belong. It makes sense that the German language has the most beautiful word for home, in the sense of the place that you belong: das Heimat. It makes sense because as both Elias (1995) and I (1994) have independently argued, historically the Germans seem to have long had an unsatisfied yearning for a place in which they belong, and have had great difficulty in managing the feeling of rejection, of not belonging and being accepted. Members of a group who feel not accepted both by foreigners and in their own group are in a position to surrender their individual identity in order to be accepted, giving rise in the German case to the principle of blind loyalty and obedience. But in any case, bimodal alienation (isolation between groups and engulfment within them) is the fundamental condition for inter-group conflict.

Under the condition of bimodal alienation, a special culture seems to develop within each group which encourages hatred and murder. There are various ways of characterizing this culture, but for my purposes here I will describe it in terms of "techniques of neutralization." This idea was originally formulated in criminology (Sykes and Matza 1957) to explain how and why teenagers engage in delinquent behavior, how a special culture develops among them which neutralizes the norms in their larger culture which oppose crime. But the idea has also been carefully applied by Alverez (1997) to the behavior of the German people in tolerating or actually engaging in genocide.

Alverez shows how each of Sykes and Matza's five techniques of neutralization can be used to explain the special culture that developed during the Nazi regime, a culture which neutralized the norms in the larger culture that forbid murder. The first technique is the Denial of Responsibility. Alverez shows that this technique in the German case usually took the form that the perpetrator was only carrying out orders from above. 2. The Denial of Injury under the Nazi regime took the form of special language which hid or disguised what was actually being done, euphemisms in which killing became "special treatment," "cleansing" (recently in the news about the massacres in Bosnia) and many other similar examples. 3. The Denial of Victim asserts that the victim actually brought on their own downfall. In the German case, it was widely believed that Jews were involved in a conspiracy to enslave the whole world, so that killing them was self-defense. Although a fabrication, many Germans, including Hitler, appeared to have believed it to be literally true.

4. Condemning the Condemners involved, in the German case, claims made by the German government and the media that the other countries that were condemning Germany were historically guilty of worse crimes, such as the treatment of blacks and Native Americans in the United States and the treatment of native peoples in the French, British and Spanish colonies. 5. In the Appeal to Higher Loyalties, German perpetrators of genocide thought of themselves as patriots, nobly carrying out their duty. 6. Finally, the Denial of Humanity is a category that Alverez himself added to those formulated by Sykes and Matza because of its special relevance to the Holocaust. Typical Nazi propaganda portrayed Jews and other non-Aryans as subhuman, filled with bestial impulses, such as the urge for destruction, primitive desires, and unparalleled evil. Although dehumanization often accompanies inter-group conflict, it seems in the German case that it was explicitly orchestrated by the government.

Any one of these six techniques can serve to encourage violence by neutralizing the norms against aggression and murder. To the extent that they are all implemented together, as they apparently were under the Nazi regime, to that extent a whole society can seemingly forget its moral values, in order to engage in wholesale slaughter. The idea of techniques of neutralization suggests the cultural foundation for collective violence. At the end of this paper I will suggest ways of reducing violence by countering techniques of neutralization. In the remainder of this section, I will focus on the issue of reducing the emotional bases of violence by dealing with shame that has gone unacknowledged.

How can spirals of unacknowledged shame and anger, which are the emotional basis of hatred and rage, be avoided or terminated when they are occurring. ? One answer may lie in the direction of acknowledgment of shame. By acknowledgment, however, I am not referring to merely verbal acknowledgment. Unfortunately, there have been very few discussions of acknowledgment in the literature on emotion. It is one of those terms like "working through" in psychoanalysis, which play a central role in professional discourse, but which are seldom defined or even illustrated through concrete examples. Since there is no literature focused directly on the acknowledgment of shame, I will resort to studies of apology, which deal with acknowledgment, if only indirectly. This literature is helpful since it places emotions and their acknowledgment within a social context. Seeking forgiveness through apology also helps to locate and conceptualize rage, because it is the opposite process: hatred and enraged behavior shatter the social bond, apology seeks to re-instate it.



Reconciliation, repairing a disruption in a social relationship, can be seen as an acknowledgment of interdependence. Viewed in this way, apologies, a direct path to repair of social bonds, expose many of the key elements in conflict and reconciliation. Describing the components of the ritual of apology may be a succinct way of unpacking the concept of acknowledgment.

The two leading theorists of apology and reconciliation are Goffman and Tavuchis. Goffman (l971) considered apology in his treatment of remedial and reparative devices. Even though his discussion of apologies is brief, it bristles with ideas (l971, p. ll3, emphasis added):


...apology has several elements: expression of embarrassment and chagrin; ...that one knows what conduct has been expected...; ...disavowal of the wrong way of behaving and vilification of the self that so behaved; ...avowal henceforth to pursue that course; performance of penance and the volunteering of restitution.

Two aspects of this long sentence catch the eye. First, in a few lines, Goffman has loaded in not one or two, but seven necessary elements for a successful apology, a monument to concision.

The other feature of this description that invites comment is that the first element of an apology that Goffman invokes is "expression of embarrassment and chagrin", which connects his argument to mine concerning shame. Moreover, placing this element at the head of the list possibly suggests that it should come first in time, or that it might be the most important condition, or both.

Goffman's discussion has proven fruitful in many directions, but it has also been criticized by Tavuchis (l991), whose discussion of apologies is much more comprehensive and detailed. Unlike Goffman, Tavuchis readily extends his analysis beyond interpersonal apology to individual-collective and collective remedial actions. He devotes chapter-length treatments to situations of apology of the one to the many, the many to the one, and the many to the many. In his treatment of interactions between individuals and groups, he covers some of the same ground as Braithwaite (l989), but in much more detail. The ceremonies of punishment and reintegration of offenders which Braithwaite describes involve the repair of bonds between the one and the many, which Tavuchis discusses in great detail. Of particular relevance to group conflict is Tavuchis's discussion of apologies of the many to many: the path to reconciliation between groups.

Tavuchis's two central complaints about Goffman's are that in the main it concerns individuals, rather than relationships, and that to these individuals an apology may be a game in which the actor is not emotionally involved (l991, l38):


...apologies (for Goffman) are conceptualized as a "set of moves" or interpersonal management ploys used by social disembodied actors trying to maximize their (questionable) moral credibility...Goffman argues that an apology entails the "splitting" of the self, whereas I underscore the necessity of "attachment" to the offense...there is no mention of what I take to be central to apology: sorrow and regret... an actor could follow all the steps described by Goffman without producing a speech act that is socially recognizable as an apology, or, its moral reciprocal, forgiveness.

Both of Tavuchis's criticisms cut close to the bone. Although Goffman uses many relational terms, his basic frame of analysis is individualistic rather than relational; it concerns a harried, anxious individual seeking to maintain her/his sense of self and status in a jungle of trying situations. The language of rules and norms, which Tavuchis also uses, is itself not quite relational, since it also emphasizes the individual as much as the social bond. Perhaps a new language of relationships is needed. In the physics of light, mathematical language provided the link between particle and wave formulations. We have no such language for linking individuals and relationships.

Tavuchis's second criticism is also just. Goffman's analysis is largely behavioral; it concerns the surface of interaction, with very little access to the interior, the meaning of events to the actors. Even Goffman's mention of embarrassment and chagrin as necessary parts of apology involves an ambiguity. He does not say that the actor should feel these emotions, but only that they should be expressed. In Goffman's world, a gesture indicating embarrassment (covering the face with a hand, for example) might be adequate, even if were merely enacted in the absence of any feeling. (In The Odyssey, when addressing an audience, Ulysses would wipe his eyes with the hem of his robe to indicate grief. The wily Ulysses would fit exactly into Goffman's scripts)

In Tavuchis's (p. 3l) description of apology, there are two essential parts, both equally important and equally necessary. One must say one is sorry, and one must also feel sorry. Without these two components, the ritual is incomplete. Tavuchis's critique of individualistic and behavioral bias locates two important flaws in Goffman's approach. To test the adequacy of some particular apology, one can invoke what I will call Tavuchis's rule: an apology will be genuine to the extent that one both says one is sorry, and actually feels sorry. His rule concerns congruence between outer and inner, as already mentioned above. An actual instance of a defective apology will provide an example.

Albert Speer, one of Hitler's chief lieutenants, was tried and convicted of crimes against humanity at Nuremberg. Unlike any of the other defendants, however, his life was spared. Instead of being executed, he spent 21 years in solitary confinement at Spandau prison, outside of Berlin. During this period, he wrote his memoirs (Speer 1970). One recurring theme is the regret he expresses about the role he played in Hitler's Germany. In some ways, the book can be read as an apology for his actions.

If the book is an apology, however, it doesn't seem to be an adequate one; it is frequently off pitch in some essential way. It is true that some of the apologetic statements sound genuine. Tavuchis (1991, p. 21) points to an excellent one. But even one false note in a text can call into question the validity of an entire apology: perhaps most of the words sound right, but does the one apologizing actually feel sorry? There are many lapses in Speer's memoirs.

In an insightful review, George Steiner (1971) describes Speer's attempts at apology as "motions, presumably sincere in their own hollow, cerebral way, of retrospective horror." Although Steiner did not provide examples to show how hollow and cerebral Speer's attempts were, they can be easily found in the text. One (Speer, 1970, p. 24) begins with a statement that almost strikes the right note:


By entering Hitler's party I had already, in essence, assumed a responsibility that led directly to the brutalities of forced labor, to the destruction of war, and to the deaths of those millions of so-called undesirable stock, to the crushing of justice and the elevation of every evil.


Even in this passage there is questionable phrasing. But it hints at some feeling of responsibility and remorse. One touch is that he heads his list of crimes with "the brutalities of forced labor," a crime in which he was directly implicated as the overseer of the armament industry.

But in the rest of the paragraph, the tone falters: "In 1931 [when he joined the Nazi party] I had no idea that fourteen years later I would have to answer for a host of crimes to which I subscribed beforehand by entering the party." Although somewhat indirect, this sentence seems to be more of an excuse ("I didn't know") than an apology. It denies his own responsibility for his actions by implying that once having joined the party, a youthful folly, blind loyalty was inescapable.

The paragraph ends with a whimper: "I did not yet know that I would atone with twenty-one years of my life for frivolity and thoughtlessness and breaking with tradition." There is a thread of self-justification and self-pity running through the entire paragraph, faint at first, but in the last sentence, dominant. Instead of expressing gratitude that his life alone was spared, of all the Nuremberg defendants, he seems to complain about the length of the prison sentence. The most shocking element however, is the terms used to describe the causes of his adherence to the Nazi Party: "frivolity, thoughtlessness, and breaking with tradition." These terms would be appropriate if he had participated in a panty raid, but not in crimes whose scope and vileness beggars the imagination.

Speer's apology fails the second part of the Tavuchis rule. Although Speer says many times that he is sorry, the way he says it suggests that he does not feel sorry. His failure occurs with both cognitive and emotional aspects. In terms of the cognitive content, self-justification is exactly the opposite of what is required in an apology, taking responsibility for one's own actions with no excuses.

Speer's apology also fails to pass the at the emotional level. The connotations of many of his words and phrases work against the expression of remorse. One example is the phrase "undesirable stock." Although Speer qualifies this phrase with "so-called," it still strikes the wrong note. He is referring to the victims of Nazi atrocities, such as the Jews and Slavs who were murdered or worked to death. To use a term from his Nazi past rather than from the present day is an appalling blunder, since it suggests that in some ways his point of view, and therefore his feelings still have not changed. Sustaining the right tone in an apology seems to require actually feeling remorse, the second part of the Tavuchis test.

My discussion of the Speer case suggests that Tavuchis's analysis is helpful in understanding the nature of apology, and therefore of acknowledgment of feelings. However, I would like to extend the description of the core ritual of apology further than Tavuchis takes it, in order to resolve an issue that he leaves unresolved. In many different passages, Tavuchis puzzles over a mystery: how can mere words resolve conflict? He notes that an apology, however fastidious, does not undo the harmful act. Tavuchis repeatedly indicates that successful apologies are like magic: using only words, one can obtain genuine forgiveness for an injurious act (l991, 122):


...although I have referred frequently to forgiveness as a crucial element in the apologetic equation, this mysterious and unpremeditated faculty has not been adequately addressed or formulated. If, as I have argued, sorrow is the energizing force of apology, then what moves the offended party to forgive? ...[the] social and psychodynamic sources [of forgiveness] have been relatively neglected.

Shame dynamics, which Tavuchis does not invoke, may speak to the issue which Tavuchis raises: what are the social and psychodynamic sources of apology and forgiveness? I will deal first with psychodynamic sources, the vicissitudes of emotions and feelings, then with social ones.

I agree with the first part of Tavuchis's rule: one must say one is sorry, or words to that effect. But in order to understand the magic of apology, it may be necessary to unpack the second part of the formula, that one must also feel sorry. What are the emotional components of feeling sorry? In the passage just quoted, Tavuchis states that sorrow, that is, grief, is the energizing force of apology. I disagree. Although feeling and displaying grief might be helpful, it may not be the main emotion required for an apology to accomplish its purpose.

I propose that an effective apology requires that the predominant emotion of the party making the apology be one of embarrassment or shame. This is difficult for the participants to see, or even for observers, because in our civilization, embarrassment and shame are so frequently and deeply disguised and denied as to be rendered almost invisible (Lynd, l958; Lewis, l971; Scheff, l990; Retzinger 1991). Although Tavuchis discusses shame at several points (as in note 4, p. 151), it does not figure prominently in his discussion. It does figure prominently in Miller's (1993) discussion of apologies; he proposes, as I do, that shame or embarrassment are the predominate emotion in a genuine apology.

Suppose, for purpose of argument, after one party has been injured by another, both parties are in a state of embarrassment or shame. Depending on the gravity of the injury, the intensity of shame may range from slight embarrassment through severe, lengthy states of humiliation. The injured party may feel helpless, rejected, powerless or inadequate, because of the treatment received: the injuring party may feel unworthy because he/she has injured the other. All of these terms have been rated as encoded references to shame (Gottschalk and Gleser, l969; Lewis, l971; Retzinger 1991). The shared mood of the two parties is bleak: they are in a state of shared embarrassment or shame.

The function of apology under these conditions is to allow both parties to acknowledge and discharge the burden of shame they are carrying with respect to the injurious act, rather than deny it. This function is difficult for the parties to be aware of in Western societies, because shame is routinely denied. Perhaps if social psychology brought unacknowledged shame to light, we might be able to understand and increase the magic of apology.

An effective apology is also difficult because it depends on a veritable symphony of verbal and non-verbal activities jointly enacted and felt by both parties. Each must coordinate their words, gestures, thoughts and feelings with those of the other. It is a dance, a pas de deux, requiring not only the right words, the lyrics, but also the right music. That is, the timing (rhythm) of the moves of each party, relative to the moves of the other, is crucial, as are the emotions displayed (melody) and felt (harmony).

This formulation in terms of emotion dynamics may remove some of the mystery from apology and forgiveness . If the lyrics and the melody of the party apologizing are right, and the attitude of the party apologized to accepting, then a dramatic mood change can occur: the parties can go from a state of shared shame to one of shared pride in a matter of minutes, from fluster, awkwardness, and emotional pain to rapport and pleasure.

My explanation of the apology/forgiveness process to this point is incomplete, however, because it has concerned only the emotional sources, not the social ones. To move toward these sources requires discussion of the nature of social relationships and social bonds at an abstract and general level.

Social relationships are difficult to describe in Western societies because human interdependency, like shame, is also routinely denied. Our public discourse is in the language of individuals, rather than relationships. A social bond may be defined in terms of the mix of solidarity and alienation (Scheff, l990; Retzinger, l991). A secure bond is a relationship in which solidarity prevails: accurate understanding of both parties of the other's thoughts and feelings, their short and long-term intentions, and their character prevails over misunderstanding or lack of understanding. In an insecure or threatened bond alienation dominates: lack of understanding or misunderstanding in these matters on one or both sides. Most social bonds are a mix, but either solidarity or alienation predominates.

As already indicated, alienation occurs in two different formats. Using Elias's (l989) "I-we" language: a secure bond requires balance between the importance of the individuals and importance of the relationship. Too much emphasis on the individual means isolation; each cannot know the other and reveal the self because they are too distant. Too much emphasis on the relationship means engulfment: each cannot know the other and reveal the self because loyalty and conformity demands that important parts of the self, basic desires, thoughts and feelings may be hidden, even from ones' self. Secrecy, deception, and self-deception go hand in hand. Modern societies tend toward individuation and isolation, traditional ones toward conformity and engulfment. Both formats are equally alienated.

The apology/forgiveness transaction signifies the removal of a threat to the social bond. In relational language, in every moment of every encounter, the bond is either being maintained, strengthened, repaired, or damaged. This is one of Goffman's (l967) central themes: every action (or inaction) by each party has an effect on the relative status and sense of self of the parties, without exception.

By verbal and nonverbal means, an effective apology is a master stroke in this scenario, a repair of a threatened or insecure bond. When one party has injured another, the bond is threatened, the parties are disconnected emotionally and/or cognitively, i.e., they are in a state of shame. A successful apology allows both parties to acknowledge and discharge the shame evoked by the injury. The apology "makes things right" between the parties, both emotionally and cognitively, it repairs the breach in the bond. The success of the action of repair is felt and signaled by both parties: they both feel and display the emotion of pride.

Bond language is needed if we are to understand and describe the process of denial and acknowledgment: acknowledgment of the state of a relationship (the degree of attunement and its accompanying emotions) leads to building or repair of bonds, denial, to damage to bonds.

Tavuchis's analysis of apology and my commentary on it suggests the complexity of the concept of acknowledgment and of its practice in real life, even at the interpersonal level. At the level of relations between nations, the issues are further complicated by the large number of participants, the vastly increased volume of discourse, and particularly by the lack of consensus which characterizes modern societies. In the theory proposed here, acknowledgment of interdependency and emotion, the state of the bond, is crucial not only for individuals, but for whole societies.


This discussion points toward several paths for conciliation between belligerent groups. My theory of conflict suggests that the foremost cause is mass alienation within and between the groups at enmity. Any steps which would decrease mass alienation would automatically lessen pressure toward conflict. In the paper on alienation which is a companion to this one, I propose that teachers need to be retrained to be aware of the way in which they reject working class and minority students. I also ask for classes on family relations which would help young people form stable families. Also in that paper I recommend reform for welfare programs to lessen rejection and shame. Young men form the bulk of combatants for inter-group and international conflict. If they could be better integrated into work or welfare, school, and family, they would be less vulnerable to pressure to fight an external enemy.

At the level of culture, to undermine the sources of inter-group conflict, we need to counter the techniques of neutralization (Sykes and Matza 1957; Alverez 1997) that are used to foment hatred and violence toward purported enemies. Although there are attempts to control the reaffirmation of hatred and hatred in the mass media, they still have not been comprehensive enough to help reduce the pressure toward violence. An obvious example is the continuing sexism and violence toward women in major films, not to mention fringe films. An expensive film like "Revenge," with major stars (Kevin Costner, Madeleine Stowe, Anthony Quinn) degrades women and encourages violence toward them, yet is being widely distributed and shown both in theaters and on TV and video. Although racism and xenophobia has been toned down somewhat, it still forms an undercurrent in many current films. It seems particularly flagrant in "action" films (such as those produced by Sylvester Stallone). Needless to say, both sexism and racism is rife in most of the old films which are
constantly being rerun on TV.

        Learning to identify and acknowledge shame in self and others is also a fundamental direction toward decreasing conflict. I have shown in this essay that alienation and unacknowledged shame are basic causes of destructive conflict, as important as material causes. Obviously material interests matter in human affairs. They are topics of quarrells. But these interests can always be negotiated, if there is no unacknowledged emotion, in a way that allows parties maximum benefit or perhaps least destructive outcomes. Unacknowledged shame figures large because it make rational negotiation of interest difficult or even impossible, given the non-rational, that is, the elements of insult and rejection when shame is not acknowledged by both parties.

The conflict in Northern Ireland can be taken as a current example of this point. This conflict has been going on for over half a century, at great expense to life, limb and pocketbook. At this writing, no resolution is in sight . The three parties most affected by the conflict are the Catholic and Protestant factions in N. Ireland, and the English police and military stationed there. The rational solution would be a scheme that would allow the sharing of power between the two factions, and the removal of the English presence. That solution would benefit all sides. It would stop the loss of life and physical damage that is occurring, and the drain on the funds that English law enforcement is costing the UK, some three billion dollars a year. Surely a way can be found to reach this goal, since all sides would gain.

Here are four suggestions. The first two begin at the local level with peace-making. In one of the papers that accompanies this one (Scheff 1997a), I suggest that all political and terrorist crimes (excepting capital crimes) be managed through "community conferences," rather than through the courts. This step would directly involve the two conflicting factions in peace-making, since they would have to arrive at a resolution of each political and terrorist crime together. This move would further benefit the peace process in that the two factions would be also put into a situation of having to share and resolve the shame that accompanies each crime, as I explain in the paper.

A second suggestion is also the local level. It is a very long step from hatred and rage to peace and forgiveness. Ultimately, when some of the smoldering rage has died down, it will probably be necessary to the three parties to the conflict to exchange apologies, in the manner outlined by Tavuchis (1991) in his discussion of group-to-group apologies. But this is the last stage of the peace process, when the social and emotional conditions will allow for conciliation and forgiveness. A prior step would be the formation of what might be called "forums of conciliation" (This idea was suggested to me by Terry O'Connell), after each community has had several community conferences in the control of political and terrorist crimes. In these local meetings in each town, first the individual factions, then a combined meeting, could have discussions of their losses and hopes for the future, building support for conciliation at the local level.

The third and fourth suggestions involve the highest levels of negotiation in the peace process. The third suggestion is to hold the primary negotiations in secret, to remove them from publicity, and from the temptation of each of the two sides to play to their home audiences. This is the way the Oslo accords were reached in the peace negotiations between the Arab states and Israel. Secret negotiations, involving the most skilled and imaginative of the negotiators from the two factions, allows them to reach creative solutions, rather than strike rigid and unchangeable postures.

Finally, the fourth suggestions, relevant to the public negotiations that should continue while the secret ones are going on, and will continue after they reach an agreement. Given the role of unacknowledged shame and insult in sabotaging rationale compromise, it is essential that all contact between the three sides be handled with exquisite courtesy, avoiding any kind of language or action that might occasion insult. Given the fact that all sides have become enormously shame-prone, from the backlog of mutual insults, this is not an easy task. But with human ingenuity being limitless, I am confident that the right personnel and practices can be found that would remove virtually unacknowledged shame and its ramifications from the peace process. Under these conditions, it might be possible for Northern Ireland or any other conflict to reach a rational compromise.



Alverez, Alexander. 1997. Adjusting to Genocide: Techniques of Neutralization and the Holocaust. Social Science History 21: pp. 139-178.

Braithwaite, J. l989. Crime,Shame,and Reintegration. Cambridge: Cambridge University Press.

Cahill, Thomas. 1995. How the Irish Saved Civilization. New York: Doubleday.

Elias, Norbert. 1995. The Germans: Power Struggles and the Development of Habitus in the Nineteenth and Twentieth Centuries. Cambridge: Polity

Goffman, Erving. l967. Interaction Ritual. New York: Anchor.

______________. l971. Relations in Public. New York: Harper.

Gottschalk, L. and G. Gleser. 1969. Gottschalk-Gleser Content Analysis Scales. Berkeley: University of California Press.

Labov, W. and D. Fanshel. 1977. Therapeutic Discourse. New York: Academic Press.

Lewis, H. l971. Shame and Guilt in Neurosis. New York: International Universities Press.

Lynd, H. l958. Shame and the Search for Identity. New York: Harcourt Brace.

Macnamara, Robert. 1995. In Retrospect: The Tragedy and Lessons of Vietnam. New York: Random House.

Miller, Ian. 1993. Humiliation. Ithaca: Cornell University Press.

Retzinger, S. M. l991. Violent Emotions: Shame and Rage in Marital Quarrels. Newbury Park: Sage.

Scheff, T. 1979. Catharsis in Healing, Ritual, and Drama. Berkeley: U. of California Press.

________. l990. Microsociology: Discourse, Emotion and Social Structure. Chicago: U. of Chicago Press.

________.1994. Bloody Revenge: Emotions, Nationalism, War. Boulder: Westview

________.1997. "Alienation, Nationalism, and Inter-ethnic Conflict." (Unpublished ms.)

________.1997a. "Local Peacemaking through "Community Conferences," (Unpublished ms.)

Scheff, T., and Retzinger, S. l991. Emotion and Violence: Shame/Rage Spirals in Interminable Conflicts. Lexington: Lexington Books.

Speer, Albert. 1970. Inside the Third Reich. New York: Macmillan.

Steiner, George. l971. Review of Albert Speer's Inside the Third Reich. New Yorker . April 14.

Sykes, G. and D. Matza. 1957. Techniques of Neutralization: A Theory of Delinquency. American Sociological Review 22: 664-70.

Tavuchis, Nicholas. l991. Mea Culpa: a Sociology of Apology and Reconciliation. Stanford: Stanford U. Press.

Whyte, John. 1990. Interpreting Northern Ireland. Oxford: Clarendon Press.