Article begins

The Ik of northern Uganda have been characterized as greedy and mean, a people whose culture of selfishness developed in response to chronic scarcity. But the relationship between hardship and human generosity is far more complex.

Few ethnographies are as controversial as The Mountain People, Colin Turnbull’s 1972 description of a marginalized African society in a state of anomie. Turnbull’s subjects, the Ik, eke out a living with a mixture of foraging and horticulture in a small mountainous region of northern Uganda. Turnbull reported shocking behaviors, including the desertion of children, mistreatment of the weak and elderly, and family members snatching food from one another. He called the Ik “the loveless people” and described them as “unfriendly, uncharitable, inhospitable and generally mean as any people can be.” Their selfish and amoral behaviors were attributable to a cultural adaptation to conditions of chronic scarcity, he concluded.

Turnbull believed that Ik mothers had little affection for their children because it was such a burden to raise them. He claimed that mothers would habitually throw their infants out to fend for themselves at the age of three, and that they would casually cast their babies down on the ground when they went about their daily work, half hoping that a predator would snatch the infant up. Accounts of Ik motherhood and other tales of “greedy” and “selfish” behavior are presented as illustrations of how the Ik had developed a culture of extreme individualism in response to perpetual hardship and scarcity. Because culture traits survive by spreading from person to person, Turnbull even went so far as to recommend that the Ik people should be rounded up in a military operation, their families broken apart and permanently displaced from their homeland. That, he thought, would eliminate the culture traits that led to the selfishness that he observed.

But our recent work with The Human Generosity Project shows that we should not conflate behavior with culture. To this day, the Ik live in conditions of scarcity, but when times are particularly difficult they report sharing more, not less. And they are just as cooperative as people in other small-scale societies. The relationship between scarcity and selfishness is clearly much more complicated than Turnbull believed it to be.

Credit: Cathryn Townsend
Photograph of two people holding hands outdoors
Ik women join hands during a festive dance.

When times are tough, having a safety net can be critical to survival. One important reason why people give to those in need, often with no expectation of repayment, is that such generosity can contribute to these social safety nets through risk pooling. Keeping the needy person going today makes them available to help you tomorrow, should you be the needy one at that point. This is the idea behind one of our species’ original and most important adaptations: central place food sharing. Evolutionary anthropologist and Human Generosity Project member Colette Berbesque and colleagues describe a pattern of “eat first, share later” among the Hadza of Tanzania: everyone who is able to forage does so, eating along the way but also bringing back to camp any surplus to share with those who were unlucky or unable to forage that day.

Some foragers also create risk-pooling systems that work on longer time scales and across larger areas than day-to-day food sharing. For example, Ju/’hoansi use small, economically insignificant gifts to establish and maintain partnerships in a system known as hxaro. Anthropologist Polly Wiessner has shown that the main function of these gifts is to establish relationships of mutual aid. When food or water shortages at one location make central place food sharing insufficient as a way of dealing with risk, people can ask their hxaro partners in distant locations, some of whom have hopefully fared better, for help.

Similarly, food-producing societies often set up risk-pooling arrangements that work on longer time scales and across larger areas than central place food sharing. Maasai and other Maa-speaking pastoralists in East Africa create risk-pooling relationships that they refer to as osotua (umbilical cord). Osotua partners agree to help each other if one of them loses livestock due to drought, disease, theft, or anything else that is inherently unpredictable. Such gifts are given freely (pesho), from the heart (oltau), and with no expectation of repayment and no creation of debt (esile). What we discovered through interviews and experimental games is that, rather than creating debt or dependency, the purpose of osotua gifts is to keep your partners and your relationships with them alive so that they will be there should you be the one in need someday.

The Ik also pool risk by sharing generously with those in need. Cathryn spent all of 2016 and portions of 2017 and 2018 among the Ik. Lee and Athena also visited the area briefly in 2016. What Cathryn found contrasted sharply with Turnbull’s claims. From her very first encounter with an Ik person (the kindly Lomeri John Mark, who is now deceased), she found the Ik to be a friendly, hospitable, and generous group of people. They suffer harsh environmental conditions and political marginalization with extraordinary grace.

Like most foragers, the Ik have a strong ethic of sharing captured by the common saying, “tomora marang’” (it’s good to share). Sharing is commonplace. If an Ik person is hungry and comes across a neighbor or friend who is eating, it is not considered a social infringement for the hungry person to simply help themselves to a portion of their neighbor’s meal. There is no need to ask; it is understood that one eats when one is hungry. In addition, many Ik believe in Earth spirits called kijawika that monitor people’s behavior, rewarding the generous and punishing the selfish. When Ik are given money and the opportunity to give any portion of it to an anonymous stranger, they are, on average, as generous as people who have played such games in societies around the world. When they are told that the recipient will be a needy person and are reminded about the kijawika, their generosity surges to an average far above what is usually seen in such experiments.

Why, then, did Turnbull observe so much selfishness among the Ik? What he failed to fully appreciate was that while he was conducting fieldwork in the mid-1960s the Ik were experiencing an unusually severe drought and subsequent famine. Their situation was made even more dire by their lack of access to rich hunting grounds that had been made into a national park (a circumstance which he did document). But he appears to have misconstrued famine behavior, including the collapse of normal social relations, as a cultural adaptation to conditions of scarcity. This attribution of allegedly selfish behavior as the outcome of culture was an interpretation based on his failure to comprehend the gravity of the famine that the Ik were experiencing at the time.

When people are starving, their behavior changes, and those changes are not attributable to any culture traits. For example, a study of semi-starvation among conscientious objectors at the University of Minnesota during World War II found that during six months of semi-starvation and a 12-week rehabilitation diet, 36 participants lost over 25 percent of their body weight and experienced such effects as lowered heart rate, anemia, fatigue, apathy, extreme weakness, depression, irritability, quarrelsomeness, social withdrawal, obsession with food, irrational hatred of others, and neurological deficits. Meanwhile, a cross-cultural survey of historical and ethnographic sources by anthropologist Robert Dirks found remarkably uniform patterns of behavior in social responses to food shortages and famines. He found that severe and prolonged famine produces “physiological slowdown” that leads to a reduction of social activities and a refocusing of energy expenditure on the quest for food. However, the early phase of social response to food shortage or famine is a period of alarm in which prosocial activities such as sharing may actually increase.

Risk-pooling systems can help buffer societies from shocks when they happen to people at different times and in different places, but they are vulnerable to collapse when everyone is in need at the same time.

Cultural institutions of risk pooling are designed to deal with limited environmental shocks. The Ik people of today share with their neighbors on a year-round basis with the explicit understanding that they are cultivating relationships that are most valuable in conditions such as particularly long dry seasons or raids by neighboring pastoralists that can empty a household’s granaries. But there is a fatalistic acceptance among the Ik that during severe famines, such as the one that Turnbull observed, many people die and there is not all that much that they can do to change their situation because they lack the resources to do so.

Historian Richard Waller has documented a similar phenomenon among the Maasai during the late nineteenth and early twentieth centuries, a period they refer to as Emutai (the wiping out). A combination of droughts and epidemics among people and livestock led to starvation, a breakdown of social relations, and warfare. But, like the Ik, when conditions improved, Maasai society recovered, complete with osotua risk-pooling relationships. Risk-pooling systems can help buffer societies from shocks when they happen to people at different times and in different places, but they are vulnerable to collapse when everyone is in need at the same time. Nevertheless, risk-pooling systems appear to be resilient even to their own collapse, re-emerging once conditions improve.

Fortunately, risk pooling is not the only way people can deal with unpredictable losses. They can also reduce the severity of those losses by taking such precautions as vaccinating their livestock or, closer to home, wearing seatbelts. People can also take steps to avoid losses entirely. Maasai and other East African pastoralists reduce their losses to carnivores by penning their animals up at night, and prudent drivers know not to venture out when roads are icy. Another common way to deal with risk is simply to be as wealthy as possible, a strategy known as risk retention or self-insurance. Among some foragers, this takes the form of food storage (although storage is not compatible with continuous mobility). Among pastoralists, it takes the form of maintaining herds large enough to support a family even if they sustain severe losses.

How much a group of people relies on these various strategies for risk management—and, thus, how generous they tend to be—depends not only on their physical environment but also on how they interact with it. Central African hunter-gatherers, who live in a tropical forest environment with relatively reliable returns to hunting and gathering, practice central place food sharing and have just as much of a reputation for sharing as do Ju/’hoansi and other Khoisan-speaking hunter-gatherers who live in the Kalahari Desert. And, although they live in the same environments, Central African hunter-gatherers have stronger traditions of sharing than their farming neighbors, who tend to focus on trade, exchange, and storage of food for the exclusive use of individual households.

Because money is a form of wealth that can easily be stored and saved, incorporation into the global market economy can also increase the extent to which people self-insure and reduce the extent to which they pool risk with others, thus reducing their generosity despite their increased wealth. Geographers Timothy D. Baird and Clark Gray have found that as Maasai households’ dependence on pastoralism decreases due to market integration they put less effort into restocking other people’s herds, and they give and receive gifts with other households less frequently. Similarly, researching cattle productivity and pastoral livelihoods among Fulani in Nigeria, veterinary epidemiologist Ayodele O. Majekodunmi has shown that risk pooling has declined and patron-client relationships have become more common as some households have become wealthy while others have remained poor.

Research in industrialized societies about whether wealth and poverty are associated with selfishness and generosity has yielded a mixed bag, perhaps reflecting the complexity of the relationships among the variables in question. For example, Angelos Stamos and colleagues found no significant relationship between prosocial behavior and socioeconomic status in an economic game and in attitudes toward charitable giving. Other studies have found both negative and positive correlations between socioeconomic status and generosity, donations to charity, trust, and helpfulness. Economists Jonathan Meer and Benjamin Priday found that the percent of income Americans give to charities tends to be flat across income levels up to a threshold of $500,000, after which it increases, despite the fact that marginal tax rates also increase as income increases. Yet charitable giving is not the only form that generosity can take. Synthesizing psychological research on social class and prosocial behavior, Paul Piff and Angela Robinson present evidence that lower socioeconomic status individuals are more likely to engage in other-beneficial prosocial behaviors whereas higher socioeconomic status individuals are more prone to engage in self-beneficial behaviors, but they also suggest that this may depend on such factors as motivation, identity, and inequality.

How might these factors influence the relationship between prosocial behavior and socioeconomic status? Lower socioeconomic status individuals might be more generous and concerned with the needs of others because they are themselves more often exposed to unpredictable risks. For them, building and sustaining networks of interdependence and mutual aid may be the best coping strategy. They are concerned with the welfare of others in their networks because they have interests in common. Given that prosocial behavior among lower socioeconomic status individuals is often a strategy for building and sustaining mutual aid networks, those with lower socioeconomic status may be more inclined to help people who have a similar social identity or people who are likely to become cooperative partners.

While scarcity that is extreme, constant, and widespread can indeed lead to selfishness, scarcity that is sporadic, unpredictable, and unevenly distributed can have the opposite effect.

Higher socioeconomic status individuals, on the other hand, can use their wealth to buffer themselves from risk. They may be less reliant on mutual aid networks and, therefore, prosocial behavior on their part might be more motivated by a desire to boost their status. Some research has shown that residing in a location with a high level of economic inequality appears to reduce generosity among higher socioeconomic status individuals. Comparing 25 countries, Dutch economists Johan Graafland and Bjorn Lous found that income inequality increases life satisfaction inequality and that both income inequality and life satisfaction inequality have a negative impact on social trust. High levels of economic inequality could be impairing the capacity of lower socioeconomic status individuals to build and maintain networks of mutual aid by undermining their sense of trust in others. But a socioeconomic class in a rich and industrialized society is not directly comparable with an entire ethnic group of hunter-gatherers or pastoralists who may live in scarce or relatively abundant environments at the margins of state-level societies. The kinship patterns, rituals, and other cooperative culture traits that we would expect to find in the latter are adaptations that serve to create extensive networks of mutual aid.

What, then, of Turnbull’s claim that the selfishness he observed among the Ik was due to their culture? As we have seen, Ik culture actually includes strong admonitions and incentives in favor of generosity. What Turnbull failed to realize, it seems, is that culture is not the only influence on human behavior. When people are starving, they may toss their generous culture traits aside in a desperate effort to survive. But when hardships occur to people at different times and in different places, the fortunate often step up and help those in need. So, while scarcity that is extreme, constant, and widespread can indeed lead to selfishness, scarcity that is sporadic, unpredictable, and unevenly distributed can have the opposite effect. And, while the wealthy may have greater ability to be generous than the poor, wealthy people also have an easier time self-insuring and so can afford to disengage from sharing relationships. Rising levels of economic inequality in the modern world complicate the situation even further.

Ever since the days of scholars Thomas Huxley and Peter Kropotkin, there has been a tug-of-war in evolutionary theory between an emphasis on competition and one on cooperation. Because both are clearly important, both must be studied. Sometimes, as when resources are especially scarce, competition can get the upper hand. But, as we have seen among the Ik, Maasai, and many other peoples, one of our species’ original and most successful adaptations has been to help those in need so that they may do the same for you someday. Of course, this is not actually a recent discovery. Inspired by what he observed among the Nuer, anthropologist E. E. Evans-Pritchard captured the idea quite succinctly more than 80 years ago: “It is scarcity not sufficiency that makes people generous, since everybody is thereby insured against hunger.”


Cathryn Townsend

Cathryn Townsend is a research fellow in the Department of Anthropology at Baylor University in Waco, Texas. Her research is on egalitarian political systems and the relationship between cooperation and adversity. She works with African hunter-gatherers and Dominican horticulturalists.

Athena Aktipis

Athena Aktipis is an associate professor and director of the Interdisciplinary Cooperation Initiative at Arizona State University. She works on cooperation across systems from human sharing to cancer biology. She is codirector of The Human Generosity Project and The Cooperation Science Network. Her forthcoming article, “Design Principles for Risk-Pooling Systems,” coauthored with Lee Cronk will publish in Nature Human Behaviour.

Lee Cronk

Lee Cronk is a human behavioral ecologist in the Department of Anthropology at Rutgers University in New Brunswick, New Jersey. His current research focuses on cooperation. He has conducted fieldwork in Kenya, Honduras, and the American Southwest. He is codirector of The Human Generosity Project and The Cooperation Science Network. His forthcoming article, “Design Principles for Risk-Pooling Systems,” coauthored with Athena Aktipis will publish in Nature Human Behaviour.

Cite as

Townsend, Cathryn, Athena Aktipis, and Lee Cronk. 2021. “Does Scarcity Lead to Selfishness?.” Anthropology News website, June 23, 2021.