Polygynous marriage can benefit women and children

Dr David Lawson — Evolutionary Anthropologist at the London School of Hygiene and Tropical Medicine — on his controversial research in Tanzania and why development policy needs to be evidence-based.

Your recently published research found there is no evidence that polygynous marriage is a ‘harmful cultural practice’ in Tanzania — Why is this such a pivotal piece of research?

Polygyny is a marriage system whereby men can marry multiple women. Historically over 80% of preindustrial societies permitted polygynous marriage, but today it is most common in sub-Saharan Africa. Polygyny is often labeled a ‘harmful cultural practice’ by development organizations, putting it in the same category as female genital cutting and child marriage. The United Nations, for example, states that polygyny contravenes a woman’s right to equality with men, and has such serious emotional and financial consequences that it ought to be prohibited. It has also been argued that polygyny is harmful at the population level. According to this account, polygyny leads to an excess of unmarried men who resort to antisocial behaviour in a desperate effort to find a marriage partner.

These positions are intuitive, but backed by ethnocentric rhetoric rather than hard evidence. They also fail to explain why, if polygyny is so harmful, is it also so common? So we set out to assess the evidence that polygyny is harmful using data from a study of more than 50 villages in northern Tanzania. Our study is unique because for the first time we were able to look at relationships between polygyny and health both across and within a set of ethnically diverse villages. Tanzania is also a great model system for this research because some ethnic groups are highly polygynous and others strictly monogamous. Our results challenge the view that polygyny is harmful and suggest that, at least in some contexts, it may serve the best interests of women and children.

How did you come to this conclusion?

First, we observed that the villages with the highest levels of polygyny also had the lowest food security and worst child health. For example, in Maasai villages up to around 40% of households were polygynous and in these villages nearly 60% of children are physically stunted, which indicates chronic malnutrition. In other villages and ethnic groups monogamy is the norm and food security and health is superior. Superficially, this pattern of results supports the idea that polygyny is harmful.

But then we took things down a level, contrasting monogamous and polygynous households within each individual community. Here our results show a very different picture. Polygynous households either did just as well or had better outcomes when compared to monogamous households within the same village. Results varied by ethnic group. For example, in the Sukuma, the most common ethnic group in Tanzania, male-headed polygynous households did particularly well with high food security and relatively healthy children. While in the Maasai, monogamous and polygynous households did equally poorly regardless of whether the household head was in a monogamous or polygynous marriage. For all ethnicities, polygynous households were also wealthier, owning more livestock and farming more land than their monogamous counterparts. From these findings we concluded the costs of sharing a husband are offset by the greater wealth of polygynous households.

Finally, we reconsidered why in the first stage of our analysis highly polygynous villages had the worse outcomes, despite the opposite pattern within villages. This time we included statistical adjustments for village differences in rainfall and access to education. Once these factors were taken into account a high prevalence of polygyny no longer predicts poor health outcomes. In fact, in the most polygynous villages children tended to weigh slightly more, suggesting they were more resilient to acute food shortages, a common occurrence in this region of Tanzania. Overall, we found no support for the idea that polygyny is harmful either at the individual or group level.


What can your findings tell us about why some cultures practice polygynous marriage?

Our results are consistent with one theory of polygyny in evolutionary anthropology, originally applied to animal mating systems — the ‘polygyny threshold model’.

Basically, this theory argues that sharing a husband can be in a woman’s strategic interest in settings where women primarily access resources via marriage and when men vary in how much wealth they have to offer. In these contexts, polygyny may be the preferred option for a woman and her family by providing equal or greater access to resources than monogamous marriage.

Of course, it remains possible that polygyny is not always in a woman’s interest, even within the sample of villages we considered. Our data can only tell us about the broad trend. In some cases, polygyny may be better understood as resulting from sexual conflict between male and female strategies, with men benefiting at the cost to individual wives and children. For example, research by the anthropologist Beverly Strassmann is more consistent with this alternative. She found children of polygynously married mothers did particularly poorly in a small-scale study in rural Mali. Polygyny itself is a diverse institution with a lot of cultural variation in associated norms of spousal recruitment, divorce, residence and resource sharing. So there is no reason to expect a universal explanation for polygynous marriage that fits all contexts.

Our findings also cast some doubt on the idea that transitions to socially imposed monogamy are driven by group-level costs of polygyny in a process evolutionary anthropologists describe as ‘cultural group selection’. After all, once differences in rainfall and access to education were taken into account, children in relatively polygynous households actually did relatively well. Instead, we argue that monogamy in this setting is becoming more common as part of a transition to having smaller families and investing more in each child. Unquestionably, the missionary movement in Tanzania has also influenced the decline of polygyny in some areas.


How does a polygynous household work in Tanzania and how common are they in comparison to monogamous households?

One in four married women living in rural Tanzania have at least one co-wife, so it is very common. But this proportion varies depending on ethnic group — of which Tanzania has over 100. The custom is also in a constant state of flux, with influences from Christianity, Islam and economic development. Typically, wives are added to a marriage sequentially, are not closely related and each co-wife lives in a separate, but often adjacent dwelling. Husbands may live permanently with one wife, or split their time between them. In other cultures, polygyny is sororal — co-wives are more typically close relatives and live together. Usually, there is an expectation that husbands will treat all wives fairly, if not necessarily equally. Of course what is fair treatment is culturally subjective.


What notions do you challenge in your research?

Our research challenges the idea that polygyny should be considered universally harmful. Within the confines of low female resource control, it may actually benefit some women, and their children, to share a husband. Prohibiting polygyny, at least without addressing the underlying subordinate cultural status of women, could be detrimental rather than beneficial for women and children. Hopefully our findings will encourage more culturally sensitive approaches to polygyny and related purportedly harmful cultural practices.

Our work also makes an important point about the nature of data used to build policy. Demonstrating that trends in large-scale aggregated datasets can mask true relationships within communities, we challenge the development sector’s reliance on large-scale national surveys to guide policy. This adds ammunition to the calls of anthropologists arguing that if development policy is to be successful it needs to do more to take into account context dependency.


Your study concerns food security and child health. But what about the wider potential of polygyny to cause harm?

This is a good question. Sharing a husband might influence other aspects of physical and mental wellbeing. But again we need to make sure that our conclusions are based on evidence. Recent studies counter simple intuition. For example, polygyny is associated with lower HIV prevalence at both national and regions levels across Africa. Work by Georges Reniers, also at the London School of Hygiene and Tropical Medicine, suggests that polygyny increases individual exposure to sexually transmitted infections, but selective recruitment of HIV-positive women into polygynous marriages where coital frequency is lower isolates transmission risks from the wider population. So the picture is complex. A recent study in Tanzania also found no link between polygyny and maternal mental health.


Can you tell us about a typical day on the project? How long did it take and how were you involved?

Like many research projects it was a massive collaborative effort, involving an international team of researchers and many years of work. I first got involved through discussions with another anthropologist involved in the research, Monique Borgerhoff Mulder at the University of California, Davis. Monique was one of the first anthropologists to study polygyny from an evolutionary perspective, transferring theoretical models from animal reproduction to human marriage systems. We decided to collaborate, and Monique brought me into contact with Savannas Forever Tanzania: Data for African Development, a Tanzanian based non-governmental organization (NGO).

Savannas Forever collected the data over a two-year period as part of a bigger project called the ‘Whole Village Project’. Teams of fieldworkers would spend several days in each village, and with the approval of village elders, survey each household and physically measure the weights and heights of children. One of the many remarkable things about Savannas Forever is their commitment to partnering with researchers like myself to really make use of the data they collect. They also presented basic summaries of the data they collect back to communities so that they can use this information to decide how best to combat the problems they face. This approach to data sharing is really unusual.

Over the last three years I have travelled to Tanzania to visit the study villages and talk with the field team to make decisions about how best to analyze the data. The next stage of the research was then presenting our work to other anthropologists and population scientists. Amazingly, I have now presented aspects of this research not only in Tanzania and London, but also in Finland, Alberta, California and New Mexico. All of this helped to refine our study. Finally, we submitted the paper for publication and after a round of revision it was accepted. It’s a long drawn out process, but its how science at this scale works.


What’s your story?

I’m a lecturer at the London School of Hygiene and Tropical Medicine, a world-leading center for public and global health research. It’s a really exciting place to work — all staff are research active, the student body is international, and there is a constant footfall of visiting academics presenting their latest findings. Our director, Professor Peter Piot, was part of the research team that discovered Ebola in the 1970s and many of our staff are involved in efforts to deal with aftermath of the recent outbreak. However, I’m a very unusual staff member at ‘The School’. This is because my expertise is in evolutionary anthropology, a field of research not usually associated with public health.

I studied biology at University College London, but nothing really clicked in place academically until I took a course in Behavioural Ecology. Behavioural ecology is the study of behaviour from an adaptive perspective. It views the behaviour of all species as strategic responses to ecological variation and the product of evolution. I quickly became fascinated by the application of this theoretical framework to human behaviour.

It took three attempts after my undergraduate degree to get funding for a PhD. I filled my time doing a Masters at the University of Liverpool. In between shifts at a bookshop, I also worked as a research assistant on two projects. One analyzing data from an intervention to increase handwashing with soap in Ghana, and another collecting saliva samples for a study of testosterone levels in Bangladeshi immigrants to London. At the time these projects seemed a bit random, but looking back they clearly shaped my latent interests in human wellbeing. When I finally got funding for a PhD in anthropology back at University College London, I was really lucky to be part of an extraordinary cohort of PhD students and post-docs, all of which have gone on to do amazing things and remain part of my professional network today.

I moved over to the London School of Hygiene in 2013 after being awarded a Research Fellowship from the UK Medical Research Council and Department for International Development. This provided funding to look at family structure and child wellbeing in Tanzania. Family structure in Tanzania, like many African nations, has a number of features generally believed to contribute to its burden of poor child health, such as high fertility, early childbearing and a high frequency of polygynous marriage. My objective has been to link up applied research that considers the health implications of alternative family structures with the more academic literature in evolutionary anthropology. This is achieved by asking the big questions about why we observe cultural variation in family structure in the first place.


What sparked your interest in anthropology?

Evolutionary anthropology extends the principles of behavioural ecology to human behaviour, with a few extra bows and whistles to account for the dynamics of cultural transmission and inheritance. So the progression from studying animal behavioural ecology to anthropology was a natural one. A background in animal behaviour is a blessing for an anthropologist. There is a knee-jerk reaction for social scientists to see evolutionary explanations as reductionist and simplistic, and therefore irrelevant when faced with the complexity of human culture. But when you start by observing some of the seemingly bonkers things that animals do, and the only available explanatory framework at your disposal is evolution by natural selection, you quickly accept that complexity can arise from deceptively simple processes. The fun part is working out how.


What advice would you give to someone keen to do your job?

Develop a thick skin and learn to love criticism. It smarts at first, but you quickly learn the best working relationships are just like the best friendships — you need to be able to trust your colleagues to tell you when you are making sense or talking nonsense, not just pat you on the back for putting the effort in. Also it is important to read beyond your own disciplinary boundaries — you are unlikely to make new insights by only looking inwards.

Other than that, nothing has been more instrumental than various mentors I’ve worked with over the years, from my PhD to today. Building these relationships is key. Good mentors point out your errors and help you recognize your own strengths. Academic progress only happens by standing on the shoulders of giants. For whatever reason, my mentors have almost exclusively been women. There is a lot of focus on the leaky pipeline in academia right now, so I’m proud to work in a field with so many great role models for both sexes.


What do you read, watch or listen to?

At the moment work is pretty all consuming. I’m submitting a grant proposal and then heading straight off to Tanzania to start data collection on a new project. When I am working I can only handle instrumental music — so it’s a lot of Trent Reznor soundtracks at the moment.

Workwise, the most influential books I’ve read in recent years include Poor Economics by Banerjee and Duflo, and Aid on the Edge of Chaos by Ramalingam. Both make a powerful case for the need to guide policy on the basis of evidence, not just good intentions. They also highlight that for interventions to be successful we need to first understand the existing adaptive strategies that disadvantaged people use to mitigate risks and safeguard their own longevity prior to external intervention. Reading these books, its impossible not to see the potential of evolutionary anthropology to inform how the international development approaches cultural diversity.

Fiction wise, I recently read and fell in love with Annihilation by Jeff Vandermeer. It’s about four female researchers — a biologist, a psychologist, a surveyor and a (cultural) anthropologist, all sent to investigate a quarantined area of a tropical jungle. But they all have amnesia, are perceiving different realities, and something wicked is lurking in the bushes. Predictably the anthropologist is a bit useless and doesn’t last very long. The biologist, however, well… she gets more of handle on things.

Further Reading:

Polygyny paper : Lawson DW, James S, Ngadaya E, Ngowi B, Mfinanga SGM, Borgerhoff Mulder M. 2015. No evidence that polygynous marriage is a harmful cultural practice in northern Tanzania. Proceedings of the National Academy of Sciences of the United States of America.

Applied Evolutionary Anthropology: Gibson MA & Lawson DW. 2015. Applying Evolutionary Anthropology. Evolutionary Anthropology: Issues, News, and Reviews 24(1): 3–14.

Poor Economics: Banerjee A & Duflo E. 2011. Poor economics: A radical rethinking of the way to fight global poverty.

Aid of Edge of Chaos: Ramalingam, B. 2013. Aid on the edge of chaos: rethinking international cooperation in a complex world.

Annihilation: Vandermeer J. 2014. Annihilation.