Aggression
Aggression
i. Psychological AspectsLeonard Berkowitz
ii. International AspectsFrederick L. Schuman
I PSYCHOLOGICAL ASPECTS
The study of aggression—here regarded as any behavior whose goal is the injury of some person or thing (cf. Dollard et al. 1939)—has long been governed by philosophical preconceptions and clouded by hopes and fears. Writers have accounted for aggressive behavior in strikingly different ways: as the manifestation of an innate destructive drive, as an inborn reaction to frustrations, or as a learned way of responding to particular situations. Many of these interpretations have clearly been influenced more by metatheoretical beliefs regarding the nature of man or religiophilosophical hopes as to what human beings should be like than by carefully controlled and precise observations. Freud, as an example, at first had maintained that aggression was a “primordial reaction” to the blocking of pleasure-seeking or pain-avoiding strivings. However, partly because of his pessimism growing out of World War i, he revised his formulation and postulated the existence of an instinctive force of destruction and death (1920). Other psychologists have advanced entirely different conceptions. Some were predisposed to deny or minimize the role of innate factors in human behavior, while for others the notion of inherent aggressiveness was incompatible with their view of man as being basically good. Whatever the exact nature of their analysis, all too often their general theoretical or philosophical assumptions resulted in a relatively extreme stand in which some factors were given very heavy emphasis and others were played down or denied altogether.
Instinct conceptions
The Freudian “death instinct.”
In Freud’s post-1920 discussion of aggressive behavior, the dominant tendency in all organic life was held to be the effort to reduce nervous excitation to the lowest possible level (1920). Just as all pleasure-seeking was supposedly oriented toward tension reduction, all organic life presumably sought death, for to die was to be free from stimulation. (Freud also proposed that death was often sought violently rather than quietly and peacefully.) But this initial striving for active self-annihilation did not find fulfillment, Freud maintained, because the death instinct, Thanatos, was opposed by the life instinct, Eros, which diverted the destructive drive from the self to others. Thus, in attacking other people the person found a release for pressures that otherwise would impel him to seek his own death.
Empirical evidence provides little support for Freud’s analysis of aggressive behavior (cf. Berkowitz 1962). To cite just one difficulty, research clearly indicates that organisms do not seek the complete elimination of excitation. There are many situations in which human beings, as well as lower animals, work for an increase in stimulation (White 1959). Death is not necessarily the inherent aim of all organic life.
Other instinct doctrines
A number of biologists as well as orthodox psychoanalysts have accounted for aggression solely in terms of some inner force. The ethologist Konrad Lorenz (1952) has suggested, for example, that excitation is built up in each instinctive center within the central nervous system and is then dissipated when the instinctive act is performed. If an animal did not engage in aggressive behavior, “action-specific energy” would supposedly accumulate within the instinctive center controlling aggression. When enough energy is built up the action pattern presumably would go off by itself, that is, there would be “vacuum activity.” In contrast to this notion of spontaneous aggression, others (for example, Scott 1958) have maintained that there is no evidence of a spontaneous stimulation for fighting arising within the body. Actual fighting, which usually involves males belonging to the same species, is relatively rare in nature; in most cases the opponents display threat ceremonies instead of coming to blows.
Exteroceptive cues. In contrast to the model based on notions of energy accumulation, many students prefer to analyze instincts as species-specific behavior patterns governed by exteroceptive stimuli, which activate and terminate the actions. External stimuli are thus considered to be important contributors to aggressive behavior. Fighting behavior in many species varies with the nature of the antagonist—whether the animal’s opponent is from its own or some other species, and if the latter, whether it is a predator or some prey (EiblEibesfeldt 1963). More than determining merely the form of the response, external cues often appear to be necessary for eliciting any kind of hostile actions. Tinbergen (1951) observed that the male stickleback fish attacked a dummy with a red spot on its belly but ignored a detailed replica of the stickleback that did not have this characteristic of the breeding male or that had the red spot on its back.
Fighting arising from competition for dominance, food, sexual partners, or territory clearly attests to the role of external stimulation in animal aggression. The aggressive activity in these cases is the product of some perceived obstacle to the attainment of a desirable goal state. Even apparently noncompetitive fighting may be explained in these terms; the combat may have been instigated in order to achieve such things as dominance or undisturbed possession of living space (cf. Berkowitz 1962, p. 17).
Internal conditions. The efficacy of external stimuli in evoking aggressive behavior, however, is probably contingent upon the presence of some suitable internal condition. In the case of the stickleback fish the necessary internal prerequisite seems to be the production of the hormones involved in reproduction. Given the required internal state, a particular stimulus evokes the aggressive response. As yet another illustration of this principle, von Holst and von Saint Paul (1962) have shown that electrical stimulation of a certain region of the fowl brain results in recognizable aggressive actions only in the presence of relevant cues, “an enemy real or artificial.” Applying this formulation to competitive fighting, we can say that the rivalry produces an emotional state creating a readiness to engage in aggressive activity and that the competitor then provides the cue that releases (or evokes) the aggression.
The frustration–aggression hypothesis
The principle just advanced is a version of the “frustration-aggression hypothesis,” which has long been used to account for aggressive behavior. Independently espoused by such writers as Freud (prior to World War I) and McDougall (1908), the hypothesis was spelled out most clearly by Dollard, Doob, Miller, Mowrer, and Sears in a now-classic monograph published in 1939. Briefly, these psychologists maintained that a frustration—“an interference with the occurrence of an instigated goal-response at its proper time in the behavior sequence”—arouses an instigation to aggression (Dollard et al. 1939, p. 7). Since this formulation has been criticized frequently, some additional points should be made clarifying and defending it. First, Dollard and his colleagues did not claim that frustration had no consequences other than aggression. A thwarting will produce instigations to many different kinds of responses. Some of these other response tendencies may be stronger than the instigation to aggression, and the aggression is not revealed openly (Miller 1941). Moreover, although the hypothesis did say that all aggressive actions presuppose the existence of frustration, a person does not have to be frustrated in order to engage in aggressive actions (Bandura & Walters 1963). A contemporary revision of the frustration-aggression hypothesis must be less sweeping and all-explanatory than the original version.
This is not to say, however, that the hypothesis must be discarded altogether, as a number of writers have insisted (e.g., Buss 1961). While we cannot deal here with all of the criticisms (cf. Berkowitz 1962 for a more complete discussion), several of the arguments against the hypothesis can be answered.
Do only some frustrations produce aggression?
To begin, several psychologists have contended that only certain kinds of frustrations give rise to aggressive responses. Threats or attacks upon the self produce aggressive tendencies, they say, but mere deprivations supposedly lead to other consequences. In a similar vein, other critics have suggested that arbitrary or unexpected thwartings lead to aggression, while less arbitrary or expected frustrations presumably do not.
Two comments can be offered in rebuttal. Dollard et al. (1939) proposed that the strength of the instigation to aggression resulting from a frustration is in direct proportion to the strength of the thwarted drive. Since the desire for self-enhancement is typically fairly strong in our society, we would expect attacks upon the self to lead to stronger aggressive reactions than, say, interference with the performance of some unimportant task. Attacks upon the self then may lead to overt aggression, while the hostile responses produced by the interruption of a task that is not relevant to the self may be too weak to be apparent. But the frustration need not even be a direct attack on the self in order to produce aggression. Buss (1963) has demonstrated that college students who were prevented by a peer from attaining a desirable goal (such as a money prize) tended to display more intense open aggression toward the peer than did a nonfrustrated control group. The thwarting was not an arbitrary one, and the allowable aggression was not instrumental to the attainment of other ends, but there was a definite aggressive reaction, if only a weak one.
The second comment deals with the matter of deprivations and arbitrary frustrations. According to the definition employed by Dollard and his colleagues, a frustration is the blocking of some on-going, goal-directed activity. A person thoroughly engrossed in his work is not frustrated just because he has been without food for a number of hours. He may be deprived of food, but there are no ongoing eating response sequences, either in his thoughts or his overt activity, that are prevented from reaching completion. His failure to eat at his regular mealtime will therefore not produce an aggressive reaction. But what if this person had been prepared to eat at a certain time and had been thinking of the food he was soon going to enjoy? Suppose his employer unexpectedly gives him a sudden job that keeps him working late at night and causes him to miss his meal. We would now expect him to become angry. Whereas some psychologists would say he has now experienced an arbitrary or unexpected frustration, in contrast to an expected frustration, the present writer maintains that only now is he frustrated, whereas formerly, before he had anticipated eating, he was not. Only now is an ongoing response sequence prevented from reaching completion at its anticipated time (cf. Berkowitz 1962, pp. 36–42).
The innate nature of the aggressive reaction
American psychologists are characteristically reluctant to refer to instinctive or innate mechanisms in accounting for human behavior. Some of the objections to the frustration–aggression hypothesis seem to be dictated by this prejudice against the notion of innate reaction patterns.
Animals and humans can be trained to respond nonaggressively to situations that ordinarily produce hostile responses. For that matter, they can also learn to act aggressively in situations where formerly they had displayed little violence. In an experiment with school children, for example, Davitz (1952) rewarded one group of youngsters for acting aggressively and competitively, while another group was rewarded for cooperative and constructive behavior. After several training sessions all of the children were frustrated when a movie they were seeing stopped and, at the same time, their candy was taken away. Observations showed that the aggressively trained group exhibited more aggression in a free-play period immediately afterwards and that the constructively trained youngsters reacted more constructively to the thwarting. Scott (1958), after reviewing several of his animal experiments that had obtained essentially similar results, concluded that aggression was the product of previous learning. Taking much the same position, others (Bandura & Walters 1963) have argued that frustration produces a heightened motivational state that enhances the strength of whatever responses the individual has learned to make in the given situation; these may or may not be aggressive in nature.
Yet the experiments just mentioned do not really invalidate the frustration-aggression hypothesis. They demonstrate that previous experience can enhance or reduce the likelihood of aggressive behavior, but they do not prove that aggression will not occur under suitable conditions in the absence of any aggression training. Indeed, several experiments indicate that animals reared in isolation, and who had not previously learned to be aggressive, can react aggressively to arousing stimuli (Eibl-Eibesfeldt 1963, p. 11). In another demonstration that prior learning is not necessary for aggressive reactions, Seay and Harlow at the University of Wisconsin Primate Laboratory thwarted some infant monkeys by separating them from their mothers (1963). Six of the eight young animals subsequently displayed some aggression against a peer-playmate—but primarily when the previously frustrated infants were in their mothers’ presence. Two aspects of this finding are noteworthy. First, and most important, monkeys of the species used (rhesus) rarely show any aggression at all during the first year of life, and these particular infants had never before been observed to act aggressively. It is very unlikely, then, that they had learned aggressive actions earlier. Second, and we shall return to this point later, the aggressive response to a frustration is clearly revealed only under certain conditions.
Inflicting injury as a goal response
Another criticism of the frustration-aggression hypothesis as advanced by Dollard and others stems from a particular philosophy of science embraced by some psychologists. In saying that aggression was a behavioral sequence whose goal was the injury of the person to whom the activity was directed, the authors of the 1939 monograph implied that this behavior was purposive or intentional. Some writers, taking the position of Watsonian behaviorists, object to the inclusion of intentionality in the definition of aggression. Intentionality usually has to be inferred, and these critics prefer to confine themselves to a strict operationism having little room for inferences. While we cannot here debate the merits and demerits of this approach, there is reason to believe that aggressive frustration reactions are frequently purposive in nature. Hokanson and his students have shown that provoked subjects who are permitted to aggress against their tormentor often display a drop in systolic blood pressure that brings their pressure level close to that exhibited by a nonaroused control group. Systolic pressure does not decline as much, however, when the angered people can carry out some activity, but do not believe they have inflicted injury on their frustrater (even though the activity is physically comparable to the aggressive response), or when the aggression is directed against someone other than the person who had provoked them (cf. Hokanson et al. 1963). If the decrease in systolic blood pressure is a sign of physiological relaxation brought about by the performance of a goal response, engaging in mere activity or aggressing against just anyone does not seem to be sufficient to produce this tension reduction; the frustrated person (who wants, and is prepared, to attack his frustrater) may be primarily concerned with injuring the person who had provoked him.
A revised frustration–aggression hypothesis
Elsewhere (Berkowitz 1962) I have suggested that the original version of the frustration-aggression hypothesis should be modified in three ways. First, I would contend that a frustration—preventing the occurrence of some goal response at its proper time in an ongoing response sequence— arouses, among other things, an emotional state, anger, that creates a readiness for aggressive acts. (The arousal state produced by a thwarting also increases the strength of the ongoing responses, whether these are aggressive or not.)
Second, it is important to make explicit what was only implicit in the formulation advanced by Dollard and his coworkers. Aggressive responses do not occur, even given this readiness, in the absence of suitable cues—stimuli associated with the present or previous anger instigators. These cues, in other words, evoke aggressive responses from a person who is “primed” to make them. The strength of the aggressive response resulting from a frustration presumably is a function of the intensity of the aroused anger and the degree of association between the available stimuli and past and present anger instigators.
By suggesting that cues are necessary to elicit aggressive actions, we can explain two propositions advanced by Dollard and his coworkers: In the absence of inhibitions, the strongest hostile responses supposedly are directed toward the perceived source of the frustration; and progressively weaker aggressive responses theoretically are evoked by objects having less and less similarity to the frustrater. In both cases, the more direct the association with the anger instigator the stronger is the aggressive reaction that is elicited.
A third necessary revision of the frustration-aggression hypothesis restricts the extent to which thwartings are employed as an explanation of aggression. Instead of maintaining that all aggression “presupposes the existence of frustration,” we now recognize that (a) suitable cues may lead to aggressive behavior by arousing previously acquired aggressiveness habits and (b) such habits may be formed through learning—for example, by observing the behavior of some adult model—without involving a thwarting (Bandura & Walters 1963). To repeat, the revised hypothesis now would claim only that frustrations create a readiness for aggressive behavior.
Hostility displacement
The revised formula discussed above has some important implications for such phenomena as displacement and “scapegoating.” In the scapegoat theory of prejudice the victim is said to be attacked primarily because he is a visible and safe target for pent-up hostility within the prejudiced individual. The aggressive “drive” supposedly “pushes” aggressive acts onto safe, available targets. Contrary to such a view, the present position contends that a target with appropriate stimulus qualities “pulls” (evokes) aggressive responses from a person who is ready to engage in such actions either because he is angry or because particular stimuli have acquired cue value for aggressive responses from him.
Several experiments by the writer and his students offer support for this analysis of hostility displacement. In one study (Berkowitz & Green 1962) it was shown that subjects who were deliberately provoked by the experimenter subsequently exhibited greater hostility toward a person who had angered them some time earlier than toward someone else who had not provoked them. By having angered the subjects previously, the former person had acquired the cue value that now enabled him to evoke hostile responses from them when they were later thwarted by the frustrater. The writer suggests that those minority groups that frequently are the victims of displaced hostility, such as Negroes and Jews, are capable of eliciting aggressive responses from thwarted people because these groups are strongly disliked, that is, they had previously aroused anger. The dislike could arise from learning that the groups have unpleasant qualities as well as from prior frustrating experiences with them.
According to this reasoning, most explanations of social prejudice are too one-sided. Typically they either explain why some people are ready to act aggressively, or they provide reasons why certain minority groups are disliked. The present formulation attempts to integrate the two sets of explanations: we have to know both (a) why some people are “primed” to act aggressively and (b) what stimulus qualities are possessed by certain groups which enable them to evoke hostile responses from the people having a readiness to behave aggressively.
Comprehensive analyses of scapegoating must deal with those people who characteristically display relatively strong hostility toward minority groups. Recent research (Berkowitz 1959) indicates that highly ethnocentric college students typically have a strong tendency to attack other people when frustrated by someone else. It is not altogether clear, however, whether this tendency is due to (a) previously acquired aggressiveness habits; (b) a proclivity to establish broad categories, especially when under stress, so that in essence the immediate frustrater is not sharply differentiated from the others; (c) intense emotional arousal; or (d) some combination of these factors.
Inhibiting aggressive reactions
When an angered person displaces hostility, he presumably does so because fear or anxiety inhibits direct aggression against the frustrater. The analysis presented here helps explain why the displaced aggression is frequently quite intense. Miller’s (1948) translation of the psychoanalytic formulation maintains, as I indicated earlier, that stimulus objects increasingly removed from the frustrater on some appropriate generalization dimension would evoke weaker and weaker aggressive responses from the angered person. Attacks upon some bystander should thus be relatively weak and never stronger than the aggression that would be directed against the anger source. At least two studies, however, have obtained a substantial departure from this prediction. In one experiment (Pepitone & Reichling 1955), angered men whose inhibitions were presumably lowered by placing them in the company of others they liked exhibited the expected aggression gradient: much stronger attacks upon their tormentor than upon other associated stimuli. However, a more strongly inhibited group of men did not show this difference; the aggression they directed against the associated stimuli was practically as intense as the aggression against the anger instigator. Essentially similar findings have been obtained in other studies. Where members of a less inhibited group generally directed the strongest aggression against the source of their anger, most of the men in a more inhibited condition actually were more hostile toward the associated stimuli than toward the anger source (Berkowitz et al. 1963).
This last finding can be explained readily by noting that the inhibition of an aggressive response is a frustration. The emotional reaction produced by the thwarting should increase the strength of the individual’s aggressive proclivities. Since there is a restraint against a direct attack on the frustrater, attacks evoked by other stimuli should be strengthened. (Dollard and his colleagues had made this prediction in their 1939 monograph.)
Expectations to aggress
Anger arousal does not in itself necessarily lead to an instigated aggressive response sequence. Stimuli associated with the anger instigator must also be present. These cues may be provided either by the external environment or the individual’s thought processes. If a person is thinking of injuring his tormentor, the aggression goal (his frustrater) is symbolically represented in his thoughts. This symbolic representation serves as a cue setting an aggressive sequence into operation, if only internally. Preventing him from attacking his frustrater would then be an additional thwarting. Suppose, on the other hand, that the angered person does not see his tormentor, or any associated stimuli, and, for that matter, does not even think of “getting even.” There would be no ongoing aggressive sequence, and an inability to attack the anger source would not be a frustration.
Berkowitz and Luehrig have obtained data consistent with this reasoning (Berkowitz 1964a), demonstrating that angered male college students who had expected to be able to attack their frustrater but were prevented from doing so subsequently exhibited a high level of aggression toward him at the end of the session. The aggression they displayed toward him was stronger than that shown by other provoked people who had been able to attack this person earlier with electric shock or who, although not having been given this earlier opportunity, had not previously expected to be able to attack him. It is not enough just to say that a “set” or expectation had been thwarted in the former group without explaining the meaning of “set.” “Set” may consist of an implicit but nevertheless ongoing chain of responses. In the present case the implicit response sequence was activated by the anticipation of the aggressive opportunity, and the inability to satisfy this expectation was thus frustrating.
Hostility catharsis
Since the time of Aristotle, social science folklore has contended that an individual can be “purged” of his emotions by displaying his feelings. Aggressive behavior should therefore weaken the instigation to further aggression (unless he is frustrated again) and should somehow make the person feel better. But although such a proposition is almost universally accepted, empirical evidence regarding this catharsis hypothesis is far from unequivocal. We cannot be sure that catharsis takes place as readily and frequently as many people seem to assume.
There are manifold problems confronting research in this area. One difficulty is that observers have often regarded the intensity of the overt aggression following an initial attack as being a good indicator of the strength of the remaining “aggressive drive.” They forget that inhibitions produced by guilt or anxiety arising from the initial aggression may weaken any subsequent expressions of hostility. But even in the absence of such inhibitions, diminutions in aggressive tendencies following an intervening experience are also not necessarily due to a cathartic drainage of aggressive “energy.” An angered person who then watches a movie or a football game may calm down and become friendlier to his frustrater, not because he has discharged his anger vicariously but because he has been so distracted by the movie or game that he does not think of the thwarting he has experienced and ceases to stir himself up. Since he does not stimulate himself, his anger dissipates. To mention one other problem, the catharsis hypothesis is frequently tested by comparing an angered group that is permitted to aggress against someone with a similarly provoked group not given this aggressive opportunity. However, as we have already seen, if measurements obtained later should suggest there is a greater level of residual hostility in the latter (nonaggressing) condition, this difference may not be due to a catharsis in the group permitted to aggress; the angered people in the nonaggressing condition may have been frustrated —assuming they had wanted to, and had been prepared to, attack the anger source—and, so, they became more strongly aroused.
Clearly, given such difficulties, we cannot definitely say now that the free expression of aggression will automatically reduce the likelihood of subsequent aggression. Indeed, several studies of children in situations similar to play therapy suggest that the expression of aggression under such permissive conditions frequently serves to increase the probability of later violence (cf. Berkowitz 1962). The permissive situation may weaken inhibitions, and the performance of aggressive actions can strengthen aggressiveness habits. Nor does the observation of other people engaging in violence generally reduce aggressive tendencies. (1) The person who watches others acting aggressively often learns to behave aggressively through modeling himself after these others. (2) The witnessed hostility may provide cues activating previously acquired aggressiveness habits. (3) The witnessed hostility may affect the observer’s judgment of the propriety of his own aggressive desires.
To demonstrate this last-mentioned possibility, in three separate experiments Berkowitz (1964b) and his students showed a filmed prize fight scene to deliberately angered college men. In some cases the aggression they watched was made to appear justified (in that a villain received his “comeuppance”), while for other men the witnessed aggression was made to appear less warranted. Since the justified fantasy aggression lowered inhibitions against aggression—as indicated by several measures—the catharsis hypothesis would predict that the angered men in this group would participate vicariously in the filmed violence and, thus, would purge themselves of their anger. But contrary to this expectation, the angered men in this justified fantasy-aggression condition later displayed stronger aggression against their frustrater than did the similarly provoked group shown the less justified aggression. If aggression was warranted on the screen, the former may have thought, it was all right to attack the villain in their own lives.
Performing an aggressive act may well be a goal response completing an ongoing aggressive response sequence, but satisfactory completion is apparently attained only to the extent that (a) the angered person himself (or perhaps someone the person associates with himself) does the attacking and (b) the frustrater (or perhaps someone associated with him) is injured. Further, we are not altogether certain as to what the effects of this completion would be. It may well produce a feeling of tension reduction, especially if the angered person had wanted to, and had been prepared to, aggress against the anger instigator but, for some reason, had not been able to do so right away. Whether the tension reduction signifies a decreased likelihood of subsequent aggression is not altogether clear, however. The aggressive act may lessen the thwarted person’s anger at the moment, and thus may lower the probability of aggression at this time. But this frustrater has also acquired cue value for aggressive responses. Much like a red flag waved in front of a bull, under appropriate conditions (such as another thwarting experience) this individual may again evoke aggressive responses from the aroused person, whether or not he was the cause of the thwarting.
The reasoning just advanced obviously differs sharply from the “drainage” conception of aggressive behavior. Contrary to the notion of a free-floating aggressive energy that may be released through many different activities (for example, attempting to master others), or in attacking a wide variety of objects, available evidence suggests that the catharsis hypothesis must be restricted in scope. Moreover, if the view given here is correct, it is not necessary to provide substitute activities in order to “drain” a supposed reservoir of pent-up emotion. Unless the thwarted person is kept aroused or is rearoused, his anger probably will dissipate with time, and the probability of aggression will decline. But even if he is angry or has developed aggressiveness habits, aggressive behavior presumably will not occur unless appropriate cues are present.
Leonard Berkowitz
[Directly related are the entries Conflict; War. Other relevant material may be found in Instinct; Psychoanalysis; Stress.]
BIBLIOGRAPHY
Bandura, Albert; and Walters, R. H. 1963 Social Learning and Personality Development. New York: Holt.
Berkowitz, Leonard 1959 Anti-Semitism and the Displacement of Aggression. Journal of Abnormal and Social Psychology 59:182–187.
Berkowitz, Leonard 1962 Aggression: A Social Psychological Analysis. New York: McGraw-Hill.
Berkowitz, Leonard 1964a Aggressive Cues in Aggressive Behavior and Hostility Catharsis. Psychological Review 71:104–122.
Berkowitz, Leonard 1964b The Effects of Observing Violence. Scientific American 210, February:35–41.
Berkowitz, Leonard; Corwin, R.; and Heironimus, M. 1963 Film Violence and Subsequent Aggressive Tendencies. Public Opinion Quarterly 27:217–229.
Berkowitz, Leonard; and Green, J. A. 1962 The Stimulus Qualities of the Scapegoat. Journal of Abnormal and Social Psychology 64:293–301.
Buss, Arnold H. 1961 The Psychology of Aggression. New York: Wiley.
Buss, Arnold H. 1963 Physical Aggression in Relation to Different Frustrations. Journal of Abnormal and Social Psychology 67:1–7.
Davitz, Joel R. 1952 The Effects of Previous Training on Postfrustration Behavior. Journal of Abnormal and Social Psychology 47:309–315.
Dollard, John et al. 1939 Frustration and Aggression. Yale University Institute of Human Relations. New Haven: Yale Univ. Press.
Eibl-Eibesfeldt, IrenÄus 1963 Aggressive Behavior and Ritualized Fighting in Animals. Pages 8–17 in Jules H. Masserman (editor), Violence and War: With Clinical Studies. Academy of Psychoanalysis, Science and Psychoanalysis, Vol. 6. New York: Grune & Stratton.
Freud, Sigmund (1920) 1950 Beyond the Pleasure Principle. Authorized translation from the 2d ed., by C. J. M. Hubback. International Psycho-analytic Library, No. 4. New York: Liveright. ? First published under the title Jenseits des Lustprinzips. A paperback edition, translated by James Strachey, was published in 1959 by Bantam Books.
Hokanson, J. E.; Burgess, M.; and Cohen, M. F. 1963 Effects of Displaced Aggression on Systolic Blood Pressure. Journal of Abnormal and Social Psychology 67:214–218.
Holst, Erich von; and Saint paul, Ursula von 1962 Electrically Controlled Behavior. Scientific American 206, March: 50–59.
Lorenz, Konrad 1952 King Solomon’s Ring: New Light on Animal Ways. New York: Crowell; London: Methuen.
Mcdougall, William (1908) 1936 An Introduction to Social Psychology. 23d ed., enl. London: Methuen. ? A paperback edition was published in 1960 by Barnes and Noble.
Miller, Neal E. 1941 The Frustration-Aggression Hypothesis. Psychological Review 48:337–342.
Miller, Neal E. 1948 Theory and Experiment Relating Psychoanalytic Displacement to Stimulus-Response Generalization. Journal of Abnormal and Social Psychology 43:155–178.
Pepitone, Albert; and Reichling, George 1955 Group Cohesiveness and the Expression of Hostility. Human Relations 8:327–337.
Scott, John P. 1958 Aggression. Univ. of Chicago Press.
Seay, B. M.; and Harlow, Harry F. 1963 Personal communication.
Tinbergen, Nikolaas 1951 The Study of Instinct. Oxford: Clarendon.
White, Robert W. 1959 Motivation Reconsidered: The Concept of Competence. Psychological Review 66:297–333.
II INTERNATIONAL ASPECTS
Power holders and decision makers in the parochial sovereignties into which mankind has been habitually divided have repeatedly resorted to armed violence against neighboring communities in pursuit of political objectives. Every such resort to force has invariably been regarded as aggression (hence as unjustified, illegal, and immoral) by its victims and, also invariably, been deemed moral, legal, and justified by the alleged “aggressors” on the grounds of “self-defense,” “preservation of the balance of power,” “national honor,” or some other plausible formula for rationalizing recourse to war.
Aggression as a concept or abstraction poses, therefore, a semantic and psychological problem rather than a problem admitting of solution by reference to traditional criteria of international law, diplomatic practice, military science, or international organization. In our time, aggression is a term of disapproval, usually limited to acts of military violence by “enemy” states whose purposes must be resisted. It is a prime article of faith, in the cult of nationalism, that aggression is always a crime committed by enemy governments and never a sin of one’s own nation-state. The obvious falsity of this dichotomy has had little or no effect on the behavior of those committed to its fallacies.
The problem is not so simple, however, as these introductory comments suggest. Whether, the propensity of human beings to resort to violence against other human beings is attributable to “instinct” or to “culture” has long been debated inconclusively. Some hold, with Sigmund Freud, that the human psyche is afflicted with a “death instinct,” which avoids suicide only by a triumph of Eros over Thanatos, or, more commonly, by deflection of aggression against the self to aggression against others. Others, along with Bronislaw Malinowski, hold that aggression is not innate but is a product of the frustration of other human aspirations.
The issue in international law
In the realm of interstate relations many people, and policy makers, confronted with the recurring tragedies of worldwide violence in the twentieth century, have earnestly striven to outlaw war and to forbid aggression by international agreement. All such efforts have thus far failed.
The “outlawry of war” was anticipated in many late medieval and early modern treaties whereby sovereignties solemnly pledged themselves to perpetual peace. Since such pledges had no discernible effect on the subsequent decisions of statesmen, the formula was abandoned in the nineteenth century. It was revived in the twentieth by the shock of World War i. The League to Enforce Peace, 1915-1919, championed American membership in a league of nations in which “aggression” was to be met with such overwhelming economic and military force that it would not be attempted.
The League of Nations Covenant of 1919 forbade recourse to war, with qualifications. The Kellogg-Briand Pact of 1928, which pledged its signatories to renounce war as an instrument of national policy, save in “self-defense,” was largely vitiated by the numerous national interpretations and reservations. Other bilateral and multilateral treaties of the 1930s reiterated the same goal. The Charter of the United Nations bound its members “to ensure, by the acceptance of principles and the institution of methods, that armed force shall not be used save in the common interest” and, further, “to maintain international peace and security, and to that end to take effective collective measures … for the suppression of acts of aggression or other breaches of the peace” (art. 1).
The assumption of the framers of the charter at San Francisco in the spring of 1945 was that a concert of powers, cooperating for common purposes, would guarantee peace and halt aggression by collective action, provided that the great powers with permanent seats on the Security Council would act unanimously on all measures of “enforcement” (cf. art. 27, the “veto” article of the UN Charter). The assumption proved false with the advent of the cold war in 1945/1946. Common action against aggression was henceforth impossible, with each contestant using real or alleged “aggression” by others as a weapon of propaganda, diplomacy, and strategy against the “enemy.”
Within the context of the logic of international law, aggression consists in resort to war or to measures of armed coercion short of formal war, undertaken in violation of treaty obligations not to resort to war or to other acts of force. In the years that followed the signing of the Kellogg-Briand Pact, national policy makers, with the support of all patriots, paid no attention to the legal duties they or their predecessors had assumed whenever prevailing concepts of “national interest” dictated an opposite course. Law is an effective guide to conduct only in organized communities whose members accept its purposes as paramount. The Western state system is not such a community.
The totalitarian states have provided numerous examples of violations of treaty obligations forbidding aggression—for example, the fascist conquest of Ethiopia in 1935-1936; the Nazi seizure of Austria and Czechoslovakia in 1938–1939; Axis intervention in Spain in 1936-1939; Hitler’s invasion of Poland in 1939; the Japanese assaults on China in 1931 and 1937 and on the United States in 1941; Stalin’s partition of Poland with Hitler in 1939; the Soviet attack on Finland in 1939-1940 and conquest of the Baltic states in 1940. However, the democracies have no better record. Witness, among recent instances, the abortive Anglo-French-Israeli attack on Egypt in 1956; India’s seizure by force of Hyderabad and Goa and attempted seizure by force in 1962 of territories in dispute between New Delhi and Peking; and the U.S. “spy flights” over the U.S.S.R. by U-2 planes (1955-1960), the Bay of Pigs invasion of Cuba in April 1961, and the “quarantine” of Cuba in October 1962. Other instances of resort to force come readily to mind, with each side accusing the other of “aggression”: the U.S. war on North Vietnam, launched February 7, 1965; the U.S. intervention in the Dominican Republic, April 28, 1965; the India-Pakistan war over Kashmir, September 1965.
The problem of definition
It is evident that decision makers in sovereign states do not abide by treaty obligations when “security” or “defense” or ambition dictates contrary conduct. This circumstance has led diplomats and legalists to cultivate the illusion that obedience to law could somehow be assured if only a more precise definition of aggression could be formulated and generally accepted. Laborious efforts at Geneva in the 1920s in the name of the League of Nations were devoid of operational results. Similar efforts by the United Nations since 1946 have been equally in vain. The most notable, albeit futile, attempt to define aggression was made by Maxim Litvinov, foreign minister of the U.S.S.R., at the London Economic Conference of 1933. On July 4 of that year he signed a “Convention for the Definition of Aggression” with envoys of Rumania, Yugoslavia, and Turkey (a few other states subsequently adhered). The signatories to this convention agreed to define the aggressor in an international conflict as that state which is the first to commit any of the following actions: (1) declaration of war upon another state; (2) invasion by its armed forces, with or without a declaration of war, of the territory of another state; (3) attack by its land, naval, or air forces, with or without a declaration of war, on the territory, vessels, or aircraft of another state; (4) naval blockade of coasts or ports of another state; (5) provision of support to armed bands formed on its territory which have invaded the territory of another state, or refusal, notwithstanding the request of the invaded state, to take on its own territory all the measures in its power to deprive those bands of all assistance or protection (art. 2). Furthermore, it was stipulated that no political, military, economic, or other considerations could serve as an excuse or justification for such acts of aggression (art. 3).
No better definition of aggression has been formulated since 1933. The futility of the enterprise is shown by the fact that Soviet policy makers in their “winter war” against Finland in 1939–1940 violated all the obligations they had so recently assumed. Other instances of violations of pledged words are innumerable in the diplomatic and military history of the twentieth century.
The dilemma
The premise of all efforts to outlaw war and to define and forbid aggression by bilateral or multilateral accords among sovereignties is that national policy makers will be deterred from misbehavior by their promises and by threats of action against them from the entire community of nations. In practice, the promises are frequently ignored, for reasons pointed out by Machiavelli more than four centuries ago. As for the efficacy of threats against aggressors in the name of collective security, the verdict of experience thus far is negative. National policy makers, dedicated to the pursuit of the national interest, whether this is defined in terms of power, pride, profit, or prestige, can always be relied upon, whatever their rationalizations, to ignore their commitments to refrain from aggression whenever they believe that resort to force will serve their purposes.
Aggression can probably never be prevented by legalistic formulae or by the artifacts of international organization and collective security, so long as the state system remains an arena of international anarchy among polities possessed of unlimited national sovereignty. Aggression will cease only when mankind reluctantly accepts the necessity of a drastic alteration of values and purposes in international relations and gives operational meaning to the ideal of world government. This goal is remote because of the universal disposition of Homo sapiens to cling to ancient ways in the face of new circumstances calling for new thought and threatening disaster if rethinking is resisted. In the absence of significant progress toward this objective, aggression in interstate relations will continue in the future, as in the past, with potentially catastrophic consequences—and all efforts to define, outlaw, and forbid recourse to force by one state against another must inevitably fail of their purpose.
Frederick L. Schuman
[See alsoCollective security; International law; War.]
BIBLIOGRAPHY
Boggs, Marion William 1941 Attempts to Define and Limit Aggressive Armament in Diplomacy and Strategy. Columbia: Univ. of Missouri Press.
Langer, Robert; and Schuman, Frederick L. 1947 Seizure of Territory: The Stimson Doctrine and Related Principles in Legal Theory and Diplomatic Practice. Princeton Univ. Press.
Litvinov, Maxim 1939 Against Aggression. New York: International Publishers. ? Speeches, together with texts of treaties and of the Covenant of the League of Nations.
Research in International Law, Harvard Law School 1939 Draft of Conventions Prepared for the Codification of International Law. Concord, N.H.: Rumford Press. ? Also published as a supplement to Volume 33 of the American Journal of International Law, 1939.
Stone, Julius 1958 Aggression and World Order: A Critique of United Nations Theories of Aggression. Berkeley: Univ. of California Press.