Nonexperimental field studies have demonstrated an association between alcohol intoxication and self-aggressive behaviors across the spectrum of lethality. Although these results are suggestive, it is not yet known whether alcohol intoxication is causally related to self-aggression. The authors therefore experimentally examined the effects of alcohol intoxication (mean blood alcohol concentration of .10) on a behavioral measure of self-aggression in men (N = 40). After consuming either an alcohol or a placebo drink, participants were provided the opportunity to self-administer shock during a task disguised as a reaction-time game, with self-aggression defined by the intensity of shock chosen. Half of the participants observed a self-aggressive model (a potential moderator of alcohol-related self-aggression). Independent alcohol and model effects were found, with alcohol accounting for over 30% of the self-aggression variance.