Aggression, Violence, Evil, and Peace
Joseph de Rivera
Although aggression, violence, and evil are interrelated, contemporary research is so specialized that it is unusual to group them together, and this chapter is unique in considering them together with ideas and research on peace. Why place such disparate fields in the same chapter? In part, it is because we study aggression and violence in order to avoid the evil they occasion and achieve peace. However, if peace were simply the absence of violence it would not require separate treatment and if peace were a completely independent field, it would be better to make use of separate chapters. Typically, there are separate chapters on aggression and pro-social behavior. Here, however, we argue that the achievement of peace rests on an understanding of aggression, violence and evil, yet requires us to go beyond that material to include not only what is usually conceived as pro-social behavior but the use of aggression in a struggle with violence and evil. Hence, we must consider these topics in conjunction with one another. The attainment of peace requires us to have an understanding of aggression, and the pitfalls of violence and evil, as well as the various paths that may lead towards a peaceful worldWe begin with aggression because, although often violent, aspects of aggression may be necessary for the achievement of peace. We consider three different ways in which aggression may be defined and four different theoretical approaches to its understanding. We will then sample the voluminous literature on forms of violence and ways in which it may be controlled. Although many of these forms are clearly related to aggression, Gandhi observed that the worst form of violence is poverty. The “aggression” behind such violence is masked, because much poverty occurs because of unjust economic and political structures. Such violence may appear unambiguously evil, yet when we turn to examine evil we will find that a difficult moral judgment is involved. Further, though most contemporary judgment considers violence to be evil, much of our striving for peace and justice involves us in violence that we fail to acknowledge as evil. This is one of the facts that requires us to base any quest for peace (valued as good) on a thorough understanding of violence and evil. In this way, we will not be naive when we finally come to consider how peace may be attained. This consideration will involve us in the exploration of four paths towards peace: Peace through strength, through negotiation, through the structural changes required by justice, and through personal transformation. Finally, we shall consider how these paths may be integrated by the concept of a culture of peace.
AGGRESSION
Aggression often leads to a violence that is evil, but it may also be involved in the attainment of peace. The term has different meanings that focus our attention on different aspects of behavior and lead to the creation of different approaches to our understanding of aggression. We may distinguish at least three different definitions:
1. Aggression as behavior intended to hurt an other, whether this intent is motivated emotionally (as by anger, pain, frustration, or fear) or instrumentally, a means to an end (as in punishing misbehavior or intimidating an other to attain one’s end.) There are two caveats to this definition. First, the “intention” to hurt may be embedded in larger intentions that have quite different meanings. Although the definition succeeds in avoiding the inclusion of hurting that is accidental or an unavoidable aspect of helping (as in washing a wound that needs to be cleaned) it does include behavior as disparate as a swat on the bottom to correct a child, and the dropping of a nuclear bomb to win a war. Second, the “other” who is hurt may be the self (as in suicide) and may or may not include animals.
2. Aggression as assertive, moving out, behavior that is aimed at getting what one desires (sometimes without regard for the wishes of others.)
3. Aggression as the assertion of dominance and removal of challenges to what one believes ought to exist.
If we work from the first definition we may view aggression as behavior that is clearly to be discouraged and socially controlled. However, the second definition is much more positive, at least in an individualistic culture, and we may want to support and even encourage such behavior, as long as it is balanced by a concern for others. The third definition raises a number of evaluative issues. It is morally “neutral” if one accepts challenges or power as a fact of life, but its very presumption of a power relationship between persons or groups may be viewed as morally repugnant. It is helpful to keep these alternative definitions in mind as we examine four major approaches to aggression: (1) A focus on aggression as learned behavior, (2) A focus on an emotional base for aggression, (3) A view of aggression as biologically based, (4) A view of aggression as embedded in conflict.
Social Learning Theory.
Focusing on aggression as behavior that results in personal injury or property destruction, Bandura (1973) has shown how people may learn such behavior by modeling the aggressive behavior of others (he does not consider the intention of the behavior.) Shown an adult striking a large inflatable “Bobo” doll, children learn the pattern of behavior. The pattern is then encouraged or inhibited by the results that follow. If the behavior is rewarded, the model is liked and chosen for emulation even when children are critical of the aggressive means that are used the amount of aggression increases; if it is punished it decreases. Models also function by suggesting the social acceptability of behavior, thus facilitating patterns of behavior that have already been learned.
When the direct punishment of aggression is itself aggressive, it models aggression at the same time it discourages it. Thus, although small amounts of non-abusive spanking can be beneficial in disciplining children between the ages of 2 and 6 (Lazelere, 1996) it is clear that the frequency of physical punishment is linearly related to the frequency of aggression towards siblings (as well as towards parents) across a wide range of ages (Lazelere, 1986). The modeling of aggression may occur in families, neighborhoods, or on TV, and in each of these cases numerous studies show that children exposed to aggressive models are more apt to engage in aggressive behavior. For example, children growing up in abusive families are apt to assault their own children (Silver, Dublin, & Lourie, 1969); higher rates of aggressive behavior occur in neighborhoods where there is a sub-culture of violence that provides models and rewards aggressive behavior (Wolfgang & Ferracuti, 1967); children exposed to a film depicting police violence show more violence during a subsequent game of floor hockey than children who had just watched an exciting film on bike racing (Josephson, 1987), the general aggressiveness of teenagers (as rated by teachers and classmates) is correlated to the amount of violence that they watched on TV when they were children (Turner, Hesse, & Peterson-Lewis, 1986), and both self-reported aggression and the seriousness of criminal arrests at age 30 is predicted by the violence level of the TV show persons watched at age 8 (as reported by their mothers 22 years earlier) (Eron, Huesmann, Lefkowitz, & Walder, 1972).
From the perspective of social learning theory, aggression is neither instinctive nor produced by frustration. It is a pattern of learned behavior that has been rewarded so that it is efficacious within a given society. Aggressive cultures assume that aggression is innate and natural, without realizing that there are other cultures where aggressive patterns of behavior do not occur, or occur with far less frequency. Of course, there are often emotional conditions that precede aggression. However, numerous studies have shown that loss, frustration, or anger only lead to aggression when an aggressive pattern of behavior has been learned and reinforced. For example, Nelson, Gelfand, and Hartmann (1969) involved children in competitive or noncompetitive play and then had them observe either an aggressive or a non-aggressive model. Those who had lost in competitive play were most prone to behave aggressively, but only when they were exposed to the aggressive model.
While we may certainly speak of aggression as a pattern of learned behavior, we may also conceptualize it as a more general social script, a program of how to act in problematic social situations. Thus, Huesmann (1986) has proposed that children learn aggressive social scripts by observing how others behave in life and on TV. Realistic violence by a perpetrator with whom the child can identify is highly salient and easily leads to fantasy and rehearsal as a way of solving problems. These aggressive scripts may be used in quite different circumstances. Huesmann, focusing on the role of scripts in interpersonal violence, points out if an aggressive script is rewarded a child learns to gain attention and get his or her ways by using such a script. Amongst middle class peers, the use of aggressive scripts is likely to result in unpopularity and poorer academic achievement. The resulting social isolation may lead to increased television viewing, and an even greater reliance on the use of aggressive scripts, eventually producing a person who habitually uses violence.
In the informational processing model proposed by Huesmann (1998), people use a heuristic search process to retrieve a script that is relevant for their situation. The use of an aggressive script will depend on how situational cues are interpreted, the availability of aggressive scripts, the normative evaluation of such a script once it is activated, and the interpretation of consequences. In regards to the last factor, he points out that if a child is beaten for aggression, the child may feel disliked rather than interpret his or her behavior as unprofitable.
Although most focus on the use of aggressive scripts by delinquents, the scripts are as available for use in international conflicts as in bullying and gang wars. McCauley (2000) has pointed out that while the least socialized are more involved in personal violence, it is the best socialized who are often involved in the inter-group violence we shall examine when we consider war. In fact, personal scripts are an aspect of the societal myths that we shall consider when we deal with the concept of evil, and Schellenber (1996) points out that interpersonal violence may be more influenced by the extent to which a society engages in war than the reverse. Thus, Ember and Ember’s (1994) analysis of the relationship between war and interpersonal violence in 186 societies, suggests that socialization for aggression and severity of childhood punishment appear to be more a consequence rather than a cause of war, and it is this socialization that is most directly related to interpersonal violence.
Aggressive scripts are available for use in any social conflict and the aggressive behavior occurs in a context where it may be perceived to be justified. If it is perceived to be justified an observer may be more apt to identify with the aggressor and, hence, more apt to model the behavior. In the context of competition, the goal of winning may entail a willingness to hurt, and this is easily conflated with a willingness to hurt in order to win. Paradoxically, the fact that context affects the meaning that we give a behavior may be a strong argument for Bandura’s behavioral definition of aggression as that which results in (rather than intends) harm. For it may be argued that bomber pilots do not intend to hurt civilians and certainly any intention to harm is embedded in the goal of carrying out a mission and attention is directed towards mundane means (the pilot who carried the Hiroshima bomb was primarily concerned that the added weight might prevent a safe take off).
There are many social forces that inhibit aggression and Bandura (1999) has extended his earlier work by examining how aggression is more likely to occur when a person is morally disengaged from the victim of the aggression. Such disengagement may occur by justifying the aggression (for example, as required by a war against ruthless oppressors to preserve world peace), by using euphemistic labels (as when the number of persons killed in a bombing raid is termed a “population response”), or by using advantageous moral comparisons (as when a torturer contrasts an individual victim’s pain with the extensive random violence inflicted by the terrorist he is attempting to apprehend). It is facilitated by displacing responsibility for the damage that is done (as demonstrated in Milgrim’s, 1974, experiments), by diffusing responsibility (as shown in studies by Zimbardo, 1995), and by increasing the distance between persons and evidence of the pain they are inflicting (Kilham & Mann, 1974). Finally, moral disengagement occurs when dehumanization prevents empathic responsiveness (see Bandura, Underwood, & Fromson, 1975). Personally, I would argue that we witness the contrast between moral engagement and moral disengagement whenever we either empathize with the struggles of an ant or step on the nuisance.
Aggression As Emotionally Based.
While Bandura focuses on aggression as a learned pattern of behavior, Berkowitz (1993) emphasizes aggression as involving meaning and motivation. In Bandura’s work on the modeling of aggressive behavior, the meaning of the behavior, the intent to harm, is implicit and taken for granted. Berkowitz (1993) emphasizes that this meaning may be crucial, and this becomes apparent if we focus on aggressive ideas and emotions. If one man tackles another in a football game, we may see the skillful, determined act of an athlete in a game or we may see a deliberate attempt to injure another person. For Berkowitz, it is only in the latter case that the model may activate the aggressive thoughts and emotional reactions that may lead to aggressive behavior. Berkowitz points out that modeling is more apt to occur when there is an identification with the aggressor, and that aggression is more apt to occur when negative emotional states exist. Berkowitz distinguishes between instrumental aggression that occurs as a means of achieving some planned end, and emotional aggression, grounded in passion and typically spontaneous and unplanned. Thus, the planned killing of a hired assassin is seen as quite different from the impulsive killing of a jealous spouse, the deliberate “taking out” of a star player as quite different from a blow thrown in temper.
Berkowitz tends to focus on emotional aggression which he sees as pushed out, impelled from within, although this impulsion can be influenced by external cues. He sees such aggression as having both a motoric component (to tighten jaw and fists, to strike out, etc) and an urge to hurt, injure, destroy. He makes two major points. First, he argues that the emotional state that underlies emotional aggression is not only anger, but all negative affect. Thus, he shows that the discomfort produced by heat, cold, noise, overcrowding, frustration and free floating annoyance, leads to the increased probability or strength of aggressive behavior. For example, when a student makes a mistake, other students behave more aggressively when they are in a hot room as opposed to a comfortable room (see Baron, 1977), students who could reward or punish other students use more punishments when they are in some pain, riots are more apt to occur hot spells (Baron & Ransberger, 1978), and domestic assaults when air pollution is high (Rotton & Frey, 1985). Also see Berkowitz (1982) and Anderson (1989).
Second, Berkowitz asserts that the emotional state consists of a network of feelings, ideas, memories and expressive motor reactions that are associated with one another so that the activation of one part of the net will activate other parts. Unpleasant memories will promote a negative mood and this will increase the probability of negative thoughts and aggressive behavior. An external cue that has an aggressive meaning (such as a weapon) may activate aggressive thoughts and, particularly if negative feelings are present, lead to an increased probability of aggressive behavior. Thus, in the classic experiment by Berkowitz and LePage (1967), angered subjects delivered more shocks to their partner when guns were present in the room as opposed to badminton rackets. (For further findings see Turner, Simons, Berkowitz, & Frodi, 1977). Even more disturbing, Berkowitz argues that cues associated with pain, frustration and suffering, aversive stimuli in general, may activate negative affect and increase the probability of aggression. Thus, Berkowitz & Frodi (1979) showed that when women university students were angered and distracted, they were more punitive when the child they were supervising was “funny looking” and stuttered. Of course the activation of negative affect does not necessarily lead to aggression. Such behavior may be inhibited by either fear of punishment or empathic concerns, and a person may learn to respond with other behaviors. However, for Berkowitz aggressive behavior is a “natural” response to negative affect that will tend to occur whenever self-control is reduced so that a person is susceptible to the influence of negative moods or external stimulation.
Biological Theories.
From a biological perspective, many species of animals exhibit aggressive behavior and we may consider a good deal of aggressive behavior as instinctually based. To the extent this is so, there are constraints on the ease with which we can modify aggressive behavior. The concept of instinct can be approached from the view of contemporary evolutionary theory or from its original conception, as used by Freud.
Behavioral evolution. Rather than view aggressive behavior as a learned response, we may conceptualize aggression as the product of evolutionary processes, behavior patterns based on genetic influences that have persisted because they have been adaptive, helping members of a species survive in specific environments. Thus, we may find aggressive patterns of behavior programmed into the nervous system because the genes that served as the basis for these programs were selected by the reproduction of the organisms that possessed them. In its original conception, instinctual behavior was viewed as a sort of drive which, like hunger, was directed towards a goal, building if it were not satisfied and becoming less and less particular about the ideal goal object until it could be satisfied by something that would not ordinarily be chosen. However, it is difficult to imagine the goal of an aggressive drive because there are so many different functions for aggression. There is the aggression involved in predation, in the defense of the young, in the struggle between males for mates etc. It seems better to consider instinctual aggression as comprised of certain behavior patterns that are available to be released by particular cues in the environment. We may then consider both internal and external factors that may influence these aggressive patterns, and how the patterns may have adaptive significance. For example, in most species males engage in more intra-species aggression than females (cite reference) and this aggression is involved in the competition to fertilize females. In some species the struggle for mates involves the establishment and defense of territory, and may function to spread members of a species out over territory, thus preserving food supplies.
When we examine the human species we find that societies vary widely in the amount of aggression. However, within any given society, males always appear more aggressive then females. Thus, observational studies find boys more aggressive than girls (Maccoby & Jacklin, 1974); when they are given stories involving conflict, teenage males are more apt to offer more violent solutions (Archer & McDaniel, 1995); and violent crimes are more apt to be committed by men (Wilson & Herrnstein, 1985). However, there may not be significant gender differences in aggressiveness within domestic disputes. Within family aggression seems as prevalent in females as in males (see Straus & Gelles, 1990).
The sex differences in aggression that are found appear related to testosterone, which appears to influence both how the brain develops and some of the physiology underlying current behavior (Rubin, 1987). Prenatal exposure to testosterone appears to increase the amount of rough and tumble play in young females (see Meyer-Bahlburg & Ehrhardt, 1982). Boy’s reports of how aggressively they would respond to provocations are related to testosterone levels (Olweus, 1986). Prisoners with higher testosterone levels violate more prison rules, especially when overt confrontation is involved (Dabbs, Carr, Frady, & Riad, 1995). However, the exact relationship between testosterone and aggression is unclear, in part because behaving aggressively appears to affect the release of testosterone and testosterone levels may be related more to dominance behavior than to aggression as intent to hurt. Mazur & Booth (1998) argue that testosterone both rises in response to a challenge to dominance and increases in winners and decreases in losers.
Alternatively, it can be argued that males gravitate towards violence because their bodies are better adopted for aggressive combat. The exercise of this advantage is a way for men to solve the existential problem posed by the fact they cannot bear children. That is, they can justify their existence by insisting on the necessity of violence and their preeminence in its exercise. (van Creveld, 2000) In humans, as in other animals, it is easy to imagine how aggressive behavior in a conflict between males could result in the reproductive advantage of stronger males and, hence that the genes of aggressive males would be more likely to be reproduced in specific environments. (Daly & Wilson, 1985). However, if we imagine early humans as existing in hunter-gatherer groups, it also seems clear that cooperative behavior within the group would also contribute to survival. Even when the sacrifices involved in such cooperation might result in a disadvantage for a particular individual and his or her genes, genes related to cooperative behavior might be preserved if the cooperation favored kin or others who reciprocated the cooperation, or if the penalty of not cooperation was high, or (in certain conditions) if the group itself benefited (See Wilson & Sober, 1994). It seems probable that the genetics favoring cooperative behavior may offset those favoring male in-group aggressiveness. However, it has been noted that boys evidence more intra-group cooperation than girls when their group is in competition (Shapira & Madsen, 1974). In his studies on the conflict between Hindus and Moslems, Kakar (1996) observes when boys and girls were asked to construct an “exciting” scene with toys and dolls identifiable as Hindu or Muslim, the boys constructed scenes of violence whereas the girls made scenes of family life. Further, Thompson (1999) has pointed out that group selection may also have favored the selection of aggressive individuals who are willing to die for their group in combat against out-groups.
Freudian theory. Freud’s psychoanalytic theory is written from a biological perspective that is based on the older idea of instincts as drives or needs. The development of his thinking about aggression is complex and has been described and critiqued by Fromm (1973). Freud ultimately postulated an instinct towards self-destruction that was neutralized by a life instinct that directed the destructive tendency outward. The former was seen as a drive towards the destruction of life that was ultimately manifested in the death of the body, but could be used in the killing for food, its metabolic break down, the killing of rivals, warfare, and the aggression of the superego against the self as in guilt and depression. By contrast, the life or “erotic” instinct sought to preserve and unite and, alloyed with the death instinct, was involved in self-defense. (See Freud, 1923).
Taken literally, few would agree with Freud’s conceptualization. However, on a metaphoric level his theory allows a rather elegant statement of a viable theoretical position, best expressed in his letter to Einstein on the cause of war (Freud, 1932/1964). Working, in the manner of evolutionary biologists today, Freud assumes that early humans lived in small groups. He postulates that these groups were initially dominated by the compelling aggression of the strongest male. This dominance could eventually be overcome by an aggressive union of weaker males. However, such a union had to be maintained by the growth of law and feelings of unity. As Freud (1932, p. 276) puts it, “Here, I believe we have all the essentials: violence overcome by the transference of power to a larger unity, which is held together by emotional ties between its members” (p. 276). As Freud surveys history he is not encouraged by what he sees. While aggression within groups is contained by the laws of enforced by group union there is no way to contain aggression between groups. Some propose the ideal of a union between groups, but a mere ideal is not sufficient to overcome the aggressive drive of independent groups, and powerful groups are unwilling to grant sufficient power to a superordinate body.
Since they are instinctually based, Freud felt it would be useless to get rid of aggressive inclinations. However, he believed that anything that increased the emotional ties between people would work against war. He distinguished two sorts of emotional ties: those directed towards a shared loved object (as towards a loved authority), and common identifications (as with a nation). At the close of his letter, Freud addresses an interesting question. Given the tremendous force of the aggressive instinct, how can there be pacifists, such as himself and Einstein, who oppose war? While he eloquently gives many reasons against war, he remains true to his biological roots. He argues that he and Einstein are obliged to be against war for organic reasons. Culture itself has been evolving with a displacement of instinctual aims. The intellect is beginning to govern instinctual life, and aggressive impulses are increasingly becoming internalized, so that some humans are developing a constitutional intolerance towards war.
In any case, contemporary psychoanalysts, from both Freudian and Jungian traditions, stress that the mature person and society must acknowledge the presence of aggressive impulses
and manage the ambivalence they see as inevitably present in human relationships (Bach & Goldberg, 1974/1983; Charny, 1982). A related view was expressed by William James (1911/1995) who, impressed by the attractiveness of war (the excitement, comradeship, self sacrifice for a higher cause), suggested that humans needed a moral equivalent (a “war’ against disease, earthquakes, poverty) if they were ever to relinquish the satisfactions offered by war.
In spite of the apparent cogency of the above arguments, it should be noted that the consensus of contemporary social scientists is that there is no instinctual press for war per se.
Thus, the group gathered at Seville to examine the problem declared, “It is scientifically incorrect when people say that war cannot be ended because it is part of human nature” (UNESCO, 1991, p. 10) .The reasoning behind such a statement is well explicated by Fromm (1973). He points out that war, like slavery, is a human institution. There is no evidence that early hunter-gather groups engaged in warfare and there was no reason for them to do so because there were no goods to plunder, no use for slaves, no territory to be defended. Far from being an aspect of “primitive” man, warfare develops along with the development of civilization. Agriculture and animal husbandry lead to surpluses and the development of specializations and hierarchies of power that than become involved in the conquest of other peoples. Fromm points out that there were and still are peaceful peoples, and evidence for a relatively high degree of civilization in a number of matrilocal societies that existed before warfare began. Note, however, that these facts do not preclude a possible instinctive base for the in-group biases that are so prevalent whenever groups must share resources. The crucial role of culture interfaces with a biological base for both cooperation and violence. Thus, although chimpanzees show a great deal of cooperative behavior, Blanchard and Blanchard (2000) point out that chimpanzees also form raiding parties and that this aggressiveness is inhibited when they are afraid.
Conflict Theory.
Rather than focuses on aggression as the response of individuals, we may examine the role it plays in the relations between people, as a way of wining conflicts, establishing dominance, or managing impression. Although this might be viewed as a sociological approach, it may be linked to both emotion and biology. In regards to emotion, de Rivera (1977, 1981) has proposed that anger is best regarded, as a response to a challenge to what a person asserts ought to exist. He argues that when persons become angry they are attempting to remove a challenge in a way that is analogous to an animal defending territory. Learned aggressive responses are recruited to serve this end. Considering aggression as an aspect of conflict is close to the biological perspective in that it considers the function of aggression in a specific environment, although the environment is a society rather than an ecology. In game theory, the choice of a competitive or mistrustful, rather than cooperative, strategy may be regarded as involving aggression in the sense that benefits to the self are sought without regard for the fact that the other is harmed. Research has shown how many situations seduce persons to play aggressively even when it is not in their best interest (See Deutsch, 1958; Scodel, Minas, Ratoosh, & Lipetz, 1959), and Wilson (1998) argues convincingly that the game theory used in the study of conflict may be enriched by utilizing the perspective of evolutionary biology.
Scheff (1999) asserts that emotional sequences play a crucial role in both interpersonal and inter-group conflict, and focuses on the roles of pride and shame as signals of solidarity and alienation. He argues that if shame is acknowledged by referring to one’s insecurity, separateness, powerlessness, then connections of solidarity and trust can be built. However, shame is often unacknowledged. When overt, such unacknowledged shame involves painful feeling with little ideation and is often signaled by furtiveness. When it is bypassed, it involves rapid thoughts that occur with little feeling, and is accompanied by hostility or withdrawal from the other in ways that mask the shame. The unacknowledged shame feeds upon itself, becoming ashamed that one is ashamed, a panic state, or a humiliated fury. A typical pattern is to mask the shame with anger at the other and Scheff sees such shame-anger loops as at the heart of destructive conflicts. He suggests that they account for the need for vengeance (restoring a deflated ego) and are at the heart of deterrence strategy and the danger of appearing weak, the underlying unacknowledged shame being euphemistically treated as “face-saving” and “status competition.”
In their examination of conflicts between individuals, groups, and nations, Rubin, Pruitt, and Kim (1994) describe how conflicts escalate in aggressiveness. This escalation often occurs in five different ways: Influence attempts move from light to heavy tactics, from persuasive attempts to threats and violence; issues proliferate from small to large so that parties become increasingly involved in the conflict and commit more resources to it; issues move from the specific to the general so that the relationship between parties deteriorate, motivation shifts from simply doing well for the self to winning and then to hurting the other, and participants may grow from few to many. Thus, the strength of the aggressive responses (from a harsh word to a physical threat), the generalization of the attack (from one aspect of behavior to a description of character), the extensity of the conflict (from a disagreement over one thing to disagreements over many) all may increase.
Seeing aggression as an aspect of conflict reminds us that at least two parties are involved and aggression increases as the parties respond to one another. The light tactics used by one party may be met by heavier tactics and these may involve psychological justifications that may lead the conflict to generalize and may deepen the other parties involvement. Such conflict spirals often involve anger and blame, fear and perceived threat to one’s image, and may rapidly lead to increasingly aggressive responses. Fortunately, such conflicts often subside, de-escalating as forbearance wins, tempers cool and apologies are made. This is particularly true when the parties to the conflict have common interests and a history of cooperation.
However, conflicts that spiral and escalate may lead to structural changes that make it difficult for the conflict to subside. These involve psychological transformations. When groups are involved, there are changes in group and community dynamics. The psychological changes that occur involve the development of negative attitudes and beliefs about the other, the development of competitive and hostile goals, and the de-individuation and dehumanization discussed above. As both Jervis (1976) and White (1984) have described, an intensely negative image of the other begins to develop, the other becoming regarded as immoral, inhuman, evil. When groups are involved, they become polarized. They become increasingly extreme in hostile attitudes, develop norms that resist compromise, and contentious group goals that contribute to in-group solidarity at the expense of the out-group. They select militant leaders, become more liable to the problems of “groupthink” (Janis, 1983) and the development of militant subgroups. The entire community may become polarized as members are forced to choose sides and neutrality becomes impossible without one’s loyalty becoming questioned.
VIOLENCE AND ITS CONTROL
Although many may be ambivalent about aggression and feel that some situations require an aggressive response, and some may favor the use of violence to gain whatever is desired, probably most humans care about others and desire the security that comes from norms against the use of violence. However, what is considered to be violent often seems to depend on the side that is favored. Thus, Blumenthal, Kahn, Andrews, and Head (1972) found that persons tend to distinguish between the violence of those they dislike and the “justified force” of groups they favor. Further, the violence which we ordinarily consider is the direct violence involved in using force to injure, damage or destroy a person or to unjustly violate a person’s rights. We tend to overlook the indirect violence of structures that create hunger. Below, we consider the many forms of direct violence as it occurs between persons and within and between societies. However, we also consider the violence that is done when human beings are distorted and prevented from developing their potential whether this occurs from the crushing effects of poverty (Gandhi, 1961) or the misuse of psychotherapeutic drugs (Keen, 2000), or the indirect or structural violence described by Galtung (1969).
Personal Violence.
Violence often occurs in an interpersonal context, as in murder or bullying. We will first consider those men who are particularly prone to violent behavior, and then examine the arenas of family violence, rape, and school bullying.
Violent men. In the United States, about 90% of violent crimes between 1960 and 1980 were committed by men (Wilson & Herrnstein, 1985). Although this percentage is changing (it is now 83%) men are more apt to be involved in violence, and some appear to be particularly prone to violence. Thus, Berkowitz (1993) cites a number of studies that present evidence for a degree of consistency across situations and across time. For example, Farrington (1989) reports on a longitudinal study of 400 working class males that lived in a section of London. The study began at about age nine. The youths and their parents were periodically interviewed, teachers rated their behavior and court records were examined. Of the one hundred youths who were most aggressive at age nine, 14% had been convicted of a violent offense by age 21 as compared with 4% of the other boys.
Above, we noted that individuals may learn aggressive scripts. These appear to be much more prevalent in some sub-cultures, are available to groups, and promoted by negative affect. Thus, Dunning, Murphy, and Williams (1988) document how the British football hooligans, who “have an aggro” as they violently attacking strangers, are working class males who grow up in families where they witness, receive, and are coached in violence until many enjoy the skills and thrills involved in its use. It should be noted that while several studies show how particular people are prone to violence over time, they also reveal the inaccuracy of prediction in that many people change to become more or less aggressive as they age. It must also be realized that initial marginal deviance can get amplified by the reactions of others so that greater problems develop (Caprara & Zimbardo, 1996; Haney, 1995).
Violence is frightening and many people seem to want to simply avoid it by locking up those prone to violence and throwing away the key. Thus, it is helpful to break the phenomena into manageable parts. When we consider how we should deal with violent persons, it seems very important to distinguish between different types and processes and to construct specific treatment methods for the types we distinguish.
What are the motivations behind violent crime? Above, we noted Berkowitz’s distinction between emotional and instrumental aggression and when we examine the experience of violent men we find complex configurations of factors. Using peer interviews of 69 men who had records for repeated violence, Toch (1969/1993) distinguished nine types of motivational processes: His most frequent classification (used to describe 28 cases) involved the promoting or defending of a self image which seemed to be constructed as a compensation for the fact that the person was not convinced of his own worth. By contrast, a different 10 cases involved the defending of the reputation of an assigned role experienced as a social obligation rather than an internal need. Other types involving a sort of self preservation or enhancement included: the removal of pressure by men feeling helpless panic or rage, the self defense of those convinced others are out to kill them, and the “norm-enforcer” who uses violence as a matter of principle. Toch contrasted all of the above types with 16 cases that involved a basic egocentrism that evidenced a complete lack of empathy with others, who were simply seen as objects. Amongst these men he distinguished exploiters who used violence when others resist, bullies who took pleasure in using terror, “self indulgers” who simply felt the world ought to let them do what ever they want, and “catharters “ who rather arbitrarily attacked to relieve frustration, depression or boredom. While Toch demonstrates that his classification can be used with some reliability, it is clear that often more than one process is involved, and his main concern is to point out that violence should not be treated as homogeneous. He also points out that once a man has been involved in violence he may develop a habit of using violence. Violence creates its own needs and reinforces the very insecurities and egocentricity that was its source. We might add that the person may define the self in a way that makes violence more probable in the future.
Baumeister and Campbell (1999) distinguish three distinct processes that they believe may be involved in an intrinsic appeal to commit violent acts (as opposed to instrumental motives for either selfish or idealistic ends.) They differentiate sadism (the pleasure of inflicting suffering or terror involved in Toch’s bullying category) from violent thrill seeking. They argue that the former is an addictive process, while the latter is more the sporadic sensation seeking of the bored (and often drunk and impulsive). They estimate that sadism may develop in about 5% of persons who are repeatedly violent and may possibly be explained by Solomon’s (1980) opponent process theory, while violent thrill seeking is motivated by boredom combined with high sensation seeking and low impulse control (and more likely to occur under the influence of alcohol). Both processes are, of course, opposed by any guilt feelings an individual may have. The third process involves threatened egoism and, at first glance, appears related to Toch’s category of self-image promoting and defending. However, Toch views his category as involving persons who are not convinced of their own worth, while Baumeister and Campbell feel that the process they are describing is more a reaction when a narcissistic view of the self is threatened. This disparity may involve some important distinctions, combined with some semantic and measurement issues. Baumeister, Smart, and Boden’s (1996) literature review seemed to show that aggressors tend to have favorable, even grandiose, views of the self, and Bushman & Baumeister’s (1998) laboratory study showed the highest aggression coming from the injured self esteem of insulted narcissists. However, the review article often involved the aggression of psychopaths and rapists (whom Toch would classify as egocentric) and narcissism also implies egocentricity. We may need to distinguish the inflated (and narcissistic) self esteem of the bully from the poor self esteem of other violent persons.
The above analyses are useful in understanding persons whose violence is not condoned by the society in which they live, but how may we understand the tolerated violence of leaders such as Hitler, Stalin, or Poi Pot, lieutenants such as Himmler or Goering, or the masses involved in publicly sponsored killings such as those that occurred in the Roman Coliseum or any number of blatant purges and massacres? While we may assume a certain amount of egocentricity, objectification and lack of empathy, there appear to be deeper motivational reasons, the sort of instinctual tendency for destruction postulated by Freud. An alternative is suggested by Fromm’s (1973) analysis of destructive character structure. Fromm distinguishes between a biologically adaptive aggression that humans share with other animals and a malignant aggression that is distinctly human and maladaptively destructive, a point discussed more fully below when we consider the nature of evil. He argues that the latter occurs in two forms: Spontaneous outbursts of destructive impulses occurring under certain conditions that promote vengeance, ecstatic destructive rituals, or the worship of destruction, and the destructive traits that are bound in certain character structures. Fromm points out that different societies are organized in ways that promote certain syndromes of traits or character structures. Feudal societies requires different personalities than capitalistic ones, cooperative societies promote different characters than competitive ones. He argues certain situations tend to produce character structures that are fundamentally destructive in that they are either sadistic or necrophilous.
Neither sadism or necrophilia are necessarily sexual and Fromm is careful to distinguish sadism from non-destructive forms of sexual pleasure. By sadism he means the passion to have absolute control over another living being; he sees forcing another to endure pain or humiliation as the manifestation of such control; and he uses case studies of Stalin and Himmler as his examples. By necrophilia he means a passion to transform that which is alive into something dead, to destroy for the sake of destruction; he presents a case study of Hitler as an example. Fromm argues that necrophilia is connected to the worship of technique and the fusion of technique and destructiveness seen in modern warfare and may be objectively measured and related to political attitudes that favor military power and the repression of dissent (See Maccoby, 1972).
Family violence. The violence occurring within families is often considered as two separate domains: child abuse, and “domestic violence” occurring between partners. Although there are obvious differences, it seems important to relate these two types of violence to each other, and to the general disruption of family management and the rather neglected topic of sibling abuse. For example, Patterson (1982) shows that the families of antisocial and abused children fail to provide consistent and effective discipline when children are aggressive, fail to monitor their whereabouts, and do not provide positive reinforcement for pro-social skills. He suggests that in such families, a problem child learns to be aggressive by attacking and dominating their siblings, and a sample of such children shows that they attack siblings at almost ten times the normal rate as well as being involved in hitting their fathers and mothers, and being hit by them, (Patterson, 1986).
One often thinks of child abuse as involving physical or sexual abuse. However, by far the most prevalent form of abuse involves neglect. Thus, Sedlak (1990) reports an incidence rate of 14.6 per 1,000 as contrasted with rates of 4.9 for physical and 2.1 for sexual abuse. Parental neglect usually occurs in situations of low family income and education, and often where there is a high level of stress and a lack of social support (Garbarino, 1991). In such situations there is probably less parental maturity, less knowledge about child development, and a greater degree of attachment disturbances. Further, mothers (and one presumes fathers also) may be quite depressed and this may well contribute to the neglect of children (See Pianta, Egeland, & Erickson, 1989). Poverty and the lack of social support also appears to be a factor in both physical and sexual abuse.
In addition, Azar (1991) points out that in order to fully understand how abuse occurs we must look at the interpersonal dynamics that occur within the social context. Her investigations of abusive mothers reveal that they often misperceive a child’s behavior. If a three year old spills a glass of milk the mother may perceive willful disobedience and may lack the skills to cope with what she perceives as a challenge to her authority. Learned patterns of aggression occurring in a situation perceived as a struggle for dominance may account for a good deal of physical abuse. In any case, in their review of theories of child abuse, Azar, Povilaitis, Lauretti, and Pouquette (1998) argue that the best way to understand child abuse is to focus on how a parent interacts with a child in a social situation that is influenced by both societal and cultural factors. The interaction will be influenced by a child who may be more or less pleasing and difficult, by a mother who may have more or less parental and social skills, impulse control, ability to manage stress, by a context that may or may not provide helpful or aggressive models, be stressful or lend social support. To some extent this overall model for understanding may also be applicable to sexual abuse. However, sexual abuse seems less dependent on societal stress and more on personality factors such as high familial dependence, psychopathy, or pedophilic tendencies (Rist, 1979) or, more generally, as having its origin within perpetrators rather than in the interaction between perpetrator and victim (Haugaard, 1988).
Unfortunately, particularly when there is an absence of social support outside the family, there are times when the only way to prevent abuse is to remove a child and place it in foster care. While this sometimes helps, studies suggest that foster parents need more training and support and should care for fewer children (Chamberlain, Moreland, & Reid, 1992).
Domestic abuse in the sense of partner abuse is often attributed to male “batterers”, and Bachman (1994) reports that Bureau of Justice statistics reveal that females experience over 10 times as many incidents of violence by intimates as males. However, it is unclear how much this data is biased by the failure of males to report being hit by their female partners. Cascardi, Langhinrichsen, and Vivian (1992) report more frequent and severe female injuries and more subsequent depression, and the U.S. Department of Justice (1995) reports about twice as many wives and girl friends killed by husbands and boyfriends as the converse (1500:700). However, telephone survey of 6,000 married or cohabiting couples (Straus & Gelles, 1990) found that as many females as males appeared to be involved in partner violence. and in as many cases of extreme violence (the superior physical strength of the males offset by the more frequent use of weapons by females). In a similar vein, surveys of lesbian couples have found as much or more violence as in heterosexual couples (Walderner-Haugrud, Gratch, & Magruder, 1997).
Regardless of the degree to which violence in perpetrated by males, it seems important to distinguish between different sorts of perpetrators. Holtzworth-Munroe (2000) distinguishes between three groups of men involved in marital violence: Those who only become violent within the family as a result of an inability to manage conflict escalation; those who have difficulty with trust issues and become overly dependent on their wives, resorting to violence when their needs are not met; and those who are anti-social and violent in all relationships. Clearly, the management of domestic abuse in each of these cases requires quite different strategies. In some cases couples therapy seems appropriate while in others it would simply prolong abuse. In some cases separation offers a solution. However, Hart (1992) reports that about 75% of emergency room visits and calls to law enforcement, and 50% of the homicides, occur after separation. Intervention programs attempting to teach men anger management and conflict resolution skills in small groups, typically report a 53% to 83% success rate (Edleson, 1996). Although the lower percentages occur when there is a longer follow up time and when success is based on the reports of the victims or arrest rates, these rates are encouraging. However, the rates are based on men who complete the programs (which last from 10 to 36 sessions) and appear to ignore differences in different types of abusers. In one evaluation, of about 500 men who contacted the program, only 283 attended the first session and only 153 completed the sessions. Programs are also available for the treatment of aggressive women (See Leisring, Dowd, & Rosenbaum, in press), though these may face the same attendance problems as the men’s programs.
Rape. Estimation of the prevalence of rape depends a great deal on how rape is defined and how the statistics are collected (Muehlenhard, Powch, Phelps, & Giusti, 1992). For example, the national incidence of reported rape and attempted rape is about half that obtained by federal telephone surveys and about 1/20 of that obtained in anonymous surveys of college students using behaviorally specific scenarios (Koss, 1992). Since 1995, instances of reported rape (as defined by the FBI) have been gradually falling, although less than the rates of other violent crime. (In 1998 the incident rate was 34.4 per 100,000 as opposed to 6.3 for murder and 360.6 for battery). However, there are a number of different types of courtship rape (Shotland, 1992) that are largely unreported, and as sexual mores have changed and women more often find themselves in intimate settings (and often under the influence of alcohol), there may be an increasing amount of date rape in which a woman is subjected to unwanted sex by a male who may or may not be aware that his advances are truly unwanted. Although college age students are at an age with maximal risk it seems clear that a high percentage of women are sexually victimized. Thus, in a well designed study involving over 3,000 college women, 15% reported that they had experienced unwanted sexual penetration because a man had used physical force or given them alcohol or drugs, and an additional 12% reported having had to resist physical force (Koss, Gigycz, & Wisniewski, 1987). Probably few of these instances were reported and many may not even have been termed “rape” or recognized as such.
On one hand, the majority of men do not seem to find rape appealing. Thus, when Malamuth (1981) asked an anonymous sample of college males if they might be inclined to rape a woman if they knew they could get away with it, 65% responded with a “not at all likely” (a 1 on a 5 point scale.) And about 75% of the 3,000 college males sampled by Koss and her colleagues reported having never engaged in instances of sexual aggression (including verbal coercion). On the other hand, 20 % of Malamuth’s sample responded with a 3 or higher, and about 9% of the Koss sample indicated that they had used physical force, alcohol or authority to obtain or attempt to obtain sex. Further, the data suggests that an additional percentage of males seem unaware that their advances were truly unwanted.
Malamuth’s initial data established that students who indicated that they might be inclined to rape if they knew they could get away with it were more apt to believe the rape myth that women really enjoyed being raped, were more apt to evidence penile tumescence when they witnessed a rape scene, and were more apt to condone violence against women. After laboratory studies (Malamuth, 1983, 1984) established they were also more inclined to deliver painful shocks to women, Malamuth (1986) correlated his measures with the type of self report measures used by Koss. His data establish that a number of measures are higher in males who report they have engaged in sexual aggression. These include a measure of reported arousal to rape, penile tumescence while reading a rape story, sexual dominance, hostility towards women, acceptance of force in sexual relations, and prior sexual experience. While any single variable is only modestly correlated with reported aggression ( the highest is a +.43 with tumescence), a combination of the variables yielded a multiple of +.67. Further, there where a number of important interactions between the variables and when these were included the multiple correlation was significantly greater, rising to +.86. It is interesting to note that 6% of the subjects, those who scored above the median on 5 variables, reported far more sexual aggression than the remaining subjects and suggest a similarity with Groth’s (1979) interviews with convicted rapists.
What sorts of background factors contribute to the making of a rapist? Malamuth (1998) presents data that shows coercive sex is most apt to be perpetrated by males who have both an orientation towards impersonal sex (often related to a violent childhood) and a hostile masculinity (involving feelings of rejection and a desire to dominate women). To some extent these risk factors may be mitigated by the ability to empathize.
If the problem of sexual aggression was only a problem of restraining 6% of males it might be fairly easily addressed. Unfortunately, it seems clear that larger percentages are influenced by sub-cultural norms in some gangs, military units, sports teams, and fraternities which suggest that what seems like rape to some is merely normal masculine action. These norms encourage the objectification of women as sexual objects, and the reinforcement of rape myths (O’Toole, 1997). For example, Buchwald, Fletcher, and Roth (1993), Lenihan and Rawlins (1994), and Boswell and Spade (1996) have reported that fraternity men are more likely to be involved in sexual violence and that some fraternities constitute sub-cultures with norms that objectify women, and an environment where drinking parties, easy access to bedrooms, and fraternal secrecy almost assure sexual aggression. Fortunately, it appears that it may be possible to create intervention programs that decrease the acceptance of the rape myths that prevail in such subcultures (Flores & Hartlaub, 1998). However, although we have been examining rape as a form of personal violence, rape may also be used impersonally as an instrument of war (Copelon, 1995). Enloe (2000) has pointed out the different ways in which such militarized rape may be used to achieve political objectives, and how it may become institutionalized. The brutality of such rape is particularly devastating to victims because they are often subsequently rejected by their own communities (Turshen, 2000).
Bullying and malicious gossip. A defining aspect of bullying is that the behavior occurs repeatedly so that there is a pattern of abuse and intimidation (Boulton & Underwood, 1992). So defined, Bernstein and Watson (1997) report that from 7% to 10 % of U.S. school children are victimized by about another 7% of the children who are active bullies. Similar percentages are found in Great Britain (Atlas & Pepler, 1998), with prevalence highest in the middle school years. The average bully appears to have normal self-esteem (See Bernstein & Watson, 1997), and the average victim less self-perceived social competence as compared with other members of their peer group (Egan & Perry, 1998). There are a number of detrimental aspects to bullying. Those who are bullied feel unsafe at school and appear to be at risk for illness, failure, and depression (Wylie, 2000). The bullies begin to think they can get what they want by using power to dominate others, something that often fails to work in adult life (Oliver, Hoover, & Hazler, 1994). And typically the majority of children who witness the bullying feel fear and fail to intervene. They may be being trained to be ineffective bystanders. Fortunately, it is possible to train teachers, students and parents in ways to intervene and such training, involving the entire community, has proven effective in Scandinavia, where programs have reduced bullying by 50% (Olweus, 1991, 1993).
Community Violence.
Although it sometimes takes personal forms, there are types of violence whose nature is more communal than interpersonal. Among such forms are riots, gang extortion and warfare, and police violence.
Riots. There are three major problems involved in the study of riots. First, while murders and suicides are carefully recorded in the medical statistics of many nations, it is no ones job to record riots and the deaths that often result from them. Thus, Richardson (1960a) noted the difficulty of obtaining statistics on the number of deaths in riots as compared to either murders or wars. The investigator must often rely on private collections painfully put together by culling news reports. Such collections are illustrated by the Lemberg Center for the Study of Violence (1968) collection of 484 racial disturbances in the United States between January, 1967-April 1968, Graham & Gurr’s (1969) more general collection covering major violence in the United States, and X’s (19 ) collection of civil disturbances in different nations between 19 and 19 . A second problem is definitional. The United States Riot Commission (1968) notes that the definition of civil disorder varies so widely that between 51 and 212 disorders was recorded by various agencies in the first 9 months of 1997. Using Richardson’s approach of counting deaths we might note that 12 of these involved deaths (2 with over 10 deaths). Gilje (1996, p.4) suggests that a riot involves a, “group of twelve or more people attempting to assert their will immediately through the use of force outside the normal bounds of law.” However such civil disorders might either be spontaneous or planned, and for many purposes it seems important to distinguish those that occur as protests from those that take place in the midst of civil war or are sponsored by a government (as in genocidal operations). Finally, a problem is posed by the many different reasons for riots. There may be little in common between the 5,000 lynchings of blacks in the United States between 1882 and 1939 (Wertham, 1966), the dozens of racial “communal” riots during the early 20th century involving struggles over contested areas or employment (Janowitz, 1969), the spontaneous “commodity” riots of dissatisfied African Americans in the 60's (Couch, 1968), Stark’s (1972) “police” riot, and the “celebration” rampages of victorious sports fans (Dunning, Murphy, & Williams, 1986).
We may distinguish at least three theoretical approaches to riots. As in his understanding of aggression in general, Berkowitz links riots to frustration and general negative affect. And the occurrence of many riots do appear to be related to factors such as high temperature (Anderson & Anderson, 1998; Baron & Ransberger, 1978), a depressed economy, and the presence of aggression eliciting cues. However, such factors seem more contributory than causal. On a psychological level, what may be more fundamental is the “relative deprivation” theory stressed by Gurr (1970). By relative deprivation Gurr means the discrepancy between what one has and what one thinks one ought to have. It seems important to note (although Gurr is not completely clear about this) that “ought” implies not simply the expectancy that one will reach a goal but a judgment about what is just, what ought to be in a moral sense. In part a social comparison process, in the sense that one ought to have what others who are regarded as similar have, probably determines this. But “regarded as similar” depends on societal norms and many a group has suffered passively for years because they have accepted a definition of inferiority imposed by a more dominant group (See Moore, 1978).
What one ought to have also depends on what regards as one’s group and this identity is often shaped by leaders who mobilize group sentiment, a factor stressed by the third theoretical approach, the “collective action theory” advanced by Tilly (1978). Tilly stresses group interests, organized by a network of social connections and mobilized by leaders who emphasize group identity. Such leadership may depend on certain structural factors in the society. Smelser (1962) emphasizes the structural strain that occurs when there is a lack of communication, deprivation and normative differences. He argues that these are conducive to riots when leaders do nothing to relieve the strain, there are insufficient channels to express grievances, and there is sufficient communication among the aggrieved to develop a generalized hostile belief system. The hostile beliefs are released by a precipitating incident that mobilizes the group for action. Building on this idea, Palmer (1972, p.155) suggests that the basic conditions for a riot may be characterized by a high degree of tension in both the social system and the role systems of participants. This weakens identity until it is restored by participation in the riot.
When we examine the empirical data on riots we find some features that are accounted for by each of the above approaches. Lieberson and Silverman (1965) compared cities that did and did not experience racial riots between 1913 and 1963. They found three structural differences. Cities with fewer riots had more racially integrated police forces, more representative forms of local government (because of district rather than city wide elections of city council and school board members), and larger percentages of self-employed blacks in retail stores, restaurants, and taverns.
Numerous riots
occurred in the black ghettos of U.S. cities in 1967 and the United States Riot
Commission (1968) examined 164 of these
disturbances occurring in 128 cities. To account for these disturbances, the
Commission stressed structural factors such as discriminatory police practices,
unemployment and inadequate housing. However, while the Commission reported an
increasingly violent social climate that was encouraged by militant groups,
they found no evidence of “conspiracy” (an absence of mobilizing leadership).
This is not to deny that there may have been the sort of underlying political
motivation stressed by Upton (1989), a desire to register an
impact in the absence of legitimate channels for protest. However, there was no
organized militant leadership. There is some evidence for the relevance of
relative deprivation rather than simple frustration in that a 1967 study by
Fogelson and Hill (reported in Skolnick, 1969) found that three quarters of
those arrested were employed and between half and three quarters in
semi-skilled or skilled occupations. However, Spilerman (1976) was unable to find
relationships between either the frequency or severity of riots and any of a number of measures of
structural strain. In part this may be due to his sample of riots (the bulk of
which were the U.S. commodity riots of 1967-68), but one would still expect his
measures of social disorganization, relative deprivation and political
representation to have some predictive power.
The failure of structural factors in predicting which cities would have the most commodity riots, together with difficulties in satisfactorily predicting which individuals would participate in them (McPhail, 1971), has led McPhail (1994) to abandon both structural strain and relative deprivation as predictors. Instead, he advocates extending Snyder’s (1979) work of examining the factors that affect the interpersonal processes that assemble a riot. Such an approach inquires into communication patterns and the motives of individuals as they assemble to engage in collective goals. McPhail argues that it is important to distinguish between collective goals that do not intend violence (although violence may result) and collective goals that do intend violence (as is the case with England’s football hooligans). It would also seem wise to consider the group emotions that can occur when group members share attitudes and have a common sentience (See Smith & Crandell, 1984). Such group emotion may play a role in the generation of crowd violence in cases where police suppositions of violence may create a self-fulfilling prophecy of violence in sports fans (Stott & Reicher, 1998).
A nuanced account of the group emotions involved in riots is provided by Kakar’s (2000) description of the protracted communal clashes between Hindus and Moslems in Hyderabad, India. After a skillful delineation of economic, political, historical, demographic, social psychological and psychoanalytic accounts, he focuses on the psychological shifts that occur at the outbreak of violence. He notes that the character of rumors begins to change, from general threats to rumors that the body is threatened by previously benign substances, how the boundaries of individuals with peaceful religious identities and a basic sense of trust, become replaced by a transcendent communal identity with a propensity for anxiety and violence, how individual behavior becomes governed by a different sort of morality, and how a history of coexistence is replaced by a history of violence. Of course these shifts are perpetrated by demagogues and much of the violence is perpetrated by gangs of young toughs, but Kaker’s point is that the entire community is caught up in altered identities, and that certain norms are still in existence that enable people to return to their traditional religious identities and live in relative harmony after the violence subsides.
Gangs and gang warfare. Groups of youth can gather together nonviolently and prior to the late 1970's such groups in the US had little use of drugs and relatively little criminal involvement (Moore, 1990 ). However, as drugs and guns became available gangs became increasingly violent and involved in crime and struggles over territory. In 1996 there were about 31,000 gangs with approximately 846,000 members in the United States (Office of Juvenile Justice and Delinquency Prevention, 1998). Gangs are a problem in many other countries as well, particularly when social control disintegrates. In Guatemala, for example, the ending of military dictatorship and police control has produced a situation where officials have reported over 500 criminal gangs with about 80 thousand members (“Guatemala has 500 gangs,” 2001). It seems likely that gangs develop whenever societies fragment and lower class males lack access to legitimate sources of power and prestige. It might be interesting to study gangs as quasi-states. Certainly delinquent gangs involve symbols of identification for group membership, territorial claims, leadership power struggles, in-group protection, and out-group antagonism (Capozzoli & McVey, 2000).
When the Soviet Union collapsed, hundreds of violent groups emerged whom Volkov (2000, p 709) notes, “…intimidated, protected, gathered information, settled disputes, gave guaranties, enforced contracts and taxed.” He argues that these entrepreneurs of violence created organizations that were essentially, “violence-managing agencies.” The more successful gradually became legitimized by becoming involved in pro-social activities and absorbed in the process of state formation.
Hill, Howell, Hawkins, and Battin-Pearson (1999) notes four risk factors that predict adolescent involvement in gangs. On the community level these include poverty, high rates of mobility and dysfunctional norms. In fact, in his examination of drug use and delinquency in 1400 different neighborhoods of New York City, Chein was able to predict the amount of delinquency in a neighborhood with a multiple correlation of +.86 by using indices of socioeconomic squalor, youth density, industrial proximity, and number of broken families. Chein argued, and to some extent was able to prove, that drug use and delinquency was due to the inability of a community to enforce functional norms, that high crime neighborhoods had a state of anomie in Durkheim’s classic sense of the term (See de Rivera, 1986). This also seems true of Hill’s second risk factor, the structure of schools that fail to engage and monitor students. It should be noted that Hill’s other factors involve family characteristics, individual personality, and the choice of friends; and there are somewhat different risk factors in rural areas (Evans, Fitzgerald, Weigel, Chvilicek, 1999).
A crucial problem created by gang warfare (and by civil war in general) is the impact which violence has on children. Kostelny and Garbarino (2001) have pointed out that in some Chicago neighborhoods, 38% of elementary school children have seen a dead body outside and 21% have had someone threaten to shoot them. They have noted how repeated violence often leads to regression, a loss of trust, sense of futurelessness, and increased aggressive behavior. They propose a series of measures to counteract these affects including home visiting and early education programs, and specific violence prevention programs at both the elementary and middle school levels.
One might think that children who have suffered the sort of violence that occurs when civil society has disintegrated, would themselves b