Physical and sexual abuse have far-reaching mental and behavioral health consequences, extending across the lifespan and, in some cases, across generations. However, empirical work in this area is limited by cross-sectional study designs, short follow-up durations, and data analytic techniques that fail to capture the nuanced developmental processes through which caregivers and children impact one another. The present study investigated the cross-lagged and bidirectional pathways between maternal childhood victimization, depression, harsh parenting, and their children’s externalizing symptoms over a 10-year period. Participants were 818 mother-child dyads prospectively identified as at-risk for family violence when children were four years old. Traditional cross-lagged panel modeling (CLPM) and random-intercept cross-lagged panel modeling (RI-CLPM) documented that maternal depression, harsh parenting, and child externalizing problems — all predicted by mothers’ early abuse experiences — exacerbated one another across time. Discrepancies between the CLPM and RI-CLPM highlighted the advantages, disadvantages, and methodological implications of each approach. Findings highlight maternal psychopathology and parenting as key mechanisms in the intergenerational impact of abuse, emphasizing the importance of trauma-informed, parent-mediated interventions for breaking long-term cycles of family dysfunction. The present findings support separating out between-person, trait-like components when interpreting cross-lagged associations, as these may confound within-person effects.