Studying intimate partner violence in a small-scale society
Violence against women is a pernicious public health problem that destroys lives all over the world. The World Health Organization reports that more than 1 in 3 women face violence from an intimate partner during their lifetime, and nearly 1 in 4 murders of women worldwide are by intimate partners. Developing effective interventions to reduce intimate partner violence requires understanding the underlying risk factors for such violence. In our study appearing in Nature Human Behaviour, we investigated some of the factors associated with intimate partner violence; we focused on how early exposures to family violence, attitudes about aggression, and decision-making authority in marriage impact women’s risk of experiencing physical abuse. We were motivated to conduct this study by prior research among Tsimane, which indicates that, on average, husbands want more children than their wives, and that wives have more children than they themselves think are ideal. While, in general, wives’ own desired family size better predicts how many children they end up having than husbands’ preferences, we sought to explore whether some men in this particular high fertility context might resort to intimate partner violence to increase their lifetime number of children.
What this study reports:
We find that intimate partner violence is associated with higher marital fertility controlling for potential confounders, and that this result holds whether we are comparing across all women, or if we examine the same women over multiple marriages, where one husband may have been abusive but the other was not. The finding is also robust to other checks that might have led to a spurious correlation between intimate partner violence and fertility. Childhood exposure to family violence is not associated with experiencing or perpetrating intimate partner violence; this finding contrasts with those from industrialized populations, where a pernicious cycle of violence often perpetuates from one generation to the next. Instead, in our study, we find that higher levels of partner violence are associated with men’s attitudes about controlling their spouse, but not with physical aggression towards other men.
What this study does not say:
This study did not explore how intimate partner violence impacts fertility, either in the short or longer term, nor did we study sexual or emotional violence (we only studied physical violence). We also did not assess the impact of intimate partner violence on the health of children of abused women, nor did we assess the children’s future prospects, which could be reduced. It would be wrong and misguided to say that evolution favors intimate partner violence, or that spousal abuse is adaptive. In some contexts, intimate partner violence could instead result in lower fertility and poorer child health, and the early death of a perpetrator due to violent reprisals from the victim’s family. The prevalence of intimate partner violence varies widely across countries, and can change rapidly over time within the same country. Evolutionary dynamics surrounding intimate partner violence remain unclear for various reasons. For example, from our interviews of women, it was clear that refined alcohol consumption was involved in many instances of violence, and one might conjecture that rates of intimate partner violence were lower in the past, without access to such alcohol. In addition, while violence is part of a broader behavioural repertoire that some men use to achieve desired goals that under some circumstances may result in having a larger family, this in no way implies that intimate partner violence is universally adaptive. Requirements for showing that a trait was and is favoured by natural selection are much more rigorous than what we do in this observational case study.
This case study highlights the fact that, though marriage is fundamentally cooperative in so many ways, there is much room for spousal conflict given the intense coordination required to raise a family and make a marriage work. Interests of husbands and wives may differ in the context of reproductive decision-making and other domains of family life. Identifying where spousal interests diverge -- whether over ideal number of children, how to divide household tasks, or how to spend money, is a necessary first step for designing effective interventions. Targeted interventions can help resolve these spousal conflicts of interest. At a broader level, changing individual and societal attitudes and expectations regarding power dynamics within intimate relationships will help reduce partner violence. Interventions must go beyond the couple itself and should include the community at large, to help shape public opinion. In prior work we have found that intimate partner violence is less common when a wife’s kin live nearby, and so extra support may be necessary to reduce isolation of wives living farther from their families.
More work is needed to better understand the causes and effects of intimate partner violence in rapidly changing indigenous populations like the Tsimane. For example, how are different forms of violence viewed by spouses and their families? In this study we found that half of all violent incidents occurred in the first five years of marriage, and so what determines how women cope in these early years, and what changes with marital duration? How might intimate partner violence change during current periods of rapid market integration? On one hand, greater access to schooling and the market economy offers the potential to empower women. On the other hand, access to refined alcohol, fungible wages, migratory wage labor, and changing expectations about desirable goods and lifestyle can all lead to spousal conflicts over the appropriate use of household resources. Unfortunately, there are few systematic studies of intimate partner violence in remote, marginalized indigenous populations, especially those undergoing socioeconomic change. But if our results reflect, in part, the experience of women in other populations, then spousal violence may be even more widespread than the frightening statistics already reveal.
The paper in Nature Human Behaviour is here: go.nature.com/2ARG0eu