The Girls Who Broke Your Heart Thread

"Sometimes I wonder if men and women really suit each other," Katharine Hepburn once said. "Perhaps they should live next door and just visit now and then." Despite the famed actress's remarks, human males and females do have a strong tendency to live together in monogamous pairs, albeit for highly varied periods of time and degrees of fidelity. Just how such behavior arose has been the topic of much debate among researchers. A new study comes to a startling conclusion: Among primates, including perhaps humans, monogamy evolved because it protected infants from being killed by rival males.

Living in pairs, what researchers call social monogamy, has repeatedly evolved among animals, although in widely varying proportions among different groups. Thus, about 90% of bird species are socially monogamous, probably because incubating eggs and feeding hatchlings is a full-time job that requires both parents. But in mammals, females carry the babies inside their bodies and are solely responsible for providing milk to young infants-and only about 5% of species are socially monogamous. That leaves most mammalian males free to run around and impregnate other females. Primates, however, seem to be a special case: About 27% of primate species are socially monogamous; and recent studies by Christopher Opie, an anthropologist at University College London, and his colleagues have concluded that social monogamy arose relatively late in primate evolution, only about 16 million years ago. (The earliest primates date back to about 55 million years.)

But why did social monogamy arise at all among mammals, including primates, given the many reproductive advantages to males having access to as many females as possible? Scientists have proposed three major hypotheses: Monogamy provides more effective parental care for infants, as in birds; it prevents females from mating with rival males, especially in species where females are widely spaced and cannot all be easily monopolized by one male; or it protects against the risk of infanticide, which is very high among some primate species, including chimpanzees and gorillas, and is often explained by the desire of a rival male to quickly return a mother to a fertile state so that he can sire his own offspring. Some researchers think that a combination of all three factors, and perhaps still others, provide the best explanation for monogamy.

Resolving this debate is important, researchers say, especially for understanding the evolution of human mating behavior. Although humans aren't completely monogamous, "the emergence of pair-bonding in humans was a major evolutionary transition, which dramatically altered the evolutionary trajectory of our species," says Sergey Gavrilets, an evolutionary biologist at the University of Tennessee, Knoxville. Many researchers think that we could not have evolved our large brains without joint parental care during the extended period of helplessness required for infant brains to grow to their full size. "Understanding the forces that drove that transition can help us better understand the causes of human uniqueness," Gavrilets adds.

Opie and his colleagues set about testing the three leading hypotheses in primates using a powerful method called Bayesian statistics. The team used previously published genetic and behavioral data from 230 primate species, representing nearly all known species such as Old and New World monkeys, lemurs, and apes, employing strict criteria to ensure that the data were reliable. For example, the team concluded that a particular species engaged in infanticide only if at least 20 publications reported the killing of infants through direct observation or as the only possible explanation. The Bayesian approach allowed the researchers to map information about primate behavior onto an evolutionary tree of the entire animal group, and thus analyze the order in which traits such as social monogamy, infanticide, and other behaviors arose over time.

As the team reports online today in the Proceedings of the National Academy of Sciences, there was a strong correlation in time between all three hypotheses-parental care, female range, and infanticide by males-and the rise of social monogamy in the roughly 60 primate species that live in pairs. However, among the three explanations, only infanticide actually preceded social monogamy in time and thus could be a driving evolutionary force, the team concludes; the other two behaviors occurred afterward and were the consequences of social monogamy and not the causes. "Our analyses clearly show that infanticide is the trigger for monogamy in primates," and likely was the trigger in humans, too, Opie says.

Why is the incidence of social monogamy in primates, 27%, so much higher than the 5% in mammals as a whole? Opie and his colleagues, who include University of Oxford psychologist Robin Dunbar-a proponent of the idea that the complex social groups typical of primates led to bigger brains-have an answer. Because the infants of primates with large brains, especially apes and humans, are helpless for longer periods of time than other mammals, they are much more vulnerable to infanticide, and thus need more protection.

Nevertheless, the reaction to the study has been mixed. "I found the paper quite convincing," says Carel van Schaik, a primatologist at the University of Zurich in Switzerland who had earlier argued for the infanticide hypothesis. "The results are very solid" for all primates. But van Schaik says that he "would be very careful to conclude from this paper that infanticide risk was also the main factor underlying human monogamy," in part because humans are not fully monogamous, as shown by studies of cultures around the world. "The current monogamy is socially imposed."

Phyllis Lee, a psychologist at the University of Stirling in the United Kingdom, agrees. "At best we engage in forms of serial monogamy," she says, pointing out that more than 60% of "traditional societies" allow men to have more than one wife. Lee adds that infanticide is a feature of many primate species that are not monogamous, "so monogamy is not the only evolutionary solution to infanticide."

Indeed, a paper to be published this week in Science looks at monogamy across all mammals and comes to a very different conclusion. Zoologists Tim Clutton-Brock and Dieter Lukas of the University of Cambridge in the United Kingdom analyzed monogamy among 2545 nonhuman mammal species. In contrast to Opie's conclusion in primates, they find in this larger sample that social monogamy arose among species where females were widely spaced and males could not monopolize several of them at once; infanticide did not seem to be a driver for monogamy among all mammals. Opie counters that wide spacing among females doesn't apply to highly social, group-living primates, so that humans, and perhaps all primates, may be unusual among mammals. If so, he says, looking at mammals across the board might mask the special features of primate evolution.

All the same, Petr Komers, an ecologist at the University of Calgary in Canada and leading proponent of the female range hypothesis for social monogamy, says he finds the authors' conclusions that infanticide was "the only possible driver to monogamy a bit surprising." Komers's own studies, like Clutton-Brock's, found that among mammals, the highest correlation was between social monogamy and species whose females stayed put in limited ranges. "Monogamy does evolve in species where infanticide is unlikely," Komers notes, such as in ungulates, or hoofed mammals. Thus no one factor is the "silver bullet" driving monogamy, Komers says, and researchers should be looking for an interplay of multiple explanations.