Cannibals and Kings

by Marvin Harris

  2

  Murders in Eden

  The accepted explanation for the transition from band life to farming villages used to go like this: Hunter-collectors had to spend all their time getting enough to eat. They could not produce a “surplus above subsistence,” and so they lived on the edge of extinction in chronic sickness and hunger. Therefore, it was natural for them to want to settle down and live in permanent villages, but the idea of planting seeds never occurred to them. One day an unknown genius decided to drop some seeds in a hole, and soon planting was being done on a regular basis. People no longer had to move about constantly in search of game, and the new leisure gave them time to think. This led to further and more rapid advances in technology and thus more food—a “surplus above subsistence”—which eventually made it possible for some people to turn away from farming and become artisans, priests, and rulers.

  The first flaw in this theory is the assumption that life was exceptionally difficult for our stone age ancestors. Archaeological evidence from the upper paleolithic period—about 30,000 B.C. to 10,000 B.C.—makes it perfectly clear that hunters who lived during those times enjoyed relatively high standards of comfort and security. They were no bumbling amateurs. They had achieved total control over the process of fracturing, chipping, and shaping crystalline rocks, which formed the basis of their technology, and they have aptly been called the “master stoneworkers of all times.” Their remarkably thin, finely chipped “laurel leaf” knives, eleven inches long but only four-tenths of an inch thick, cannot be duplicated by modern industrial techniques. With delicate stone awls and incising tools called burins, they created intricately barbed bone and antler harpoon points, well-shaped antler throwing boards for spears, and fine bone needles presumably used to fashion animal-skin clothing. The items made of wood, fibers, and skins have perished, but these too must have been distinguished by high craftsmanship.

  Contrary to popular ideas, “cave men” knew how to make artificial shelters, and their use of caves and rock overhangs depended on regional possibilities and seasonal needs. In southern Russia archaeologists have found traces of a hunters’ animal-skin dwelling set in a shallow pit forty feet long and twelve feet wide. In Czechoslovakia winter dwellings with round floor plans twenty feet in diameter were already in use more than 20,000 years ago. With rich furs for rugs and beds, as well as plenty of dried animal dung or fat-laden bones for the hearth, such dwellings can provide a quality of shelter superior in many respects to contemporary inner-city apartments.

  As for living on the edge of starvation, such a picture is hard to reconcile with the enormous quantities of animal bones accumulated at various paleolithic kill sites. Vast herds of mammoth, horses, deer, reindeer, and bison roamed across Europe and Asia. The bones of over a thousand mammoth, excavated from one site in Czechoslovakia, and the remains of 10,000 wild horses that were stampeded at various intervals over a high cliff near Solutré, France, testify to the ability of paleolithic peoples to exploit these herds systematically and efficiently. Moreover, the skeletal remains of the hunters themselves bear witness to the fact that they were unusually well-nourished.

  The notion that paleolithic populations worked round the clock in order to feed themselves now also appears ludicrous. As collectors of food plants they were certainly no less effective than chimpanzees. Field studies have shown that in their natural habitat the great apes spend as much time grooming, playing, and napping as they do foraging and eating. And as hunters our upper paleolithic ancestors must have been at least as proficient as lions—animals which alternate bursts of intense activity with long periods of rest and relaxation. Studies of how present-day hunters and collectors allocate their time have shed more light on this issue. Richard Lee of the University of Toronto kept a record of how much time the modern Bushman hunter-collectors spend in the quest for food. Despite their habitat—the edge of the Kalahari, a desert region whose lushness is hardly comparable to that of France during the upper paleolithic period—less than three hours per day per adult is all that is needed for the Bushmen to obtain a diet rich in proteins and other essential nutrients.

  The Machiguenga, simple horticulturalists of the Peruvian Amazon studied by Allen and Orna Johnson, spend a little more than three hours per day per adult in food production and get less animal protein for this effort than do the Bushmen. In the rice-growing regions of eastern Java, modern peasants have been found to spend about forty-four hours per week in productive farm work—something no self-respecting Bushman would ever dream of doing—and Javanese peasants seldom eat animal proteins. American farmers, for whom fifty-and-sixty-hour work weeks are commonplace, eat well by Bushman standards but certainly cannot be said to have as much leisure.

  I do not wish to minimize the difficulties inherent in comparisons of this sort. Obviously the work associated with a particular food-production system is not limited to time spent in obtaining the raw product. It also takes time to process the plants and animals into forms suitable for consumption, and it takes still more time to manufacture and maintain such instruments of production as spears, nets, digging sticks, baskets, and plows. According to the Johnsons’ estimates, the Machiguenga devote about three additional hours per day to food preparation and the manufacture of essential items such as clothing, tools, and shelter. In his observations of the Bushmen, Lee found that in one day a woman could gather enough food to feed her family for three days and that she spent the rest of her time resting, entertaining visitors, doing embroidery, or visiting other camps. “For each day at home, kitchen routines, such as cooking, nut cracking, collecting firewood, and fetching water, occupy one to three hours of her time.”

  The evidence I have cited above leads to one conclusion: The development of farming resulted in an increased work load per capita. There is a good reason for this. Agriculture is a system of food production that can absorb much more labor per unit of land than can hunting and collecting. Hunter-collectors are essentially dependent on the natural rate of animal and plant reproduction; they can do very little to raise output per unit of land (although they can easily decrease it). With agriculture, on the other hand, people control the rate of plant reproduction. This means that production can be intensified without immediate adverse consequences, especially if techniques are available for combating soil exhaustion.

  The key to how many hours people like the Bushmen put into hunting and collecting is the abundance and accessibility of the animal and plant resources available to them. As long as population density—and thus exploitation of these resources—is kept relatively low, hunter-collectors can enjoy both leisure and high-quality diets. Only if one assumes that people during the stone age were unwilling or unable to limit the density of their populations does the theory of our ancestors’ lives as “short, nasty, and brutish” make sense. But that assumption is unwarranted. Hunter-collectors are strongly motivated to limit population, and they have effective means to do so.

  Another weakness in the old theory of the transition from hunting and collecting to agriculture is the assumption that human beings naturally want to “settle down.” This can scarcely be true given the tenacity with which people like the Bushmen, the aborigines of Australia, and the Eskimo have clung to their old “walkabout” way of life despite the concerted efforts of governments and missionaries to persuade them to live in villages.

  Each advantage of permanent village life has a corresponding disadvantage. Do people crave company? Yes, but they also get on each other’s nerves. As Thomas Gregor has shown in a study of the Mehinacu Indians of Brazil, the search for personal privacy is a pervasive theme in the daily life of people who live in small villages. The Mehinacu apparently know too much about each other’s business for their own good. They can tell from the print of a heel or a buttock where a couple stopped and had sexual relations off the path. Lost arrows give away the owner’s prize fishing spot; an ax resting against a tree tells a story of interrupted work. No one leaves or enters the village withou
t being noticed. One must whisper to secure privacy—with walls of thatch there are no closed doors. The village is filled with irritating gossip about men who are impotent or who ejaculate too quickly, and about women’s behavior during coitus and the size, color, and odor of their genitalia.

  Is there physical security in numbers? Yes, but there is also security in mobility, in being able to get out of the way of aggressors. Is there an advantage in having a larger cooperative labor pool? Yes, but larger concentrations of people lower the game supply and deplete natural resources.

  As for the haphazard discovery of the planting process, hunter-collectors are not so dumb as this sequence in the old theory would suggest. The anatomical details in the paintings of animals found on the walls of caves in France and Spain bear witness to a people whose powers of observation were honed to great accuracy. And our admiration for their intellects has been forced to new heights by Alexander Marshak’s discovery that the faint scratches on the surface of 20,000-year-old bone and antler artifacts were put there to keep track of the phases of the moon and other astronomical events. It is unreasonable to suppose that the people who made the great murals on the walls of Lascaux, and who were intelligent enough to make calendrical records, could have been ignorant of the biological significance of tubers and seeds.

  Studies of hunter-collectors of the present and recent past reveal that the practice of agriculture is often forgone not for lack of knowledge but as a matter of convenience. Simply by gathering acorns, for example, the Indians of California probably obtained larger and more nutritious harvests than they could have derived from planting maize. And on the Northwest coast the great annual migrations of salmon and candlefish rendered agricultural work a relative waste of time. Hunter-collectors often display all the skills and techniques necessary for practicing agriculture minus the step of deliberate planting. The Shoshoni and Paiute of Nevada and California returned year after year to the same stands of wild grains and tubers, carefully refrained from stripping them bare, and sometimes even weeded and watered them. Many other hunter-collectors use fire to deliberately promote the growth of preferred species and to retard the growth of trees and weeds.

  Finally, some of the most important archaeological discoveries of recent years indicate that in the Old World the earliest villages were built 1,000 to 2,000 years before the development of a farming economy, whereas in the New World plants were domesticated long before village life began. Since the early Americans had the idea for thousands of years before they made full use of it, the explanation for the shift away from hunting and collecting must be sought outside their heads. I’ll have more to say about these archaeological discoveries later on.

  What I’ve shown so far is that as long as hunter-collectors kept their population low in relation to their prey, they could enjoy an enviable standard of living. But how did they keep their population down? This subject is rapidly emerging as the most important missing link in the attempt to understand the evolution of cultures.

  Even in relatively favorable habitats, with abundant herd animals, stone age peoples probably never let their populations rise above one or two persons per square mile. Alfred Kroeber estimated that in the Canadian plains and prairies the bison-hunting Cree and Assiniboin, mounted on horses and equipped with rifles, kept their densities below two persons per square mile. Less favored groups of historic hunters in North America, such as the Labrador Naskapi and the Nunamuit Eskimo, who depended on caribou, maintained densities below .3 persons per square mile. In all of France during the late stone age there were probably no more than 20,000 and possibly as few as 1,600 human beings.

  “Natural” means of controlling population growth cannot explain the discrepancy between these low densities and the potential fertility of the human female. Healthy populations interested in maximizing their rate of growth average eight pregnancies brought to term per woman. Childbearing rates can easily go higher. Among the Hutterites, a sect of thrifty farmers living in western Canada, the average is 10.7 births per woman. In order to maintain the estimated .001 percent annual rate of growth for the old stone age, each woman must have had on the average less than 2.1 children who survived to reproductive age. According to the conventional theory such a low rate of growth was achieved, despite high fertility, by disease. Yet the view that our stone age ancestors led disease-ridden lives is difficult to sustain.

  No doubt there were diseases. But as a mortality factor they must have been considerably less significant during the stone age than they are today. The death of infants and adults from bacterial and viral infections—dysenteries, measeis, tuberculosis, whooping cough, colds, scarlet fever—is strongly influenced by diet and general body vigor, so stone age hunter-collectors probably had high recovery rates from these infections. And most of the great lethal epidemic diseases—smallpox, typhoid fever, flu, bubonic plague, cholera—occur only among populations that have high densities. These are the diseases of state-level societies; they flourish amid poverty and crowded, unsanitary urban conditions. Even such scourges as malaria and yellow fever were probably less significant among the hunter-collectors of the old stone age. As hunters they would have preferred dry, open habitats to the wetlands where these diseases flourish. Malaria probably achieved its full impact only after agricultural clearings in humid forests had created better breeding conditions for mosquitoes.

  What is actually known about the physical health of paleolithic populations? Skeletal remains provide important clues. Using such indices as average height and the number of teeth missing at time of death, J. Lawrence Angel has developed a profile of changing health standards during the last 30,000 years. Angel found that at the beginning of this period adult males averaged 177 centimeters (5′ 11″) and adult females about 165 centimeters (5′ 6″). Twenty thousand years later the males grew no taller than the females formerly grew—165 centimeters—whereas the females averaged no more than 153 centimeters (5′ 0″). Only in very recent times have populations once again attained statures characteristic of the old stone age peoples. American males, for example, averaged 175 centimeters (5′ 9″) in 1960. Tooth loss shows a similar trend. In 30,000 B.C. adults died with an average of 2.2 teeth missing; in 6500 B.C., with 3.5 missing; during Roman times, with 6.6 missing. Although genetic factors may also enter into these changes, stature and the condition of teeth and gums are known to be strongly influenced by protein intake, which in turn is predictive of general well-being. Angel concludes that there was “a real depression of health” following the “high point” of the upper paleolithic period.

  Angel has also attempted to estimate the average age of death for the upper paleolithic, which he places at 28.7 years for females and 33.3 years for males. Since Angel’s paleolithic sample consists of skeletons found all over Europe and Africa, his longevity estimates are not necessarily representative of any actual band of hunters. If the vital statistics of contemporary hunter-collector bands can be taken as representative of paleolithic bands, Angel’s calculations err on the low side. Studies of 165 !Kung Bushman women by Nancy Howell show that life expectancy at birth is 32.5 years, which compares favorably with the figures for many modern developing nations in Africa and Asia. To put these data in proper perspective, according to the Metropolitan Life Insurance Company the life expectancy at birth for non-white males in the United States in 1900 was also 32.5 years. Thus, as paleodemographer Don Dumond has suggested, there are hints that “mortality was effectively no higher under conditions of hunting than under those of a more sedentary life, including agriculture.” The increase in disease accompanying sedentary living “may mean that the mortality rates of hunters were more often significantly lower” than those of agricultural peoples.

  Although a life span of 32.5 years may seem very short, the reproductive potential even of women who live only to Angel’s 28.7 years of age is quite high. If a stone age woman had her first pregnancy when she was sixteen years old, and a live baby every two and a half years thereafter,
she could easily have had over five live births by the time she was twenty-nine. This means that approximately three-fifths of stone age children could not have lived to reproductive age if the estimated rate of less than .001 percent population growth was to be maintained. Using these figures, anthropological demographer Ferki Hassan concludes that even if there was 50 percent infant mortality due to “natural” causes, another 23 to 35 percent of all potential offspring would have to be “removed” to achieve zero growth population.

  If anything, these estimates appear to err in exaggerating the number of deaths from “natural” causes. Given the excellent state of health the people studied by Angel seemed to enjoy before they became skeletons, one suspects that many of the deceased died of “unnatural” causes.

  Infanticide during the paleolithic period could very well have been as high as 50 percent—a figure that corresponds to estimates made by Joseph Birdsell of the University of California in Los Angeles on the basis of data collected among the aboriginal populations of Australia. And an important factor in the short life span of paleolithic women may very well have been the attempt to induce abortions in order to lengthen the interval between births.