. 2
( 15)


production that could allow for the surplus in fodder and grains needed to
feed the animals. Thus we may conclude that plant domestication with
primitive farming was probably a precondition for the domestication of
animals. In the ¬rst instances of domestication of animals, which date to

Herodotus, Histories, IV:46. Quoted in John Gardiner-Garden, Greek Concep-
tions on Inner Asian Geography and Ethnography from Ephoros to Eratosthenes,
Papers on Inner Asia no. 9 (Bloomington: Research Institute for Inner Asian
Studies, 1987), p. 5.
Lewis Henry Morgan, Ancient Society, ed. Leslie A. White (Cambridge, Mass.:
Harvard University Press, 1964), p. 290.
S. I. Vejnshtein, “The Problem of Origin and Formation of the Economic-
Cultural Type of Pastoral Nomads in the Moderate Belt of Eurasia,” in The
Nomadic Alternative, ed. W. Weissleder (The Hague: Mouton, 1984), p. 127.


between 7500 and 6000 b.c. in the area of the Fertile Crescent, the animals
were kept as a nutritional complement to agricultural products. Some of
the animals “ for example, the ox, the onager, and the dog “ were then used
for other purposes, such as a means of transportation or as protection for
domestic animals against predators.
With characteristic insight, Owen Lattimore emphasized the importance
of the oasis economy for the evolution of Inner Asian steppe nomads.
He hypothesized that early domestication was possible in areas where the
natural environment was equally favorable to agriculture and to animal
husbandry. In the steppe oases, where large herbivores captured in the
steppe could be kept and fed, people gradually learned how to use them,
and eventually moved out into the open steppe, thus becoming “specialized
pastoralists.”11 Lattimore attributed the causes that ignited this process and
“pushed” the ¬rst nomads into the steppe to an economically more ef¬cient
adaptation to the natural environment of the steppe.12
Although Lattimore™s displacement theory is not supported by archaeo-
logical evidence, archaeologists have emphasized the importance of agri-
cultural production in the oases, which could also spark revolutionary
changes in economic patterns, social organization, and cultural develop-
ment. For example, the colonization of oases was at the root of what has
been called the Bactria-Margiana Archaeological Complex, as well as of the
later “Oxus Civilization” of Central Asia.13 In terms of the development of
conditions suitable for the advancement of pastoralism, the oasis environ-
ment is thought to have been conducive to the appearance of a mixed
farmer-pastoralist economy because the proximity of grasslands imposed
fewer restrictions on stock raising than did valley agriculture, where an
imbalance between humans and animals could be disastrous.14 According
to some theories, the oasis dwellers who specialized in stock breeding even-
tually separated themselves from their original environment and became
nomadic pastoralists.15 Yet these nomads retained close ties with farming
communities, upon which they remained to an extent dependant for agri-
cultural and handicraft products.
Extensive archaeological studies have made it clear that the line that sep-
arated early pastoral and farming communities, at least to the late Bronze

Owen Lattimore, Inner Asian Frontiers of China (Boston: Beacon Press, 1962
[1940]), pp. 158“63.
Ibid., pp. 63“64, 409“12.
Fredrik T. Hiebert, Origins of the Bronze Age Oasis Civilizations in Central Asia,
Bulletin 42 (Cambridge, Mass.: American School of Prehistoric Research, 1994).
A. Khazanov, Nomads and the Outside World (Cambridge: Cambridge
University Press, 1984), p. 89.
Shinobu Iwamura, “Nomad and Farmer in Central Asia,” Acta Asiatica 3 (1962):

Age, between the second and the ¬rst millennium b.c., was not neatly
de¬ned, and even specialized pastoral nomads are known to have engaged
in agriculture.16 In the Central Asian steppes, the ¬rst mixed pastoralist-
agriculturalist communities appeared following a period during the
Paleolithic in which a sparse population of hunters of large game (“mega-
fauna”) dominated the human landscape. Organized into small societies,
these communities were characterized by “relative stability, embodied in
nomad base camps, and intellectual progress, re¬‚ected in a large number
of prestige, symbolic innovations from statuettes to symbolic marks.”17 Pas-
toral cultures appeared ¬rst in the western Eurasian steppes, west of the
Urals, in the mid-third millennium b.c. These pastoral communities are
identi¬ed by their distinctive mound burials (kurgan).18
From the mid-third millennium b.c., the northern regions of Central
Eurasia, east of the Urals, were transformed by the shift from an economy
of predation to an economy of production. The steppe regions became pop-
ulated with diversi¬ed communities of Neolithic hunters and ¬shermen as
well as Bronze Age pastoralists and agriculturalists. Possibly because of a
climatic desiccation that affected soil productivity, a general transition to
more pronounced forms of pastoralism occurred in the steppe and semi-
desert areas of Eurasia.19 These environments created conditions favorable
to the breeding of animals, and agriculture could also be practiced. Pas-
toralists occupied the higher alpine pastures, such as those in the T™ien-shan
and the Altai regions, whereas along the lower course of the Amu Darya,
in Central Asia, animal breeding co-existed with irrigated agriculture
modeled after the system of irrigation of the Khorezmian civilization, at the
northeastern end of the Mesopotamian world.
Although herders became gradually more mobile and the aridization
of the climate made agriculture more problematic in several areas, this evo-
lutional trajectory did not necessarily mean the abandonment of agricul-
ture. The more common picture in central Asia during the ¬rst half of
the second millennium b.c., was the development of settled agro-pastoral

S. Vainshtein, Nomads of South Siberia. The Pastoral Economies of Tuva (Cam-
bridge: Cambridge University Press, 1980), pp. 145“65.
V. M. Masson, and T. F. Taylor, “Soviet Archaeology in the Steppe Zone: Intro-
duction,” Antiquity 63 (1989): 779“83.
These have been identi¬ed with a “macro-culture” known by the name of Yama
culture; see Natalia I. Shishlina and Fredrik T. Hiebert, “The Steppe and the Sown:
Interaction between Bronze Age Eurasian Nomads and Agriculturalists,” in The
Bronze Age and Early Iron Age Peoples of Eastern Central Asia, ed. Victor Mair
(Washington: Institute for the Study of Man, 1998), 1: 224“25.
P. M. Dolukhanov, “Pal©o©cologie de l™Asie centrale aux ages de la pierre et du
bronze,” in L™Asie centrale et ses rapports avec les civilisations orientales, des
origines à l™age du fer, M©moires de la Mission Arch©ologique Fran§aise en Asie
Centrale (Paris: Diffusion de Boccard, 1988), 1: 215“17.

societies that appear to have wielded considerable political and military
power. In addition to the aforementioned climatic changes, the interaction
between steppe peoples and more advanced agricultural cultures in the
oases of Central Asia and an internal evolution toward greater economic
specialization seem to have played an important role in the formation of
mobile pastoral societies, such as those of the early Andronovo period
(1900“1750 b.c.).20
According to Khazanov, the evolutionary pattern in the formation of
pastoral nomads has four phases: (1) sedentary animal husbandry, (2)
semi-sedentary pastoralism, (3) herdsman husbandry or distant pastures
husbandry, (4) semi-nomadic pastoralism and pastoral nomadism proper.21
David correlates these four stages with as many types of archaeological cul-
tures, thereby proposing an evolutionary development.22 The ¬rst phase is
represented by the early horse breeders, evidence for whom has been found
in the forest-steppe zone of southern Russia, at the site of Dereivka. Prim-
itive horseback riding, presumably a development in the steppe between the
Ural and the Volga in the mid-third millennium b.c., characterizes the
second phase, resulting in the increased mobility of these early pastoral
communities. The third phase, during the second millennium b.c., corre-
sponds to the ¬‚ourishing of the bronze culture in the steppe region and the
emergence of wheeled vehicles pulled by horses. Covered wagons provided
transportation and shelter during migratory moves, and light chariots may
have been used in warfare and for herd control. The fourth phase, from the
beginning of the ¬rst millennium b.c., corresponds to the emergence of
ancient nomads, when horseback riding had already evolved into a mature
stage of development. It is during the third phase, therefore, that we may
assume that horses began to be ridden, but how widespread this was, and
how important it was for the general social and economic life of these agro-
pastoral communities, is moot. These data today have to be reconsidered
in light of new evidence that places the earliest form of horseback riding in
the late fourth millennium b.c.

The Horse

The role of the horse in the transition from agro-pastoralism to fully devel-
oped mounted pastoral nomadism has been considered crucial. In particu-
lar, horseback riding allowed different herding strategies, making it possible

Shishlina and Hiebert, “The Steppe and the Sown,” 1: 231.
Khazanov, Nomads and the Outside World, pp. 19“25.
T. David, “Peuples mobiles de l™eurasie: contacts d™une p©riph©rie ˜barbare™ avec
le monde ˜civilis©,™ à la ¬n de l™Age du Bronze et au 1er Age du Fer,” in L™Asie
Centrale et ses rapports, pp. 159“68.


for fewer people to control larger herds, and, by allowing increased mobil-
ity, leading to expansion of the political and cultural horizons of early pas-
toralists.23 The horse is an animal that is notoriously dif¬cult to tame, and
according to some, the ¬rst equid to be domesticated was not the horse,
but the more docile onager.24 Nonetheless, the large number of horse
remains recovered at the site of Dereivka (4200“3700 b.c.), in the south
Russian Steppe, leaves no doubt that the domestication of the horse prob-
ably began in the ¬fth“fourth millennium b.c.25 Among horse remains
found at the Dereivka site, evidence of tooth wear caused by a hard bit,
dating from before the invention of the wheel “ therefore ruling out the
hypothesis that the horses had been hitched to carts “ indicates that the
Dereivka horses were not only bred but also ridden.26 The ¬nding of cheek
pieces made of deer antlers with holes drilled in them supports this con-
clusion. It is also based on the assumption that hard bits were in circula-
tion and that their use was generalized (bit wear was found on the tooth
of a single horse). Horseback riding is also assumed to have been developed
to control large herds of horses. It is not clear, however, if this evidence suf-
¬ces to prove that the horse was actually ridden, since horses might have
been used as draft animals even in the absence of the wheel.27 Even if the
¬rst horse breeders actually mounted the horse, the communities remained
predominantly agricultural, also raising pigs, cattle, and sheep. Although
the horse was the most important of the Dereivka animals, it remained so
within the economic context of early agro-pastoralists.28

David Anthony, “The Opening of the Eurasian Steppe at 2000 bce,” in The
Bronze Age and Early Iron Age Peoples of Eastern Central Asia, 1: 94“113.
J. F. Downs, “Origin and Spread of Riding in the Near East and Central Asia,”
American Anthropologist 63 (1961): 1193“1203.
Dmitriy Yakolevich Telegin, Dereivka: A Settlement and Cemetery of Copper Age
Horse Keepers on the Middle Dnieper, ed. J. P. Mallory, trans V. K. Pyatkovskiy
(Oxford: B.A.R., 1986); Marsha Levine, “Dereivka and the Problem of Horse
Domestication,” Antiquity 64 (1990): 727“40.
David Anthony and Dorcas Brown, “The Origin of Horseback Riding,” Antiq-
uity 65.246 (1991): 22“38.
A striking example of this use of the horse comes from a bronze ¬gurine adorn-
ing the handle of a dagger from the Rostonska burial, near Omsk, showing a
horse bridled at the mouth pulling a human being on a pair of skis; see E. N.
Chernykh, Ancient Metallurgy in the USSR (Cambridge: Cambridge University
Press, 1992), p. 228. This use of the domesticated horse, although attested later
than Dereivka, must have been possible during the Dereivka period and demon-
strates that tooth wear caused by a bit, in the absence of corroborating evidence
such as petroglyphs and other visual representations, is not necessarily evidence
of horseback riding.
Claudia Chang and Perry A. Tourtellotte, “The Role of Agro-pastoralism in the
Evolution of Steppe Culture in the Semirechye Area of Southern Kazakhstan

Nonetheless, these early domesticators played an important role in the
selection of the species. The Dereivka horses are not signi¬cantly different
from those recovered thirty-¬ve hundred years later, at Pazyryk, in the Altai
Mountains of Kazakhstan, although they are very different from the smaller
wild horses. The type of bones proves that human-controlled selection took
place, and, whether or not they invented riding, these early communities
must be given credit for their high level of specialization in breeding.29
In northern Kazakhstan a settlement of the fourth“third millennium b.c.
has been excavated where 99 percent of all animal remains recovered
belong to horses, indicating that those people “ who lived in large, semi-
subterranean houses “ specialized in horse breeding.30 At this site cheek
pieces have also been found, but the economic and social characteristics of
this settlement do not suggest a mobile lifestyle.
A conservative interpretation would date a signi¬cant impact of early
horseback riding on western and Central Asia to between the mid-third
and early second millennium b.c.31 The early horse-riding communities,
however, were not properly nomadic. Although some communities were
more or less mobile, riding in wheeled carts to follow their herds, their
pastoralism cannot be de¬ned as a regular cyclical migration seasonally
alternating among different pasture grounds; rather, this was “herder
husbandry” or at most semi-nomadism.32 These communities also depended
on agricultural production and had settlements; the migrations of some
groups documented by archaeological data were most likely permanent dis-
locations due to causes that could have ranged from pasture exhaustion to
climatic changes to external threats.33

during the Saka/Wusun Preriod (600 bce“400 bce),” in The Bronze Age and
Early Iron Age Peoples of Eastern Central Asia, 1: 266.
V. I. Bibikova, “On the History of Horse Domestication in South-East Europe,”
in Telegin, Dereivka, pp. 163“82.
A. P. Derevyanko and D. Dorj, “Neolithic Tribes in Northern Parts of Central
Asia,” in History of Civilizations of Central Asia, 1: 185.
M. A. Littauer and J. H. Crouwel, Wheeled Vehicles and Ridden Animals in
the Ancient Near East (Leiden: Brill, 1979), pp. 45“47, 65“68. The development
of horseback riding as a specialized activity, by the way, was a long process
that reached completion with widespread diffusion in Inner Asia of the saddle
and the stirrup at a still unde¬ned time that might have fallen within the ¬rst half
of the ¬rst millennium a.d.; see A. D. H. Bivar, “The Stirrup and Its Origin,”
Oriental Art, n.s. 1, 1 (1955): 61“65. On the stirrup, see also Lynn White,
Jr., Medieval Technology and Social Change (Oxford: Clarendon Press, 1962),
pp. 14“28.
Khazanov, Nomads and the Outside World, p. 93.
Earlier authors tended to see such large-scale migrations as evidence of the exis-
tence of pastoral nomads in the Central Asian steppe as early as the third mil-


The Aryan terminology that appears in a fourteenth-century b.c. Hittite
treatise on horsemanship illustrating the training of the chariot horse
suggests that such training may have been developed by steppe Indo-
European peoples, perhaps the Iranian ancestors of the Achaemenian
dynasty.34 What does seem clear is that most improvements in the training
and domestication of the horse were achieved by a people who were already
familiar with animal breeding and who had been specializing in this eco-
nomic activity, although they still practiced farming. It is also possible that
the steppe environment allowed contacts among early pastoralists that
favored the spread of horse-training techniques. Nevertheless, the transi-
tion to actual pastoral nomadism as practiced by horseback riders was
probably not completed until the beginning of the ¬rst millennium b.c., and
the ¬rst Scythian mounted archers appear on the scene only in the tenth or
ninth century b.c.35

Andronovo™s Chariots

Climatic changes may have led to the increased mobility of the steppe
people starting in the third millennium b.c. In the Bronze Age the techno-
logical level of the people of the steppe region was greatly advanced by the
widespread introduction of metal artifacts into all branches of production,
leading to the emergence of groups skilled in metallurgy who moved about
in wheeled vehicles.36 The earliest wheeled vehicles in the Eurasian steppes
were heavy wagons, dated to 2900 b.c. and attributed to the Yamnaya
culture, located on the lower Dnieper.37 Only much later, in the late third
and early second millennium b.c., do wheeled vehicles appear east of the
Urals, in connection with the spread of the Andronovo people. The people
of this widespread and singularly successful Central Asian culture were
adept at animal husbandry, and their craftsmen had mastered the art of

lennium b.c. Cf. Raphael Pumpelly, ed., Explorations in Turkestan. Expedition
of 1904. Prehistoric Civilizations of Anau, 2 vols. (Washington: Carnegie Insti-
tution, 1908).
Pentti Aalto, “The Horse in Central Asian Nomadic Cultures,” Studia Orientalia
46 (1975): 4“7. On the linguistic evidence for the word “horse,” see Juha
Janhunen “The Horse in East Asia: Reviewing the Linguistic Evidence,” in The
Bronze Age and Early Iron Age Peoples of Eastern Central Asia, 1: 415“30.
Franz Hancar, Das Pferd in praehistorischer und frühistorischer Zeit (Wien:
Herold, 1956), pp. 551“63.
V. Dergachev, “Neolithic and Bronze Age Cultural Communities of the Steppe
Zone of the USSR,” Antiquity 63.241 (1989): 801.
Anthony, “Opening of the Eurasian Steppe,” 1: 102“103.


bronze metallurgy. These metalworkers were able to manipulate alloys
so that the quality of the bronze would be harder or tougher according to
the speci¬c function of the weapons and tools that they made. Later,
Andronovo people, because of their long-distance migrations, may have
played an important role in the development of oasis economy, a point sug-
gested by similarities in the nomadic ceramics from distant areas.38
The broader utilization of mineral ores from multiple independent met-
allurgical sources and the expansion of the use of wheeled vehicles and
bronze objects are all signs of economic development. Yet the concomitant
abundance of weapons indicates that there were increasing tensions among
various communities:39 During this “second epochal type of culture,” after
the Neolithic revolution, it seems that “[t]he struggle for forcible redistrib-
ution of pasture and accumulated wealth [gave] rise, at a certain stage, to
a type of militarization of society that found expression and progress in the
production of weapons.”40
Moreover, chariots, mostly used for war, should be distinguished from
the four-wheeled wagons and two-wheeled carts used to transport people
and goods. Though based on pre-existing models of wheeled vehicles, the
war chariot seems to have been developed by the agro-pastoralists of the
Andronovo culture.41 The chariots were light and fast; they had spoked
wheels and a rear axle supporting a box in which normally no more than
two warriors could either stand, kneel, or sit.42 Recent discoveries have
revealed fully formed chariots with spoked wheels of the Sintashta-
Petrovka culture, and these may date to as early as 2026 b.c.43 These are
technically and conceptually very similar to chariots found both in western
Asia “ at the Lchashen site in the Caucasus “ and in East Asia, at the Shang
royal site of An-yang. However, according to the expert opinion of Littauer
and Crouwel, the Sintashta-Petrovka chariots had a gauge and especially a
wheel nave that were too narrow, resulting in a very unstable structure that
could not have been ef¬cient for hunting, racing, or ¬ghting. The inef¬-
ciency of this type of chariot is made even more evident by the probable
availability of horseback riding, which clearly was a superior means of

Hiebert, Origins of the Bronze Age Oasis Civilization in Central Asia, p. 135.
V. M. Masson, “Les cultures anciennes d™Asie moyenne: dynamique du d©veloppe-
ment, occupation des aires ©cologiques, rapports culturels,” in L™Asie centrale et
ses rapports, p. 34.
Masson and Taylor, “Soviet Archaeology in the Steppe Zone,” p. 780.
V. F. Gening, “Mogil™nik Sintashta i problema rannikh indoiranskikh plemen,”
Sovetskaia Arkheologiia 4 (1977): 53“73.
Stuart Piggott, The Earliest Wheeled Transport: From the Atlantic Coast to the
Caspian Sea (Ithaca: Cornell University Press, 1983), p. 95.
David Anthony and Nikolai B. Vinogradov, “Birth of the Chariot,” Archaeology
48.2 (1995): 38.


transportation, herd control, and warfare. Hence Littauer and Crouwel
imply that chariots, which originated in the Near East, where a continuous
line of development can be seen from four-wheeled carts to two-wheeled
carts to light chariots, were taken on by the nomads predominantly for the
symbolic of accompanying the dead to their burial place. In other words,
the “prestige value” that the chariot enjoyed in the Near East prompted its
construction in the steppe, not its “workaday” usefulness.44
The Andronovo people™s unquestioned economic superiority propelled
this culture across the Eurasian steppe from the Urals to South Siberia
whether by horseback or by chariot, and numerous studies indicate that the
chariot was imported into China from the west, through Central Asia, pos-
sibly around the thirteenth century b.c.45 Although no de¬nite evidence has
emerged yet, it is plausible that the Andronovo culture™s contacts with the
eastern part of Central Asia, and especially its interaction with the archae-
ological context of northwestern China (present-day Sinkiang), may be
responsible for the introduction into China of the chariot, whose western
origin is doubted only by few. These contacts are attested to by the archae-
ological evidence, including similar bronze artifacts such as axes, celts
shaped as spades, and other implements.46
The earliest Chinese chariots to have been found were discovered in
burials of the Shang dynasty at An-yang; buried with the chariots were
their horses and drivers, who served as sacri¬cial victims. This type of
vehicle was used by the aristocracy for display, for hunting, and in war. It
was made of a central pole, with one horse harnessed on each side, and a
box “ typically rectangular or oval; a spoked wheel was at each end of an
axle attached crosswise to the rear end of the central pole. The chariot

M. A. Littauer and J. H. Crouwel, “The Origin of the True Chariot,” Antiquity
70 (1996): 934“39.
Edward L. Shaughnessy, “Historical Perspectives on the Introduction of the
Chariot into China,” Harvard Journal of Asiatic Studies 48.1 (1988): 189“237;
Stuart Piggott, “Chinese Chariotry: An Outsider™s View,” in Arts of the Eurasian
Steppelands, ed. Philip Denwood, Colloquies on Art and Archaeology in Asia no.
7 (London: Percival David Foundation, 1978), pp. 32“51; Littauer and Crouwel,
Wheeled Vehicles and Ridden Animals in the Ancient Near East. For a detailed
study of the Chinese chariots of the Shang dynasty, see Magdalene von Dewall,
Pferd und Wagen im fruhen China (Bonn: Habelt, 1964), pp. 109“77. See also
Robert Bagley, “Shang Archaeology,” in Cambridge History of Ancient China,
ed. Michael Loewe and Edward L. Shaughnessy (Cambridge: Cambridge Univer-
sity Press, 1999), pp. 202“208.
E. E. Kuzmina, “Cultural Connections of the Tarim Basin People and Pastoral-
ists of the Asian Steppes in the Bronze Age,” in Bronze Age and Early Iron Age
Peoples of Eastern Central Asia, 1: 63“93; Ke Peng, “The Andronovo Artifacts
Discovered in Toquztara County in Ili, Xinjiang,” in The Bronze Age and Early
Iron Age Peoples of Eastern Central Asia, 2: 573“80.

appears in China already fully formed.47 There seem to have been no other
wheeled vehicles, such as wagons or carts, pulled by cattle or equids, in use
in China before the introduction of the chariot. Later, during the Chou
dynasty, chariots were a common feature of the funerary inventory of the
richest tombs, as well as forming the core of both the Chou and foreign

Further Cultural Developments

Much of what happened in the second millennium b.c. is still open to ques-
tion: we see the depopulation of some areas of Central Eurasia, and more
contacts take place between settled farmers and mobile herders. Moreover,
the movement of people, formerly occurring from west to east, seems to be
partly reversed as we can also track an east-to-west movement, possibly as
part of a more general migration phenomenon radiating from the Sayano-
Altai region. Around the late second millennium b.c. the progress of Central
Asian peoples in metallurgy was stimulated by the so-called Seimo-Turbino
transcultural complex. The Seimo-Turbino became consolidated as a cul-
tural phenomenon, including both pastoralists and mobile Neolithic hunters
of the forest, in the seventeenth century b.c. Chernykh places the point of
“departure” of these fast-moving people in the Sayano-Altai region, further
east than the eponymous Uralic sites that lent their names to the complex.
In this region the encounter of pastoral steppe cultures with metal-working
forest people gave rise to a metallurgically advanced, extremely mobile,
warlike society. From this part of Inner Asia the Seimo-Turbino people
spread westward, a movement well documented by Chernykh based on
metallographic analysis.48
The Rostovka site, on the Irtysh River, is representative of the eastern,
or Siberian, variety of the Seimo-Turbino complex. Here bronze produc-
tion, consisting of tin bronze and associated with the ancient mines of the
Rudny Altai Mountains, was mostly comprised of weapons such as sock-
eted axes, socketed spearheads, and dagger-knives. These tin bronzes
eventually reached the Urals, evidence of the westward motion of the
Seimo-Turbino people. A further clue to the Altai region as the original
home of the Seimo-Turbino people is found in their iconography, which
includes animals, such as the wild sheep, typical of the Altai and T™ien-shan
regions. As they moved west they came into contact with the Andronovo
people, and they may have disappeared as a separate cultural unit by the
¬fteenth century b.c. It is possible, however, that the Seimo-Turbino met-

Yang Pao-ch™eng, “Yin-tai ch™e-tzu te fa-hsien yü fu-yüan,” K™ao-ku 1984.6:
546“55; K™ao-ku 1984.6: 505“09.
Chernykh, Ancient Metallurgy in the USSR, pp. 215“31, chapter 9.


allurgical phenomenon played an important role in the formation of
Karasuk metallurgy in the Sayano-Altai region and western Mongolia.
From around the twelfth to the eighth century b.c. a new culture, known
as Karasuk, came to dominate the region of South Siberia, the Yenisei and
Minusinsk Basin, and the Altai extending as far as western Mongolia.49 Like
their neighbors in northern China, the Karasuk people had a mixed
economy, which, although mainly based on livestock, also relied on agri-
culture and other supporting activities.50 Findings of antelope and deer
bones suggest extensive hunting by the Karasuk, whereas cattle and horse
remains indicate that animal husbandry was their main productive activity.
During the Karasuk period improved metallurgic technology resulted in
important innovations, among them the bronze bit, which greatly enhanced
the possibilities offered by horseback riding.
This vast cultural complex extended its in¬‚uence and contacts to north-
ern China, and the Karasuk metal inventory presents many analogies with
the bronzes of the so-called Northern Zone complex. For instance, we ¬nd
a type of knife with a hunched blade, similar to the “foreign” bronze knives
found at An-yang and widespread across northern China, and similar to
the Chinese daggers of the “Ordos” style, with a narrow guard. The pick-
axes display tubular sockets for hafting such as those of the Northern Zone,
though the blade™s pointed cutting edge may have been derived from a
Shang prototype.51 These similarities indicate that the Northern Zone of
China was in contact with a wide cultural area and possibly functioned as
a clearinghouse for new technical developments into and out of China.

Early Nomadic (Scythian-type) Cultures
in the Eurasian Steppe

The Karasuk people lived in felt tents, traveled in hooded carts, ate a variety
of dairy products, and adapted remarkably well to a mobile way of life.52
Yet “true” early pastoral nomads, that is, pastoralists moving with their
herds according to a ¬xed seasonal cycle, appear only in the late Bronze
and early Iron Age, a phenomenon that brought about a great expansion
across Central Eurasia of mounted warlike nomads. The emergence of this
new anthropological type is attested to by the iconography of tenth-century
b.c. Iran and ninth-century b.c. Assyria and is con¬rmed by Assyrian and

S. V. Kiselev, Drevniaia Istoriia Iuzhnoi Sibiri (Moscow: Nauka, 1951).
A. P. Okladnikov, “Inner Asia at the Dawn of History,” in Cambridge History of
Early Inner Asia, ed. Denis Sinor (Cambridge: Cambridge University Press, 1990),
pp. 85“88.
The metallurgical cultures of northern China will be discussed in the next chapter.
Okladnikov, “Inner Asia at the Dawn of History,” pp. 94“95.

Greek sources of the ninth and eighth centuries b.c. who assign these groups
names such as Cimmerians, Scythians, and Sakas.53
These ethnonyms are associated with the pastoral nomadic peoples who
inhabited the region of the Pontic Steppe north of the Black Sea and Central
Asia from the eighth century b.c. From the eastern to the western parts of
the Eurasian steppe region, these early Iron Age peoples shared a cultural
universe that was remarkably homogeneous, at least at the level of their
material development and artistic expression. These “early nomads” or
Scytho-Siberian peoples, as they are sometimes called by archaeologists,
engaged in pastoral nomadism as their primary economic activity and thus
their livelihood was based on cattle, sheep, and horses. Most prominent in
this society was the aristocratic class of mounted warriors, specially trained
as light archers, that held a privileged position over other groups.
Herodotus™s description of the Scythians delineates the social hierarchy of
the early pastoral nomads: the Royal Scythians at the top of the ladder and
the commoners, identi¬ed as “agricultural,” “nomadic,” or “free” Scythi-
ans, below them. In such a martial society, weapons were produced in abun-
dance and were buried with the warriors. The horse was essential not only
for herding and in battle but also for the nomads™ technological develop-
ment. Horse-harness components constituted a large portion of the metal-
lurgical production of nomadic cultures. The horse was also integral to the
nomads™ belief system, and horse sacri¬ce played a prominent role in funer-
ary rites. Another distinctive element of their culture was the “animal-art
style” of nomad metalwork.
Archaeologists have adopted several of these traits for classifying early
nomadic cultures. In particular, the presence of the so-called Scythian triad
“ weapons of bronze and iron, horse gear, and artwork in the “animal style”
“ in the funerary inventory has been regarded as the common denomina-
tor of steppe nomadic cultures, which are also often identi¬ed with the
typical grave mounds (kurgan) where horse and horseman were buried
together.54 However, scholars still have no de¬nite explanation for the

The ¬rst historically documented steppe nomads are the Scythians, treated exten-
sively in Greek historiography from the ¬fth century onward. John Gardiner-
Garden has written several other works on Scythian historiography, published by
the Research Institute for Inner Asian Studies (Bloomington). These are Apol-
lodoros of Artemita and the Central Asian Skythians (Papers on Inner Asia no.
3, 1987); Herodotos™ Contemporaries on Skythian Geography and Ethnography
(Papers on Inner Asia no. 10, 1987); Ktesias on Central Asian History and
Ethnography (Papers on Inner Asia no. 6, 1987); and Greek Conceptions on Inner
Asian Geography and Ethnography from Ephoros to Eratosthenes (Papers on
Inner Asia no. 9, 1987).
However, it is clear that funerary rites varied widely, and the strong presence of
local or regional cultural elements makes burial type a less valuable diagnostic


appearance of the mounted nomadic culture of the steppe around the end
of the second and beginning of the ¬rst millennium b.c. This lack of con-
sensus is all the more remarkable because of the gap of several centuries
between when the prerequisites of pastoral nomadism were achieved “ the
dairying techniques, use of animal-driven wheeled transport, and horse-
manship available in areas of Central Asia, the Kazakh Steppe, and Siberia
by the second half of the second millennium b.c.55 “ and the actual appear-
ance of the nomadic cultures, in the early ¬rst millennium b.c.56
Most scholars tend to privilege “internal” factors, such as overpopu-
lation, aridization, or simply an increase in the degree of specialization
and division of labor between agriculturalists and pastoralists, rather than
external ones, such as invasions or cultural contacts, for the nomads™
appearance. According to some scholars, pastoral nomadism evolved
naturally from advanced pastoralism, and was the result of both larger
herd size and the accumulated experience of a more progressive, mobile
pastoral economy. Climate changes may have led to a reduction in arable
land, as a result of which these formerly sedentary land cultivators
and cattle breeders were obliged to become nomads. Some scholars believe
that an essential contribution to the evolution of pastoral nomadism
came from the forest hunters, who borrowed animals from their sedentary
neighbors and then, after they began to use the horse, moved into the
Yet another theory holds that, in the late second to early ¬rst millennium
b.c., as a result of overpopulation, the cattle breeders and agriculturalists
of the oases gave rise to groups of pastoralists who herded their animals
into the surrounding deserts, which then became the “barbaric periphery
of the agricultural oases.”58 As already discussed, this view is commonly
associated with the “theory of displacement” (the term is Khazanov™s),59
enunciated by Lattimore, according to which pastoral nomadism emerged
as an effect of the “push” on marginal populations, already settled on the
edge of the steppe and in the oases, exercised by the expansion of seden-
tary agriculturalist societies.60 This type of “impact-response” relationship

Khazanov, Nomads and the Outside World, p. 94.
See Vejnshtein, “Problem of Origin and Formation of the Economic-Cultural Type
of Pastoral Nomads in the Moderate Belt of Eurasia,” in The Nomadic Alterna-
tive, pp. 127“33.
Ibid., pp. 130“31.
Mariana A. Itina, “The Steppes of the Aral Sea in Pre- and Early Scythian Times,”
in Foundations of Empire: Archaeology and Art of the Eurasian Steppes, ed. Gary
Seaman (Los Angeles: Ethnographics Press/University of Southern California,
1992), p. 50.
Khazanov, Nomads and the Outside World, p. 89.
Lattimore, Inner Asian Frontiers of China, pp. 328, 412.

between agriculturalists and underdeveloped “frontiersmen” is, however,
not supported by current archaeological research.61
Finally, according to Gryaznov, in the eighth century b.c. some tribes in
different parts of the steppe took to nomadism as it gradually became a
more rewarding economic activity. What resulted was a generalized increase
in aggressive warfare among the inhabitants of the steppe, aimed at secur-
ing territory suf¬cient to support nomadic herding. In search of booty and
land, these tribes then attacked the sedentary peoples. Settled communities
were thus compelled to turn to a nomadic life themselves, when conditions
permitted, in order to effectively defend themselves. Their agricultural pro-
duction was considerably reduced, preserved only at the tribes™ winter
pastures.62 Rudenko, on the other hand, proposed a far more gradual
transition, spanning several centuries, to “true” pastoral nomadic status.63
Most scholars today would agree that mature pastoral nomadic cultures
were built on the achievements in both technology and social and political
organization of previous agro-pastoral peoples. Pastoral nomadic commu-
nities, moreover, often appear to be part of larger social con¬gurations
based on economic diversi¬cation, according to the possibilities offered by
the particular environment in which they found themselves.
That there were multiple population shifts is uncontested. As a result of
climatic changes and a sharp rise in aridity, Bronze Age inhabitants of the
steppe started moving south, following the riverways, in search of pasture.64
These movements were not synchronous, and reverse movement seems also
to have occurred, creating a complex picture of intersecting streams and

For a critique of the popular “diffusionist” theory, according to which the state
emerged among the nomads only in conjunction with the formation of the
state among the agricultural societies, see Lawrence Krader, “The Origin of the
State among the Nomads of Asia,” in The Early State, ed. Henry J. M. Claessen
and Peter Skalník (The Hague: Mouton Publishers, 1978), pp. 93“107.
M. P. Gryaznov, The Ancient Civilization of Southern Siberia (New York: Cowles
Book Co., 1969), pp. 131“32.
S. Rudenko, Kul™tura naseleniia tsentral™nogo Altaia v skifskoie vremia (Moscow:
Nauka, 1960), p. 197.
Scholars have noticed that by the mid-second millenium b.c. some areas of north
and south Kazakhstan, Semirechiye, and northeastern Kirghizstan were appar-
ently abandoned, only to become populated again during the early Saka period.
The southern Ural steppe, which was densely populated in the Bronze Age, was
also deserted by the end of the second millennium; it was repopulated in the sixth
century b.c. by people from Kazakhstan and Central Asia. See V. A. Alekshin,
“Problème de l™origin des cultures arch©ologiques du n©olithique et de l™‚ge du
Bronze en Asie centrale (d™après les rites fun©raires),” in L™Asie centrale et ses rap-
ports, pp. 255“64. See also Leonid T. Yablonsky, “Some Ethnogenetical Hypothe-
ses,” in Nomads of the Eurasian Steppes in the Early Iron Age, ed. Jeannine
Davis-Kimball et al. (Berkeley: Zinat Press, 1995), p. 243.

currents responsible for the genesis of the material culture of the steppe
nomads, especially in metallurgy, and for the development of the nomads™
artistic taste.

Chronology and Distribution

Scholars have believed for some time that the earliest foci of a “Scythian”
culture could be found in the west, that is, in the region of the Volga River,65
and that mounted nomadism was “imported” into the eastern steppe across
Central Asia.66 Although a logical assumption, given that a number of essen-
tial innovations (horseback riding, wheeled vehicles, metallurgy) entered
Central Asia from the western and southwestern ends of the steppe, several
decades of archaeological work, mainly by Soviet archaeologists after
World War II, have made it increasingly clearer that at some point the
process may have been led by the eastern steppe regions, including South
Siberia, Tuva, the Sayano-Altai region, and western Mongolia.
Today the consensus tends to privilege Central Asia as the place of origin
and dispersal of the Scytho-Siberian cultures.67 Archaeological features of
the nomadic Scythian culture were ¬rst recognized in the large kurgan
burials of the Altai and T™ien-shan regions of the sixth to fourth century
b.c. Kiselev™s work in the Minusinsk Basin and Yenisei Valley in southern
Siberia revealed a culture that replaced the Karasuk around the seventh
century b.c., known as Tagar.68 These were still semi-nomadic people who
essentially continued the traditions of the Karasuk culture, although in the
Yenisei region Tagar society supported an aristocracy similar to that of the
In the Altai region and in Tuva early nomads appear possibly as early as
the ninth century b.c. Gryaznov™s chronology locates the beginning of the
Altaic-Scythian period in the ninth century b.c. This was followed by the
Maiemir period (seventh to ¬fth century b.c.), by the Pazyryk period (¬fth
to third century b.c.), and, ¬nally, by the Shibinsk (Shibe) period (second

Karl Jettmar, Art of the Steppes (New York: Crown Publishers, 1967), p. 215.
Supported by linguistic evidence, Heine-Geldern has argued in favor of the thesis
of an Indo-European migration that took place from the Pontic region to eastern
Asia during the ninth and eighth centuries; see R. Heine-Geldern, “Das Tocharen-
problem und die Pontische Wanderung,” Saeculum 2 (1951): 225.
Esther Jacobson, The Art of the Scythians (Leiden: Brill, 1995), pp. 29“39. For
a general introduction to the history of Scythian and Sarmatian tribes, see A. I.
Melyukova, “The Scythians and Sarmatians,” in Cambridge History of Early
Inner Asia, pp. 97“117.
Kiselev, Drevniaia istoriia Iuzhnoi Sibiri, pp. 302“303.
Gryaznov, The Ancient Civilization of Ancient Siberia, p. 217.


century b.c. to ¬rst century a.d.). From the eighth“seventh century onward
two different groups in the Altai can be identi¬ed by their respective burials:
the kurgan and the stone box. The chronology of the early nomadic cul-
tures of the early Iron Age in the Altai that is generally accepted today
follows Gryaznov™s model closely and is divided into an early stage (eighth
to sixth century b.c.), a middle stage (¬fth to third century b.c.), and a late
stage (second century b.c. to ¬rst century a.d.).70 The chronological upper
limit of the early nomads has found con¬rmation in the work carried out
by Gryaznov and Grach at Arzhan, in Tuva. This monumental burial site,
dated at the earliest to the ninth or, more possibly, to the eighth century
b.c., is synchronous with the appearance across the steppe region of Inner
Asia of a unitary cultural layer, that of the early Iron Age, which stretched
from the Pontic area to the eastern Altai.71 In Tuvan archaeology, this ¬rst
period was also followed by a middle period “ slightly earlier than else-
where “ dated to the seventh and sixth centuries b.c., and by a late period,
dated from the ¬fth to the third century b.c. At the end of its late stage of
early nomadic (“Scythian-type”) evolution, Tuva entered the “Hunno-
Sarmatian” period common to vast parts of Central Eurasia and generally
identi¬ed with the arrival of new pastoral nomadic cultures from the east
in concomitance with the expansion of the Hsiung-nu empire. Data from
wooden remains from the early nomadic sites (Arzhan, Tuetka, Pazyryk,
and Shibe), based on C14-calibrated analysis adjusted to dendrochronolog-
ical calculations (based on tree rings), con¬rm Gryaznov™s chronology. The
data also show that, given the con¬rmed early dating of the Arzhan kurgan
(tenth to ninth century b.c.), the animal style typical of nomadic steppe art
was probably developed as a native tradition, and not as an adaptation of
Near Eastern motifs.72
In analyzing the cultural evolution of the early nomads, we must keep
in mind that the phenomenon was not linear and that it had a vast range
of regional variations based on each group™s adaptation to different eco-
logical conditions and different forms of economic development. Not all
Bronze Age agro-pastoral communities became nomadic; indeed, some
peoples migrated to river valleys and took up plough agriculture.73 More-
over, early nomadic communities often showed continuity with preceding

Nikolai A. Bokovenko, “History of Studies and the Main Problems in the Archae-
ology of Southern Siberia during the Scythian Period,” in Nomads of the Eurasian
Steppes in the Early Iron Age, pp. 255“61.
For a map of all the early nomadic (Scythian-type) cultures of Eurasia, see M. P.
Grjaznov, Der Großkurgan von Arzan in Tuva, Südsibirien (München: Verlag
C. H. Beck, 1984), p. 77.
Mark E. Hall, “Towards an Absolute Chronology of the Iron Age of Inner Asia,”
Antiquity 71 (1997): 863“74.
Yablonsky, “Some Ethnogenetical Hypotheses.”

Bronze Age steppe communities, in both their cultures “ funerary rites,
pottery, and metallurgy “ and their physical aspects. During the Scythian
period a plurality of cultural traditions commonly coexisted in the same
general area. For instance, among the nomadic people of the seventh“sixth
century b.c. in Khorezm, in the regions between the lower Syr Darya and
the Amu Darya, different types of funerary rites and burial methods were
practiced within a fairly consistent cultural context suggesting contacts
further east, with the steppes of Kazakhstan and southern Siberia.74 Perhaps
the most plausible explanation for this phenomenon is that the nomadic
communities formed in the Altai, Tuvinian, and Kazakh Steppes moved
west in waves of varying speed and intensity and gradually mixed with the
local people, thus generating multiple cultural combinations.

Social and Economic Development

In general, the passage from a society whose economy is mixed to a society
whose economy is dominated by a nomadic element is accompanied by pro-
found changes in social structure and organization.75 Eventually, nomadic
specialization was instrumental to the formation of more complex types of
integration among peoples whose economic bases were varied and depen-
dent upon local environmental characteristics.76 The evolutionary trajectory
of the early nomads proceeded in two directions. Internally, nomadic-
dominated societies attained greater social strati¬cation, typi¬ed by larger
burials and funerary inventories marked by an abundance of prestige goods.
The aristocratic warrior class is also likely to have had access to whatever
surplus may have been produced under their rule, including farm products,
metal tools and weapons, and perhaps even trade revenues. Externally,
nomadic-dominated societies tended to have more contacts with neigh-
boring communities, sometimes over great distances. The relationship the
nomads established with these communities was one of commercial
exchange and economic symbiosis but also one of latent hostility, which
sometimes resulted in con¬‚ict. As the early nomads spread across Central
Eurasia, their most advanced communities established themselves in a posi-
tion of supremacy over non-nomadic peoples. In the northern Caucasian
Steppe, for example, where signs can be found of a Scythian culture as early

Leonid T. Yablonsky, “Material Culture of the Saka and Historical Recon-
struction,” in Nomads of the Eurasian Steppes in the Early Iron Age, pp. 201“
G. E. Markov, “Problems of Social Change among the Asiatic Nomads,” in
Nomadic Alternative, p. 306.
This case is argued in particular in Chang and Tourtellotte, “Role of Agropas-
toralism,” 1: 264“79.

as the eighth century b.c., the nomads clearly came to dominate the local
sedentary population.77
The formation of a military aristocracy, organized into politically author-
itative clans and lineages, occurred, naturally, at different times in different
places. The Arzhan complex is so large and impressive that there can be no
doubt that this was the tomb of a very powerful man.78 The creation of
centers of tribal unions is visible particularly from the seventh“sixth century
b.c. onward. For instance, the emergence of tribal unions in the southern
Ural Steppe and Volga-Don inter¬‚uvial zone between the sixth and the
fourth century b.c. can be deduced from the partition of the cemeteries into
sectors, of which some were reserved for the members of a military elite,
including, possibly, those with a “supra-tribal” position (i.e., a class of
noblemen within a confederation of tribes).79 The group of kurgans found
on the Ilek River, where larger and more complex burial structures are set
apart from the others, may be one example of this social structure. Indeed,
this particular site has been taken as evidence of the formation of a tribal
union that had its center here, and whose aristocratic chiefs were buried
with greater pomp.80
In the Altai region, the earliest burials are much simpler than the kurgans
of the Pazyryk stage.81 More important, the Pazyryk culture presents a

Vladimir Petrenko, “Scythian Culture in the North Caucasus,” in Nomads of the
Eurasian Steppes in the Early Iron Age, pp. 5“22.
The complex, a circular mound built over an intricate wooden structure, includes
seventy burial chambers whose sizes vary between 15 square meters and 150
square meters. At the center, two smaller cells are located within a larger enclo-
sure, presumably the place of rest of the king and queen. Various noblemen, atten-
dants, and a total of 160 horses were also buried in the structure, which has
yielded an abundance of weapons, jewelry, and bronze decorations in the animal
style. See Grjaznov, Großkurgan von Arzan in Tuva, Südsibirien; Nikolai A.
Bokovenko, “The Tagar Culture of the Minusinsk Basin,” in Nomads of the
Eurasian Steppes in the Early Iron Age, p. 302.
Marina Moshkova, “Sarmatians, Concluding Remarks,” in Nomads of the
Eurasian Steppes in the Early Iron Age, pp. 185“88.
Vladimir Dvornichenko, “Sauromatians and Sarmatians of the Eurasian Steppes:
The Transitional Period from the Bronze Age,” in Nomads of the Eurasian
Steppes in the Early Iron Age, p. 106.
The Pazyryk culture is one of the better studied early nomadic “kurgan” cultures
of Central Asia. Its chronology was recently re-worked by Hiebert, who con-
cluded that some Pazyryk burials could have been as early as the ¬fth century
b.c., and that similar kurgans in the same region date earlier and later, thereby
showing a continuous tradition. See Fredrik T. Hiebert, “Pazyryk Chronology and
Early Horse Nomads Reconsidered,” Bulletin of the Asia Institute, n.s. 6 (1992):
117“29. The classic work on Pazyryk is Sergei I. Rudenko, Frozen Tombs of


variety of burial sites, indicating the co-existence of different social groups.
The older funerary assemblage was comprised of, essentially, bronze
weapons, with rare jewelery in bronze or gold, but the burial goods found
in tombs of the later stage display better bronze-casting techniques; more-
over, iron was used more broadly, particularly for horse gear, and gold pro-
cessing was much more sophisticated.82 Chinese silk has also been found
in several burials, con¬rming that contacts, however indirect, occurred
between the Pazyryk culture and China.83
Western Mongolia certainly belonged to the same cultural horizon as the
Altai and Tuva regions. The Pazyryk and Uyuk cultures ¬nd a parallel here
in the site at Ulangom, dated to the ¬fth to the third century b.c. In its
central and eastern regions, Mongolia was also home to a completely dif-
ferent ethno-cultural group. This cultural complex is known as the slab-
grave culture after the type of burial practiced, in which simple pits were
lined with slabs of stone and not surmounted by a moundlike structure.
The physical type of this group, distinctly Mongoloid, is also very different
from the Europoid “Saka” people of the Altai. Nevertheless, in the early
Iron Age these two distinct cultural and anthropological areas shared
elements of material culture ranging from the shape of their arrowheads
to psalia and bridle bits, animal-style motifs, and the so-called deer stones
(large stone slabs engraved with stylized deer and anthropomorphic
motifs).84 This is further evidence of the vitality of a “steppe civilization”
where diverse metallurgical, artistic, and possibly spiritual components
were rapidly transmitted, exchanged, and absorbed from community to
community. The very rapidity with which these elements spread encourages
us to surmise that many groups had reached a fairly similar degree of eco-
nomic, technological, and social development and that contacts among
them had intensi¬ed steadily over time.
One factor that possibly affected the further political and economic
development of nomadic societies was the rise and spread of iron metal-
lurgy. Iron began to be used in Central Asia around the early ¬rst millen-
nium b.c. Early centers of iron metallurgy were located in the southern
(Anau) and Ferghana (Dal™verzhin) regions. This new technology soon

Siberia: The Pazyryk Burials of Iron Age Horsemen (London: Dent and Sons,
Nikolai A. Bokovenko, “Scythian Culture in the Altai Mountains,” in Nomads
of the Eurasian Steppes in the Early Iron Age, pp. 285“95.
For silk found in a recently excavated Pazyryk burial, see N. V. Polosmak, “The
Burial of a Noble Pazyryk Woman,” Ancient Civilizations from Scythia to Siberia
5.2 (1998): 125“63.
V. Volkov, “Early Nomads of Mongolia,” in Nomads of the Eurasian Steppes in
the Early Iron Age, pp. 324“25.


expanded to other areas, including the steppe region. According to some
scholars, iron metallurgy contributed to greater political centralization,
sharper class strati¬cation, and the formation of larger and more closely
integrated socio-economic units.85
These larger units comprised a variety of economic activities. For
instance, it is clear that across the steppe region of northern China, although
pastoralism was their main activity, the nomads also hunted, whereas the
regional economy included agricultural communities and trade among the
different groups.86 As already mentioned, to assess each nomadic group™s
economic basis we must consider its speci¬c environment and contacts with
neighboring communities. The image of the “pure” nomad is often mis-
leading when applied to groups in this early period; thus the cattle breed-
ers of the Altai appear to have had some semi-permanent settlements where
they engaged in primitive agriculture. The tools found in their dwellings
indicate that they lived in centers that supported a range of supplementary
economic activities, from hoe farming to hunting, leather processing, craft-
ing of bone and horn, and metallurgy.87
The presence of sedentary or semi-sedentary communities within reach
of the nomads affected the exchange economy, with implications at both
the social and the economic levels. For example, the settlements of the Itkil
culture in the trans-Ural forest steppe region, dated between the seventh
and the third century b.c., have yielded numerous remains of metallurgical
production. Other metallurgical centers have been found also on the Kama
and Belaya Rivers and in the forest-steppe area of the Volga River. The arti-
facts discovered at these sites include such typical nomadic metalwork as
weapons, harnesses, cauldrons, jewelry, and mirrors, and seem to have been
produced by the craftsmen of these sedentary communities for a nomadic
market.88 Evidence of the close integration of different cultural and socio-
economic groups has been discovered in the region to the south of Ferghana
Valley and in the highland valleys of the Altai mountain range, where
between the ¬fth and third century b.c. the burials of nomadic and settled
peoples display no differences in their funerary artifacts.89
In conclusion, an assessment of the economic base of the early nomads
needs to take account of several factors. Among these are the intensity and
frequency of contacts among neighboring communities within a region.

G. A. Koshelenko, “L™Asie Central au d©but de l™age du fer: le problème des rela-
tions ext©rieures” in L™Asie centrale et ses rapports, pp. 171“72.
Nicola Di Cosmo, “The Economic Basis of the Ancient Inner Asian Nomads and
Its Relationship to China,” Journal of Asian Studies 53.4 (1994): 1092“126.
Bokovenko, “Scythian Culture in the Altai Mountains,” pp. 285“95.
Moshkova, “Sarmatians, Concluding Remarks,” in Nomads of the Eurasian
Steppes in the Early Iron Age, pp. 185“88.
Yablonsky, “Some Ethnogenetical Hypotheses,” pp. 241“52.

Even more important is the complex socio-political organization, which was
likely to have been based on a hierarchy of lineages (especially within the
nomadic aristocracy), of communities, such as farming settlements, and of
social groups “ characterized by status (free people, servants, or slaves) and
by economic activity (e.g., metallurgy or other type of craftsmanship).

Material Culture

Metal objects played a critical role in the material culture of the early
nomads. By the eighth century b.c. metalwork achieved a high degree of
similarity across Central Eurasia. From the eighth“seventh century b.c.
onward, a larger and more varied inventory of weapons dominated funer-
ary assemblages. Common features signifying the prestige of the deceased
were the design of horse harness articles, artifacts in the animal-style tra-
dition, weapons, bronze cauldrons, and mirrors. Among the weapons, the
most common were the bow and arrow, although few bows have survived.90
Arrowheads were divided into two main categories, tanged and socketed;
other weapons included daggers, swords, and spearheads.91 Among the
Sakas of the Pamir and T™ien-shan regions we ¬nd iron daggers and bimetal-
lic daggers (iron blade with a bronze hilt) already in the necropolises of the
early period (eighth to sixth century b.c.), accompanied by bronze daggers
and bronze or iron arrowheads. Iron daggers continue to be found in the
large kurgans of the Pazyryk period.92 At the time, horse gear included bits,
cheek pieces, bridle cockades, girt buckles, and strap plates; bits were made
of bronze and had joined mouthpieces, and cheek pieces were made of
bronze, bone, and horn.
In the burial structures of the Saka peoples of Central Asia and Ka-
zakhstan of the seventh and sixth century b.c. we ¬nd knives made out of
iron. Initially iron was used primarily for making utilitarian goods rather
than luxury items.93 Later the use of iron became more widespread, and
more objects, such as horse accouterments, were made from the metal.94
Prestige items, typically in bronze and precious metals, included ritual and
ornamental objects besides weapons. Indeed, among the most valuable
pieces of bronze production found throughout the nomadic world were
cauldrons that had a ritual function connected with animal sacri¬ces,

E. McEwen, “Nomadic Archery: Some Observations on Composite Bow Design
and Construction,” in Arts of the Eurasian Steppelands, pp. 188“202.
For an inventory of Scythian armament, see Bourchard Brentjes, Arms of the
Sakas (Varanasi: Rishi Publications, 1996), pp. 17“42.
Yablonsky, “Material Culture of the Saka,” pp. 201“39. 93 Ibid.
Dvornichenko, “Sauromatians and Sarmatians of the Eurasian Steppes,” pp.

possibly to cook the meat.95 All of this is evidence of a mature martial
society, with horses and riding as elements of prime economic and social
The material culture also included imported items. Sauromatian bronze
helmets and scale or plate armor not of local production appear in the Volga
River region and southern Ural Steppes in the ¬fth“fourth century b.c.,
showing an increase in the exchange economy among neighboring com-
munities.96 Likewise, the Sarmatians acquired jewelry by trade with the
region of Tanais (the river Don) and Phanagoria on the Bosporus.97
The nomadic animal art style is open to endless variations, yet preserves
a remarkably unitarian aspect. The animals most often represented are
mountain goats, elk, birds of prey, and boars; large felines are a favorite
subject. Animal-style motifs decorate the handles of weapons and knives,
metal plates, buckles, and horse gear.
Finally, gold and jewelry acquired greater relevance in funerary assem-
blages after the sixth century b.c. In a Saka burial of the Iron Age, exca-
vated in eastern Kazakhstan, pieces of sheet-gold decoration are sewn on a
chieftain™s clothing. Advanced techniques in gold manufacturing, such as
incrustation and granulation, are evidenced by artifacts, including a three-
dimensional ¬sh made from sheet gold, with its eyes and ¬ns of inlaid
turquoise, and its body adorned with granulation.98


The material culture and social organization of the early pastoral nomads
of Eurasia have impressed researchers for their high degree of similarity
across an immense territory. Archaeologists and art historians have stressed
in particular the ritual and social signi¬cance of the horse, the abundance
of weapons, and the artistic vocabulary dominated by the “animal style.”
Moreover, the formation of early nomadic cultures cannot be disassociated
from technological advances, especially with respect to the horse and
chariot, which increased the nomads™ mobility and made nomads militar-
ily superior to their settled neighbors. Their upbringing in a pastoral setting,
where they acquired riding and shooting skills, and their social need to
organize themselves into militarylike parties for seasonal migrations and
hunting, made pastoral nomads into natural warriors.

These appear throughout Central Eurasia, from the Ordos region of northern
China to western Central Asia. See Miklos Erdy, “Hun and Xiongnu Type Caul-
dron Finds throughout Eurasia,” Eurasian Studies Yearbook (1995): 5“94.
Vladimir Dvornichenko, “Sauromatians and Sarmatians of the Eurasian Steppes,”
pp. 105“16.
Moshkova, “Sarmatians, Concluding Remarks,” pp. 185“88.
Yablonsky, “Material Culture of the Saka,” p. 211.


Under these circumstances, and with evidence of increasing aridization
of the steppe region, pastoral nomadism remained a successful adaptation
to the Inner Asian grassland environment, which allowed for not only the
subsistence of the nomads but also their evolution into larger and more
complex societies. Nomadic social formations retained a characteristically
martial outlook and produced an aristocratic class whose main occupation
seems to have been the practice of war. The emergence of such a class, prob-
ably linked to competition for pasture and to the need for defense during
seasonal migrations, was a major element in the expansion of the political
power of pastoral nomads over settled or semi-settled communities. In other
cases, relations between nomads and agriculturalists depended upon less
violent forms of economic and cultural exchange.
In a secondary stage of the development of nomadic societies, the type
of items that conferred prestige on their owners, previously dominated by
weapons, came to include decorative objects. This period is characterized
by an elite class that cherished the rare and beautiful ornaments, such as
golden plaques decorated with animals in the round and inlaid with pre-
cious stones, that signi¬ed wealth and status rather than just military
prowess. This evolution reveals a shift not only in the taste but also in the
social function of the nomadic elites. Together with evidence of trade, espe-
cially the importation of works of art from nearby sedentary communities,
the precious art objects in the funerary inventory of the nomads suggests
that the aristocracy de¬ned itself no longer exclusively as military leaders,
but as performing a range of commercial and political functions resulting
in the accumulation of wealth in the form of precious metals and jewelry.
The formation of a leadership that controlled the sources of wealth, such
as trade and production, was arguably a necessary condition for the cen-
tralization of power and military expansion on nomadic polities. As we will
see in the next chapter, a similar trend can also be observed among the pas-
toral nomads of northern China.

chapter two

Bronze, Iron, and Gold
The Evolution of Nomadic Cultures on the
Northern Frontier of China

Introduction: The Northern Complex

Scholars have long recognized that a cultural frontier, understood as an area
of contacts among carriers of different material cultures, existed to the
north of China as early as the Shang dynasty.1 The origin of this cultural
complex, its connection with China and areas in Central and northern Asia,
and the characteristics of the separate cultural enclaves recognizable within
it have been objects of much debate. Yet two critical questions remain unan-
swered: When do we begin to see a clearly delineated frontier between
China and the north? More importantly, how do we de¬ne the northern
China™s frontier has been often understood as an ideal line dividing two
ecological zones: the steppes and deserts of the north and the farmland of
the south. Although this line may have shifted north or south in response
to climatic variations over time, from the viewpoint of human agency this
interpretation of the frontier remains fundamentally static and tells us little
about cultural exchange and political interaction.3

For an early and still excellent analysis of the northern frontier, see William
Watson, Cultural Frontiers in Ancient East Asia (Edinburgh: Edinburgh Univer-
sity Press, 1971), pp. 96“124.
This chapter is based in part on Nicola Di Cosmo, “The Northern Frontier in
Pre-Imperial China,” in Cambridge History of Ancient China, ed. Michael Loewe
and Edward L. Shaughnessy (Cambridge: Cambridge University Press, 1999),
pp. 885“966.
Kathryn Linduff has suggested that the relationship between the Central Plain
and the northern cultures emerging in the early Bronze Age evolved into one of
core-periphery, thus surmising a relationship of dependency of the “frontier”

Until the third century b.c. “ when a clearly demarcated political bound-
ary between the north and China emerged with the formation of the
Hsiung-nu empire (209 b.c.) “ the northern frontier of China remained
extremely ¬‚uid. However, at least three interconnected but independent
processes played roles in de¬ning the northern frontier: one ecological and
economic, another cultural, and the last political. In this chapter, I will
examine the ¬rst two of these processes, that is, the formation of pastoral
nomadism in northern China, and the distinctive traits of the resulting cul-
tures. The evidence for the analysis of the economic and cultural contexts
is archaeological and is based on the divergence between the Chinese and
Northern Zone™s discrete metallurgical traditions. Archaeological evidence
also shows that a series of “ecological frontiers” between different modes
of production, social organizations, and adaptations to the environment
had already developed by the mid-second millennium b.c.4

Metallurgy: The First Frontier

From the beginning of the second millennium b.c., beyond the core area of
the Shang civilization, lay a broad belt of cultural transition between the
Central Plain culture and the Bronze Age cultures of Central Asia and South
Siberia. The Shang civilization was in close contact with this intermediate
zone, and several Shang sites display features that can be immediately rec-
ognized as alien and intrusive. However, the process by which these
adjacent but distinct cultural zones were formed is still unclear. The debate
has focused on whether this cultural zone, regarded as a transitional area
between the Sinitic East and the Inner Asian complex of Mongolia, South
Siberia, and Central Asia, was formed through Chinese cultural diffusion
to the northwest or, instead, was in¬‚uenced by contacts with the North and
West.5 It is now clear that the Northern Zone (pei-fang ti-ch™ü) of China “
largely in today™s Inner Mongolia and Liaoning, and in the northern areas
of Shansi, Shaansi, and Hopei “ was already an independent cultural unit
during the Shang dynasty and that it acted as a ¬lter and link between the

upon the core. See her “The Emergence and Demise of Bronze-Producing Cul-
tures Outside the Central Plain of China,” in The Bronze Age and Early Iron Age
Peoples of Eastern Central Asia, ed. Victor H. Mair (Washington: Institute for
the Study of Man, 1998), 2: 619“43.
Traditional dyadic divisions between “steppe and sown” are becoming more
suspect; see J. P. Mallory, “A European Perspective on Indo-Europeans in Asia,”
in The Bronze Age and Early Iron Age Peoples of Eastern Central Asia,
1: 175“201. Recent approaches privilege a less rigid demarcation.
Max Loehr, “Weapons and Tools from Anyang, and Siberian Analogies,”
American Journal of Archaeology 53 (1949): 126“44; W. Watson, Frontiers of
Ancient China, pp. 54“56.

Central Plain and Central and northern Asia.6 Even before the Shang
dynasty, what is now northern China was home to cultures distinct from
those in the core area of Central Plain civilization. These early cultures, dis-
tributed over a broad area and dating back to the late third and second
millennium b.c., were responsible for the development of a closely knit
metallurgical network across northern China. Of these early northern cul-
tures, the most revealing are the Ch™i-chia culture in the northwest, the Chu-
k™ai-kou culture in the north-central sector, and the Lower Hsia-chia-tien
culture in the northeast.
The Ch™i-chia culture is the earliest Bronze Age culture discovered within
the territory of present-day China and has been dated to the late third mil-
lennium b.c.7 Based on typological comparison of the pottery, the Ch™i-chia
culture is regarded as a continuation of the Neolithic cultures that devel-
oped in today™s Ning-hsia and Kansu provinces. Although its main sites are
located in Kansu,8 the Ch™i-chia culture was broadly distributed, extending
north and east into Inner Mongolia, the upper Yellow River Valley, and the
upper Wei-he and Huang-shui River Valleys. Connected with earlier
Neolithic cultures, such as the Ma-chia-yao, during the ¬rst half of the ¬rst
millennium b.c., the Ch™i-chia people displayed cultural traits that were
among the most advanced in China. Their bronze production was exten-
sive, and they progressed from forging copper tools (knives, awls, chisels)
to casting objects (knives and axes) in open molds to more complex casting
using composite molds (mirrors and socketed axes).9

Lin Yun, “A Reexamination of the Relationship between Bronzes of the Shang
Culture and of the Northern Zone,” in Studies of Shang Archaeology, ed. K. C.
Chang (New Haven: Yale University Press, 1986), p. 272. On the relationship
between Xinjiang and neighboring cultures an excellent overview is provided by
Chen Kuang-tzuu and Fredrik T. Hiebert, “The Late Prehistory of Xinjiang in
Relation to Its Neighbors,” Journal of World Prehistory 9.2 (1995): 243“300.
For a general survey of Northern Zone bronzes and their archaeological impor-
tance, see also Wu En, “Yin chih Chou ch™u te pei-fang ch™ing-t™ung ch™i,” K™ao-
ku hsüeh-pao 1985.2: 135“56; T™ien Kuan-chin, “Chin-nien-lai te Nei Meng-ku
ti-ch™ü te hsiung-nu k™ao-ku,” K™ao-ku hsüeh-pao 1983.1: 23.
For a comprehensive examination of the Ch™i-chia culture, see Corinne Debaine-
Francfort, Du N©olithique à l™Age du Bronze en Chine du Nord-Ouest: la culture
de Qijia et ses connexions (Paris: Editions recherche sur les civilisations, 1995).
That is, the sites of Huang-niang-niang-t™ai, Ch™in-wen-chia, and Ta-he-chuang.
See Hu Ch™ien-ying, “Shih-lun Ch™i-chia wen-hua te pu-t™ung lei-hsing chi ch™i
yüan-liu,” K™ao-ku yü wen-wu 1980.3: 77“82, 33; and Hsieh Tuan-chü, “Shih-
lun Ch™i-chia wen-hua,” K™ao-ku yü wen-wu 1981.3: 79“80.
An Zhimin [Chih-min], “The Bronze Age in the Eastern Parts of Central Asia,”
in History of Civilizations of Central Asia, vol. 1: The Dawn of Civilization: Ear-
liest Times to 700 B.C., ed. A. H. Dani and V. M. Masson (Paris: Unesco, 1992),
p. 322.


The Ch™i-chia was a sedentary culture based on agriculture, with stock
breeding as an important activity. Some sites yield evidence of horse domes-
tication (for example, Ta-he-chuang and Ch™in-wei-chia), and pigs feature
prominently among animal remains and funerary sacri¬ces.10 The sudden
appearance of advanced bronze metallurgy and the domestication of the
horse in northern China strongly suggest that the Ch™i-chia people had
extensive contacts with other cultures, especially those in the north and
west, since no similar achievements are documented among the eastern pre-
Shang cultures.
Some scholars have hypothesized that bronze metallurgy in the north-
west preceded the advent of metallurgy in Central China and thus that the
origins of the Ch™i-chia culture may be found in cultural processes taking
place in the west.11 A close connection may have existed between the Seimo-
Turbino cultural complex and northwestern China, which would explain
the transmission of South Siberian metal artifacts to the Ch™i-chia cultural
area; evidence for such a connection rests mainly on socketed axes, handled
knives, and the hafting method of handled awls and knives that have been
unearthed at Ch™i-chia sites.12 One ax excavated at Hsing-lin, in eastern
Kansu, is a local casting but exhibits traits “ such as a single loop on one
side of the hafting edge “ characteristic of the eastern type of the Seimo-
Turbino socketed ax. A bronze knife, also found at Hsing-lin, is closely
related to knives from Rostovka and Sopka, and the geometrical decora-
tion on fragmentary handles of daggers and knives recalls Seimo-Turbino
motifs. Finally, the hafting method (a metal blade inserted in a bone handle)
used for an awl and a knife found at separate Hsi-ning sites is most similar
to that used for metal awls and knives with bone or wood handles found
in Minusinsk sites of the Okunevo period. Moreover, the same method was
used by the Seimo-Turbino people. Although the precise manner in which
the Ch™i-chia culture was connected with the western cultural horizon that
included South Siberia, Central Asia, and Mongolia is unclear, that they
were in contact seems highly likely.

K™ao-ku hsüeh-pao 1974.2: 29“62; K™ao-ku hsüeh-pao 1975.2: 57“96.
An Chih-min, “Shih-lun Chung-kuo te tsao-ch™i t™ung-ch™i,” K™ao-ku 1993.12:
1110“19. It is assumed by some that metallurgy in areas to the west of Ch™i-chia,
such as Sinkiang, may go back to the early second millennium b.c. The evidence
comes from cutmarks on logs recovered from the cemetery of Ku-mu-kou, which
are too deep and clean to have been produced by anything other than a metal
tool; see Wang Binghua, “A Preliminary Analysis of the Archaeological Cultures
of the Bronze Age in the Region of Xinjiang,” in Between Lapis and Jade: Ancient
Cultures of Central Asia, ed. F. Hiebert and N. Di Cosmo, Anthropology and


. 2
( 15)