Domestic violence may increase family size

A study in the journal Nature Human Behaviour finds that women in a remote Amazon Basin culture who are brutalised by their partners tend to give birth to more children compared to their unmolested peers — a finding that has been described as an “important empirical test of the potential adaptive function of intimate partner violence”.

The study, led by Jonathan Stieglitz of the Institute for Advanced Study, Toulouse, France, mapped the reproductive patterns of 105 women in Bolivia’s Tsimané culture across an average of 18 years of marriage.

The results showed that, with the exception of extremely young brides, women who were victims of intimate partner violence (IPV) in the previous year had about a 10 to 15% higher chance of having a child compared to those who had not recently been abused, with the disparity rising with age.  

The findings seem sure to rekindle a debate that holds rape to be a deeply rooted evolutionary adaptation geared to increasing male reproductive success. That argument was made in 2000 by Randy Thornhill and Craig Palmer in their highly controversial book A Natural History of Rape: Biological Bases of Sexual Coercion, which was well received by some scientists and robustly and loudly contested by others, including many feminists.

That said, Stieglitz, who describes himself as a human ecologist rather than an evolutionary biologist, is adamant that it is “wrong and misguided” to interpret his finding in this framework. In fact, he says, there’s no indication his findings have anything to do with rape. They may simply reveal IPV as a way some men use to win an argument over how many kids to have.

The Tsimané are a subsistence culture in Bolivian lowlands east of the Andes. One reason they were chosen for the study was that they have an extremely high birth rate — an average of nine children per woman. That means that even though the study only involved 105 adult females, the researchers were able to track them through 740 births.

Unfortunately, their rate of marital violence — defined as any deliberate physical, sexual, or psychological harm — is high, with a full 85% of the women reporting at least one incident of IPV in their lifetimes, and many reporting an average of one per year. That’s why Stieglitz’s team looked at births per year, rather than each woman’s total number of children. That way, they expanded a data set of 105 women to 1905 total years of marriage, with 740 births.

Intriguingly, Stieglitz says, in this non-industrialised culture there was no sign that IPV was related to whether or not the men had been exposed to family violence during their childhoods … or to violent media. Nor was their IPV related to their level of aggressiveness toward other men.

In addition, the Tsimané civilisation is historically peaceful and has no strong history of male social dominance.

Instead, Stieglitz says, there is a disparity among how many children husbands and wives want, with wives, not surprisingly, wanting fewer kids than their husbands.

It is this marital conflict, he suggests, that probably drives the finding, with husbands resorting to domestic violence in order to assert their wills and get what they want — not necessarily by raping their spouses, but by coercing them to do as their husbands want. In fact, the study found, husbands who believed they should exert control over their wives were twice as likely to engage in spousal abuse.

For those who want to interpret this as a sign that evolution has placed male violence in our genes, Stieglitz has several important caveats. First, he says, this study looked only at the frequency of births, not the health and subsequent reproductive success of the children.

“There is evidence,” he says, “that closer spacing between births can lead to poor health outcomes for children, which may influence the children’s later reproduction.”

Also, he notes, in some cultures, abusers could get their genes culled from the evolutionary herd due to “early death” from “violent reprisals” by an abused wife’s family.

In a second paper in the same issue of the journal, Elizabeth Pillsworth, an anthropologist at California State University, Fullerton, calls Stieglitz’s result an “important advance” in understanding the underlying causes of marital violence, adding: “This suggests that the credible threat of violence, based on experience, may be sufficient to manipulate women’s sexual or other behaviours in marriage.”

Leslie Cannold, an ethicist at Griffith University in Australia, adds that increasing social mobility may help women reduce such problems by allowing them to choose partners who do not resort to violence to force them to accede to their wills.

“Physical coercion is the oldest and most successful strategy humans have to get what they want,” she says. In societies where women can respond by choosing partners who don’t do this, she suggests, men might themselves respond by altering their own behaviour.

Stieglitz says that his findings could apply to many things other than the number of children a couple has, including disputes over chores, money, and other household issues, any of which might be a source of conflict that might tempt men to use violence to assert control. “Identifying where spousal interests diverge, is a necessary first step,” he says.

Interventions, he adds, start with trying to change societal attitudes.

But other, more subtle, interventions might also be useful.

“In prior work we have found that intimate partner violence is less common when a wife’s kin live nearby, he says, “so extra support may be necessary to reduce isolation of wives living farther from their families.”

Meanwhile, Cannold says, even if the study is viewed in evolutionary terms, it doesn’t mean male violence is hard-wired and unchangeable.

“Conservatives love this sort of explanation, which gets men off the hook,” she says. “However, this misunderstands the key concept of evolution, which is about changes in response to environmental inputs.” If women change the type of partners they want, she says, these changes can be “quick.”