Constrained by bearing costly young that mothers nurture by themselves, gorillas, chimpanzees, and orangutans breed more slowly. Orangutans hold the record, with intervals between births as long as eight years. Across the Great Apes, the average is closer to six years. Once weaned, these offspring provision themselves. But human children, born even more helpless than other apes, also mature more slowly and remain dependent far, far longer. When the anthropologist Hillard Kaplan surveyed the literature from every foraging society for which he could find quantitative data, he calculated that it takes roughly 13 million calories to rear a human baby from birth to nutritional independence at around age 18 or older. The anthropologist Karen Kramer has come up with similar estimates for a Maya horticultural society. Long before her first child was self-sufficient, the Mayan mother typically bore another.91 Even though human babies are unusually fat at birth (three times fatter than expected for a mammal of their size) and take far longer to become nutritionally independent, hunter-gatherer mothers routinely produce them at three- to four-year intervals, almost twice as fast as the six- to eight-year intervals typical of other apes.92 Such hyperfertility would have been feasible only if mothers in ancestral populations had been able to count on alloparental assistance.
A simple comparison between primates with and without assistance reveals a clear pattern. In species with shared care (that is, help carrying infants but no provisioning), infants still grow faster and their mothers breed again after shorter intervals.93 Presumably this is because mothers save energy, are free to forage more efficiently, and are better fed themselves.94 On average, mothers with help wean their young at an earlier age and conceive again sooner. Provided it was safe to turn their infants over to another group member, mothers with willing caregivers in their group breed faster and consequently produce more offspring who reach reproductive age.
From this broad comparative perspective, some curious demographic patterns in the Primate order start to make sense. One reason that leaf-eating colobine monkeys living in tightly knit kin groups with relatively relaxed female dominance relations breed faster than other monkeys is that they can afford to take advantage of offers by other females to carry their babies. When alloparental baby-carrying includes provisioning as well, benefits from daycare are magnified further. With care by both mothers and others, and with infants buffered from starvation around the time of weaning, such full-fledged cooperative breeding means infants survive in spite of being weaned early.
Mothers in a range of creatures produce costly young, but none more costly than a human infant. Nor does any other animal, even other apes (who also have slow life histories), take anything like so long to mature.95 Yet humans living under “natural” conditions (by gathering and hunting) breed faster than other apes. Colonizers par excellence, anatomically modern humans spread out of Africa and then migrated around the globe to Europe, Asia, Australia, and eventually to North America, South America, and the Pacific. The only other primates to routinely share both care and provisioning of young in this way and as a consequence to breed faster and to rapidly colonize new habitats are the callitrichids. So how can such broad comparisons inform the way we think about childcare in the genus Homo?
ALLOPARENTS ARE CRITICALLY IMPORTANT IN HUMANS TOO
Historians of the family like Stephanie Coontz, along with anthropologists, psychologists, and social workers, have long been aware that, across time and in diverse locales, infants born into poverty, at low birthweight or premature, or to a teenage or unmarried mother tend to do better cognitively, emotionally, and physically if they grow up in extended families. Whether alloparental interventions involve older siblings, grandmothers, or other kin, or just a particularly interested mentor, a vast cross-disciplinary literature attests to the fact that mothers with more social support are more responsive to their infants’ needs. The greater the risk factors, the more evident do correlations between alloparental support, maternal sensitivity, and child well-being become. As Coontz puts it, “Children do best in societies where childrearing is considered too important to be left entirely to parents.”96
It is hard to imagine babies at greater risk than those born to desperately poor women in eighteenth-century Europe—an era when depositing infants in foundling homes or abandoning them outright was rampant. Tellingly, the availability of support from matrilineal kin to help the mother and reduce the opportunity costs of caring for her child played a bigger role in the mother’s decision to keep rather than abandon her baby than did actual income.97 Three hundred years later, the perception that social support—in the form of available childcare—is going to be hard to obtain leads women in industrialized nations like Germany and the United States to postpone childbirth or decide against having children altogether.98
Evidence from high-risk groups in the United States underscores how much social support matters. The presence of a grandmother in the same household with a teenage mother, or just frequent visits from a grandmother, increases the chance that infants will forge more secure attachments to their young and inexperienced mothers.99 Babies born to unmarried, low-income teenagers who grow up with a grandmother in the household also tend to test better on cognitive development—perhaps because they have spent less time alone or feel more secure.100 Similar correlations are reported for low-birthweight infants born to teenage mothers. Having a grandmother on hand early on (typically the maternal grandmother) was correlated with improved health and cognitive outcomes three years later.101
Again and again, the mother’s perception of social support and the infant’s sense of security (perhaps in response to stronger signals of maternal commitment) seem to matter more than any actual improvement in material resources available to the mother-infant pair. In a randomized controlled trial carried out by David Olds and his colleagues at the Prevention Research Center for Family and Child Health at the University of Colorado in Denver, trained nurses were sent to the homes of first-time expectant mothers. They made six or seven visits during pregnancy, followed by 21 visits in the period between birth and the child’s second birthday. Modest as such intervention may seem—little more than every so often having another woman offer social support and mentoring—it was correlated with a cascade of beneficial outcomes detectable as long as 15 years later. When matched with similar mothers not visited by nurses, the children of visited mothers grew up emotionally more responsive, were less likely to exhibit emotional vulnerability when exposed to fearful stimuli, learned language sooner, and had higher Mental Development Index scores than children in the control group. Children of visited mothers were also significantly less likely to be abused by their mothers.102
Supportive interventions have produced similar outcomes in other cultures. For example, visits to a Brazilian maternity ward resulted in mothers’ increased willingness to feed their babies exclusively with breast milk, and mothers who continued to receive visits after they returned home were more likely to continue breastfeeding irrespective of their socioeconomic status.103 The tougher that times become, and the more that childrearing competence is compromised, the more pronounced the psychological benefits from alloparental support seem to be.
Even though social scientists have long been aware of such correlations, and mothers clearly feel the need for social support, the evolutionary rationale for links between perceived support, maternal decision-making and behavior, and the emotional well-being of children went unexplored. Relevant studies were rarely undertaken with past survival and fitness concerns in mind. The most extensive and methodologically sophisticated psychological studies were almost invariably undertaken in Western countries where people are socially and spatially separated into nuclear families, live in houses with walls and access to modern medicine, and no longer have to worry all the time about eventualities like their children being eaten. Emotional and cognitive benefits to extended families were noted, but there were few opportunities to link alloparental support to actual child survival. Yet from an evolutionary pe
rspective, child survival was the currency that mattered.
We failed to consider the profound impact of older siblings, grandmothers, uncles, or the mother’s lovers in worlds where more than half of all infants born would starve, be murdered or eaten, or succumb to accident or disease before they matured. Only at the end of the twentieth century, as findings by human behavioral ecologists and sociobiologists started to come in, did it become clear that in foraging societies with high rates of infant and child mortality—societies like those our ancestors evolved in—support from alloparents not only improved health, social maturation, and mental development, it was essential for child survival.
THE PENNY DROPS
By the last quarter of the twentieth century, a handful of human behavioral ecologists and sociobiologists, aware of the occurrence of shared care and cooperative breeding in some other animals, began to entertain suspicions about collateral kin. But only since about 1999 has sufficient evidence been amassed to allow us to consider these disparate findings in an evolutionary perspective and to interpret their impact.
In the mid-1980s a young doctoral candidate in anthropology, now a pediatrician, Paul Turke, became sufficiently impressed by sociobiological research on “helpers at the nest” in monkeys and other animals to want to find out if helpers affected the reproductive success of humans as well. Together with the sociobiologist Laura Betzig, Turke went out to study the relation between family composition and reproductive success among Pacific islanders on Ifaluk atoll. What this husband-wife team discovered was that parents whose firstborn was a daughter actually produced more surviving children than parents whose firstborn was a son because (Turke hypothesized) daughters are more active in caring for younger siblings than sons are in that society.104
About this same time, a fellow sociobiologist, Mark Flinn, found a similar correlation between alloparental assistance and maternal reproductive success among Caribbean villagers in Trinidad. Mothers with nonreproductive helpers on hand had higher reproductive success than those without.105 Daughters proved the most helpful, but having any helper in the household, male or female, was still correlated with increased child survival. Shortly afterward, Kristen Hawkes noticed something odd about grandmothers among the Hadza people she was studying. Her discovery would provide the catalyst for her fellow anthropologists to begin to think in new ways about the evolutionary significance of women past childbearing age.
Hawkes has been a pioneer in the study of foraging strategies among hunter-gatherers, and she had gone out to the eastern rift valley of Tanzania to study one of the last remaining such groups. She and her team were among the first fieldworkers to measure just how much food different members of a Hadza group contributed to the daily diet. Along with James O’Connell, an archaeologist, and the human ethologist Nick Blurton Jones, Hawkes followed Hadza men, women, and children as they foraged, counting and weighing every edible item that each man, woman, and child brought back. Day after day, they trudged along as women collected berries and nuts or hacked at the ground with their digging sticks to pry out starchy tubers from underneath the sun-baked surface. They trotted after men when they went off hunting—or at least when they attempted to, for Hadza men’s predilection for reputation-enhancing big game like eland meant that hunters rarely succeeded. Eland weigh 500 kilograms or more and are, relative to the leanness typical of most wild game, deliciously marbled with fat. Yet these most desirable of ungulates are also widely dispersed, elusive, and more difficult to bag than common prey like hares or tortoises. Most days the men came home empty-handed, and it was food gathered by women day to day that kept children fed.
Hawkes and her colleagues also noticed something else. The first gatherers to leave camp in the morning and the last to return in the evening, as well as those who ended up carrying the heaviest loads, were not (as one might expect) young women in their prime. Nor were they the mothers with hungry children waiting back at camp. Rather, the most dedicated food-gatherers were the leathery-faced old women, long past their prime. In a landmark paper titled “Hardworking Hadza Grandmothers,” the researchers described great-aunts and grandmothers who, far from taking advantage of their no-longer-child-burdened “golden years” to put their feet up, were working harder than ever.106
For children in these foraging groups, having a grandmother or great-aunt helping to feed them was correlated with faster growth rates.107 In times of food shortage, it was also correlated with a higher likelihood of survival.108 Turke’s reports from Ifaluk atoll, Flinn’s from Trinidad, and now these findings from hunter-gatherers in Tanzania all pointed to intriguing parallels among cooperative breeders. Whether older sisters, grandmothers, or great-aunts, in every study it was alloparents willing to help who permitted mothers to produce more children likely to survive. Impressed by these discoveries, I became convinced that humans, like many birds and mammals, must have evolved as cooperative breeders, and by 1999 I was saying so.109 Since then, the case for cooperative breeding has only grown stronger, as researchers collected and analyzed data from larger populations, including horticultural as well as foraging societies. These bigger sample sizes quickly began to yield highly significant results.
Only a tiny fraction of humanity still lives by gathering plant foods and hunting with spears—or, in a few cases, nets—as people in African forests have done for tens of thousands of years and as the Aka still do.110 But even as foragers have come to rely at least in part on trade with their settled neighbors, or on occasional employment by them, they continue to rear children in the traditional way, and with good reason. By 2000, the anthropologist Paula Ivey Henry had discovered that among the Efe the number of alloparents a baby had at one year of age was correlated with how likely the child was to be alive at age three.111 That same year, a reanalysis of old medical records showed that even among settled, horticultural peoples, alloparents were critical for child survival.
Tantalized by findings such as those from the Efe, and by Hawkes’s suspicions about the role of Hadza grandmothers, two British anthropologists, Rebecca Sear and Ruth Mace, dusted off records from one of the most ambitious studies ever undertaken on maternal and child health in a traditional society before the introduction of modern medicine. Between 1950 and 1980 researchers from the United Kingdom Medical Research Council had monitored the nutritional status of mothers and the growth rates of their children among Mandinka horticulturalists in The Gambia, West Africa. Of 2,294 children in their sample, 883, nearly 40 percent, died before age five. As Sear and Mace pored over the old records on growth rates and child mortality, they asked themselves questions about family composition that the medical researchers had not thought to analyze before. They already knew that if a mother died before a child was weaned, it was bad news. But this time they asked who else, besides the mother, mattered to a child’s survival?
The results from their reanalysis of the Gambian data were stunning. If the child had older siblings (especially sisters) or if the child’s maternal grandmother was living nearby and was herself past reproductive age, the child’s probability of dying before age five fell from 40 percent to 20 percent.112 Not surprisingly, mothers were critical for survival during the first two years of life while the baby was still dependent on breast milk. After age two, however, by which time Mandinka children are usually weaned, the presence of a mother no longer had any measurable effect on child growth or survival. Apparently, compensatory care by allomothers was sufficiently good that the physical condition of weanlings was unaffected by the death of their mothers. Thus Mandinka referred to anyone plump as being “fat as an orphan.”113
From the perspective of a Mandinka child, having an older sister on hand to babysit or a maternal grandmother to provide extra food as well as care was, literally, a lifesaver. Yet the presence of the biological father, paternal grandparents, or an elder brother had no measurable impact on child survival. If paternal loss ushered a stepfather into the picture, however, a child’s chances of survival plummeted.114 Ot
herwise, as the researchers bluntly put it, “Fathers make absolutely no difference to child anthropometric status or survival”—provided allomothers were on hand.115
In later chapters I will consider these findings more broadly, including specific contexts where fathers do matter very much, and where a child’s older siblings, aunts, uncles, and especially grandmothers may have negative as well as positive impacts on child well-being. But for the moment, my point is simply that for primates generally and for humans there are circumstances when alloparents can be as important, sometimes more important, than parents. Frankly, this was not something social scientists had expected to find, and it became apparent only because the mortality rates among Mandinka children, especially in the months and years right after weaning, were so high.
High as they seem, child mortality rates among Gambian horticulturalists at the middle of the twentieth century were not atypical for African populations before the introduction of modern medicine. They are high compared with rates at the end of the century but are within the range of mortality statistics reported for various wild primates, for nomadic hunters and gatherers, and presumably for our Pleistocene ancestors as well. The best available data for Hadza, Ju/’hoansi, or Aka foragers indicate that 40 to 60 percent of children in these populations—and more in bad times—died before age 15.116 Given that child survival is the single most important component of maternal reproductive success, if allomaternal involvement reduced mortality by even a small amount, over generations the evolutionary implications would be significant.117 And if these heretofore unacknowledged benefactors actually managed to cut child mortality in half—as in the Mandinka case—their evolutionary impact would have been enormous.
Mothers and Others Page 12
