Between 1986 and 2006, the Lord's Resistance Army abducted an estimated 50,000 to 80,000 children in Northern Uganda. Boys were forced into combat. Girls faced sexual violence and forced labor. More than half of those taken were children, experiencing trauma during critical developmental years. The vast majority eventually escaped or were freed when the conflict ended.

Decades later, survivors have married and started families. Some married other survivors. Some married people who were never abducted. The question is: does exposure to violence in childhood impact violence in the household as an adult?

The Missing Piece: Looking at Both Partners

Most research on conflict and domestic violence focuses on individuals. Studies examine whether men who experienced combat are more violent toward their wives, or whether women in conflict zones face higher abuse rates (Cesur and Sabia, 2016; Stojetz and Bruck, 2023; Saile R, 2013; La Mattina, 2017). But this approach misses something fundamental: violence happens in relationships, and both partners bring their histories into the household.

To understand violence after conflict, we need to examine how partners' trauma histories interact, not just whether someone experienced trauma themselves. This bilateral approach reveals patterns that completely disappear when looking at survivors alone.

Data from 214 couples in Northern Uganda's Kitgum district, where abduction rates were exceptionally high, shows that the couple’s combined history may matter. In randomly selected villages, 47% of men and 38% of women had been abducted at some point. In a region where abduction was so widespread, the data show that people didn’t consider abduction status when choosing a partner. Instead, they looked for the same things people consider everywhere, including education level and family wealth. This lack of sorting on abduction status created an opportunity to compare four types of couples: those where neither partner was abducted, only the woman, only the man, or both.

What the Bilateral Framework Reveals

Looking at women's abduction status alone tells you almost nothing about their risk of intimate partner violence. But examine the couple together, and a clear pattern emerges.

Abducted women married to non-abducted men face substantially higher rates of physical violence. These women experience more than half a standard deviation increase in nearly every type of physical abuse compared to women in households where neither partner was abducted. They are significantly more likely to experience severe forms of violence, including kicking, punching, and violence with weapons.

This elevated risk largely disappears when both partners share a history of abduction. Women married to men who were also abducted show violence rates similar to couples where neither partner was abducted. The protective effect appears to operate through increased prosociality. Men who were abducted demonstrate significantly more empathy toward women with similar trauma histories, measured through incentivized behavioral games where they allocated real money to recipients of different types.

For men, the pattern differs. Men who were abducted face higher rates of victimization, particularly verbal abuse, regardless of their partner's history. Importantly, abducted men are not more likely to perpetrate violence. Their lower perceived social status in the community may reduce the social costs of victimizing them.

Violence Across Generations

The bilateral framework matters for children too. Looking at individual parents' abduction status shows no clear pattern. But examine the household as a unit, and the risk becomes apparent.

Children in homes where either parent was abducted are 17-20 percentage points more likely to be hit by their mothers. This holds whether the mother herself was abducted, the father was abducted, or both were abducted. Among formerly abducted mothers specifically, the increased violence is concentrated on daughters.

Fathers' use of physical discipline shows no relationship to abduction, regardless of which parent experienced trauma. The result underscores that violence transmission operates through specific parent-child dyads that only become visible when examining household-level trauma dynamics.

Why This Matters for Policy

These findings have direct implications for how we design interventions in post-conflict settings.

First, programs need to assess household trauma dynamics, not just screen individuals. A woman's risk depends on her partner's history, not just her own. A child's risk depends on patterns at the household level. Interventions that treat survivors in isolation miss the relational context that determines actual violence patterns.

Second, we need to recognize that men can be victims of intimate partner violence, particularly men with conflict trauma. Current programming overwhelmingly focuses on women as victims and men as perpetrators. This framework doesn't fit the evidence. Men with trauma histories face real victimization risk, driven partly by stigma and reduced social status.

Third, the protective effect of shared trauma for women suggests that couple-level interventions might be particularly effective. Abducted men already show more prosocial behavior toward partners with similar histories. Programs that build on this foundation, perhaps through couples counseling or shared support groups, could strengthen protective factors.

Finally, interrupting intergenerational transmission requires attention to mother-child relationships in trauma-affected households. However, this could vary depending on the cultural context as Acholi mothers are the traditional disciplinarians of children.

Beyond "Violence Begets Violence"

The conventional wisdom is simple: violence exposure creates violent people. But the evidence shows something more complex. The same childhood trauma produces different outcomes depending on household composition. Who you marry matters as much as what you experienced.

This isn't to minimize the real harms of conflict exposure. Abduction creates lasting mental health challenges, reduces social status, and increases stress. But these individual-level effects don't translate directly into household violence. They are mediated by partners' responses, which depend on partners' own trauma histories.

The bilateral framework reveals patterns that simpler models miss. It shows that some combinations of trauma histories create elevated risk while others provide protection. It demonstrates that gender patterns are more nuanced than typical perpetrator-victim frameworks suggest. And it highlights that violence transmission across generations operates through household-level dynamics that require household-level solutions.

Twenty years after Northern Uganda's conflict ended, its effects persist in intimate relationships and parenting practices. But these effects are not uniform, and they are not inevitable. Understanding how trauma histories combine within households is essential for designing interventions that actually match the patterns we observe. Programs that ignore these relational dynamics risk missing the survivors who need support most, and missing opportunities to build on the protective factors that already exist in some households.

Looking at couples together, not individuals alone, changes what we see and what we can do about it.

About the Author

Miranda Lambert is a Ph.D. candidate at Texas A&M University

Her research interests are in Development Economics, Experimental and Behavioral Economics, and Health Economics. To learn more about her research, visit: https://sites.google.com/view/mirandalambertecon/

Social Media Handles:

• LinkedIn
• X
• Bluesky
• Mail