Data for this study were taken from the 1988 General Social Survey. The full probability sample consisted of 1,481 English- speaking persons 18 years old or older. Because certain questions were asked of only a subsample of the total, 737 cases were available for analysis in this study. Models were tested through logistic regression with the use of the SPSS statistical package. The analysis is presented in three stages that correspond to the theoretical model, which conceives of gun ownership as dependent upon civil and religious punitiveness, civil punitiveness as dependent upon religious punitiveness, and religious punitiveness as dependent on fundamentalism. Research results suggest that among some Americans, gun ownership is a punitive response to crime, which is rooted in religious beliefs. Whether the links between religious punitiveness and guns are direct or are filtered through support for punishment administered by civil authority, however, varies by race. In contrast to African- Americans, the ownership of firearms among whites is related to their attitudes toward civil punitiveness, not religious punitiveness. Moreover, except for the effect of punishment of sin, the only religious variable that taps punitiveness, religion does not seem to have a direct role in the formation of whites' attitudes toward civil punitiveness or their tendency to own guns. The white and black models had only one relationship common to both. Evangelicals of both races were more likely to believe that sin must be punished. At odds with previous research is the lack of a direct link from fundamentalism to civil punitiveness. Among African-Americans, this relationship is negative. This suggests that, controlling for other relevant variables, fundamentalists are not inordinately punitive in civil matters. The results clearly show that any research that focuses on the role of religion in civil matters must recognize the importance of race, not just as a control variable in the analysis, but as an indicator of cultural experiences that influence the nature of belief systems and the relationships between those beliefs and socially relevant attitudes and behaviors. 4 tables, appended variable descriptions, and 32 references