Catherine Perlès
Even if apes were more omnivorous than is generally believed, man would remain unique among primates for the diversity of his diet. Among the factors contributing to the distinctiveness of human alimentary behavior are the variety and complexity of the underlying economy and the highly socialized, not to say ritualized, nature of human feeding.
The study of prehistoric man’s alimentary behavior and its gradual transformation should therefore be an important topic. Despite this importance, however, the current state of research leaves students of prehistory in a paradoxical situation: while we are capable of discussing the fabrication of, say, stone tools in minute detail, the best we can do when it comes to the history of eating is to propose a sketchy and still largely hypothetical portrait.
Chemistry and Prehistory
Some day, perhaps, chemical analysis of fossilized bones will make a full history of the prehistoric diet possible. Bony tissue contains dietary “markers” such as trace elements and carbon isotope ratios.1 At the present time, however, these advanced methods are mostly still in the experimental stage. There are formidable technical problems to overcome, because many factors can alter the chemical composition of bones after burial. Despite these difficulties, promising results have already been obtained, particularly in regard to fairly recent prehistoric populations. For example, studies of certain North American tribes have demonstrated a regular seasonal alternation between food drawn from the land and food drawn from the water. By contrast, in studies of Danish tribes of the Mesolithic era, no such complementary seasonal cycles have been found to differentiate coastal from inland sites. Apparently the two types of sites were occupied by different groups: fishermen in the one case, hunters in the other. Elsewhere, among agricultural populations, investigators have pointed to the existence of dietary differences between individuals of different social status.
Because of the many technical issues yet to be resolved, these results must be regarded as preliminary. Hence chemical analysis of bones, although an exciting avenue of future investigation, is still a long way from providing us with a complete picture of the prehistoric diet. If we want to trace the history of human eating back to the beginning, we must for the time being rely on traditional archaeological methods, with all their well-known shortcomings.
Those methods are particularly problematic when it comes to the study of diet, since plant matter, like all organic substances, decays in the soil. It is virtually certain, however, that primitive man derived most of his calories from plants, except in extreme latitudes, where the environment was quite special.2 Indeed, although man, like other primates, is omnivorous, the symbolic importance attached to meat does not reflect its actual nutritional importance. Much of this essay is concerned with meat consumption, but this has to do more with the survival of vestiges of that activity than with its actual importance in the prehistoric diet. But this emphasis also stems, in my view, from the fact that prehistorians themselves wittingly or unwittingly ascribe too much importance to meat-eating in their various theories of “hominization.”
The First Hominids: Fierce Hunters or Carcass Snatchers?
To this day, meat, especially meat obtained by hunting wild animals, enjoys a symbolic importance out of all proportion to its place in the human diet. This no doubt accounts for the decisive role traditionally ascribed to hunting in theories of hominization from antiquity to the present.3 And it probably also explains the hyperbole in recent efforts to refute such theories by showing that the first hominids were scavengers.
In deposits left by the first hominids (Australopithecus and Homo habilis) in East Africa some 1.5 to 2.5 million years ago, we find substantial numbers of animal bone fragments together with primitive stone implements. The richest sites, located beside lakes and rivers, contain the remains of animals ranging in size from turtles to elephants along with hundreds of chips left from the cutting of stone tools. On the basis of such evidence, these hominids were long considered to be the first primate hunters. By definition they were also permanent bipeds. But bipeds are not particularly well adapted for hunting, particularly when the absence of any long-range projectile weapon made it impossible to hunt from ambush. It was therefore assumed that the males of these species must have cooperated in the hunt.4 Since the remains of the prey were gathered together at certain sites, these were held to be “campsites” and adduced as evidence for mutual assistance and complementarity between males and females.5
Some four to five million years ago, the story goes, diminished rainfall in East Africa reduced the extent of jungle and expanded the savanna, depriving man’s ancestors of many plant resources and forcing them to adapt by eating more meat, which required them to band together in organized hunts. This organized hunting led in turn to improvements in communication skills and intellectual faculties and to a gendered division of labor. In other words, the type of social and family organization regarded as typically human allegedly originated with hunting.6 It was therefore ostensibly because Australopithecus became a hunter that he developed human characteristics.
What kind of hypothesis is this? Is it a historical fact or what some ironically call a “scenario,” nothing more than a reflection of our own preconceptions? In fact, this reconstruction of the prehistoric past is based on reasoning by analogy.
An American prehistorian, L. Binford, was the first to ask what proof there was that the animal bones and stone implements were left by the same individuals.7 Perhaps the animal remains were left by wild carnivores and the tools by hominids. After more than a decade of debate, most scholars, particularly in the United States, accept Binford’s hypothesis that hyenas gathered most of the carcasses and the hominids arrived to pick over the remains only after the predators were sated. Others, however, hold that the first hominids were sufficiently well organized to snatch freshly killed prey, with considerable meat still left on the bones, and carry it off to sites where stone tools were cached.8
In my view, although research in recent years has yielded solid results, the interpretation of those results suffers from methodological errors and is just as prone to ideological bias as the previous interpretation.9 A broader view precludes seeing our earliest ancestors as miserable scavengers incapable of killing on their own the animals they ate.
Toward a Rehabilitation of the Early Hominids
In order to avoid circular reasoning, one must distinguish two questions. First, who brought the carcasses to the sites where they were deposited? And, second, how were those carcasses obtained?
The fact that animal remains are found together with stone tools and chips is one whose importance should not be underestimated. It is difficult to explain the abundance of both tools and chips if the sites in question were merely the lairs of hyenas picked over by hominids in search of leftover meat and marrow. Carnivores, moreover, do not normally amass bones out in the open. When they tear a carcass apart, they systematically disperse its parts. If they have young to feed, they take portions of the dead animal back to their dens. Thus dispersion is followed by selective accumulation.
When bones are amassed out in the open, therefore, there has to be some reason to explain why the same site was repeatedly reused. At first glance, nothing in the setting (by the side of a lake or river) accounts for this reuse, which is unusual for animal predators but typical of later human hunters. Potts’ hypothesis that the presence of tools accounts for the reuse of these sites explains both their character and location. If Potts is correct, most of the carcasses must have been brought to these places by the hominids themselves rather than by hyenas.
Second, analysis of what portions of carcasses are present on the site unambiguously demonstrates that no secondary scavenging was involved. One systematically finds the richest meat-bearing portions (leg, shoulder, etc.). The hominids therefore must have had access to freshly killed animals and not to what was left over after other carnivores had eaten their fill. Indeed, this “rehabilitation” of the early hominids has met with unexpected corroboration in the form of recent findings concerning the behavior of chimpanzees in the wild. Some are skillful hunters: in certain seasons male chimps will hunt smaller monkeys and other prey and eat up to 200 grams of meat per day.10
Consequently, it seems difficult to deny that the first hominids were predators in the broad sense. Still, animal protein was probably just a complement to a diet based mainly on plants. There are two kinds of evidence for this assertion: the wear patterns found on the teeth of early hominids, which indicate a plant-based diet, and the range of territory in which they operated. From this we conclude that Homo habilis was an opportunistic omnivore.
The Development of Diversified Hunting
There is no evidence that this alimentary strategy was significantly modified in Africa over the next several hundred thousand years. When Homo erectus replaced Homo habilis some 1.5 million years ago, the density of animal remains that we find at surviving sites actually tended to decrease. The bones are also more fractured, which suggests a more intensive consumption of marrow.
By contrast, when Homo erectus began to establish himself in more temperate regions, the greater seasonal contrasts and corresponding decrease in plant resources must have led to a greater, though by no means exclusive, nutritional role for meat. In Europe throughout the Paleolithic period we find archaeological evidence for a diversified meat diet, which very early came to depend on an ability to kill large animals.
The variety of animals consumed and the distribution of their ages suggests that hunters went after isolated animals rather than attacking entire herds en masse. Small groups of hunters made regular (though not necessarily daily) forays in search of whatever game they could find. But diversified hunting does not mean random hunting. All hunters observe potential prey even when they are not hunting. On every expedition decisions had to be made: for example, whether to follow the tracks of the first small game encountered or hold out for meatier prey or for something easier to capture. Obviously such choices depended on the day and the season, on experience with previous hunts, on the availability of stored meat, and on the group’s knowledge of the abundance of game in the area and of plant resources at that time of year.11
Gathering plants was in fact an integral part of hominid alimentary strategies. Because few traces of vegetal foodstuffs have survived, it is difficult to evaluate their consumption either qualitatively or quantitatively. It is nevertheless fair to assume that plants were important both nutritionally and economically, particularly after the domestication of fire opened up a whole new range of resources, namely, plants that, while toxic in their raw state, could be consumed after cooking.12
It was some 500,000 years ago that regular domestic use of fire began to have a profound effect on the hominid diet and doubtless on social behavior related to eating as well. Many carnivores have a taste for cooked meat (a natural by-product of forest fires). Yet only man had the ability to eat cooked meat regularly and thus to move from feeding to cooking and ultimately to a whole range of culinary techniques. It is striking to observe that the earliest traces of cooking are contemporary with the earliest vestiges of campfires.13 Charred bones have been found apart from the remains of the fire itself, indicating that the bones once held grilled meat and were not simply parts of animal carcasses used as fuel. Although grilling was sporadic more than 500,000 years ago, it became common once fire was truly domesticated.
The social impact of cooking was no doubt more immediately apparent than were its nutritional implications. It encouraged commensalism, or eating with others, and led to greater division of labor within the group, all of whose members participated in a regular round of activities associated with eating. The result was a more complex group organization.
Specialized Hunting in the Late Paleolithic Period
Toward the end of the middle Paleolithic era we detect the first signs of organized hunting, as individual animals began to be separated from herds by the concerted efforts of groups of hunters. Systems of subsistence based on the exploitation of a selected species of animal became commonplace. Among the species hunted in this fashion in one region or another were reindeer, horses, bison, aurochs, and even mammoths. This new type of hunting called for a socioeconomic structure and range of techniques quite different from those associated with opportunistic hunting. Its economic objectives were also different.
Opportunistic hunting had involved individuals or small groups and required little preparation other than daily maintenance of weapons. Many kinds of weapons and tools were needed to kill and prepare the diverse prey. By contrast, driving herd animals into pits took large numbers of people (often including women and children), usually drawn from groups of families residing separately for much of the year. These periodic assemblies for the hunt suggest that there must have been some means of spreading information and some form of social integration or expanded social structure extending beyond the domestic or residential unit. Hunting herd animals (often migratory) in this way required extensive preparation: pits had to be dug, and large numbers of specialized weapons had to be assembled, along with tools for slaughtering, butchering, and treating hides. In compensation for all this preliminary effort, the hunt could be expected to yield considerable quantities of meat, lard, marrow, hide, sinew, and hair. These reserves of food and materiel might be sufficient for several weeks or months. Hence this type of hunting was more demanding socially and technically as well as more risky (if no animals were captured) but more rewarding if successful. Broadly speaking, it was in areas that practiced this form of hunting that Paleolithic art also developed: sanctuaries, movable objects, and embellishments and ornaments of various kinds. This art, incorporated into communal rituals, no doubt encouraged integration and collaboration. But the art itself was in turn encouraged by the stockpiling of food reserves, which allowed groups to stay put for extended periods.
Collaborative hunting also implied a need to preserve meat after slaughter. The cold, dry climate of the late Paleolithic made this task easier. Meat was dried, smoked, or frozen in pits dug in the earth (in this preglacial climate the subsoil was permanently frozen). Such pits, covered with mammoth tusks, have been found in central and eastern Europe. Elsewhere, whole carcasses, weighted down with stones, were deposited in lakes that froze over in winter, thus preserving the meat.
Stored meats and other foods could be eaten either dry or boiled. Water was boiled either by tossing heated rocks into vessels of wood, bark, or hide, vestiges of which have been found near campfires, or by heating it directly over the fire in animal bellies or skins. Large pieces of meat were also grilled on spits. Traces of spits made of bone have survived. Smaller pieces could be grilled directly on the stones used to line hearths in this period.14
The late Paleolithic—a time when primitive man fished for salmon, captured birds, hunted animals in large groups, and developed the ability to depict his prey in art—in many respects marks the apogee of the hunter-gatherer way of life.
Diversity of Diet in the European Mesolithic
As the glaciers began to recede some 8,000 years before the birth of Christ, a moist, temperate climate returned to the middle latitudes of Europe. Tundra, steppes, and prairies gradually gave way to forest. The larger meat-bearing animals were supplanted by smaller prey more difficult to hunt in the dense forest. But other resources became more abundant: not only plants but also fish, mollusks, and birds. The Mesolithic diet was much more diverse than that of the late Paleolithic. Dozens of species from a variety of biotopes could be exploited from a single site. Nevertheless, we must be careful not to confuse diversity with abundance.
If deer, boar, and roe deer had been truly abundant, would Mesolithic man have added rabbit, birds, and snails to his diet? If a few plant species had supplied all of his energy needs, would he have scoured the area around his campsite for fruits, vegetables, and nuts? The very mobility of certain Mesolithic groups, which settled briefly in an area to exploit its resources before moving on, suggests conditions well short of abundance.
Farther north, in game-rich forests close to the sea or to rivers and lakes filled with fish and attractive to migratory birds, we do find traces of more permanent dwelling sites. Here abundant seasonal resources sustained life and made it possible to accumulate reserves. Hence although the Mesolithic tribes living along the Mediterranean were quick to learn agricultural and husbandry techniques imported by immigrants from the Near East, their northern cousins clung for a much longer time to their ancestral ways, even as they began to trade with the sedentary peoples to the south, exchanging fish, game, and gathered plants for ceramics, the meat of domestic animals, and polished stone tools.
Agriculture and Husbandry: Liberation or New Problems?
Although wild cereals had been harvested earlier, it is not until the Neolithic period that we see, first in the Near and Middle East and later in Europe, the beginnings of the economic “revolution” that became the basis for the traditional European diet. This revolution consisted in the cultivation of cereal crops (especially wheat and rye) and the raising of livestock, especially sheep, goats, cows, and pigs. It was at this time that wheat was first used to make both leavened and unleavened bread. The number of millstones that have been found and the care taken with them inside the dwelling indicate just how important a staple grain was from the outset.
There is a tendency, however, to overestimate the benefits of the ensuing dietary transformation. Because all the major western civilizations were founded on foods domesticated in this period, we are apt to view it through rose-colored lenses. In the temperate and northern parts of Europe, cereals and meat from domestic animals were still supplemented by resources from the wild, with the result that people in those parts of the world enjoyed an ample, balanced diet. By contrast, in Mediterranean Europe and the Near and Middle East, little use was made of wild plants or animals. In the early stages of the transition to the Neolithic, in fact, those potential sources of food appear to have been deliberately abandoned, perhaps for ideological reasons: turning his back on the wild, man concentrated on the domestic. The bulk of his diet consisted of cereals, legumes, beef, and mutton. Overreliance on cereals probably had negative effects on health. Dietary deficiencies were probably not the only cause of poor health, however. The growing density of sedentary populations created a propitious environment for epidemics.
Many scholars believe that a second dietary revolution followed close on the heels of the first: this involved the use of earthenware pottery, which allegedly made it possible to cook food in boiling water. Porridge made from various grains became popular as a result. Undue nutritional and dietetic importance has been ascribed to this innovation, however. For one thing, pottery was by no means indispensable for boiling food. As we saw earlier, this method of cooking was already practiced in the Paleolithic era. For another, it now appears certain that in several parts of the world the first ceramics were seldom used for cooking (this is true of Greece and Pakistan, for example). Pottery was used as cookware by these same peoples, but for several centuries they continued to prefer such traditional cooking methods as grilling, baking on heated slabs, or braising in a pit filled with heated stones. Meanwhile, innovations were made in the use of fire: raised hearths were constructed in dwellings and the first closed ovens (similar to traditional bread ovens) were built. Thus “pottage” (in the etymological sense, meaning food cooked in an earthenware pot) did not catch on quickly everywhere. It may have taken hold sooner in colder areas, where it was nutritionally important to preserve fat in the diet, but this remains to be proven. Ceramic containers were in any case used for preparing and storing fermented drinks, which were probably drunk on festive and ritual occasions.
Symbolism and the Origin of Cooking
Various technical and economic developments of the Paleolithic period made for a variety of new methods of preparing and preserving food. Many foods were available to satisfy man’s dietetic requirements, and it is likely that cultural preferences, or tastes transmitted from generation to generation, made their appearance at this early stage. Thanks to more numerous and better-preserved artifacts from the Neolithic period, we can be certain that cultural preferences existed by then. Accordingly, it is significant that in Europe hunting regained importance toward the end of the Neolithic. The prey was symbolically significant—the deer—and the renewed interest in hunting coincides with the first evidence of social inequalities and the emergence of elites. The fact that the deer was not domesticated was a true social and cultural choice, and the transfer of wild deer outside of their native habitat confirms the ideological importance attached to the “wild.”15
Eating, which was at first a response to individual needs, thus gradually became a key element of group structure, a mark of identity, and a symbolic means of expressing thought. Evidence exists of complex culinary creations in the great empires of the Middle East.16 These impressive states no doubt provided work for a new type of specialist: the cook. By contrast, there is little evidence of culinary sophistication or elaborate cookery in hunter-gatherer societies: preparations were simple, and the use of direct cooking (either by grilling meat or heating it on coals) made it difficult to mix ingredients to achieve new flavors. Menus were quite monotonous.
By contrast, we do find a number of hunter-gatherer societies that were quite adept at combining various substances for the purposes of healing and working magic. These preparations possessed certain features of true culinary art: they were complex combinations of ingredients for other than strictly nutritional purposes. I am therefore tempted to view these practices as forerunners of the more sophisticated culinary techniques that seek to achieve new flavors by means other than mere cooking. The consumption of fermented beverages probably also originated in rituals of one kind or another.
1. For example, the strontium/calcium ratio normally decreases with increasing meat consumption, while the ratio of carbon-13 to carbon-12 can indicate what types of plants were eaten. See, for example, E. Wing and A. Brown, Paleonutrition: Method and Theory in Prehistoric Foodways (New York: Academic Press, 1979).
2. This accounts for the interest in measuring the strontium/calcium and carbon-12/carbon-13 ratios in order to determine the proportion of animal and vegetable matter in the diet.
3. W. Stoczkowski, Anthropologie naïve, anthropologie savante: De l’origine de l’homme, de l’imaginaire et des idées reçues (Paris: Centre National de Recherche Scientifique, 1994) p. 215.
4. R. Ardrey, The Hunting Hypothesis (New York: Atheneum, 1976).
5. G. L. Isaac, “The Food-Sharing Behaviour of Proto-Human Hominids,” Scientific American 238 (1978): 90–108.
6. G. Mendel, La Chasse structurale (Paris: Payot, 1977).
7. L. Binford, Bones, Ancient Men, and Modern Myths (London: Academic Press, 1981).
8. R. Potts, Early Hominid Activities at Olduvai, Foundations of Human Behavior (New York: Aldine de Gruyter, 1988).
9. For instance: (1) conflation of two distinct questions, namely, who assembled the carcasses and how were they obtained?; (2) the use, by Binford and others, of so-called residual analysis, where only remains different from what carnivores normally leave in their own lairs (the “residue”) are attributed to the activity of hominids; from this choice it follows that the hominid cannot be seen as a hunter, because everything characteristic of the behavior of hunters (selection of prey, choice of the portions of the carcass containing the most meat or easiest to dismember, etc.) is automatically attributed to the carnivores; (3) contrary to the views expressed by R. J. Blumenschine in “Carcass Consumption Sequences and the Archaeological Distinction of Scavenging and Hunting,” Journal of Human Evolution 15 (1986): 639–59, and in Early Hominid Scavenging Opportunities, International Series 283 (Oxford: British Archaeological Reports, 1986), interpretations based in part on analysis of carcasses found at excavation sites do not distinguish between hunters and scavengers but between rapid access to the carcass (whether by hunting or theft of fresh kill) and delayed access (secondary scavenging). Rapid access leads to consumption of the best meat, while the poorer leftovers go to the latecomers. But for an exception to the ideological bias that often characterizes the method, see Potts, Early Hominid Activities.
10. C. Boesch and E. Boesch, “Hunting Behavior of Wild Chimpanzees in the Tai National Parks,” American Journal of Physical Anthropology 78 (1989): 547–73; G. Telecki, The Predatory Behavior of Wild Chimpanzees (Lewisburg, Penn.: Bucknell University Press, 1973).
11. S. J. Mithen, Thoughtful Foragers: A Study in Prehistoric Decision Making (Cambridge: Cambridge University Press, 1990).
12. A. B. Stahl, “Hominid Dietary Selection before Fire,” Current Anthropology 25 (1984): 151–68; Stahl, “Plant-Food Processing: Implications for Dietary Quality,” in Foraging and Farming: The Evolution of Plant Exploitation, ed. D. R. Harris and G. C. Hillman (London: Unwin-Hyman, 1989), pp. 171–93.
13. C. Perlès, Préhistoire du feu (Paris: Masson, 1977).
15. Although deer are easy to domesticate, untamed deer straight from the wild were transported by boat to islands where previously there had been no deer. See J.-D. Vigne, “Domestication ou appropriation pour la chasse: Histoire d’un choix socioculturel depuis le Néolithique,” in Exploitation des animaux sauvages à travers le temps, XIIIe Rencontres Internationales d’Archéologie et d’Histoire d’Antibes, IVe Colloque International de l’Homme et l’Animal, October 15–17, 1992 (Juan-les-Pins, France: Editions Association pour la Promotion et la Diffusion des Connaissances Archeologiques, 1993), pp. 201–20.
16. J. Bottéro, “Le plus vieux festin du monde,” L’Histoire 85 (1986): 58–65; Bottéro, “La plus vieille cuisine du monde,” L’Histoire 187 (1995): 80–85.
