Principles of Archaeology

Linda Gosner

Principles of Archaeology

Paper 3: Grandmother Theory November 8, 2010

Grandmother Theory

Although humans and many of our primate relatives experience terminal fertility and menopause at similar ages, the human species is distinct because women often survive in good health for many decades beyond menopause. Likewise, both male and female humans reach full maturity later than other primates and have higher fertility rates. Even our genetically closest primate relatives, chimpanzees, have quite different life histories: their young mature more quickly than human children do, females become frail shortly after menopause, and males and females reach “old age” at approximately 33 years and have much shorter average life spans than humans (Hawkes 2003:389; Hawkes 2010). The grandmother theory, conceived by Kristen Hawkes and other collaborators in the Department of Anthropology at the University of Utah, seeks to explain these phenomena in terms of human evolution. Hawkes proposes that early in human history the most vigorous elder women began to help provision food for their grandchildren. In relieving their daughters of some of this burden, they allowed the young mothers to wean offspring earlier and to have children more frequently. Thus, older, infertile females –grandmothers—could increase their daughters’ fertility. In Darwinian terms, “more vigorous longer-lived females would allocate less to their own current reproduction. But sufficient benefit from the help of elders could compensate that cost, resulting in selection for greater longevity” (Hawkes 2003:388). Hawkes has found supporting evidence for the grandmother theory both in her ethnographic observation of modern hunter-gatherers and in the archaeological record. Her approach is part of the larger field of behavioral ecology, which examines “behavior in terms of selection….looking at how variations in behavior relate to variations in survival, reproduction, and the rearing of offspring” (Shennen 2002:22).

Some argue that the long human lifespan is a recent development correlating with the advent of modern medicine (O’Connell et al. 1999:476). They cite the fact that life expectancy in the United States, western Europe, and Japan is around 80 years, while the lifespan of people living in traditional settings with limited access to modern health care is often under 40 years. Because life expectancy is determined from the average age at death, this variability can be accounted for by high infant mortality in traditional settings, not by variation in maximum human lifespan. Hawkes notes that 72% of women who reached adulthood in 19th century France (an agriculturally-based society), lived past the age of terminal fertility and often much longer than that, surviving into old age. The same is true for populations living as hunter-gatherers and not dependent on domesticated foods. These include the !Kung of southern Africa, the Ache of South America, and the Hadza of northern Tanzania (Hawkes 2003:390).

During the mid-1980s, Hawkes, with her colleagues J.F. O’Connell (also her husband), and N.G. Blurton Jones, conducted an ethnographic study of the Hadza people in order to evaluate whether the grandmother theory could be responsible for Hadza women’s typically long postmenopausal lifespans. Because human ancestors were hunter-gatherers, as the Hadza are, studies of Hadza behavior can also be used to infer how early humans might have behaved in terms of food sharing and social organization. This study traced between 200 and 300 Hadza from September 1985 through mid-July 1986, focusing on the “relationship between variation in mothers’ and grandmothers’ foraging returns and children’s weight changes” (Hawkes et al. 1997:557).

Hawkes observed that Hadza men and teenage boys spent much of their time hunting at night near water holes, practiced “encounter hunting” during the day, occasionally joined women’s foraging expeditions, and also spent time collecting honey when the group was living at a camp near abundant beehives. The food that the men provided to the group was irregular; women’s foraging activities were more constant and dependable as a caloric source (Hawkes et al. 1997:554). Non-nursing mothers’ foraging time accounted for 46% of the variation in their children’s weight changes, while nursing mothers’ foraging time accounted for only 13%. The researchers observed that mothers reduced the amount of time they spend foraging after they gave birth in order to care for their newborn, but increased foraging time gradually during their nursing period as the child grew older. When one child was weaned and a new child was born, grandmothers provided the additional foraging support necessary to feed the weaned child (Hawkes et al. 1997:559).

Hawkes and her colleagues based the definitions of “grandmother” on behavioral criteria rather than on genetic links to a mother, defining grandmothers as “senior women who consistently feed, tend, and otherwise support the children of younger women” (Hawkes et al. 1997:554). The researchers observed that weaned children assisted in foraging, but they usually targeted seasonal fruits and other plants that required little processing. Adults concentrated on finding tubers, plants that are available year round, but too difficult for children to dig up, process, and cook. This division of labor allowed the Hadza to successfully inhabit regions where children would not be able to live if they were entirely responsible for their own food intake, as is the case with non-human primate young (Hawkes et al. 1997:560). Grandmothers were able to remain a valuable part of the group, and retain their “fitness” even when they were no longer having children of their own. In other words, a “trade-off” occurred at some point during the women’s lifetimes; they switched their attention from reproductive effort to “somatic effort toward increased survivorship later” (Hawkes et al. 1997:563). This pattern is consistent with the grandmother hypothesis.

These ethnographically observable phenomena among the Hadza correlate well with the grandmother hypothesis, but it is difficult to prove that these behaviors occurred early in human history. It is also hard to demonstrate that they were responsible for distinctively human aging characteristics and for post-menopausal survival. An older hypothesis, the so-called hunting hypothesis, had a different explanation for the late maturity and high fertility of humans as compared to other primates. Proposed by S.L. Washburn, the hunting hypothesis was influential in human evolution studies beginning in the 1950s (Hawkes 2003:381). According to this theory, early humans became increasingly reliant upon hunting large game for survival. They became bipedal in order to hold tools in their hands for hunting, and this complex task required larger brains. For humans to develop large brains, their children had to be more dependent on their mothers for longer periods of time while their brains developed and were later in reaching maturity than their quadruped relatives. This meant that mothers had to take care of their children longer and depend upon their mates to do the hunting, which enabled women to have more offspring. By contrast to the grandmother hypothesis, which posits the cooperation of the grandmother as essential to the survival of the children, this theory points to the development of the nuclear family as the essential unit for survival.

The hunting hypothesis is based on the idea that the earliest humans were hunting large game, but recently uncovered finds and re-analyses of old evidence suggest that they may have been getting their meat predominantly through scavenging (Hawkes 2003:382). Beginning around 2 million years BP, H. ergaster (the first successful species of our genus, Homo, and a direct human ancestor) underwent several morphological changes. Its members reached larger size, matured later, and developed smaller teeth and jaws and a narrower thorax by comparison to contemporary australopithecines. While these features were initially attributed to more frequent hunting and meat consumption, there is no archaeological evidence of big game hunting until 500,000 BP. Nor does isotopic analysis of bones show increased meat consumption among H. ergaster. Other developments, which the hunting hypothesis assumes to have occurred contemporaneously, are now known to have occurred at different times: bipedalism predates big game hunting and the development of large brains by 1.5 million years. Lastly, assumptions about the importance of the nuclear family might have been based more on 1950s ideals than prehistoric realities. Hawkes and her colleagues suggest that the grandmother theory better explains the morphological changes in H. ergaster than does the increasingly obsolete hunting hypothesis (Hawkes 2003:383).

O’Connell, Hawkes, and Jones’ work on H. erectus attempts to show the applicability of grandmother theory, and the evolution of long human lifespans even before the development of fully modern humans. It is important to note here that H. ergaster and H. erectus are sometimes considered to be the same species; the distinction between the two is sometimes debated. O’Connell et al. cite recent re-analyses of damage to animal bones that show either no evidence of hominid alterations or show passive hominid scavenging. H. erectus seem to have targeted the marrow of long bones and the contents of the brain cavity of animals. Neither would have provided a stable source of food for juveniles and certainly would not have yielded enough food for the daily caloric requirements of the group as a whole (O’Connell et al. 1999:464). Thus, the hunting hypothesis is inappropriate.

General correlations between dental eruption stages, adult body size, and estimates of longevity based on extant fossils indicate that H. erectus lived longer and matured later than other hominoids, though their fertile period was likely the same as the others’. Australopithecines, on the other hand, had life histories similar to those of modern chimpanzees (O’Connell et al. 1999:469-470). These changes in H. erectus may have been caused by alterations in the environment: the general cooling over the past 3 million years would have eliminated fruit and other food resources in the winter. This meant that children and juveniles could no longer collect this food during the cold months. As a result, adults would have had to collect the food resources more consistently available to them – notably tubers –with more frequency in order to feed themselves and the youngsters alike. Though it is hard to trace tuber consumption in the archaeological record, O’Connell et al. cite a variety of factors suggesting high tuber consumption: H. erectus’ human-like digestive anatomy, increasing evidence of fire and digging tools, and increasing geographical range of H. erectus into arid climates where tubers were one of the primary foods available (1999:473-474).

This scenario is consistent with the grandmother hypothesis – increasing dependence on tubers would have necessitated adult participation in food acquisition for the young. Aging H. erectus women who had no young children could aid in feeding their daughters’ offspring, allowing the grandchildren to wean earlier and the daughters to become pregnant again sooner. Selection for vigorous post-menopausal women would have begun to change the life histories of H. erectus to one more equivalent to that of fully modern humans. Here, the authors assume that a grandmother would have more interest in her daughter’s offspring than in her son’s; a grandmother could be assured of her genetic link to a daughter’s children, while a genetic connection to a “daughter-in-law’s” children is unreliable (O’Connell et al. 1999:477).

Though Hawkes’ ethnographic observations of the Hadza provide convincing evidence for the grandmother hypothesis at work in the modern-day world, evidence for the grandmother hypothesis among H. erectus is less clear. More concrete proof for reliance on tubers would strengthen the authors’ argument; such proofs could perhaps include analysis of H. erectus dentition and testing for strontium and calcium isotopes to understand their diet (O’Connell et al. 1999:475). As more examples of skeletal remains are found, opportunities for such studies may increase. Certain inherent problems will always be present, however, including the inaccuracy of age estimates for adult skeletal remains, the absence of large population samples, and the absence of elderly osteological remains in the archaeological record (such bones are poorly preserved) (O’Connell et al. 1999:476).

Other possible scenarios can explain trends in human life histories. One is Frank Marlowe’s “patriarch hypothesis,” which suggests that human longevity after menopause is a byproduct of selection for longevity in males rather female “grandmothers.” Marlowe argues that males who lived past their peak vigor could still be successful reproducers because of the power and status they acquired during their lifetime; if they reproduced late in life, they would pass on genes that extended the male life span to their offspring. Marlowe believes that these genes were not part of the Y chromosome, so they were passed on to female and male offspring alike. Such a man’s children, of either gender, would live longer. These longevity genes were unrelated to those that dictate the age of menopause, Marlowe argues, and menopause continued to take place at the same time in humans as in our primate relatives (Marlowe 2000:27-30, 38).

Marlowe, like Hawkes, cites examples from the Hadza: the hunting skills of Hadza men, their hunting reputations, and their daily food returns do not increase in the years after their prime. However, older men between 55 and 60 years of age often remarry and have a new family of children with a young wife (Marlowe 2000:33). Marlowe makes no mention of studies that show increasing evidence for birth defects among children of old or elderly fathers (for example: Rabin 2007). Such defects are presumably genetic disadvantages and weaken Marlowe’s argument from a Darwinian perspective.

Today, fewer and fewer people are coming to the defense of the hunting hypothesis to explain human longevity after menopause and other distinctive characteristics of human life history. Nevertheless, both the grandmother hypothesis and the patriarch hypothesis will require more substantial proof. Hawkes’ and her colleagues’ ethnographic work with the Hadza is quite convincing (though it could be strengthened if they used a genetic rather than a behavioral definition of the grandmother), but supporting evidence from the archaeological record should continue to be sought and studied using modern scientific methods of analysis. Likewise, the patriarch hypothesis argument could be bolstered with more genetic studies on which genes promote longevity and how – and whether -- they correspond to gender.

Works Cited:

Hawkes, Kristen

2010 How grandmother effects plus individual variation in frailty shape fertility and morality: Guidance from human-chimpanzee comparisons. PNAS 107 (2):8977-8984.