The Horse, the Wheel, and Language: How Bronze-Age Riders From the Eurasian Steppes Shaped the Modern World
Page 25
Figure 10.8 From our 1998 data: bevel measurements of never bitted, occasionally bitted, and frequently bitted horse teeth plotted against age. All domesticated horses had precisely known ages; all feral horses were aged by examining entire mandibles with intact incisor teeth. The line excludes feral horses and horses aged ≤3 yr. and includes only bitted horses. After Brown and Anthony 1998.
Very young horses with newly erupted permanent premolars do display natural dips and rises on their teeth. New permanent premolars are uneven because they have not yet been worn flat by occlusion with the opposing tooth. We had to exclude the teeth of horses two to three years old for that reason. But among the 105 measurable P2s from mature equids that had never been bitted, Pleistocene to modern, we found that a “natural” bevel measurement of more than 2.0 mm is unusual (less than 3% of teeth), and a bevel of 2.5 mm is exceedingly rare (less than 1%). Only one of the 105 never-bitted teeth had a bevel measurement greater than 2.5 mm—a single tooth from the Leisey equids with a mesial bevel of 2.9 mm (the next-nearest bevel was 2.34 mm). In contrast, bevels of 2.5 mm and more occurred in 58% of the teeth of mature horses that were bitted.18
A bevel of 3 mm or more on the P2 of a mature horse is evidence for either an exceedingly rare malocclusion or a very common effect of bitting. If even one mature horse from an archaeological site shows a bevel ≥3 mm bit wear is suggested, but is not a closed case. If multiple mature horses from a single site show mesial bevel measurements of 3 mm or more, they probably were bitted. I should stress that our method depends on the accurate measurement of a very small feature—a bevel or facet just a few millimeters deep. According to our measurements on 178 P2 teeth of mature equids the difference between a 2 mm and a 3 mm bevel is extremely important. In any discussion of bit wear, precise measurements are required and young animals must be eliminated. But until someone finds a population of mature wild horses that displays many P2 teeth with bevels ≥3 mm, bit wear as we have defined it indicates that a horse has been ridden or driven.19
INDO-EUROPEAN MIGRATIONS AND BIT WEAR AT DEREIVKA
Many archaeologists and historians in the first half of the twentieth century thought that horses were first domesticated by Indo-European–speaking peoples, often specifically characterized as Aryans, who also were credited with inventing the horse-drawn chariot. This fascination with the Aryans, or Ariomania, to use Peter Raulwing’s term, dominated the study of horseback riding and chariots before World War II.20
In 1964 Dimitri Telegin discovered the head-and-hoof bones of a seven- to eight-year-old stallion buried together with the remains of two dogs at Dereivka in Ukraine, apparently a cultic deposit of some kind (see figure 11.9). The Dereivka settlement contained three excavated structures of the Sredni Stog culture and the bones of a great many horses, 63% of the bones found. Ten radiocarbon dates placed the Sredni Stog settlement about 4200–3700 BCE, after the Dnieper-Donets II and Early Khvalynsk era. V. I. Bibikova, the chief paleozoologist at the Kiev Institute of Archaeology, declared the stallion a domesticated horse in 1967. The respected Hungarian zoologist and head of the Hungarian Institute of Archaeology, Sandor Bökönyi, agreed, noting the great variabity in the leg dimensions of the Dereivka horses. The German zoologist G. Nobis also agreed. During the late 1960s and 1970s horse domestication at Dereivka was widely accepted.21
For Marija Gimbutas of UCLA, the domesticated horses at Dereivka were part of the evidence which proved that horse-riding, Indo-European–speaking “Kurgan-culture” pastoralists had migrated in several waves out of the steppes between 4200 and 3200 BCE, destroying the world of egalitarian peace and beauty that she imagined for the Eneolithic cultures of Old Europe. But the idea of Indo-European migrations sweeping westward out of the steppes was not accepted by most Western archaeologists, who were increasingly suspicious of any migration-based explanation for culture change. During the 1980s Gimbutas’s scenario of massive “Kurgan-culture” invasions into eastern and central Europe was largely discredited, notably by the German archaeologist A. Häusler. Jim Mallory’s 1989 masterful review of Indo-European archaeology retained Gimbutas’s steppe homeland and her three waves as periods of increased movement in and around the steppes, but he was much less optimistic about linking specific archaeological cultures with specific migrations by specific Indo-European branches. Others, myself included, criticized both Gimbutas’s archaeology and Bibikova’s interpretation of the Dereivka horses. In 1990 Marsha Levine seemed to nail the coffin shut on the horse-riding, Kurgan-culture invasion hypothesis when she declared the horse age and sex ratios at Dereivka to be consistent with a wild, hunted population.22
Brown and I visited the Institute of Zoology in Kiev in 1989, the year after Levine, learning of her trip only after we arrived. With the cheerful help of Natalya Belan, a senior zoologist, we made molds of dozens of horse P2s from many archaeological sites in Ukraine. We examined one P2 from Early Eneolithic Varfolomievka in the Caspian Depression (no wear), one from the Tripolye A settlement of Luka Vrublevetskaya (no wear), several from Mesolithic and Paleolithic sites in Ukraine (no wear), many from Scythian and Roman-era graves (a lot of bit wear, some of it extreme), and those of the cult stallion and four other horse P2s from Dereivka. As soon as we saw the Dereivka cult stallion we knew it had bit wear. Its P2s had bevels of 3.5 mm and 4 mm, and the enamel on the first cusp was deeply abraded. Given its stratigraphic position at the base of a Late Eneolithic cultural level almost 1 m deep, dated by ten radiocarbon dates to 4200–3700 BCE, the cult stallion should have been about two thousand years older than the previously known oldest evidence for horseback riding. Only four other P2s still survived in the Dereivka collection: two deciduous teeth from horses less than 2.5 years old (not measurable), and two others from adult horses but with no bit wear. So our case rested on a single horse. But it was very clear wear—surprisingly similar to modern metal bit wear. In 1991 we published articles in Scientific American and in the British journal Antiquity announcing the discovery of bit wear at Dereivka. Levine’s conclusion that the Dereivka horses were wild had been published just the year before. Briefly we were too elated to worry about the argument that would follow.23
It began when A. Häusler challenged us at a conference in Berlin in 1992. He did not think the Dereivka stallion was Eneolithic or cultic; he deemed it a Medieval garbage deposit, denying there was evidence for a horse cult anywhere in the steppes during the Eneolithic. That the wear looked like metal bit wear was part of the problem, since a metal bit was improbable in the Eneolithic. Häusler’s target was bigger than bit wear or even horse domestication: he had dedicated much of his career to refuting Gimbutas’s “Kurgan-culture” migrations and the entire notion of a steppe Indo-European homeland.24 The horses at Dereivka were just a small piece in a larger controversy. But criticisms like his forced us to obtain a direct date on the skull itself.
Telegin first sent us a bone sample from the same excavation square and level as the stallion. It yielded a date between 90 BCE and 70 BCE (OxA 6577), our first indication of a problem. He obtained another anomalous radiocarbon date, ca. 3000 BCE, on a piece of bone that, like our first sample, seems not to have been from the stallion itself (Ki 5488). Finally, he sent us one of the bit-worn P2s from the cult stallion. The Oxford radiocarbon laboratory obtained a date of 410–200 BCE from this tooth (OxA 7185). Simultaneously the Kiev radiocarbon laboratory obtained a date of 790–520 BCE on a piece of bone from the skull (Ki 6962). Together these two samples suggest a date between 800 and 200 BCE.
The stallion-and-dog deposit at Dereivka was of the Scythian era. No wonder it had metal bit wear—so did many other Scythian horse teeth. It had been placed in a pit dug into the Eneolithic settlement between 800 and 200 BCE. The archaeologists who excavated this part of the site in 1964 did not see the intrusive pit. In 2000, nine years after our initial publication in Antiquity, we published another Antiquity article retracting the early date for bit wear at Dereivka. We were disappointed, but by then Dereivka
was no longer the only prehistoric site in the steppes with bit wear.25
Figure 10.9 Horse-related sites of Eneolithic or older age in the western and central Eurasian steppes. The steppe ecological zone is enclosed in dashed lines.
(1) Moliukhor Bugor; (2) Dereivka; (3) Mariupol; (4) Matveev Kurgan; (5) Girzhevo; (6) Kair Shak; (7) Dzhangar; (8) Orlovka; (9) Varfolomievka; (10) Khvalynsk; (11) S’yezzhe; (12) Tersek; (13) Botai
BOTAI AND ENEOLITHIC HORSEBACK RIDING
The oldest horse P2s showing wear facets of 3 mm and more are from the Botai and Tersek cultures of northern Kazakhstan (figure 10.9). Excavated through the 1980s by Victor Zaibert, Botai was a settlement of specialized hunters who rode horses to hunt horses, a peculiar kind of economy that existed only between 3700 and 3000 BCE, and only in the steppes of northern Kazakhstan. Sites of the Botai type, east of the Ishim River, and of the related Tersek type, west of the Ishim, contain 65–99.9%/horse bones. Botai had more than 150 house-pits (figure 10.10) and 300,000 animal bones, 99.9% of them horse. A partial list of the other species represented at Botai (primarily by isolated teeth and phalanges) includes a very large bovid, probably bison but perhaps aurochs, as well as elk, red deer, roe deer, boar, bear, beaver, saiga antelope, and gazelle. Horses, not the easiest prey for people on foot, were overwhelmingly preferred over these animals.26
Figure 10.10 A concentration of horse bones in an excavated house pit at the Botai settlement in north-central Kazakhstan, dated about 3700–3000 BCE. Archaeozoologist Lubomir Peske takes meas urements during an international conference held in Kazakhstan in 1995 “Early Horsekeepers of the Eurasian Steppe 4500–1500 BC.” Photo by Asko Parpola.
We visited Zaibert’s lab in Petropavlovsk, Kazakhstan, in 1992, again unaware that Marsha Levine had arrived the year before. Among the forty-two P2s we examined from Botai, nineteen were acceptable for study (many had heavily damaged surfaces, and others were from horses younger than three years old). Five of these nineteen teeth, representing at least three different horses, had significant bevel measurements: two 3 mm, one 3.5 mm, one 4 mm, and one 6 mm . Wear facets on undamaged portions of the Botai P2s were polished smooth, the same kind of polish created by “soft” bits in our experiment. The five teeth were found in different places across the settlement—they did not come from a single intrusive pit. The proportion of P2s exhibiting bit wear at Botai was 12% of the entire sample of P2s provided, or 26% of the nineteen measurable P2s. Either number was just too high to explain by appealing to a rare natural malocclusion (figure 10.11). We also examined the horse P2s from a Tersek site, Kozhai 1, dated to the same period, 3700–3000 BCE. At Kozhai 1 horses accounted for 66.1% of seventy thousand identified animal bones (others were saiga antelope at 21.8%, onager at 9.4%, and bison, perhaps including some very large domesticated cattle, at 2.1%). We found a 3 mm wear facet on two P2s of the twelve we examined from Kozhai 1. Most of the P2s at Botai and Kozhai 1 did not exhibit bit wear, but a small percentage (12–26%) did, consistent with the interpretation that the Botai-Tersek people were mounted horse hunters.27
Figure 10.11 Three horse with bit wear from the Botai settlement. The photos show extensive postmortem damage to the occlusal surfaces. The undamaged middle tooth showed smooth enamel surfaces but had a significant wear facet, like a horse ridden with a “soft” bit of rope or leather.
Botai attracted the attention of everyone interested in early horse domestication. Two field excavations by Western archaeologists (Marsha Levine and Sandra Olsen) have occurred at Botai or Botai-culture sites. The original excavator, Victor Zaibert, the Kazakh zoologist L.A. Makarova, and the American archaeozoologist Sandra Olsen of the Carnegie Museum of Natural History in Pittsburgh all concluded that at least some of the Botai horses were domesticated. In opposition, the archaeo-zoologists N. M. Ermolova, Marsha Levine, and the German team Norbert Benecke and Angela von den Dreisch concluded that all the Botai horses were wild.28 Levine found some pathologies in the Botai vertebrae but attributed them to age. Benecke and von den Dreisch showed that the Botai horses exhibited a narrow range of variability in size, like Paleolithic wild populations. The ages and sexes of the Botai horses were typical of a wild population, with a 1:1 ratio between the sexes, including all age groups, even colts and pregnant mares with gestating fetuses. Everyone agrees that whole herds of wild horses were killed by the Botai people, using herd-driving hunting techniques that had never been used before in the Kazakh steppes, certainly not on this scale. Were the hunters riding or on foot? Native American hunters on foot drove bison herds over cliffs before the introduction of horses to the Americas by Europeans, so herd driving was possible without riding.
Sandra Olsen of the Carnegie Museum concluded that at least some Botai horses were used for transport, because whole horse carcasses were butchered regularly over the course of several centuries in the settlement at Botai.29 How would pedestrian hunters drag eight-hundred-pound carcasses to the settlement, not just once or twice but as a regular practice that continued for centuries? Pedestrian hunters who used herd-driving hunting methods in the European Paleolithic at Solutré (where Olsen had worked earlier) and in the North American Plains butchered large animals where they died at the kill site. But the Botai settlement is located on the open, south-facing slope of a broad ridge top in a steppe environment—wild horses could not have been trapped in the settlement. Either some horses were tamed and could be led into the settlement or horses were used to drag whole carcasses of killed animals into the settlement, perhaps on sleds. Olsen’s interpretation was supported by soil analysis from a house pit at Botai (Olsen’s excavation 32) that revealed a distinct layer of soil filled with horse dung. This “must have been the result of redeposition of material from stabling layers,” according to the soil scientists who examined it.30 This dung-rich soil was removed from a horse stable or corral. The stabling of horses at Botai obviously suggests domestication.
One more argument for horseback riding is that the slaughter of wild populations with a 1:1 sex ratio could only be achieved by sweeping up both stallion-with-harem bands and bachelor bands, and these two kinds of social groups normally live far apart in the wild. If stallion-with-harem bands were driven into traps, the female:male ratio would be more than 2:1. The only way to capture both bachelor bands and harem bands in herd drives is to actively search and sweep up all the wild horses in a very large region. That would be impossible on foot.
Finally, the beginning of horseback riding provides a good explanation for the economic and cultural changes that appeared with the Botai-Tersek cultures. Before 3700 BCE foragers in the northern Kazkah steppes lived in small groups at temporary lakeside camps such as Vinogradovka XIV in Kokchetav district and Tel’manskie in Tselinograd district. Their remains are assigned to the Atbasar Neolithic.31 They hunted horses but also a variety of other game: short-horned bison, saiga antelope, gazelle, and red deer. The details of their foraging economy are unclear, as their camp sites were small and ephemeral and have yielded relatively few animal bones. Around 3700–3500 BCE they shifted to specialized horse hunting, started to use herd-driving hunting methods, and began to aggregate in large settlements—a new hunting strategy and a new settlement pattern. The number of animal bones deposited at each settlement rose to tens or even hundreds of thousands. Their stone tools changed from microlithic tool kits to large bifacial blades. They began to make large polished stone weights with central perforations, probably for manufacturing multi-stranded rawhide ropes (weights are hung from each strand as the strands are twisted together). Rawhide thong manufacture was one of the principal activities Olsen identified at Botai based on bone tool microwear. For the first time the foragers of the northern Kazakh steppes demonstrated the ability to drive and trap whole herds of horses and transport their carcasses into new, large communal settlements. No explanation other than the adoption of horseback riding has been offered for these changes.
The case for horse management and riding at Botai and Kozhai 1 is based o
n the presence of bit wear on seven Botai-Tersek horse P2s from two different sites, carcass transport and butchering practices, the discovery of horse-dung–filled stable soils, a 1:1 sex ratio, and changes in economy and settlement pattern consistent with the beginning of riding. The case against riding is based on the low variability in leg thickness and the absence of riding-related pathologies in a small sample of horse vertebrae, possibly from wild hunted horses, which probably made up 75–90% of the horse bones at Botai. We are reasonably certain that horses were bitted and ridden in northern Kazakhstan beginning about 3700–3500 BCE.
THE ORIGIN OF HORSEBACK RIDING
Horseback riding probably did not begin in northern Kazakhstan. The Botai-Tersek people were mounted foragers. A few domesticated cattle (?) bones might be found in some Tersek sites, but there were none in Botai sites, farther east; and neither had sheep.32 It is likely that Botai-Tersek people acquired the idea of domesticated animal management from their western neighbors, who had been managing domesticated cattle and sheep, and probably horses, for a thousand years before 3700–3500 BCE.
The evidence for riding at Botai is not isolated. Perhaps the most interesting parallel from beyond the steppes is a case of severe wear on a mesial horse P2 with a bevel much deeper than 3 mm, on a five-year-old stallion jaw excavated from Late Chalcolithic levels at Mokhrablur in Armenia, dated 4000–3500 BCE. This looks like another case of early bit wear perhaps even older than Botai, but we have not examined it for confirmation.33 Also, after about 3500 BCE horses began to appear in greater numbers or appeared regularly for the first time outside the Pontic-Caspian steppes. Between 3500 and 3000 BCE horses began to show up regularly in settlements of the Maikop and Early Transcaucasian Culture (ETC) in the Caucasus, and also for the first time in the lower and middle Danube valley in settlements of the Cernavoda III and Baden-Boleraz cultures as at Cernavoda and Kétegyháza. Around 3000 BCE horse bones rose to about 10–20% of the bones in Bernberg sites in central Germany and to more than 20% of the bones at the Cham site of Galgenberg in Bavaria. The Galgenburg horses included a native small type and a larger type probably imported from the steppes. This general increase in the importance of horses from Kazakhstan to the Caucasus, the Danube valley, and Germany after 3500 BCE suggests a significant change in the relationship between humans and horses. Botai and Tersek show what that change was: people had started to ride.34