The facilitation of aggression by aggression: Evidence against the catharsis hypothesis.

Abstract

Ninety male subjects were either attacked or treated in a more neutral manner by a male confederate. On a subsequent maze-learning task, one third of the subjects shocked the confederate, one third observed as the experimenter shocked the confederate, and one third waited for a period of time during which the confederate was not shocked. Finally, all subjects shocked the confederate as part of a code-learning task. Subjects who had been attacked and had shocked the confederate during the maze task delivered shocks of greater intensity on the code task than did subjects in the other two conditions, and the former subjects also experienced a greater reduction in diastolic blood pressure than did the latter. The results contradict the hypothesis of aggression catharsis and are discussed in terms of feelings of restraint against aggressing that a subject experiences after committing an aggressive act. The idea of aggression catharsis is popular both in conventional wisdom and in psychological writing. Certainly it seems to make sense that attacking an enemy allows the attacker to purge himself of hostility and anger toward his victim. Furthermore, in both psychoanalytic theory (e.g., Schafer, 1970) and the frustration-aggression hypothesis (Bollard, Doob, Miller, Mowrer, & Sears, 1939), the notion is expressed that aggression leads to a release of aggressive drive or instigation, so that further acts of violence are rendered less probable. This point of view stipulates a fairly simple cause-effect model: instigation produces aggressive arousal, which motivates an aggressive response, which in turn reduces arousal. The end result is a declining level of aggression. Largely because of this cause-effect assumption, however, the catharsis hypothesis has encountered troubles on empirical, methodological, and theoretical levels. Although some investigators have reported results that support the notion of catharsis (e.g., Doob, 1970; Feshbach, 19SS; Thibaut & Coules, 1952), many others have made findings that contradict it (e.g., deCharms & Wilkins, 1963; This study was supported by National Science