lithics, 33 ivory points
source: After G. Haynes 2002. Used with the permission of Cambridge University Press.
source: After G. Haynes 2002. Used with the permission of Cambridge University Press.
of miles from the cache site (Tankersley 2002). In fact, apart from mammoths, it is remarkable how rarely the fossils of extinct Quaternary animals are found in thoroughly convincing association with prehistoric human artifacts (see table 7). The lack of associations is especially noteworthy because Clovis points are well represented in the fossil record, particularly in the plow zone of the Midwest.
To date no Clovis points or other early human artifacts are known to be associated with ground sloth remains. One famous claim of such an association at Gypsum Cave (Harrington 1933) was based on stratigraphy and has not been verified by radiocarbon dates (Heizer and Berger 1970). The wooden darts or shafts from Gypsum Cave yielded radiocarbon dates at least 8,000 years younger than the sloth dung in which they were embedded. No prehistoric artifacts have, to my knowledge, been found in Rampart Cave. (Clovis artifacts are rare in caves in general, and none have been discovered in any Grand Canyon caves.) The fossil record thus yields no direct evidence of hunters having killed ground sloths, much less of young people using the animals for target practice, as I have suggested. Convincing camel kill sites also have yet to be discovered. (As discussed in chapter 7, hopes for such a site at Tule Springs were dashed.) The many southwestern fossils of horses and mastodons found to date also have no Clovis or other archaeological associations. Only rarely does extinct North American megafauna turn up in archaeological sites. Before seizing upon that fact as evidence that humans did not kill off American proboscideans, it is worth noting that "The United States contains more megamammal killsites than there are elephant killsites in all of Africa—a land mass that is much larger than the United States" (G. Haynes 2002b, 183; emphasis in original).
The record regarding bison is somewhat different. Clovis hunters clearly hunted bison at Murray Springs and other sites in the West, leaving their points with the bones. Overall, there are far more associations of human hunters with bison than with mammoths or mastodons. The vast majority of bison kill sites, however, are younger than Clovis and feature Folsom points, especially in the High Plains to the east of Arizona and the Rocky Mountains. Some archaeologists take this as a further argument against overkill. As my archaeologist friend Jim Hester put it in about 1965, voicing a view widely held by archaeologists, "How can you invoke overkill to explain extinction of the mammoth when . . . we have many more younger sites . . . associated with bison kills? They stretch over thousands of years, from Folsom on to Midland. In contrast, there are only a few Clovis associations with mammoth. Yet the bison survived and mammoths are gone. If people did it, the field evidence is backwards." The answer to Jim's question is important. In the fossil record a catastrophic event that results in extinction may not last long enough to accumulate appreciable evidence. If hunting is sustained over thousands of years, a great deal of field evidence of past bison hunts can be expected, but if other plant and animal resources are limited, bison as the only large and widespread prey is not as vulnerable to extinction. In this "arms race," bison become wilier at avoiding hunters, who in turn learn better hunting techniques. After European contact in the late 1400s, disease swept through Native American populations. In a century bison began to spread into the southeastern United States and into New York State, where they had not been for thousands of years. Their numbers dropped again as European settlers began to hunt them and the Anglo frontier pushed buffalo back west.
As for cave paintings, according to Gary Haynes, "There are no known cave paintings, portable artwork, carved figurines, or petroglyphs that clearly and unambiguously portray Clovis-era images" (G. Haynes 2002b, 158). In contrast, extinct mammals, especially mammoths and woolly rhinoceroses, along with living species, are magnificently portrayed in the artistic galleries drawn by Paleolithic people in Europe. They are depicted not only on rock shelter or cave walls and ceilings but also on ivory or bone artifacts. Such ancient art displays superb knowledge of animal behavior and morphology. For example, a female reindeer is portrayed sniffing a newborn, as reindeer and caribou do to determine their infants' parentage (Guthrie 2005). Some of the finest drawings, such as those in Chauvet Cave in France (Chauvet, Deschamps, and Hillaire 1996), are over 30,000 years old. With the exception of a widely reproduced rock drawing of a putative proboscidean in Utah and perhaps one other, there is nothing in the New World to suggest a lengthy association with mammoths and other extinct species. American rock art portrays living animals only and is considered mainly postglacial, and therefore post-Clovis, in age.
Cultural associations with extinct animals have also been difficult to find in other areas outside Afro-Eurasia, the continents of human origin. Australia's Willandra Lakes region, for example, is rich in archaeological remains, especially those of Holocene age, and in paleontological sites with extinct fauna. To my knowledge, however, there is no overlap between the two, though the human presence at Lake Mungo goes back over 30,000 years. At Lancefield, in the state of Victoria, an incredibly rich deposit of 10,000 extinct kangaroos is associated with charcoal originally radiocarbon dated at 26,000 years and now considered older. Investigators Richard Wright and Dizzy (Richard) Gillespie found no decisive cultural association and no indisputable evidence of butchering of the kangaroos. Similarly, few definite cultural associations with extinct species are found in Madagascar. The same is true of the Mediterranean islands and Pacific islands, including Hawaii, where the Polynesians arrived only about 1,500 years ago (Martin and Steadman 1999, 24, 26).
David Steadman, Greg Pregill, and Dave Burley (2002) have described a unique association at the Tongoleleka site on the island of Lifuka in the Tongan archipelago. There, bones of an extinct megapode or bush turkey, Megapodius alamentum, occur adjacent to those of a giant extinct iguanid (Brachylophus, undescribed sp.) as well as pottery of the pioneering Lapita culture and bones of introduced chickens (Gallus). The associated fossils are virtually the same age: weighted mean average of calibrated radiocarbon ages in calendar years are 2,780 to 2,750 for six dates on the chicken bones, 2,840 to 2,760 for six dates on the iguana, and 2,950 to 2,780 for eight dates on the megapodes. This is a fabulous find, nailing the association of extinct animals with human activity, exactly what is not easy to uncover 10,000 years earlier in Clovis time in North America. Perhaps, with time, more such associations will be found, although very few have turned up in a quarter century of remarkable success at discovering new taxa of insular fossils, many of extinct land birds.
Only on the South Island of New Zealand are prehistoric human artifacts abundantly associated with extinct fauna. There the overlap between evidence of human occupation and bones of 10 species of moa teetering on the edge of extinction covers less than 100 years (Holdaway and Jacomb 2000; Worthy and Holdaway 2002).
As far as direct archaeological evidence is concerned, therefore, the most we can say for the Americas is that Clovis hunters overlapped with mammoths and bison and at least on occasion they hunted both. If human artifacts are not found with bones of ground sloths, horses, and extinct camelids, and rarely with mammoths, why should archaeologists pay attention to the view that people forced the extinctions (Meltzer 1993)? Zooarchaeologist Don Grayson, for example, concludes from the lack of camel kill sites that people rarely or never killed camels (Grayson 1991). Even if the extinctions did happen in Clovis time, perhaps Clovis hunters had little or nothing to do with them. After all, fewer than 70 mammoths are all that can be accounted for at some 20 known Clo-
vis sites (see table 7). The number seems trivial in terms of the loss of a continental population not only of mammoths but also of many other large mammals that must have totaled many millions. Nevertheless, for three reasons I would suggest that the lack of kill sites is actually supportive of the overkill model.
First, if there are few contemporaneous associations of extinct megafauna with Clovis artifacts, there are fewer with post-Clovis artifacts. Although Folsom points, for example, are only a few hundred years younger than Clovis points, archaeologists have not found them in a clear-cut association with any extinct species beyond taxa of bison. There are no unambiguous records of mammoths, horses, camels, ground sloths, or other extinct genera of mammals anywhere in North America younger than those of Clovis time. A camel (terminal date 10,080 ± 179) once thought to be associated with Folsom artifacts is actually not well associated. In fact, the more usual pattern is an older layer of extinct megafaunal remains with few, if any, artifacts, plus a younger layer with much evidence of people and what they ate, but no extinct species (Martin 1986).
At a minimum, then, the archaeological evidence indicates that by the time Folsom points came into use, about 10,700 radiocarbon years ago, most or all of the great mammals were already gone. Except for extinct taxa of bison and possibly California Condors in West Texas, there were no Holocene (postglacial) mammoths, ground sloths, or other extinct megafauna. Similarly, Australia (45,000 to 55,000 years ago) and Madagascar (500 BC to 1500 AD) have revealed few, if any, cultural artifacts of pioneering peoples in association with remains of megafauna on the edge of extinction (Martin and Steadman 1999, 40).
Second, even if spears were used to kill, for example, ground sloths, stone spear points would not necessarily be found in association with Shasta ground sloth bones. They would have been recovered and used again. This may be a partial explanation for the paucity of apparent kill sites of animals other than bison. As for the number of bison kill sites, perhaps these are attributable in part to the fact that bison were commonly killed in a herd or group. Or perhaps by Folsom time the location of quality stone for making tools was so well known that much less care was taken to recover points than in the days of the mammoth hunters.
Third, the rate of extinction would have determined the archaeological visibility of the event. An event that occurs in a very short period (a few tens or hundreds of years throughout a continent, as I posit for the American extinctions) will scarcely be detectable in the fossil record (Martin 1973; Mosimann and Martin 1975). Rapid and massive extinction of large animals, for whatever reason, would have left virtually no trace. I believe the impact of the first human invaders was so sudden and severe, and the opening and closing of the Clovis window so rapid, that we may never find many kill sites. Indeed, I believe we are fortunate to have any (G. Haynes 2002b), especially in open areas where sedimentation (and hence a fossil record) is scant. If a stratigraphic alignment of the first human arrivals with the last presence of an about-to-be-extinct fauna is difficult to detect only 3,000 years ago on a small island like Li-fuka in the Tongan archipelago, it is small wonder that few mammoth kill sites 13,000 years old have turned up in America and there are as yet no megafaunal kill sites in Australia (some consider Cuddie Springs to be an exception; Wroe and others 2004). Only under unusually favorable circumstances does a very careful excavation reveal the field evidence archaeologists have long demanded.
This also explains the lack of art: the animals vanished while the Clo-vis people were on the move. The lack of any artistic depictions of these animals is an argument in favor of a very short temporal overlap between them and the first hunters. In the Old World, where the overlap was much longer and the extinctions fewer and more gradual, art is abundant.
In addition, quite apart from the overkill model, it is always particularly fortuitous to discover creatures that died and fossilized right at the moment of extinction, especially if that extinction was almost instantaneous, and especially for taxa that were relatively rare to begin with. The most we can expect is some chronological control on when the last populations were alive. For mammoths, horses, ground sloths, and others in western North America, we may have the right millennium of extinction (between 13,500 and 12,500 calendar years or 11,500 and 10,500 radiocarbon years). By leaning on negative evidence, we may narrow the chronology down in at least a few cases, such as those of the Columbian mammoth, the Shasta ground sloth, and Harrington's extinct goat, to what may prove to be the right century or two (around 10,900 to 11,100 radiocarbon years). If all 32 extinct North American genera were struck down within a few decades, the event is beyond resolution by current dating techniques.
Those who deny overkill often turn to natural causes to account for the extinctions. Such an approach has major problems. Detailed fossil pollen and macrofossil plant records are continually appearing for habitable parts of the planet and various dating methods, especially radiocarbon dating, are available for chronological comparisons. If large mam mals thoughout North and South America disappeared simultaneously around 11,000 radiocarbon years ago as a result of some extraordinary climatic shock, nothing similar is apparent outside the Americas, not even in the West Indies. Field evidence of climatic or other forcing of megafau-nal death is even harder to uncover than kill sites. Where are the "freeze sites," if some cold shock wave is involved? Would such a shock kill off plateau- or montane-ranging species before they could descend to adjacent lower and warmer elevations, like the Mojave Desert in Arizona and California, watered by the Colorado River? In any case, to match the global "deadly syncopation" of Ross MacPhee, the killer climatic change would have to have struck Australia long before the Americas and Madagascar, New Zealand, and the Pacific islands long after. If "killer cold" exterminated large mammals in South America around 10,500 radiocarbon years ago, might we not expect some concurrent losses of megafauna in Australia, South Africa, and New Zealand?
Most fossils are proof of nothing more than an organism's existence and death. In very few cases can we infer an unambiguous cause of individual mortality, much less of its extinction as a species. This makes it difficult to extract definitive answers from the fossil record. Those who favor the climate theory can argue that the scarcity of kill sites indicates that people had little to do with the extinctions. Those who favor overkill can argue that a mere decade of human impact in any new region would have been enough to entrain the extinctions but highly unlikely to be reflected in the fossil record (Mosimann and Martin 1975; Mithen 1997).
The case of the dinosaurs makes an interesting comparison here. If the main evidence for dinosaur extinction were fossils alone, few if any paleontologists would have dreamt that that extinction was sudden. Even with the discovery of the Chicxulub Crater in the Yucatan, which coincides in age with the Cretaceous-Tertiary boundary and makes the case for a catastrophic extinction of Mesozoic biota, some paleontologists have been slow to abandon their gradualist views, and not all have done so. The Alvarez model implies that the dinosaurs all died out in less than a year at the end of the Cretaceous, as the result of major disruption of the atmosphere following a severe extraterrestrial accident, such as an asteroid impact. Many geologists regard such a conclusion as robust. But sections of continental deposits dating from the end of the Cretaceous that could potentially harbor the last bones of non-avian dinosaurs are few. The most common dinosaur in the fauna, Triceratops, is, as one would expect, the one found closest to the boundary. Dinosaurs simply are not common enough to trace easily to the boundary. The same may be anticipated for at least some of the 32 genera of mammalian extinctions in the Americas in near time. As paleontologists have discovered, the fossil record may not be good enough to incorporate all of the species lost when a catastrophe occurs.
In the Southwest, twig figurines are particularly likely to be associated with extinct animal remains. The associations are not contemporaneous, so they tell us little about the extinctions, but they do shed interesting light on past human attitudes toward the extinct beasts. I first began focusing on these associations in 1969 at Stanton's Cave. Bob Euler's particular interest in the cave centered on the numerous split-twig figurines that he and others found there. Each figurine was 4 to 6 inches in length and had a head, neck, and legs attached to a body, all constructed from a single willow twig. The figurines looked like some kind of small ungulate, either a mountain sheep or a deer (see plate 17). Some even had a slender twig or a splinter inserted through the midsection, very likely symbolizing a spear.
Bob's National Geographic Society grants to investigate the cave were triggered in part by rapid loss of these figurines from it. By the time his excavations began, at least 75 figurines had found their way to museums, some via well-intentioned individuals who sought to keep them safe. The problem was that both the "rescuing" and the looting destroyed context. With no knowledge of the figurines' provenience (exact location within the cave), a crucial piece of information in site analysis, it was impossible to reconstruct any pattern that might help determine their meaning or function. Undoubtedly any figurine arrangements originally left on the cave floor had been the first to disappear. However, Bob's team discovered some figurines in clusters of up to five that had been carefully cached between flat rocks by human hands. There were a total of over 160 figurines, most of which radiocarbon dated at 3,000 to 4,000 years.
Interestingly, Stanton's Cave showed no evidence of ever having been lived in, despite its apparent suitability. That seemed strange. Bob and his field team discovered no hearths, no kitchen middens containing bone scraps, no stone knives or scrapers—not even any potsherds. A few domestic artifacts did appear, but only at or just beneath the floor of the cave. These scraps were probably associated with the Pueblo II (1050 to 1150 AD) Kayenta Anasazi ruin on a terrace at the mouth of South Canyon, a few hundred yards upstream. They were not associated with the figurines.
Although I do not pay much attention to prehistoric artifacts, the
figurines haunted me. Surely they had symbolic value. What were they made for? Why were so many found in Stanton's Cave? Had they somehow been associated with the abundant remains of Harrington's extinct goat, which were found in quantity in the cave and were at least 6,000 to 30,000 years older? If there was a connection—if, for example, the figurines had intentionally been placed near the remains—did their mak ers believe that these remains were of an existing creature, or did they recognize that they were of something different, something extinct?
Bob suggested the former, and raised the possibility that the cave had been a special place, a sacred site: "It is generally agreed that [the figurines] represent some form of magico-religious ritual, the twigs inserted through the body having been representations of spears that functioned to insure success in the actual hunt" (Euler 1984, 9). I wondered if the figurine makers hoped to find some of the mysterious animals, the extinct goats, still alive. Perhaps they imagined that they might resurrect them. For his part, Steve Emslie thought these people were expert hunters and trackers who immediately knew the remains were not of living animals but of ancestral ones (personal communication, November 2001). Their presence would have made a cave—a symbolic or accepted entrance to the total darkness of their underworld and the ancestors—a sacred place, a place for shrines.
The mystery was clarified when Emslie discovered an unvandalized cave in the Grand Canyon rich with figurines associated with the much older remains of extinct animals. He called the new site Shrine Cave (Emslie, Euler, and Mead 1987). His team also discovered five other relatively undisturbed caves high in vertical cliffs of the Grand Canyon. Em-slie reported, "Radiocarbon dates indicate cultural use of these sites between 4300 to 3700 B.P. . . . Unlike most other Archaic sites in the Grand Canyon, these contain numerous rock cairns as well as cairns built partially or entirely of indurated packrat-midden fragments of late Quaternary age. Archaic artifacts, including split-twig figurines, appear to be deliberately associated with fossil material of extinct mountain goats and other vertebrate remains" (Emslie, Mead, and Coats 1995). In Gypsum Cave, Harrington (1933) reported an intimate association of Archaic artifacts with extinct fauna. Although he thought they were contemporary, as we have seen, radiocarbon dates indicated otherwise. Judging by the quality of the artifacts, including painted throwing sticks, Gypsum Cave could have been the most sacred sanctuary of all, a place where Archaic people venerated animals extinct for thousands of years.
In the Grand Canyon, Archaic habitations or campsites are scarce. Perhaps the rugged landscape that made the canyon unsuitable for dwelling enhanced its value for ceremonial purposes. To date, hundreds of split-twig figurines and dozens of cairns have been found in Grand Canyon caves, many of them difficult of access. The sheer number of these apparently sacred sites is remarkable given the paucity of dwelling places in the canyon at the time the figurines were created, or indeed at any time prior to widespread Pueblo Il-age occupations.
Twig figurines of animals have also been recovered from Archaic sites in Nevada, Utah, and Colorado. These are mostly younger and probably served a different function, as they are associated with habitation sites, rather than with extinct animal remains. Those from Cowboy Cave in Utah, for example, while the same age as those from the Grand Canyon and westward, are less carefully made and appear more secular than sacred. They are associated with a variety of other artifacts, including Gypsum Cave points, presumably used in hunting deer and mountain sheep.
The split-twig figurines could tell us nothing concrete about the extinction of Harrington's mountain goat in the Grand Canyon. They were thousands of years too young to have had anything to do with the extinct animals of the Pleistocene. This did not mean, however, that they were completely tangential to the question. The positioning of cairns, twigs, or effigies near dung or bones of the extinct goats was ample evidence of prehistoric people's inordinate interest in these goats (an interest shared by investigators such as Larry Coats, Steve Emslie, Bob Euler, Dick Harington, Jim Mead, and Norrie Robbins). For whatever reason—a focus on the hunt, or simple fascination—our ancestors were strongly drawn not only to living large animals, but also to the remains of extinct ones. Occasionally, extinct animal remains are found in much younger archaeological sites, as at Casas Grandes in northern Chihuahua. There the "rock shop," a small room about 1,000 years old, contained many teeth of mammoths, Paleozoic fossils, and semiprecious stones (DiPeso, Rinaldo, and Fenner 1974).
The intense interest of prehistoric people in Quaternary fossils is especially evident in the Mediterranean region. Adrienne Mayor (2000) credits Austrian paleontologist Otto Abel with the idea that the Cyclops of Homeric times may have originated in the discovery in caves in Sicily of crania of fossil dwarf mammoths: their medial nasal cavity suggests one large eye socket. On a similar front, a major preoccupation (approaching an obsession) with depicting large mammals appears 30,000 years ago as an important component of Old World Paleolithic art (Guthrie 2005).
When I started reviewing literature for this chapter, especially Emslie, Mead, and Coats (1995), I was surprised by the strong suggestion of "megafauna worship" by Archaic people. There may be a sociobiologi-cal reason for this attraction. For at least two million years in Africa, the interaction between evolving humans and large animals would have been dynamic, intense, and durable. A deep emotional reaction to large animals would have been incorporated into our genome. Our Pleistocene ancestors hunted or scavenged large mammals, and some of them—lions, hyenas, leopards—returned the favor. It would not surprise me if some residue of these ancient terrors and traumas lingered in our dreams.
Was this article helpful?