The revise the paper of the Eurasian nomad history

INTRODUCTION Emma C. Bunker

I N Til E ·\STE Rl': E URASIAN STE PPES,l away from the newly developed industrial areas, re- minders of the ancient hunting and herding

peoples accent the landscape. Neolithic petro- glyphs of animals and humans abound in the mountains of northern China and Inner Mongo- lia," Horse hoof prints carved into the surface of sacred rocks on Binder Uul in northeastern Mongolia (fig. Bl ) are antecedents of the hoof prints on the hilt of a later bronze knife (see No. 62). Bronze Age slab graves still stand on the side ofa Mongolian hill (fig. B2), while a nearby deer- stone recently adorned with ribbons (fig. B3) sig- nals a living reverence for the past. Other deer- stones in northern Xinjiang' indicate the com- plex cultural interconnections that existed

Fig. B2. Bronze Agc slab all\!ongonmor't, Tov almag, eastern Mongolia.june, 1996. Photo by Emma C, Bunker.

Fig. B3, Deerstone with modern votives in Heruii aimag, eastern Mongolia.june. 1996. Photo b)' Emma C. Bunker.

Fig. Bl. Boulder with Neolithic pictographs at Binder Uul , Hentii aima g, eastern l\lon golia,June. 1996. Photo by Emma C. Bunker.

between the many ancient Eurasian steppe peo- ples. Clusters of stone-ringed pit-graves belong- ing to the ancient Xiongnu peoples mark the grassy Mongolian uplands. A gazelle skull recently placed on an ancient grave (fig. B4) is a haunting reminder of the power of the past.

Connoisseurs of Asian art have long admired the artifacts of the herding and hunting peoples that inhabited the eastern Eurasian steppes be- yond the northern frontiers of ancient China between 1500 BC and the second century AD. Popularly known as "Ordos Bronzes," after the Ordos region where they were first discovered and acquired, these artifacts were useful, P0rL- able objects-personal ornaments, horse gear, tools, and weapons-richl y decorated with intri-

Fig. B4. Gazelle skull on a Xiongnu grave atjargalr' khnau.Hentii aimag, eastern Mongolia.june, 1996. Ph oto by Emm a C. Bunker.

cate geometric, zoomorphic , and vegetal motifs , the visual embodiments of clan and group affilia- tions, and the supernatural world that governed people's lives.

The Arthur M. Sackler collections include more than five hundred eastern Eurasian steppe artifacts. None are from controlled excavations but the majority are said to have been acquired in north China. ~Ian}' pieces were purchased from well-known collections formed in China during the first third of the twentieth century, a time when enthusiasm for collecting such works nour- ished in B~ijing (see below pp. 99-111 for the his- tory of the collection ). These small artifacts held a special fascination for western missionaries and diplomats who saw them as representations of non-Chinese elements absorbed into the greater Chinese world. As such, they could yield clues to the unique sinicizing process seen by scholars as determining the development of Chinese culture since Neolithic times . These little artifacts were readily available from grave robbers and dealers who sold them inexpensively to collectors eager to own a piece of the past.

Earl y publications of ancient bronzes belong- ing to the eastern Eurasian steppe peoples had resulted in art-historical confusion. Known pri- marily through chance finds and curiosity shop discoveries, rather than through a rchaeologically validated disco veries, the artifacts frequentl y had provenance but not provenience : ' and thus remained archaeological orphans without ante-

8

cedents. As a consequence, theories based on these unexcavated pieces had limited value," ''''estern archaeological literature traditionally described these artifacts as Animal Style,' Ordos," Scythe-Siberian," Sino-Siberian ," and most re- cently nomadic," all imprecise terms without his- torical or archaeological basis.

The term "An imal Style," which conjures up romantic visions of glorious horsemen adorned with zoomorphic symbols galloping over the \'<15t expanse of the Eurasian steppes, has unfortu- nately become universally accepted for the art of the Eurasian steppe peoples. First coined by Mikhail I. Rostovtzeff ." Animal Style has been used to explain the transmission of certain zoo- rnorphic motifs to most remote places, as if the style were a disease, leaving zoo m orp h ic designs in the art of other cultures. This has obscured the real art-historical picture. Furthermore, use of the term Animal Style suggests that only one style go verned the an of all the regional steppe groups, when in fact no general style existed. Instead, recent archaeological discoveries pro- vide evidence of numerous distinctive regional cultures throughout the steppe zone.

In the past, unexcavated artifacts belonging to the eastern Eurasian steppe peoples have been iso lated aesthetically and removed from their original cultural context, rendering them close to meaningless . Archaeologically informed reconstructions demonstrate the original func- tion of an artifact. Artifacts can broaden our knowledge of a culture onl y if th e specific func- tion of the piece and the regional culture to which it belonged are known . Because of careful excavation we know that the two gilded wooden eagle plaques recovered from Tomb 2 at Basha- dar in the Alt.ai Mountains of southern Siberia originally decorated a saddle (fig. B5). Photo- graphic documentation from excavation sites in north Ch in a sh o w that man y small bronze plaques, previously thought to be harness deco- ration, were originally personal ornaments cast in large sets to adorn belts , necklaces, and pectorals (fig. B6). Considering an artifact as an isolated work of art robs it of its original purpose and sig- nificance. Connoisseurs value an artifact from

the eastern Eurasian steppes that is more finely cast than one that is not. To the person who orig- inally used that artifact, its shape and iconogra- phy may have been more important than the quality of workmanship. If the original context of each artifact is preserved, we may be able to reconstruct its function and meaning and pre- serve some of its integrity, otherwise we risk mak- ing serious mistakes,

Eastern Eurasian steppe bronzes are not just delightful bits of the past, but reflect major cul- tural traditions that are misunderstood, in part, because pastoral economies are often presented as backward and at a lower level of development than those of settled agricultural folk with whom most scholars empathize. The eastern Eurasian steppe peoples belonged to a verr different cul- tural sphere from that of the carl}' dynastic Chinese with whom they are often contrasted.

The cultural traditions of the Eurasian steppes had far-reaching impact not only on ancient China but also on the West. The steppe peoples frequently served as intermediaries between the eastern and western reaches of Eurasia, and their artifacts reflect complex cul- tural contacts not always documented in ancient texts ." In lieu of ancient road maps, these arti- facts remain to mark the passage of people and their artistic heritage.

Recent archaeological research now provides new informat.ion about. the many peoples of t.he

eastern Eurasian steppes and offers a more sci- entific approach to the study of their artifacts, an approach understood by scholars both in Asia and the West. This volume will attempt such a scientific approach based on all the available information-archaeological, historical, ecologi- cal , iconographic, stylistic, metallurgical, and anthropological.

PURPOSE OF TInS STUDY

One of the aims of this volume is to provide a cul- tural context for the bronze artifacts in the Arthur M. Sackler collections. Since the majority of the Sackler artifacts come from ancient China's Northern Zone, this will be the main area of concentration. The artifacts are com- pared to recently published excavated material made available by archaeologists in the People's Republic of China, Mongolia, Buryatia, Tuva, Siberia, and Russia. This material is presented in the Archaeological Overview written by Dr. Katheryn M. LindufT (see pp. ism.

The publication of these new discoveries makes possible the identification of specific regional styles and the establishment of a relative chronology into which the Sackler artifacts can be placed. Detailed comments on the develop- ment of individual artifactual types and their decoration are reserved for the catalogue entries, Chapters 1-10. Chapter 11 contains works that are often considered steppe artifacts but should

Fig. B5. Reconstruction of saddle ornaments from Tomb 2 OIl Bashadar, Altai Mountains. 61h celllury BC. The Hermitage State Museum. 5 t, Petersburg, Russia. Photo by Emma C. Bunker.

Fig. B6. Grave Ii4. Yuhuangmiao, Yanqing county. Beijing district. Photo courtesy ofJin Fengyi.

not be asso ciated with the ancient eastern Eurasian ste ppe peoples.

The manufacturing and plating techniques used to produce artifacts associated with the east- ern Eurasian steppe peoples are di scussed in Appendix 2. When po ssible , the specific manu- facturing process for each Sackler artifact is sug- gested in the entry. Forgeries in the Sackler col- lections and the reasons for rejecting them as genuine antiquities are discussed in Appendix 3, in keeping with Dr. Sackler's insistence on a sci- entific evaluation of all published artifacts. These decisions are based on both art- historical and sci- entific research.

The bronzes presented in the catalogue are divided into four br oad temporal ph ases on the basis of recent archaeological discoveries. Th e first phase, the Bronze Age, dates from about 1500 BC to the ninth century BC, roughly con - temporaneous with dynast ic China's Shang and ''''estern Zhou periods. The second phase, the Iron Age, ext ends from the ninth into the third century BC, roughly contemporaneous with the Eastern Zhou period, while the thi rd phase, the Xiongnu period, dating from the late third cen - tury Be through the first century AD, and the fourth, th e Xianbei period, first through third centuries AD, are equivalent to the Western and Eastern Han periods. These distinction s are used to divide the objects in the catalogue. These are then subdivided into chapters based on their relationship with specific regional developments. Each chapter contains a short introductory essay explaining why the pieces within it are plac ed in a particular time period and region. The Sackler material was not acquired systematic ally, so spe- cific types that are particularly characteristic of certain regions and time periods ma y be lacking in the collection; these will be presented in the Archaeological O verview, Most of the Sackler bronzes belong to the vario us cultures of ancient China's No rthe rn Zone, but som e also displ ay features that connect them with more distant Eurasian steppe cultures in Mongolia and south- ern Siberia.

No comprehensive publication deals with the overall cultural history of the eastern Eurasian

10

Fig. B7. View of grasslands from Binder Uul , Hcntii aimag, eastern Mongolia.june, 1996. Photo by Emma C. Bunker.

Fig. B8. Sand dunes in the desert ncar Dunhuang, Gan su province, October, 1987. Photo by Emma C. Blinker.

Fig. B9. Tianshnn (Heavenly Mountains), Xinjiang province, Ma)', 1995. Photo b)' Emma C. Bunker.

-

steppe peoples between 1500 BC and the third century AD. Instead, there are numerous studies devoted to regional developments that are not always understood within the larger picture of the Eurasian steppes. The regionalism that dic- tated the shape of the ancient cultures prevails in modern scholarship. The Chinese tradition of regional ownership and publication of excavated material has not encouraged the study of related material from diverse locations. Archaeologists in Russia and Mongolia have only sparse knowledge of the material excavated in China, and western scholars have only a partial knowledge of archae- ology in Russia and Mongolia. The synthesis pre- sented here is by no means complete.

It is hoped that the more scientific approach to the bronze artifacts of the ancient eastern Eurasian steppe peoples and to the artistic sym- bols that enhance their meaning will inspire fur- ther research in a field still very much in its infancy.

THE LAND

The eastern Eurasian steppe is a vast arc of eco- logically varied land embracing northern and northwestern China. It stretches from Heilong- jiang, Jilin and Liaoning provinces in the east, through southern Inner Mongolia, northern Hebei, Shanxi and Shaanxi provinces to the south, through Ningxia into southeastern Gansu from where ancient trade routes led west by way of the Hexi corridor through Xinjiang province in eastern Central Asia. It is referred to in China as Beifang (Northern Zone) : ' Above it, the land extends northward through eastern Mongolia to Lake Baikal in Buryatia, and to the northwest through western Mongolia to the Altai Moun- tains, and farther northwest to the Minusinsk Basin of southern Siberia.

This huge area is not ecologically homoge- neous, as it includes extensive grasslands (fig. B7), deserts (fig. B8), mountains and forests (fig. B9), rivers and lakes. The local environments sup- ported a varied range of economic activit)', such as stock breeding, hunting, fur and livestock trading, fishing, and agriculture. In many instances, there is evidence for the existence of pastoralism along

11

with some agriculture, always difficult in the steppes because of a short growing season.

China's Northern Zone is divided into west- ern and eastern regions by the Taihang Moun- tains that run north-south along the western bor- der of Hebei province. The western region is characterized by grasslands suited to large-scale herding that ultimately developed into nomadic pastoralism and livestock trading, while the mountainous, forested areas immediately east of the Taihang Mountains in Hebei province were more suited to hunting, trapping, and the mar- keting of furs and leather. To the east, the fertile soil of the Liao River miley known as the Dongbei in Chinese archaeological literature, sustained agriculture and stock breeding but not large- scale transhumant herding. The Dongbei , equiv- alent to the area that used to be called Man- churia, includes present-day Liaoning, Jilin, and Heilongjiang provinces.

The inhabitants of the eastern Eurasian steppes were culturally different from their sedentary agricultural Chinese neighbors, who tilled their terraced farmlands (fig. BrD). The vast grasslands, abundant with wildlife, encour- aged a less settled way of life. The Great Wall,

Fig. B1O. View or terraced farmlands in southern Ningxia, june, 1990. Photo by Emma C. Bunker.

begun in the Warring States period (475-221 Be) and extended in late r periods, stands as a symbol of these distinctive differences that play an important role in the history of eastern Asia."

THE PEOPLE

The ancient peoples of the Eurasian steppes led highly structured lives govemed by the seasonal demands of the herding, hunting, livestock trad- ing, and agricultural activity specific to each re- gion. Although man y groups were culturally and technologically interconnected, their material re- mains indicate distinct regional characteristics.

The common denominators among these groups were their non-urban existence; primary economic reliance on animals, both wild and domestic; the exploitation of the horse for trans- portation ; a profici en cy in metalwor king; small personal weapons with integrally cast hilts; metal personal ornaments to indicate status and clan affiliation; and distinctive dress-coats, trousers , and prominent belts hung with tools and small weapons (fig. B6, also fig. 82.1) . Animals pro- vided essentials of wool, leather, and fur for cloth- ing and shelter; meat, butter, milk , and cheese for food ; and dung for fuel.

The distinctive cultural differences between sedentary and pastoral peoples have su rvived into the p resent, and are reflected in current ten- sio ns between the Han Chinese and their non-

Chinese northern neighbors in Inner Mongolia, where open ste ppe and spatial freedom have his- toricall y been the norm. " Th e traditional Confucian belief of superiority exhibited by the Han Chinese has colored the history of the Northern Zone in the past, and continues to exert influence today. .

During the second and early first millennia BC, the northern peoples were basicall y seden- tary pastoralists, practicing settled animal hus- bandry, hunting, and some agriculture. By the middle of the first millennium BC, so m e regional groups had adopted nomadic pastoralism or transhumance, the seasonal migration in search of water and pasture , and this change in lifestyle is reflected in the design and dec oration of their artifacts, Nevertheless, each transhumant group belonged to a large r group with a settled home base where the mobile herde rs returned to shel- ter their herds during wint e r.

It could be dangerous to equate the lives of ancient peoples with that of peoples in the steppes today, but the remaining monuments, passages in ancient Chinese lite rature, and recently published archaeological di sco veries suggest that th e basic condition s of life today on the eastern Eurasian steppes have changed little over the last two thousand years.

According to Huan Kuan in the Yantielun (Disco u rses on Salt and Iron) , written in the first

Fig. Bit. Erecting a framework for a ger.

Fig. B12. Occupied grrs ill the Tianshan, XiJliiang province, Mar. 1995. Photos by Emma C. Blinker.

Fig. B13. Herding sheep in the foothills of the Tia nshan. Xi njia n g, June. 1995. 1'11 010 by Emm a C . Bun ker.

century BC, the Xiongnu lived in "Qiollg1u tents" (do m e-sh ap ed 11Uts). " The term "Qiollg1u tent" refers to a gel' (fig . B11 and B12), a round trellis tent whose portable framework of wooden poles is felt-covered. It is frequently called a )'Ul'/, a Turkish word that actually means "home terri- tory" or "camp site" rather than simply "tent"." The construction could on occasion be mounted on a wheeled cart for transport. This portable architecture," ideally suited to pastoral mobility, continues in use today. People still graze their herds on the grasslands of northwestern China (fig . B13) and Mongolia (fig . BI4), moving sea- sonally for the benefit of their herds. The horse is still a prized possession (fig. BI5), and ready mounts are available tied near every gel: Displays of horsemanship, such as horse racing and "goat- grabbing," a local version of Buzkashi,· are sports that originated in antiquity and are still enor- mously popular today.

Felt-making is one of the most notable north- ern crafts and may have been invented by the peoples of the Eurasian steppes." Used both as clothing and to cover the gel" frame, felt plays an important part in native culture in Mongolia. Made by spreading, welling, beating, and rolling large quantities of sheep's wool, felt is a dense, windproof non-woven ' fabric, Rolling the felt wrapped around a log is often accomplished with the help of a horse.

Sima Qian in the Shiji wrote that the popular alcoholic drink among the Xiongnu, was fer- mented mare's milk. " The taste for fermented

13

mare's milk continues today in Mongolia. Known as airag in Mongolian, rujiu in Chinese, kumis in Turkish, and o.\)'gnln in Greek, it was also a favorite beverage of the Scythians mentioned by Herodotus."

In order to understand the artifacts from the Northern Zone, it is necessary to identify where, when, why, and how each artifact was made. Fortunately, the publication of recent archaeol- ogy in the People's Republic of China, Mongolia, Buryatia, Tuva, Siberia, and Russia enables the scholar to focus on specific local styles, artifactual types, and their manufacturing techniques.

The artifacts of the Northern Zone were not created for purely ornamental purposes by indi- vidual artists who expressed their own taste or feelings. The artifacts that characterize each regional group were to a great extent deter-

Fig. BI·1. Herding caule in Tov aimag, eastern Mongolia.june. 1996. PhOlO by Emma C. Bunker.

. t

Fig. B15. Mongolian horseman , Hemii aimag, eastern Mongolia.june, 1996. Photo by Emma C. Bunker.

Fig. B16. Bronze Chariot fining from Rujiazhuang, Baoji CQUIll}'. Shaanxi. IOlh-9lh centuries BC. After LII & Hu 1988. pl. XXIV:!.

mined by the land and the ecology. Each region supported a specific economic complex that yielded a specific range of artifacts traceable in the archaeological record. Within each region, people produced artifacts designed to meet their specific needs and developed an artistic vocabu- lary representing their beliefs, a combination that rendered the paraphernalia of each group unique. The mythological requirements of each cultural group and the types of local fauna often dictated the zoomorphic motifs. Every motif was a recognizable symbol imbued with auspicious meaning that was understood by the people who used and viewed it. In a society apparently with- out written records, people learned their history, value systems, clan recognition, religious beliefs , and heroic codes of ethics by pictorial and oral accounts. In a society where much depended on complex hierarchical relationships within clans, and between various groups, art served as a pow- erful binding force.

Archaeological activity over the last four decades challenges us with a wealth of new infor- mation. It is now possible to formulate a tentative

picture of how the ancient cu ltures o f the eastern Eurasian steppe peoples developed, but enor- mous gaps still remain.

mSTORICAL REFERENCES Ancient Chinese texts contain frequent refer- ences to various non-Chinese peoples located on the northern frontiers throughout the late sec- ond and first millennia Be. The concerns of the writers were primarily political or geographic. and it is often not clear whether the names given are what the northern peoples called themselves, Chinese transcriptions of their foreign names, or merely generic terms. The basic importance of early literary references is their significance as historical tradition."

Oracle-bone inscriptions" found in the vicin- ity of Anyang in Henan province indicate the presence of duo fang (the many alien states) beyond the northern frontiers with whom the Shang interacted during the late Bronze Age." Border clashes with these peoples apparently increased with the expansionist policies of Wu Ding, the powerful thirteenth-century BC Shang ruler. The fact that the closest sources for copper, tin, and salt were located to the northwest of Anyang could well have precipitated conflict be- tween the Shang and northwestern fang groups.

After the conquest of the Shang in 1045 BC by the Zhou, hostilities continued." The Zhou heartland, situated in the Wei River valley at Haojing near Xi'an in Shaanxi , had more con- tact with the pastoral groups on their northwest- ern borders than did the Shang. Occasionally, these pastoral groups were depicted in Zhou art. A bronze chariot Iiuing from a Western Zhou cemetery at Rujiazhuang, Baoji county, Shaanxi (fig. 816) shows a man wearing typical northern clothing-trousers, soft boots, ajacket decorated with two stag figures, and a wide belt."

The recently discovered Duo You ding in- scription records the sudden attack on the Zhou in the late ninth century BC by the Xianyun, whose home territory appears to have encom- passed modern Ningxia, Gansu, and northern Shaanxi." The Xianyun fought in chariots and not on horseback, but their sudden aggressive-

•

.

ness ma y have been kindled by a chain of events begun farther west due to the adoption and spread of mounted warfare and transhumance early in the first millennium Be. (See the Archae- ological Overview (or a discussion of this topic. )

In 771 BC the Quamong from the Ordos region forced thc Zhou to move their capital eastwards from Shaanxi to Luoyi near Luoyang in Henan, signaling the end of Zhou central authority. The following period, known as East- ern Zhou (771-22 1 BC), was characterized by a rise of regional power among a large number of feudal states which owed only a nominal alle- giance to the Zhou dynasty.

New peoples, the Rong and Di, are recorded during this period in the north and northeast, but they are clearly described as fighting on foot." Early literary references note the presence within the Zhou sphere of non-Zhou peoples who emphasized hunting and animal husbandry. The importance of hunting is also suggested by the p resentation of tiger and leopard skins to the Duke of Jin by the Duke of Wuzhong.3 1 The Wu zhong, another name for the Shanrong, lived north of the state ofYan in northern Hebei.

Marriage alliances between the elite of the northern Chinese states and various northern groups occurred during the Spring and Autumn period. The Di tribes replaced the Rong as the major non-Chinese peoples north of the Jin state by 662 BC,:'~ and served as mercenaries in the armies of the Jin.'\,1 The king of Zheng took a OJ wife in 636 BC," and Duke Xian ofJin (reigned 675-651 BC) had three Rong wives." His son , Duke Wen ofJin, spent twelve years in exile with the Di peoples."

The legendary rescue of Duke Mu of Qin in 645 BC by a band of "savages who had drunk wine and eaten a good horse" described in the Shi]!suggests an exposure to northern rituals that were incomprehensible to the Chinese." The cus- tom of men swearing brotherhood by drinking wine mixed with their own blood and the blood of a horse was a practice found among many peo- ples of the Eurasian steppes, including the Scythians in the Black Sea area described by Herodotus."

15

Trade was important to both the northern peoples and the dynastic Chinese. According to the Zhan guo ce (An nals of the Warring States), the most important commodities for the State of Zhao were "the horses ofTai [Dai] , the coursers of [the] Hu, and the jades of the K'unluns (Kunlun Mountains] ." N The need for good horses was a steady impetus for trade between the Chinese and their northern neighbors since the introduction of the chariot through contact with peoples in the Northern Zone during the late Shang period."

During the fourth century BC, the term "Hu" replaced "Rong" and ''OJ'' in reference to the northern peoples, indicating the appearance of new groups practicing mounted warfare." This change coincides with mention of the Rouzhi (Yuezhi) and the Xiongnu in Chinese historical texts, and the decision of King Wuling of Zhao, traditionally dated 307 BC, to have his troops adopt "barbarian dress" and the art of archery on horseback." The Chinese sources make a clear distinction between the Hu and the Xiongnu."

The major sources for information on the pastoral peoples during the last third of the first millennium BC are the Shiji, begun by Sima Tan (d ied 110 Be) and finished by his son Sima Qian (?145-?86 BC),H and the Hall Shu, compiled about AD 80 by Bangu and based primarily on the Shiji and on research done by Bangu's father Banbiao (AD 54) .1;

According to the Shiji, the Rouzhi were the dominant group on China's northwestern fron- tiers during the late fourth and third centuries Be.~6 They are described as pink and white in color, "a nation of nomads, moving from place to place with their herds." " "dressed in felt and furs.? " and living "in an area between the Qilian or heavenly mountains and Dunhuang.'?' Their location in what is present-day Gansu gave them access to the Central Asian Saka tribes farther west, to which they ma y have been in some way culturally related.

The supremacy of the Rouzhi over the local steppe confederacy was superseded by the Xiongnu at the end of the third century BC, when the Xiongnu recaptured their previous grazing lands in the Ordos region after the fall of

the Qin Dynasty in 206 BC. Shortly thereafter, the Xiongnu unified al1 of the eastern Eurasian steppe peoples into a vast steppe empire that rivaled the newly formed Han Empire of China during the last two centuries of the first millen- nium BC. According to the Shijl; the Xiongnu conquered the Rouzhi (Yuezhi) in the west, the Donghu in the east, and the Dingling in the north, as well as other groups to the north and west." The importance of the Xiongnu during the Western Han period was so great that Sima Qian devoted a whole chapter in the ShY-i to them, recording many details about their pas- toral existence."

"The animals they raise consist mainly of horses, cows, and sheep, but include such rare beasts as camels (see No. 233), asses, mules, and wild horses (see No . 224). They move about in search of water and pasture and have no walled cities or fixed dwellings, nor do they engage in any kind of agriculture. Their lands, however, are divided into regions under the control of various leaders. They have no writing. The little boys start out by learning to ride sheep and shoot birds and rats with a bow and arrow. Thus, al1 the young men are able to use a bow and act as armed cav- alry in time of war. It is their custom to herd their flocks in t~mes of peace and make their living by hunting, but in times of crisis they take up arms and go off on plundering and marauding expe- ditions. For long range weapons they use bows and arrows, and swords and spears at close range. If the battle is going well for them they will advance, but if not, they will retreat, for they do not consider it a disgrace to run away. (T h ey eat) the meat of domestic animals, and wear clothes of hide, or wraps of felt or fur,'?"

"On the death of his father, a son will marry his stepmother, and when brothers die, the remaining brothers will take the widows for their

16

own wives, In burials, the Xiongnu use an inner and outer coffin, with accessories of gold, silver, clothing, and fur, but they do not construct grave mounds or plant trees on the gra\'e, nor do they use mourning garments. When a ruler dies, the ministers and concubines who were favored by him and who are obliged to follow him in death often number in the hundreds of thousands."

Modern archaeology does not substantiate this description of wholesale immolation in the tombs, nor the total lack of fixed dwelling places. Sima Qian's description of the Xiongnu is an early description of transhurnance, in which tribes migrate seasonally in search of pasture and water, and return in the winter to permanent quarters where their livestock can have shelter from the harsh winter storms. The existence of Xiongnu settlements is confirmed by recent dis- coveries at Ivolga southeast of Lake Baikal." The mention in the Hall Shit of a Xion gnu crop fail- ure in 89 BC also indicates a settled element of the population."

Trade was an important issue between the Xiongnu and the Chinese during the Western Han period (206 BC-9 AD). The list of desirable imports recorded in the Yantielun (Discourses on Salt and Iron) includes mules, donkeys, camels, and horses of al1 colors and types, plus furs, rugs, carpets, and all manner of semi-precious stones." By contrast, numerous descriptions of Han gifts to the Xiongnu Shall)'1t (ruler) are recorded in the Sh y-i. In 192 BC the Shall)'1t Modun received an imperial princess, silk, brocade, and a gold- ornamented belt of leather." The Han Chinese established markets on the northern frontiers in 152 BC to cater to their Xiongnu clients as part of the Heqin policy.';" A princess was sent in 152 BC and another in 33 BC.'" In 51 BC gifts from the Chinese included a belt, clothing, a sword with a jade guard, gold, and bolts of fabric."

.

NOTES

I. The word "ste p pe ," meaning 6'l~\SSr plains in Russian, has been synon}'mons with the vast bell of marginal land that stretches across northern Eurasia from the Great Wall in northern China to thc Carpathian Mountains in eastern Europe since the publication of Karl.Jetllllar's landmark study, DieJnilrl'll SI'''JIN'IIl~'i1I(1'/; Baden-Baden, HUH . Thc English translation , A/t oj thr SI"NN~\, appeared in 19117. Some regions arc nor true steppc hut taiga (forest) and desert steppe, Thc desig- nation "Eurasian steppes" must be carefully applied, sincc "Eurasian" is sometimes misinterpreted by Chinese-speaking scholars as "European."

2. Gai Shan lin I981l; Gai Shanlin 1989. 3. ~Iu Shunying et "I. 19~J.I. p, 131, nos. 330-31. .1. Thc Ordos is only a small part of thc eastern Eurasian

steppes, First used by Minns 19·12, p. 30, the tenn is not appropriate as a general designation .

5. These terms should not bc (on fused. "Provenience" here refers to thc lind spot of an object, while "prove- nance" refers to an object's ownership history, frequently quite dilfercnt from its provenience. See Ella l~m6.

Ii. On the problem of unexcavated artifacts, see Bunker I!189.

7. Coined by Rostovtzeff 1922: sec also Rosiovtzeff 192!l. 8. ~linns 1!).J2. p. :10. 9. Frequently used to describe artifacts bclonging to

certain semi-nomadic grollps in the Eurasian stcppes between thc Sth and lst centuries BC, sec.Jacobson 1995, p. 2, this term has led somc Chinese scholars to describe Nonhern ZOliC artifacts as Scythian, in spite of the fact that thc nearest Scvthians lived thousands of miles to the west, '

10. Salmon}' 1933. Thc term is now obsolete. II. The term "nomadic" docs not mean pastoral, but

refers to pastoral nomadism or transhumance, the sea- sonal migration in search of pastures. Recent archaeol- 06')' reveals that while many eastern Eurasian steppc peoples were pastoral. they lIIight be trunshumant, settled or both. dcpcnding on the environment.

12. Rostovtzeff 1!122. 13. Scc So 8.: Bunker 1995. for a discu ssion of thc cultural

relationship between ancient dynastic China and its northern non-Chinesc neighbors.

1·1. Lin 'till 1986, p . 24S. and Di Cosmo 1997. 15. See Di Cosmo 1994 fOl';m cxtensh'c discussion of

ancicnt Inncr A~ian economics.

16. Williams Im16. 17. McGm'em 1939, pp. 44--15. For thc )rllllirlllll, sce

Loewe 19!13. IS. Andrcws 1979, pp. lO-22.

19. Yn ''ing-shih 1967, p . 40. Sec also Kriuko\' 8.: Kur}'lc\' 1992.

20. For discussion of the sport, sce Azo}' 19S2.

17

21. Barber 1991, pp . 215-22; Blllkell H179, pp . 7-8. 22. \\'almn W!13, \01. 2, p. 143. 23. Book IV. 2. Sec also :-.IcGm·ern 1939, pp. ,1·1- ·15.

2·1. For a h istory of the Northern Zone peoples, sec Di Cosmo 1997; Prusek 1971; Prusek 1966.

25, Keightlc}" 19i8, p. 13·\. Oracle honc inscriptions arc at present the most reliable source for Shang history. Many purported Shang texts, such as the Shang S"" (S/rijillg), wC\'C not written down until milch later ami reflect later traditions, Sec Shaughnessy 19!J3.

26. Prusek 1971, pp. 31)...t6; Chang 1980, pp . 2-18-59.

27. Shaughnessy 1991. I). 286. 1045 BC is thc date calcu- lated for the conqlJcst.

28. Lu & Hu Hl88, pl. XXIV; I. 29. Shaughnessy 19S:l--S5, p. (is.

30. Prusek 1971, p. 215. 31. Oi Cosmo 1997: Prusek 1971, p. 217 . 32, OJ Cosmo 1!197. 33. Lewis 19!JO, p. 271, no. 25, pp. 278-79, no. 103. 34. Prusek 1971, p. 158. 35. Wal~on 1989, p. 21. ss. Prusek 1971, pp. 15!1-(jO. 37. Prusek 19i I, pp. 131-33. 38. IV, 711. 39. Crump 1979, p. 324. ·10. Extensively discussed in Shaunghnessy 1988.

Sec also Linduff 1994: So & Bunker 1995, pp. 26-27.

41. Crump 1979, pp. 223-225. 42 . Walmn 191)1, \'01. 2, p. 159. 43.l'l1Isek 1971, p. 225; further discussed in Bunker 1992.

+1. Hulscwc Hl93, pp . -105-6. 45 . Hulsewe 1993, pp . 12~1-30. 46 . Watson 1961, \ '01. 2, pp. 16D-61;.Jcllmar 1967, p. 143. 47. Walmn 1961, \'01. 2. pp. 2(j7-68. 48. Prusek Hl71, p. 225 . 49 . Watson 1993, \'01 . 2, p . 234.

50. Watson 196 I , \"01. 2, r- 168. 51. Wal~on 196 I, \"01. 2, chapter 110, pp. 155-92. 52. \\'al~on 1961, \'01. 2, pp. 155-56. For thc association of

fclt and fur with the peoples of the Northern Zone in later history, sec Honcy 1992, pp. 3-0\.

53. \ \Tal~on 1993, \'01. 2. p. 130.

54. WaL~on 1961. \ '01. 2. p. 164. 55. ~Iin;ye\' 1996, p. 75. 56. ~lcGO\'cl'n 1939, p . -1:12, no . 29. 57. Galc 1931, pp . 14-15. 58. Biclcnstein 195-1, pp . 87-88. 59. Sec So & Bunker 1995, pp. 25-26. (iO. \\'al~on Hl61, \"01. 2, p. Ii-l. 61. Bic1enstein 1954, p . 91.