Male and female subjects were presented with a detailed description of a date in which the male used low or moderate force to obtain sex after the female began to protest either early, moderately, or late during foreplay. Her protest consisted of only pleading or pleading plus a physical struggle. The 'true' experiment was analyzed by means of AOV and path analysis. Subjects were more likely to blame the woman and to perceive her as desiring sex with low force and late onset of protest. The man was viewed as more violent and the incident more likely to be viewed as rape when there was more force, more protest, and earlier onset. Attitudes toward women was a significant predictor of all dependent variables and no overall sex differences were found. The study concluded that the attributions of the male's violence and the female's desire for sex are definitional components of rape and intervening variables, caused by the manipulated variables and in turn causing the perception of rape. Because experimental manipulations were shown to affect more than one cognition, an argument can be made for developing causal models, including situational and cognitive intervening variables in predicting final attributions. The benefits of using 'weak' manipulations are discussed. Tabular data and 40 references are provided. (Author abstract modified)