Two waves of longitudinal data were used to examine the sequencing between violent victimization, violent behavior, and gun carrying in a high-poverty sample of African American youth. Multivariate logistic regression results indicated that violent victimization T1 and violent behavior T1 increased the likelihood of initiation of gun carrying T2 when examined separately (by 132% and 91%, respectively). However, only violent victimization T1 was a significant predictor of initiation of gun carrying T2 after controlling for violent behavior T1. More nuanced analyses uncovered no significant difference in the likelihood of initiating gun carrying when comparing offensive versus defensive gun carriers. The theoretical and policy implications of these findings are also discussed. (Contains 3 tables and 2 figures.)