Intergenerational Transmission of Violence


The notion that family violence persists across generations is pervasive among clinicians, researchers, and the general public. Although many people expect consistent intergenerational transmission of violence (ITV), many scholars have questioned the supposed inevitability of transmission. Phenomena such as partner violence and child abuse clearly lead to myriad negative outcomes for many victims, including subsequent victimization due to involvement in relationships with violent partners, as well as perpetration of violence toward others, including partners and children. Estimates of the likelihood of ITV across generations vary widely, and researchers have found several risk and protective factors that alter the rates of transmission. Ultimately, the majority of people exposed to family violence during childhood are not involved in partner violence or child abuse as adults.

Transmission of family violence across generations may occur via several mechanisms. Social learning theory indicates that children learn to be perpetrators and/or victims of violence through exposure to their parents’ expressions of violence. According to attachment theory, child abuse leads to insecure attachment between parents and children; changes in the child’s internal working model result in later relationship difficulties and inadequate care for one’s own children. Another possible explanation is that family violence during childhood results in increased stress and negative life events; during adulthood, high stress and limited resources lead people to use violence. Assortative mating suggests that people select mates similar to themselves, increasing the risk of becoming involved in partner violence for people who are already predisposed. Some researchers point to features with genetic components shared by parents and children that predispose both to family violence, such as antisocial traits, alcoholism, and impulsivity. Some traits shared by parents and children may not be passed genetically but instead may be learned during childhood, such as violence approval, poor emotion regulation, deficits in social information processing, and hostile attributions about interpersonal relationships.

ITV research typically employs one of three methodologies, with inclusion of control samples varying among studies. First, many researchers examine the rates of violence in the childhoods of adults currently involved in family violence as perpetrators or victims. Alternatively, researchers begin with a sample of adults who experienced violence in their families of origin, then investigate rates of family violence during adulthood. Less commonly, researchers take a sample of children with varying family violence histories and follow them into adulthood. This latter prospective approach avoids reliance on retrospective recall of participants, which can be prone to error and bias. However, prospective studies are costly in terms of money, time, and researcher effort. Typically, retrospective studies result in higher estimates of transmission rates than prospective studies. Use of self-report measures produces much higher rates of violence than reliance on substantiation by government agencies, which can result in large variations in transmission rates.

Joan Kaufman and Edward Zigler illustrated how the same transmission data can be presented in different ways, resulting in substantially different estimates of transmission. For example, using parents’ abuse histories as the starting point, a 1979 study by Rosemary Hunter and Nancy Kilstrom found an 18% rate of transmission; that is, of parents with an abuse history, only 18% abused their own infants. If current abuse had been the starting point instead, these same data would have shown a 90% transmission rate because 9 of the 10 parents currently abusing their infants had been maltreated as children.

Transmission of Partner Violence

Sandra Stith and colleagues conducted a meta-analysis of marital ITV, combining the results of 39 separate studies. They found an average correlation of r = .18 between witnessing partner violence as a child and perpetrating partner violence as an adult; this link was stronger for men (r = .21) than women (r = .11), indicating that boys who witness partner violence are more likely than girls to become perpetrators. There was a small correlation (r = .14) between witnessing partner violence and becoming the victim of partner violence as an adult; this link was stronger for women (r = .18) than men (r = .09), suggesting that girls who witness partner violence are more likely than boys to become victims. Although not included in that meta-analysis, evidence from other studies is mixed as to whether adults resemble their same-sex parent more than their opposite-sex parent in terms of violence perpetration and victimization. Because the base rate of violence is higher in dating relationships than in marriages, the transmission rate may be somewhat higher as well, but there have been no meta-analyses to date that compare dating and married couples. Among the factors that can increase likelihood of partner ITV are antisocial behavior, receipt of harsh parenting during childhood, experiencing abuse as a child, depression, substance abuse, attitudes condoning violence, and general relationship conflict.

Transmission of Child Physical Abuse

Kaufman and Zigler estimated that one third of abused children grow up to become abusive parents. Subsequent studies have found both higher and lower rates ranging from less than 10% to more than 40%, depending on factors such as study samples, methodology, and definitions of abuse and violence. Regardless of exact rates, the bulk of the literature is clear that having a history of child abuse consistently increases the likelihood of later perpetration of child abuse, but the majority of people abused as children do not go on to maltreat their own children. Researchers have looked for factors that cause some parents to break the cycle of violence and others to continue the cycle. Several protective factors have been found to decrease the likelihood of violence transmission, such as stable relationships, nonviolent partners, receipt of emotional support, involvement in psychotherapy, and stable home environments. Risk factors that increase the likelihood of violence transmission include young parental age, mental illness including depression and posttraumatic stress disorder, substance abuse, child illness or disability, poor parenting, financial stress, and other forms of violence in the home. In addition, children who experience more severe abuse, more frequent acts, and more injuries are more likely to go on to abuse their own children.

Transmission of Child Sexual Abuse

Because most perpetrators of sexual abuse are men, transmission studies have focused on men as perpetrators and women as mothers of sexually abused children. The largest longitudinal study following sexually abused boys into adulthood found that less than 12% became perpetrators of sexual abuse against children (most of the victims were outside their families). Looking retrospectively at known child sexual abusers, studies have found an average of 28% were sexually abused as children. One of the largest risk factors that appears to increase risk of transmission is exposure to other forms of family violence. In terms of female victims, there is a higher rate of sexual abuse among children of sexually abused mothers than those of nonabused mothers. Contact with the mother’s abuser appears to increase children’s risk of being sexually abused, indicating that in many families, the same person may be responsible for transmission across generations.

Multiple Forms of Family Violence

The 2000 meta-analysis by Stith and colleagues found that witnessing partner violence and experiencing child abuse in the family of origin had similar impacts on subsequent adult partner violence. In terms of child physical abuse and sexual abuse, extant research indicates that all forms of family violence do appear to lead to some increase in child maltreatment in the next generation. The extent of the increased risk is modest, however, and there are numerous factors that can increase and decrease the likelihood of ITV.

Transmission Over Time

Although a history of family violence is one of the greatest risk factors for perpetration and victimization as an adult, intergenerational transmission is far from certain. The majority of people exposed to violence as children later break the cycle of violence. It should also be noted that the rates of violence found to date may not hold true for future cohorts; as the rates of family violence decline over time, the rates of ITV may also change in future generations.

Next post:

Previous post: