SEVEN
Were There Scavengers at Torralba?
Healthy debate about the hunting capacity of Lower and Middle Paleolithic foraging peoples continues as strongly now as it did more than two decades ago (González Echegaray and Freeman 1998). The multilevel sister sites of Torralba (Fig. 7.1) and Ambrona (Fig. 7.2) in the province of Soria on the high Spanish Meseta, excavated since the 1960s under the direction of F. Clark Howell (the second more recently reinvestigated by M. Santonja), have been prominent among European Acheulean foci of this discussion, and are probably familiar to most readers because their abundant faunas contain several individual elephants of a very large mid-Pleistocene species, Elephas (Paleoloxodon) antiquus, although other large herbivores such as horses, rhinos, and wild oxen are also present, and in the case of the horse, are as numerous as elephants in some levels (see Howell et al. 1992; Freeman 1994; González Echegaray and Freeman 1998 for recent reviews of these sites and additional bibliography).
The evidence that the human presence at Torralba and at Ambrona is related to the presence of the animals, and that humans actually manipulated the animal remains at both sites, is quite convincing. Preservation is excellent for a mid-Pleistocene site, and several kinds of evidence converge to support that conclusion. Occupations—or, if you prefer, episodes of utilization of the site and deposition of tools and bones—are multiple, rapid, short-term accumulations, sometimes subject to some disturbance or reworking; but usually that disturbance has reoriented materials without destroying their associations. These are not simply lag deposits: that transport of bones and artifacts was apparently not extensive for most levels is shown by the lack of spatial separation of refit lithics and the near-articulated positions of some skeletal elements. While the deposits suggest relatively rapid though sporadic accumulation of the archeological materials, each of the levels distinguished contains the bones of several individual animals of different species, whose carcasses were apparently all in utilizable condition at essentially “the same time”—that is, they could have all been processed at once, and sometimes at least the evidence suggests that to have been the case (body parts of different animals are intermingled, tools or flakes apparently from the same core are found with different animals). Where they can be determined for single levels, age profiles are characteristic of catastrophic accumulations, and the only appropriate catastrophe, given all else that we know of the environment, was pretty certainly human-related. Stone tools (and shaped wooden ones, recovered mostly as plaster casts incorporating patches of charred wood) occur right among the animal remains, and in several cases are intimately juxtaposed to bones in tight physical association. The body parts of some species, including the elephants, have been rearranged selectively in ways that cannot be explained by geological processes or the behavior of the animals themselves (Fig. 7.3). Many of the bones bear gross macroscopic traces of deliberate flaking (Figs. 7.4, 7.5), hacking (Fig. 7.6), and abrasion; some show microscopic traces that have been interpreted as characteristic of slicing with sharp-edged implements; and others are partly charred. The nature of the artifacts supports the suggestion of hunting and/or butchering: cleavers, handaxes, scraping tools, and sharp-edged slicing implements are present among the stone tools, and the wooden implements include one meter-long shaft with a sharp wooden point, that in any other Acheulean site would be interpreted unquestioningly as a spear (Fig. 7.7). Statistical tests provide evidence of consistent relationships between artifacts and all the major animal species represented, and certain kinds of tools and specific body parts are regularly related in abundance in ways that cannot be explained except as the result of cultural choice.
FIGURE 7.1. Overview (from north) of the Torralba excavations (1962); checkerboard of three-meter square excavation units visible in right foreground
FIGURE 7.2. Ambrona, overview (from north) of area opened in 1981; standing walls separate excavated three-meter squares; on-site museum at right
Despite this accumulation of evidence, some scholars, principal among them L. R. Binford, have denied that the animals and the stone tools at these two sites have anything to do with each other. To these critics, the faunal remains are the result of long periods of accumulation of the bones of individuals that died natural deaths from age, disease, carnivore predation, or local disaster, and the artifacts attest the ephemeral presence of hominids passing through the landscape for unrelated purposes. On occasion, of course, the critics admit that these hominids made use of the meat they could scavenge from animals recently dead of natural causes, but they assert that ability to hunt, immobilize, and kill these huge beasts was beyond their limited cultural (technological and organizational) capacity. In fact (although this caricatures their positions) the advocates of a “scavenging phase” of hominid food procurement seem to take the oversimple position that all dietary needs for meat protein were satisfied by scavenging, while those on the other side in the debate seem to take the equally unrealistic stand that no meat was ever scavenged from naturally dead carcasses.
FIGURE 7.3. Bone alignment in Ambrona Lower Unit, Area I (1963); elephant femora and tusk, with perpendicular radio-ulna
FIGURE 7.4. Flaked elephant bone from Torralba
FIGURE 7.5. Flaked juvenile tusk tip from Ambrona (possible point)
FIGURE 7.6. Hacked elephant long bone from Torralba
I find both polarized assertions untenable. In this chapter I continue to insist that the patterns of evidence from Torralba and Ambrona convincingly document animal butchery. But I will go further to assert that while Lower and Middle Paleolithic hominid foragers in mid-latitude Europe may have scavenged (and probably did so) from freshly dead carcasses whenever the opportunity presented itself, scavenging alone could not have provided the regular supplies of digestible animal food that were absolutely required for survival. Consequently, from the time of the earliest hominid presence in such latitudes, foragers had to actively seek, catch or immobilize, and dispatch the prey on which they fed. I have become convinced of this position after listening to decades of debate, and I insist that though it is moderate it is not really a middle ground: the idea of a purely “scavenger phase” of hominid foraging for Lower and Middle Paleolithic peoples in mid-latitude Europe is, I firmly believe, simply untenable. Of course, I could be wrong. But my reading of the literature (though I certainly do not claim to know it all) does not convince me that I am. Please understand that I make no claim to present anything like a comprehensive review of the relevant literature, nor do I pretend to firsthand familiarity with the relevant African analogues; I hope to learn a great deal in that respect from the presentations of colleagues such as Profs. Fisher and Frison, who have that personal experience.
The difficulty of demonstrating the “hunting” position is manifest. None of the evidence from Torralba or Ambrona that I have cited above—or from any other mid-Pleistocene site I know—can be used to prove hunting, unless perhaps the spearpoint is thought to do so. Some other European Acheulean sites have yielded wooden spears: Lehringen and Schöningen (Thieme 1996, 1997) are among the sites where such pieces were recovered. Another kind of evidence, perhaps the best for hunting to date, comes from Boxgrove, where Mark Roberts reports finding a spear wound in a horse scapula, in a context that includes other evidence suggestive of human hunting (Roberts 1996). Even that case, which seems quite compelling, might conceivably be challenged, claiming it resulted from the use of a sharpened lever to pry muscles and bone apart during the butchering process, rather than from weapon use. But conclusive evidence of hunting is difficult to obtain throughout prehistory, even in sites where no one has ever doubted that bone accumulations resulted from deliberate hunts. Most hunted animals die of poison, infection, internal bleeding, or destruction of vital organs. Only rarely does a weapon point become embedded in bone. Due to the decay of soft body parts (and at least the shafts of most hunting weapons), it is very difficult to prove that any butchered carcass from a Paleolithic site was actually killed by humans. The evidence for hunting remains mostly indirect. Such factors as an accumulation of carcasses of animals of several species, with individuals from all active age ranges, in approximately their expectable proportions in living herds, might under certain circumstances provide evidence of hunting. (Such “catastrophic” distributions can, of course, also be produced by natural die-offs, whose causes would first have to be excluded.)
FIGURE 7.7. Plaster cast of wooden point, with associated elephant pelvis, from Torralba
In short, it is hard to imagine that direct evidence of hunting as a deliberate and consistent cultural pattern, of the sort that would silence all objection, will ever be obtained from any mid-Pleistocene site. Much of the debate about hunting or scavenging must continue to be based on indirect evidence, on theoretical considerations, and on the study of living creatures. However, not everything is possible. There is evidence in the anatomy and physiology of modern humans, the skeletal anatomy of our earlier ancestors, the behavior of our close primate relatives, and the nature of past environments that constrains the possible food-procurement alternatives that must have been available to mid-Pleistocene foragers.
HOMINID DIETARY REQUIREMENTS: THE EVIDENCE OF ANATOMY AND PHYSIOLOGY
Our anatomy (and that of our hominid ancestors, as far as the skeletal evidence goes) reflects a dietary pattern that includes the ingestion of meat. In the anatomy of their dentition and digestive apparatus, humans today are adapted to be omnivores—to digest meat as well as certain kinds of plant foods. To judge from the fossil record, hominids have been omnivores for the last two and a half to three million years at least. While we and our ancestors lack the anatomical equipment of specialized carnivores, we have long been facultative carnivores—able to use meat as a dietary staple, supplemented by other foods, whenever it became available.
Protein from meat has many dietary advantages: it builds muscle rapidly, and is a high-quality source of energy; weight for weight, it takes less meat to provide a given amount of usable protein than it does nuts, seeds, or the other good vegetal protein sources. If prey stomach contents and organ meats are eaten, a hominid can survive completely without the need for vegetal foods, in most environments. A strict vegetarian must spend much more time and energy foraging to satisfy dietary requirements than does an omnivore who balances vegetal intake with meat protein. In these respects, a totally meatless diet even if practical would be an inferior diet for hominids. Since the 1950s many paleoanthropologists have recognized that the consumption of meat was an important part of the adaptive niche developed and occupied by the ancestors of modern Homo sapiens sapiens.
In fact, a strict vegetarian diet is not simply impractical, but impossible, for foraging groups of Homo sapiens. Without modern pharmaceuticals, modern humans (and there is no reason to believe that their close mid-Pleistocene ancestors and relatives were significantly different in this regard) must ingest regular quantities of animal food in order to survive. We require regular supplies of trace quantities of cobalamine (vitamin B-12) which is unavailable (at least in adequate quantity) from any wild vegetable source (see Berkow 1992—it may, however, be possible to get needed quantities from some cultivated legumes). Unless this substance is regularly ingested, anemia and relatively rapid death ensue. Modern strict vegetarians, even those who use legumes as a dietary base, must acquire cobalamine (in vitamin supplements, for example) or they develop what medicine knows as “vegan anemia.”
Ingesting carrion is a risky way to meet this dietary need. It is no accident that most specialist scavengers have digestive systems that tolerate toxins produced during the decay of meat. Bacteria present in the ground or in the guts and on the skins of animals spread rapidly as a carcass decays, and the decay process is faster in warmer climates. Bacterial metabolism may be associated with the production of toxic chemical by-products. Hominids lack the scavenger’s physiological defenses against many of these spoilage microorganisms. Consequently, meat that has decayed beyond the initial stages is not food but poison to us, and there is no reason to believe that such was not the case for our earlier hominid relatives as well. Our ability to acquire resistance to such meat-spoilage bacteria or carcass contaminants as Streptococcus, Staphylococcus, Salmonella, Shigella, Escherichia, and Clostridium botulinum (which can be ingested from contaminated carrion) is limited and in some cases nonexistent (see Burrows 1963). Once ingested, some of these bacteria, or their by-products, remain in the system for weeks, and contact even with resistant individuals will infect others. We also know that ingestion of some animal parts, such as brains or offal, risks exposure to viral infections like kuru or such sicknesses as mad cow disease (a fact that may be less important to earlier hominid evolution, since death or debilitation from the resulting disease may take quite a long time).
How did early hominids obtain animal food, once its consumption entered their survival strategies? Aside from trade with or raiding of their neighbors for food, or outright cannibalism, before domestication there were only three ways. One, of course, is the acquisition of meat by scavenging from carcasses abandoned by predators, who had either fed to satiation or been driven away from their kills. The second is the deliberate hunting of mobile animals. These views are often polarized, their proponents suggesting either that all meat protein consumed was obtained by hunting or that, at least for the first millions of years of hominid history, none of it was. This polarization is unnecessary and unreasonable. In the last analysis all would agree that hominids obtained some meat protein in yet a third way: by catching and eating small, relatively immobile animal resources such as insects and larvae, eggs and fledging birds, snails, tortoises, newborn mammals, and in general any creature weak enough or slow enough that it could not escape. The debate about hunting really centers on the hominid ability to take larger, fleeter prey. L. R. Binford (1981, 1983, 1984, 1985, 1987) has been a vocal exponent of the viewpoint that well into the Upper Pleistocene hominids were unable to hunt large game. However, his seems an extreme position.
BEHAVIORAL EVIDENCE FROM LIVING PRIMATES
Primate ethology suggests that facultative hunting, even of relatively large animals, is not beyond the range of behaviors we might expect from an early hominid. There are relevant data from the behavior of living, free-ranging primates, who require no language, no very complex organizational skills, nor any sophisticated technology to capture and kill other animals. By now, everyone knows that our closest primate ancestors, the chimpanzees, are facultative hunters who by cooperative action are able to capture and devour such quick-moving creatures as monkeys or young antelope. Among chimps and baboons, both of whom are facultative hunters and apparently enjoy meat protein, the extent of scavenging is apparently small.
According to an early study (on baboons, certainly less capable than chimpanzees) by S. W. Washburn:
all of the information that we have on primate hunting . . . suggests that these animals will take eggs, young birds, and other small, living creatures, but that they do not scavenge. The evidence shows that hunting small, easily captured prey is far simpler and more nearly universal than is scavenging. Besides, scavenging from large carcasses when carnivores are nearby can be exceedingly difficult. . . . It is said that it is easy to drive a lion from its kill, but this is only true in areas where lions are accustomed to being hunted and are trained to stay away from man.
Finally, scavenged meat is a rare occurrence by comparison with meat that is easy to hunt. In an area in Africa in which there are many lions, there are still hundreds of baboon per lion; so that if the baboons were to try to live on scavenged meat, they would have a hard time getting an ounce of meat per day per baboon. The most minimal hunting of easily killed animals is more rewarding than this. (Washburn 1968: 342)
Carrion in edible condition is a rare commodity, as anyone who does much hiking in the wilderness will realize. Even road kill, much more abundant along U.S. highways than carnivore kills, would not provide enough accessible food to support a small group of human foragers. Barring a cataclysm, live animals are always more abundant than recent kills.
Craig Stanford (1995: 261) observed that chimps in the Gombe reserve show little interest in scavenging. Geza Teleki also studied chimpanzee hunting there, and with the exception of stealing bushbuck carcasses from baboons, seconds or at most minutes after they were captured—the chimps had often observed (and been excited by) the hunt in progress—he noted no examples of true scavenging. In fact, in experiments, chimps would not treat carcasses of species they normally preyed on, offered them by the experimenter, as food. He comments: “in view of these observations I am puzzled that the original notion of a scavenger phase in hominid evolution has recently regained popularity among scholars concerned with the hominid fossil record. . . . Evidence to the contrary has been appearing regularly over the same span of years . . . but with little apparent effect” (Teleki 1981: 329).
BEHAVIOR OF NON-PRIMATE SCAVENGERS
Other observations of relevance come from the study of non-primate mammals who are known as efficient scavengers. The spotted hyena is probably Africa’s best-known (and best-adapted) mammalian scavenger. Hans Kruuk spent three and a half years studying their behavior in Serengeti National Park and the Ngorongoro Crater. He found that even in the Serengeti, where the proportion of scavenging was relatively high, 68 percent of animals eaten by hyenas were hunted and killed by hyenas. The reason that the importance of hyena hunting had not previously been observed is that hyenas are nocturnal hunters (Kruuk 1972: 111). Hyenas lurking about lions feeding from a carcass were in fact often not scavengers awaiting their chance at the meat, as would popularly be thought, but hunters themselves that the lions had driven from their own kills. Jackals and wild dogs, too, kill much of the meat they feed on, though each will also scavenge, as will lions. In fact, in Africa each of the major carnivores kills or scavenges depending on opportunity (Schaller and Lowther 1969: 325).
Most dedicated scavengers have developed special abilities that facilitate the utilization of kills made by other animals. They locate carrion using highly developed senses. The eyesight of buzzards and other flying carrion-eaters is extremely acute. Terrestrial scavengers often rely on especially sharp olfaction. All specialized scavengers move rapidly about a great deal of territory in their search for food. In contrast, hominids are not particularly mobile, and neither they nor any of their closest relatives possess the visual or olfactory acuity of a competent scavenger.
Hominids are, and to judge from our relatives, the apes, probably always were, more intelligent and their behavior more flexible than the social carnivores. It is relatively easy for higher primates—as for social carnivores—to hunt and kill small animals, or the young or incapacitated individuals of medium-sized species. It is much harder and more dangerous for them to drive large predators away from their kills. While today it is said not to be particularly difficult for hunters to drive large predators from their kills, Homo sapiens sapiens is considerably better equipped in intelligence, foresight, weaponry, social communication, and cooperation than were our earlier ancestors. Early hominid scavenging would almost certainly have had to be passive scavenging—waiting until a carcass was abandoned by its predators, rather than driving them off.
THE AVAILABILITY OF CARRION: THE SERENGETI
One must consider other lines of evidence before deciding on the relative likelihood of a scavenging adaptation. Studies of the availability of edible meat from scavengeable carcasses have produced other data of importance to this question.
Schaller and Lowther (1969: 325–30) reported the results of a brief study undertaken in the Serengeti, during the dry season. In two transects, one across the plains, the other in woodlands, they found the remains of several carcasses, some neonates, and two sick or crippled animals. They concluded that while meat-eaters might survive over the short term by killing and eating disabled beasts as well as feeding on carrion from predator kills, scavenging alone would not provide a sufficiently regular and predictable food supply for survival over the long run. For long-term survival, a carnivorous hominid group would have had to combine scavenging with the killing of incapacitated or very young animals.
The suggestive study of carcass availability (in the same general region) undertaken by R. Blumenschine (1987) extends these results. He found that the potential of the Serengeti for scavenging from predator kills depended on the nature, size, and density of herbivore species; the nature of locally abundant carnivores; whether the area is riparian forest (where water is generally available, lions abound, and spotted hyenas are rare) or open grassland (with dry-season water holes); the time of year; and the extent of competition among predators for meat.
One of the principal predators in his study, the spotted hyena (Crocuta crocuta), can crush open and chew long bones and skulls of middle-sized animals, digesting the bones as well as their contents, and leaving little for the scavenger. All carnivores except the spotted hyena soon stripped the flesh from kills but left long bones and skulls intact. Where there were few spotted hyenas, or whenever long bones and skulls were abandoned whole, a scavenger who could smash the bones would find a good source of protein therein. As one might expect, the carcasses of larger animals by and large provide more potentially usable food for scavengers than do smaller animals. In the southern plains of the study area, where large, migratory herds were present in the dry season, seasonal scarcity of water forced herbivores to congregate around waterholes. This could result in a surplus of food for predators: even spotted hyenas, who usually leave nothing for other scavengers, under those circumstances would sometimes satiate themselves and abandon carcasses that still had meat on them. Usable carcasses of medium-sized animals (the size of adult wildebeests or zebras) would have been relatively more abundant at those times; according to Blumenschine, even if most meat had been stripped from carcasses, intact marrow bones and skull contents would nevertheless be more available for a scavenger in such conditions.
In sites such as Olduvai, on the Plio-Pleistocene boundary, Blumenschine estimates that large herbivore biomass should have been greater than it is today. Speculating that saber-tooth cats, supposed to be less complete consumers of large carcasses, might have been relatively common during the Early Paleolithic in riparian woodlands (where hyenas are comparatively rare) led him to suggest that marrow bones and meat on large carcasses would probably have been even more readily available at such times and places. So Blumenschine sees considerable potential in that environmental setting for an early hominid opting to get animal protein by scavenging.
In a later paper (1989), Cavallo and Blumenschine report that tree-stored leopard kills would add to the meat available for hominid exploitation, though only prey smaller than antelope-sized would persist for more than about an hour (a pattern unlike that of carcasses on the ground). While he tells us that carcasses of larger animals on the ground persisted for several hours or up to as many as four days, small animals usually being devoured in minutes or hours, Blumenschine provides no information on the rapidity of putrefaction, or the toxicity of tissues at various periods after death. These are essential questions about the time carcasses would provide tissues in digestible condition. Nor are absolute quantities of usable tissue reported. Just how much reliance could be placed on scavenged meat protein as a dietary mainstay in the Serengeti situation remains unclear. And, useful as Blumenschine’s paper is for modeling hominid behavior in one sub-Saharan African region at the Plio-Pleistocene boundary, none of his observations is directly applicable to the Spanish case.
Selvaggio (1998a) reports that spotted hyenas cache parts of the carcasses of animals they have obtained in the shallows of Lake Macat (in the Ngorongoro crater) and may abandon them for as much as one or two days; she observed that the meat appeared, at least on superficial observation, to remain fresh during that time, and would have been available to scavenging hominids. She rightly suggests that hominid scavenging of meat or bone from such caches could account for the accumulations of cutmarked bones at some African Plio-Pleistocene sites. However, because of the size discrepancy of the animals from Selvaggio’s caches and those from Torralba/Ambrona, hyena caching does not seem to be a viable explanation for the bulk of those Spanish accumulations.
CARRION AVAILABILITY IN MID-LATITUDE EUROPE
The Spanish Meseta, where most Acheulean sites (such as Torralba and Ambrona) are found, is a mosaic of ecosystems having little in common with the Serengeti (or the Ngorongoro crater). There are today some 30 species of ungulates in the Serengeti, including many antelopes, while there were no more than a third this total number in mid-Pleistocene Spain. Large herbivores in both areas include(d) elephants and rhinos. Authors differ in their estimates of animal populations, but all counts agree that animals are extremely abundant in the Serengeti. There, in moist savanna, large herbivore biomass may run from 8,000 to 10,000 kilograms per square kilometer (Delaney and Happold 1979: table 11.14)—biomass for all herbivores is substantially greater—and in woodlands runs perhaps a fifth as high. Again, when all herbivores are considered, the biomass is larger, rising to some 5,000 kilograms per square kilometer (Hendrichs 1970). (As one might expect, these average values fluctuate with the seasons, and there are longer-term fluctuations as well.) When migratory ungulates are present on the Serengeti plains, there may be 220 of them per square kilometer, while in woodlands, the density is less than a tenth of that (Houston 1979: 268). There are about a half dozen middle-sized to large carnivores. Their numbers, according to the best survey, are relatively high—there may be 7,000 of them in the reserve as a whole, and biomass for the five largest predators is 14–16 kilograms per square kilometer (Schaller 1972: 454). (Schaller, incidentally, notes that adult rhino and elephant are too large to be manageable prey for lions.) The number of available carcasses for use by scavengers has been estimated at one per 33 square kilometers in plains areas in the wet season—this drops to one per 300 square kilometers in the dry season—and one per 412 square kilometers in the woodlands (Houston 1979: 268). Although these averages do not take into account dry-season aggregation around waterholes, they do not suggest that life as an exclusively scavenging hominid with dietary reliance on a regular supply of meat would have been particularly easy.
In Europe, in contrast, ungulate biomass was always substantially smaller. Though precise estimates are impossible due to human interference with ecosystems, figures on the order of 500 to 1,000 kilograms per square kilometer for mixed woodlands and 3,500–5,000 kilograms per square kilometer for mid-latitude grasslands seem as large as is reasonable (see, e.g., Bourlière 1964). In the European case, there were at any period about as many species of large carnivores as in the Serengeti; among them lions, wolves, bears, and hyenas (and earlier, saber-teeth) were the principal figures. But their numbers were a small fraction of those in the African area. By the time sites like Torralba and Aridos were occupied, saber-teeth were either very rare or completely absent from the Spanish landscape. No contemporary (mid-Pleistocene) carnivore in Spain was large enough to attack adult elephants of the size of the E. antiquus or the rhinos from Torralba and Ambrona (nor is hyenid water-caching of body parts of these large animals a serious possibility).
Seasonal scarcity of water (or moisture in the form of snow) was not a limiting factor for large mammals. The lean season would have been the winter, as it is today. Winter cold, deep snow, scarcity of edible vegetation, and difficulty traveling imply reduced carrying capacity for herbivores, particularly during glacial phases. Climate was extremely severe, with much colder winters than at present, when Torralba and Ambrona were utilized. These critical factors limited the size of animal populations and the internal variety of communities. There were never as many large herbivores as there were in the Serengeti, nor as many different kinds of them. Limiting herbivore populations restricts the number of local predators they can support. Consequently, fresh carcasses of animals killed by carnivores must always have been rarer, harder to find, and further between than in the African case. If kills were scattered more sparsely over the landscape, hominids would have had to travel considerably farther to find carcasses than in the Serengeti case. The density of usable carcasses could not have been more than a tenth to a fifth as great in Spain. It is highly unlikely that specialized scavenging could have been a viable adaptive strategy in these environmental conditions.
The winter scarcity of plant food would have affected hominids as much or more than it did true vegetarians—hominids cannot digest grasses, whether fresh or dried as hay, as can a specialized herbivore. Since hominids do not hibernate, eating was a yearlong necessity. Food storage is one possible way out of this dilemma, but there is absolutely no evidence that storage of any kind was practiced until much later in the Spanish Paleolithic. There are no potential storage facilities in any Spanish Acheulean site. While pits have been found in rare Mousterian excavations, there is not the slightest evidence that they were used for long-term food storage (or that they were used to store anything edible other than meat). Under the circumstances, meat protein would have been a dietary necessity, and if it could not regularly be obtained by scavenging it must have been procured by hunting. If there ever was a scavenging phase of human subsistence, it seems probable that it had long vanished before the colonization of mid-Pleistocene Europe.
In short, while the simple answer to the question posed by the title to this chapter is yes, there were almost certainly scavengers at Torralba (and Ambrona), I suggest that the animals found in these two sites must almost inevitably include many that were deliberately killed by mid-Pleistocene human hunters.
POSSIBLE HUNTING METHODS
At one time it was very generally accepted that early hominids, camped in places like Olduvai Gorge, obtained some animal food by killing tortoises and other small animals including neonates and young of larger mammalian species. In fact, even though there has more recently been a tendency to explain the remains of these animals as the result of death by natural causes, kills by hominids remain quite plausible, and well within the behavioral repertoire of our living higher primate relatives. On the other hand, the hunting of adults of such huge creatures as the Torralba/Ambrona elephants seems an impossibly daunting challenge for hominids with such rudimentary technologies as those indicated by the Acheulean evidence. Hunting elephants is not without danger to the modern gunman armed with a large-bore rifle; by what possible means could such large and powerful creatures have been immobilized and killed by our primitive ancestors? One suspects that such ethnocentric and a priori ruminations underlie much of the argument for a long-enduring scavenging phase of hominid evolution. To the contrary, however, the literature of exploration and reports of observations of African hunting methods suggest that big game was until recently being hunted successfully with very rudimentary equipment.
Techniques observed in use to capture game by skillful elephant-hunting groups without firearms included dropping heavily weighted spears from trees, spearing them repeatedly with small-diameter spears, stabbing them from below with large metal spears (in all these cases, the weapon point could be smeared with poison), hamstringing them with swords from horseback, cutting off their trunks, catching them in footsnares or pitfalls, and net hunting. We know of no pitfalls earlier than the Upper Paleolithic, however, and no technique requiring a metal point or blade was then available. Of this list, only the use of wooden spears, footsnares, and net-hunting remain as possibilities. But there is one other method that is widely reported and that would have been both possible, suitable, and devastating: the use of fire either to surround and burn animals or to drive them into situations, such as the mucky shallows of lakes, from which they could not readily escape. Sir Samuel Baker observed this technique before 1890, seeing large numbers of elephants and other animals driven into a narrowing ring of fire, which left them half suffocated by smoke, badly burnt, and often blinded. The technique was so effective in taking whole herds that Dr. G. Schweinfurth, who saw it employed by the Azande (1873), feared that its repeated use would lead to the extermination of the species (Cloudsley-Thompson 1967; R. Carrington 1962). Henry Stanley (1890: 339) reported seeing vast heaps of bones of slaughtered game in a circle some 300 yards in diameter on the shores of Lake Albert. The bones included remains of animals of many kinds, from elephants to bushbuck. He attributed this accumulation to the familiar practice of ringing the animals with fire. The use of fire drives has long seemed to me to be the most probable technique for trapping and immobilizing elephants along the shores of the Ambrona lagoon or the riverbanks of the Spanish Meseta. It would account for the simultaneous occurrence of bones of several other species in the accumulations better than any alternative I can envision.
DIRECTIONS FOR FUTURE RESEARCH
While I am as sure of my ground as I may be given the current state of knowledge, I am under no illusion that this essay or this symposium will resolve the hunting/scavenging debate to everyone’s satisfaction. I hope that papers by other participants, particularly Fisher, Frison, and West, will help clarify the issues. But I anticipate that much further research will be needed to arrive at any resolution. I believe that we can now see some of the directions that this research must take.
It is in my opinion fruitless to hope to find convincing evidence of hominid hunting in marks left on bones. If every bone recovered bore convincing butchery marks, that would still not indicate that humans had hunted the butchered animals. Even finding an apparent spear point embedded in bone would probably only result in arguments over the function of the artifact type; as we all know, most stone artifacts that could be used as spear points could equally well have been used as knives. The best associations between stone tools and animal remains will remain unsatisfactory as evidence for human hunting.
Before the study of Paleolithic butchery methods can progress, we must also clarify our thought about the study of butchering traces or any other kind of tool marks on mid-Pleistocene bone. Ancient bone isn’t fresh bone; pristine and unaltered tool marks or other traces of working or utilization on mid-Pleistocene bone are not to be expected, and when they do occur are likely to be both rare and debatable.
We must also abandon the assumption, held by many Paleolithic prehistorians, that any indication whatever of carnivore involvement in altering a bone assemblage completely rules out any human agency. Wolves will scavenge carcasses hunted by humans today, and did so in the recent past. Carnivores fed on and gnawed remains of carcasses left behind after humans removed what they wanted from hunted game. Why then do some prehistorians/paleoanthropologists find it so hard to conceive that carnivores could have behaved in like fashion earlier in the Pleistocene? More subtly, we must recognize that while residues of carnivore behavior may sometimes mimic apparent results of human behavior so closely as to confuse or mislead us, it is equally true that patterned human behavior can mimic carnivore activity: for example, technology may limit the hunter, or cultural choice dictate that only very young or incapacitated animals be taken as prey.
The study of modern analogues remains the most productive single line of approach to a resolution of the hunting/scavenging debate, but in future, data gathering must be more systematic and more precisely controlled, and comparisons subjected to reasonable constraints. We must learn more about the abundance of large mammals, about the ways they may be hunted, about the behavior of predators and scavengers, and about what they leave behind in places like the African grasslands and forests where one can still observe elephants and other large mammals in interaction under largely naturalistic conditions. It is essential to complement observations of carcass availability with chemical and bacteriological analyses, to show that the apparently available caracasses would have been digestible and not harmful to hominids. We must also learn to avoid the Bushman pitfall: the fallacy of assuming that all elephants, environments, scavengers, or hunters must be identical to the one case we are familiar with or the one that has most recently been popularized. African analogues are never going to be a perfect fit to the European data, but as long as the differences are recognized and their effects correctly evaluated, the comparisons and contrasts we find will be increasingly enlightening.
Advancing our knowledge of the range, limits, and development of early hominid subsistence and environmental utilization will require the convergence of several different lines of evidence. Better paleoenvironmental data, refined calibration of the duration of accumulations, an increasingly scrupulous examination of artifacts and associations, more attention to details of site context, more careful calculation of minimum individual estimates and mortality profiles, and tighter control in the study and comparison of modern analogues to ensure their relevance are all needed. More research on the staging of hominid and carnivore alteration of bones, where both are present, is sorely needed. Recent work of this sort by Blumenschine and Selvaggio (1991; Selvaggio 1994, 1998b) is an excellent beginning, but only a beginning. None of these investigations will resolve the hunting/scavenging discussion, and some will only bear on it indirectly, but if all are taken together they will certainly provide a more accurate understanding of individual cases, and well-analyzed cases will lead in the aggregate to more realistic and reliable reconstructions of the socio-economic behavior of our early ancestors.
REFERENCES
Baker, S. W. 1890. Wild Beasts and Their Ways: Reminiscences of Europe, Asia, Africa, and America. Vol. 1: 28–148. London, Macmillan.
Berkow, R., ed. 1992. The Merck Manual, 16th ed. Rahway, Merck Research Laboratories.
Binford, L. R. 1981. Bones: Ancient Men and Modern Myths. New York, Academic Press.
———. 1983. In Pursuit of the Past. London, Thames and Hudson.
———. 1984. Faunal Remains from Klasies River Mouth. New York, Academic Press.
———. 1985. Human Ancestors: Changing Views of Their Behavior. Journal of Anthropological Archaeology 4: 292–327.
———. 1987. Were There Elephant Hunters at Torralba? In The Evolution of Human Hunting, ed. M. Nitecki and D. Nitecki, 47–105. New York, Plenum.
Blumenschine, R. 1987. Characteristics of an Early Hominid Scavenging Niche. Current Anthropology 28: 383–407.
Blumenschine, R., and M. Selvaggio. 1991. On the Marks of Marrowbones by Hammerstones and Hyenas: Their Anatomical Patterning and Archaeological Implications. Proceedings of the 11th Congress, Union International des Sciences Pré- et Protohistoriques, Mainz, 1987.
Bourlière, F. 1967. The Natural History of Mammals, 3rd ed. New York, Alfred A. Knopf.
Burrows, W., ed. 1968. Textbook of Microbiology, 19th ed. Philadelphia, W. B. Saunders.
Carrington, R. 1962. Elephants. London, Penguin.
Cavallo, J., and R. Blumenschine. 1989. Tree-Stored Leopard Kills: Expanding the Hominid Scavenging Niche. Journal of Human Evolution 18: 393–99.
Cloudsley-Thompson, J. L. 1967. Animal Twilight, 11–50. London, G. T. Foulis & Co.
Delaney, M., and D. Happold. 1979. Ecology of African Mammals. London, Longman Group Ltd.
Freeman, L. 1981. The Fat of the Land: Notes on Paleolithic Diet in Iberia. In Omnivorous Primates: Gathering and Hunting in Human Evolution, ed. R. Harding and G. Teleki, 104–65.
———. 1994. Torralba and Ambrona: A Review of Discoveries. In Integrative Paths to the Past: Paleoanthropological Advances in Honor of F. Clark Howell, ed. R. Corruccini and R. Ciochon, 597–637. Englewood Cliffs, Prentice-Hall.
González Echegaray, J., and L. Freeman. 1998. Le Paléolithique Inférieur et Moyen en Espagne. Grenoble, Jerome Millon.
Hendrichs, H. 1970. Schätzungen der Huftierbiomasse in der Dornbuschsavanne nördlich und westlich der Serengetisteppe in Ostafrika nach einem neuen Verfahren und Bemerkung zur Biomasse der anderen pflanzenfressenden Tierarten. Säugetierkundliche Mitteilungen 18: 237–55.
Houston, D. 1979. The Adaptations of Scavengers. In Serengeti, Dynamics of an Ecosystem, ed. A. Sinclair and M. Norton-Griffiths, 263–86. Chicago, University of Chicago Press.
Howell, F. C., L. Freeman, K. Butzer, and R. G. Klein. 1992. Observations on the Acheulian Occupation Site of Ambrona (Soria Province, Spain), with Particular Reference to Recent (1980–1983) Investigations and the Lower Occupations. In Die altest Besiedlung Europas, ed. G. Bosinski.
Kruuk, H. 1972. The Spotted Hyena. Chicago, University of Chicago Press.
Mann, A. 1981. Diet and Human Evolution. In Omnivorous Primates: Gathering and Hunting in Human Evolution, ed. R. Harding and G. Teleki, 10–36. New York, Columbia University Press.
Roberts, M. 1996. “Man the Hunter” Returns at Boxgrove. British Archaeology 18.
Schaller, G. 1972. The Serengeti Lion. Chicago, University of Chicago Press.
Schaller, G., and G. Lowther. 1969. The Relevance of Carnivore Behavior to the Study of Early Hominids. Southwest Journal of Anthropology 25: 307–41.
Schweinfurth, G. 1873. The Heart of Africa. London, Sampson Low, Marston, Low & Searle.
Selvaggio, M. 1994. Carnivore Tooth Marks and Stone-Tool Butchery Marks on Scavenged Bone: Archeological Implications. Journal of Human Evolution 27: 215–28.
———. 1998a. The Archaeological Implications of Water-Cached Hyena Kills. Current Anthropology 39(3): 380–83.
———. 1998b. Evidence for a Three-Stage Sequence of Hominid and Carnivore Involvement with Long Bones at FLK Zinjanthropus Olduvai Gorge, Tanzania. Journal of Archaeological Science: 191–202.
Stanford, C. 1995. Chimpanzee Hunting Behavior and Human Evolution. American Scientist 83: 256–61.
Stanley, H. M. 1890. In Darkest Africa. New York, Charles Scribner’s Sons.
Teleki, G. 1981. The Omnivorous Diet and Eclectic Feeding Habits of Chimpanzees in Gombe National Park, Tanzania. In Omnivorous Primates: Gathering and Hunting in Human Evolution, ed. R. Harding and G. Teleki, 303–43. New York, Columbia University Press.
Thieme, H. 1996. Altpaläolitische Wurfspeere aus Schöningen, Niedersachsen: Ein Vorbericht. Archäologisches Korrespondenzblatt I 26: 377–93.
———. 1997. Lower Paleolithic Hunting Spears from Germany. Nature 385: 807–10.
Washburn, S. W. 1968. Comment in Discussion of Part VII. In Man the Hunter, ed. R. Lee and I. DeVore, 342. Chicago, Aldine.