The main meta-analysis included 30 experimental studies that used between-subjects designs, male confederates, and male subjects who were social drinkers. Studies using other designs or subject populations were integrated with meta-analytic procedures when possible and were summarized descriptively when not. Evidence from the main meta-analysis indicated that alcohol caused aggressive behavior. In addition, alcohol appeared to influence aggressive behavior as much or more than it influenced other social and nonsocial behavior. Several theories or explanations of intoxicated aggression are discussed, and certain suggestions are offered to guide future alcohol and aggression research. If a balanced placebo design is not used, a control condition should be added to a placebo design. The alcohol versus control comparison may provide the best estimate of the effects of alcohol on aggression. Those who use antiplacebo and placebo conditions should take special steps to ensure that beverage manipulations are successful. Attention should also be paid to variables that can potentially moderate alcohol effects, and researchers should continue to test and develop theories of intoxicated aggression. 81 references and 4 tables (Author abstract modified)