Cultures and Ethnic Groups West of China in the Second and First Millennia B.C.

by Jettmar Karl

(Original print at Asian Perspectives, XXIV (2), 1981)

We have seen the external diffusionist "Western Origins Theory' discarded except for a few very isolated traits." In this sentence, proclaimed during the Conference on the Origins of Chinese Civilization (Berkeley, 1978), Meacham (1983: 170-171) seemed to express the predominant opinion.

But the question remians whether these "very isolated traits" indicate ethnic movements of considerable political and demographic dimension. Perhaps we can appreciate their full importance by studying the counter-current diffusion of technical achievements of Chinese origin toward the West. They may be explained by trade, or as goods brought back by withdrawing invaders. Arguments will be presented in chronological sequence up to the end of the Sandai period. We should keep in mind that the better-explored areas (Soviet-ruled Middle Asia and south Siberia) lie more than 3000 km west of the Sandai territories. (Archaeological work done there rests upon broad and solid foundations, incomparable with the hastily done excavations in eastern Siberia.) Finds between the two areas, in Tibet, Sinkiang, Chinghai, and Kansu (now called Central Asia in the Sovient literature), are scanty, despite all the efforts of the last few years. The mechanism of contact will remian mostly unexplained.


The Neolithic period of the western steppes (with very few and very late indications of a food-producing econmony) was included in the Kel'teminar culture. Along the uppercourse of the Yenisei an Aeneolithic culture called Afanasievo was observed and dated in the first centuries of the second millennium B.C. It was followed by a period known as the Bronze Age of the Steppes, when people lived from farming and stock-breeding, a period which is summarized under the name Andronovo culture (Jettmar 1966a). Loehr believed that the Afanasievo and Andronovo cultures of the steppes might have stimulated metallurgy in CHina (Loehr 1949). According to the unanimous opinion of our Soviet colleagues, the cultures of militant horsemen came into existence only at the beginning of the first millennium B.C.

Research in recent years has led to two main conclusions. First, the Neolithic cultures of the steppes of Central Asia and Kazakhstan (Vinogradov and Mamedov 1975) do not give the impression of great radiation from their centers. In this respect they are clearly different from those existing in the steppes west of the River Ural (i.e., in Europe) during the fourth and third millennia B.C. Since the Mesolithic substratum in the East is rather archaic (Matjusin 1976), and since transition to a productive economy takes place much later than in the steppes between the Urals and the River Dnieper (Yamno or Kurgan Culture, Merpert 1977), it is an error to look for sources of the Kurgan Culture in the East (cf. Gimbutas 1975--as far as Lake Balkash!).

Secondly, the name Andronovo Culture ought to be used with great caution in the future, inasmuch as it suggests a uniform ethnic substratum of Indo-Iranians or Iranians, which cannot possibly he maintained (Itina 1977). Only marginal groups influenced by the Timber Grave Culture, for example, the Novokumak complex between the Rivers Ural and Tobol (Sintasta, Gening 1977; Smirnov and Kuz'mina 1977) made use of chariots and perhaps spoke an Indo-Iranian idiom. Farther to the East were peasants evolving from a local substratum, for example, in East and Central Kazakhstan (Cernikov 1960).

The Soviet archaeologists have observed cultures of superlocal importance that are arranged in two zones north and south of the steppes belt, that is, in the forest zone and its marginal areas, as well as in the oasis regions of the south and southeast marginal of Middle Asia.


In northern Afghanistan and the bordering parts of the Soviet Union, in the regions called Margiana and Bactria in antiquity, burials and setlements, fortresses and cult places have been excavated in recent years (Sarianidi 1973; 1974; 1976; 1977a; 1977b; Sarianidi, Terekhova, and Cernykh 1977; Askarov 1977). Numberous cemeteries, alas, have been robbed in recent times (cf. Amiet 1977; Ghirshman 1977; Tosi and Wardak 1972). The excavations give evidence of a group of cultures apparently beginning in the third millennium B.C. (earliest radiocarbon dates 2280 B.C). They cannot have come into existence due to the expansion of tribes from Turkmenia and Gorgan eastward only, as was at first believed, for they contain elements betraying contact with civilizations in the interior and on the northwest margin of the Iranian Plateau. Amiet (1977) quite rightly lays stress on an Elamian component. On the other hand, affinities to India, to Harappa, and to contemporary cultures as well as later ones cannot be overlooked.

The so-called Margiano-Bactrian Archaeological Complex was surrounded by a similarly dynamic periphery. This became evident in Bactria itself (P'jankova 1974; Mandel'stam 1968), in the Bukhara region (Kuz'mina 1958; Askarov 1963), and in Ferghana where a hoard was found as early as 1894 (Khak). This hoard contained a pin with a head representing a milking scene (Zadneprovskij 1962:52-55).

It is tempting to connect the Margiano-Bactrian archaeologcial complex (and many bronzes which came as stray finds from neighboring areas (Kuz'mina 1966) with East Asia: Amiet even proposes to derive the so-called Nestorian seal-amulets (Hambis 1954) from the stamp-seals of Bactria.

In later research (and not under the immediate impact of the rather revolutionary radiocarbon dates of bronze finds in Southeast Asia), fresh attempts will certainly be made to trace Western influences on the rising metallurgy of China. It will then be necessary to analyze this complex. I will not attempt it here, since the situation is not as yet clear. I will confine myself to the remark that all earlier refutations of such influences (which were based on an evaluation of the scanty Andronovo material) are to be abandoned. Here in the south, types are represented that might have been of importance to Archaic China, for example, one-edged knives and knife-shaped picks to be inserted into a wooden staff. If the Caucasian dagger published by Hajek really was found in China, it certainly belongs in this context as does the strange dagger from Rasit (Pogrebova and Clenova 1970).


IN the northern zone, the progressive cultures with incipient metallurgy are Krotovo and Samus' (between the Rivers Ob and Irtys) as well as Okunev in the Minusinsk Basin, an island of steppe on the Upper Yenisei surrounded by forested mountains (Molodin 1977). Examples of pottery with affinities to these cultures are said to have been observed in Inner Mongolia on Chinese territory (Clenova 1977; KK 1964[1]: 2). Engravings where found on stone slabs which were used as building- material in burials of the Okunev Culture. Two stylistic groups can be discerned: a realistic one, representing bulls and men with bird-masks, and a schematic one, the essential motifs of which are a horned mask with three eyes and a symbol of the sun (a ring with four tips). Neither of these two styles is known in other cultures of the forest zone. The realistic style could possibly be derived from the south of Central Asia (Matjuscenko 1977). The schematic one may possibly have a relation to "cryptic magic," which was presumed to have been the background of some painted designs of the Yangshao Culture in Kansu (Chang 19777:110-132; cf. Formozov 1969: 109, 194).

The affinities are significant, since there is a Mongolid component of Central Asian origin in the Okunev population (brachycephalic skulls) (Ivanova 1966). Within the framework of this northern complex, a center of bronze casting and metal trade arose that later influenced Eastern Europe. Large ceremonial knives with ends decorated by elaborate figures, and sometimes even with a whole scene, come from important sites (Seima, Turbino) west of the Urals. (Clenova 1972, pl. 69).

Such designs may ultimately depend on the Margiano-Bactrian archaeological complex (Khak!) for their origins. On the other hand, an affinity of these knives to the long ceremonial knives of the Shang dynasty (cf. Karlgren 1945) has been maintained again and again (clenova 1972:135-138), and for good reason. Socketed spearheads with two loops (non-existent in the south) and socketed celts belong to the same complex.

The existence of a trade and imgration zone along the southern margin of the Taiga could have been the basis of parallel developments and borrowings that allowed scholars to suppose the existence of a Ural- Altaic linguistic family.

In the beginning of the second millennium B.C. or perhaps even before this, Europids as well as the Okunev people lived in the Altai (Alekseev 1961). The Afanasievo Culture must be attributed to this strain of people, since it was (long ago) suggested to have Western affinities. As far as we know, peoples of Western appearance could also have come from the south, from the active periphery of the Margiano-Bactrian archaeological complex.


In the second half of the second millennium B.C. the Margiano-Bactrian archaeological complex went through further development. Perhaps the ceremonial axes and elegant socketed celts, which Sarianidi (1977b) illustrated, belong to a relatively late phase, that is, the end of the second millennium B.C.

Horses and Chariots

Astonishingly, there are no hints of intense horse-breeding activity in Bactria. This makes even more important the fact that in vast parts of the steppes and the adjacent mountain ranges (in the foreland of the Karatau, in the Pamirs, in the Mongolian Altai) rock carvings of chariots have been discovered during the last few years (Lattauer 1977). There is a tendency to connect them with the eastward expansion of at least one of the cultures summed up under the name Andronovo. In ornamentation as well as in the form of the chkeekpieces there is a clear affinity with the Mycenaean Shaft-Graves (smirnov and Kuz'mina 1977).

Considering the destruction of most cemeteries in Afghanistan (where only ceramics and metal objects were spared), the apparent absence of horse bones, horse gear (with bone cheekpieces), and carts is not decisive. Soviet scholars are convinced that the custom of depositing chariots in the graves of the Shang rulers came from the West, as well as the ceremonial significance of the chariot itself. The finds of Sintasta, where the wheels are standing in furrows carefully dug into the soil of the grave-chamber (exactly as in China), as well as the conventionalized rock carving, confirm this thesis, which was also accepted by von Dewall (1964).

The Karasuk Culture

After the long phase of "Andronovo Type" cultures, the Karasuk Culture was formed in southern Siberia. The pottery shows affinities to several groups in Middle Asia and Kazakhstan (Grjaznov 1966; 1970). Metal implements and weapons depend on models developed in the Sandai neighborhood, represented by the oldest stratum of the so-called Ordos bronzes. In one of my first articles (1950), I tried to bring evidence to defend Kiselev, who upheld this thesis after being a scientific adviser in China (Kiselev 1949; 1960). At that time, I believed that the expanding civilization of China had led to a displacement of barbarian neighbors.

Today, we must consider something that puzzled the late Soviet physical anthropologist G.F. Debec, according to one of his last letters. The skulls of the Karasuk necropoles are very similar to those of the much earlier Okunev burials (Clenova 1977). This fact perhaps solves the riddle of the "flight" into the Minusinsk Basin so far away: There were traditional connections with this area, which was surrounded by mining districts. Therefore, we find a certain continuity between Okunev-Samus' metallurgy (reflected in Seima-Turbino) and that of Karasuk. The exact date of the displacement cannot be fixed.

Meanwhileit has become fashionable to state that the "Central Asian," that is, Ordos-Karasuk metallurgy (cernykh 1976) had largely influenced the armament system of East European tribes of the tenth and ninth centuries B.C. (Terenozkin 1975).


Tribes participating in the battles that led to the shift of power from Shang to Chou certainly reveal cultural traits of East European origin, either because they were immigrants themselves or because they had accepted innovations of immigrants. As I have shown several times (Jettmar 1966b; 1970), this becomes perfectly clear from the material expanded on by von Dewall (1966). She did not acknowledge it herself before the finds of Pai-fu, Ch'ang-p'ing, a northern suburb of Peking (LL 1976 [4]), where made. Three burials in wooden chambers belong to the transition period from Shang to Chou, according to the radiocarbon date (1120 B.C.) At that time the region belonged to the state of Yen. The victorious Chou installed a new dynasty there, whose symbol could be a prototype of the "tamgas" used during later periods by the peoples of the steppes (Vajnberg and Novogorodova 1976). The daggers are of a form known from the Ordos region, and may have belonged to the guards accompanying the new rulers.

Of decisive, importance are four horn psalia or cheekpieces (KK 1976: fig.18, 4) with three perforations, the middle one perpendicular to the others, which were found together with horse bits made of bronze in grave M2. They correspond to a type known in the Early Iron Age (Belogrudovsk and the Early Cernolessk cultures) of the forest steppes in the Dnieper regions (Terenozkin 1958; 1965; Grakov 1977). The date is twelfth to ninth centuries B.C. A similar cheekpiece was found in a slab-grave in Transbaikalia; among the so called Ordos bronzes there is a replica in metal (Jettmar 1970).

The process can be seen as follows: In the generation of the Chou conquerors some troops of the army used metal snaffles (which continue) and horn prongs or bone as crossbars instead of the Shang-time bridle (cheek-plate and plaited leather forming the bit). The latter innovation was abandoned for a while and replaced once more by cheek-plates. Remembrances are preserved in the decorations. Only after several intermediary forms (von Dewall 1966) did the technically superior combination of metal toggles with only tow openings and metal snaffle appear.

Warrior Attacks and Royal Burials

Toward the end of the ninth century B.C. the Chou dynasty had to cope with the attacking Hsien-yun. Prusek (1971) tried to show that this attack aimed at the heart of the kingdom and was executed by mounted warriors. These events should be compared with the almost contemporary invasions of the Cimmerians and the Scythians in the Near East, with which Prusek suggested some kind of connection

With the excavation of the Arzan kurgan in Tuva this suggestion becomes substantial. The kurgan is not only the greatest but also the very oldest princely burial of the Asian steppes (except perhaps for the totally deprived mausolea of Tagisken). The federation that buried their king together with about 160 horses in the middle of his attendance of high-ranking and richly ornamented persons much have had their pasturelands in western Mongolia. The king had tribes under his rule whose ethnic differentiation is reflected by varying systems of horse gear (Grjaznov and Mannaj-ool 1972; 1973a; 1973b; 1974; 1975). Forms not as yet known have been found there.

In the plan of the burial, rituals of manifold origin were combined. Among them was the exposure of the dead above ground in wooden troughts or chests (as was done until only recently by the Kafirs of the Hindukush, and also in the Tarim Basin until the Han period). The soil under the structure was left intact. The central chamber was surrounded by a model palace-town. The tradition of such palace-towns, with rings of surrounding habitations (3 zones, partite in 9 sections), is documented at its earliest in the Margiano-Bactrian archaeological complex (Sarianidi 1977a). A much later monument of this kind of is Koj-Krylgan-kala in Choresmia (Tolstov and Vajnberg 1967).

The model of the palace-town (diameter 80m, height up to 3m) was made of wooden beams and was included in a round platform of stone. One of the boulders was recognized as a fragment of a "stag-stele." The kurgan itself can be dated in the seventh century B.C. at the latest (on account of the weapons and ornaments showing archaic animal style); the stag-stele had already been broken and was treated carelessly, so it must be considerably older, and the style of decoration as well. Moreover, the tribes united here into a confederation certainly had an independent "prehistory" allowing the formation of separate systems of horse gear, meaning that mounted warriors already existed in Mongolia before, say, the tenth or ninth centuries B.C.

Terenozkin (1976) has collected plausible arguments of an immigration of the Pontic Scythians coming from the East. Kyzlasov (1977) thinks that their ancestors were a Saka tribe living in east Kazakhstan.

In any case we now know that the mobilization of the population of the eastern steppes and their transition to a life as mounted warriors began earlier than was thought before. It is convincing that the impact of the Hsien-yun was a consequence of this mobilization. On the other hand, if the contact took place very early, then we can explain how some stylistic patterns rooted in certain regions of China were incorporated in the archaic Animal Style. Surely the curled animal belongs to this East Asian heritage.

Continuing East-West Interaction

The contacts between East and West which began here did not break off. In this way Animal Style motifs of Near Eastern origin were mediated to the Far East (Artamonov 1971a). In such as interply, tribes with Europid features and Mongolids form a zone of manifold and competing cultures belonging together in essential characteristic aspects. Such cultures side by side are clearly to be seen in Tuva (Grac 1967; 1975). Western Mongolia was in the hands of Europid groups as far as we know (Volkov 1974). Evidently the famous kurgans of the High Altai (Rudenko 1953; 1970) were created by tribes who had their winter grazing-grounds farther south on what is today Chinese territory. It is not possible to demonstrate relations in great deal in this paper, but two examples may be mentioned.

First, in the Pazyryk kurgan no. 2, one of the horses had a kind of frontlet carved in horn andn painted red and yellow (Jettmar 1964: pl. p. 91). It is clearly derived from a metal prototype already attested during the Shang Period. Second, perhaps due to Chinese influence, objects made of cast iron were used in various areas of the steppes during a short period during the fifth century B.C. (Jettmar 1970)

Yueh-chih -- The Far Eastern Heritage

Perhaps many of the existing affinities are difficult to understand, since sites of the Yueh-chih (for a while the most important and most troublesome neighbors of the Chinese) have not as yet been identified in their old eastern homelands (Maenchen-Helfen 1945; 1957).

Soviet archaeology may be of some help to us here. In Khalcajan, a dynastic sanctuary of the Kushans, a relief of high artistic value displays a solemn act of alliance. The men of one of the participating groups were characterized by Pugacenkova (1971) as follows:

There is a marked, artificially inflicted deformation of the skull, particularly noticeable in side-view: the occiput is flat, the receding forehead bulges in a triangle over the bridge of the nose, lending a rather stern look to the faces of the men, young and aged alike. The faces are thin, the straight nose is not large. The dark eyes of average size have no trace of epicanthus, but their corners are outlined in black towards the temples; this, coupled with the high cheek-bones, lends the faces a certain "Tartar" look. The specific trimming of the black hair does not occur among other ancient peoples: the hair, bound by a strap, rises over the forehead, is combed away from the ears and clipped below them: the small arrow-shaped moustache frames the upper lip: there are side-whiskers.

This group apparently formed the light cavalry, without any armor, as seen on another relief.

The men of the second group have none of this "Tartar" look: long mustaches fall over clipped pointed beards. When one of them is depicted among the riders, he as well as his horse wears heavy armor. A lance ias seen in the horseman's hand.

Pugacenkova compared the sculptured heads of the first group (evidently those who were in a somewhat superior social position) with the face on the coins of Heraos, who was of Yueh-chih origin and who is considered to be the ancestor of the early Kushan kings. She thinks that fraternization between the nomad invaders and the Bactrian nobility of Iranian origin belonged to the political aims of Heraos, and that his success in this respect is depicted here.

In my opinion, the main scene could even portray a double marriage which united a local dynasty with that of Heraos. This would explain how the Bactrian language prevailed over the idiom of the immigrants.

In any case, it seems certain that the traditional appearance of the Kushan nobility is splendidly presented here. The representations of horsemen that can be seen in applique on felt-canvas found in the Pazyryk kurgan no. 5 bear certain similarities. The scene, may times repeated, shows a rder before a goddess on a throne--a motif well known in Scythian contexts. The fifth kurgan is of course much earlier than Khalcajan (fourth century B.C.?), and the differences are considerable, but one sees the black hair (without cap), the mustache (here twirled upward) and the flying mantilla. The rider sits on a horse with crenelated mane (Maenchen-Helfen 1957).

A second motif decorating these canvases is even more informative, as it shows the fight of a sphinx (with antlers) against a fantastic bird. The outlines of the bodies are traansformed into curvlinear designs with scrolls, which are very well known from Chinese textiles of the Han period (cf. Lubo-Lesnicenko 1961). In spite of this, the decor was certainly not made by Chinese, as the choice of material (felt) shows. Up above, the sphinx wears his mustache in the same "national" fashion as the horseman before the goddess. (Pugacenkova could not explain the fact that among the noblemen of Heraos a sphinx was to be seen on the reliefs of Khalcajan.)

In her dissertation, Pause recognized that the allegedly Chinese embroidery found in this kurgan (which formed a design of double tendrils with twigs of flower-buds and birds) does not correspond to Chinese prototypes of this period (see Lubo-Lesnicenko 1961, pl. XLIX). Similar observations could be made about the pictures of swans (executed in felt) that decorated the canopy of the funeral chariot, which had four wheels and was, according to Littauer (1977), a variation on the Chinese model.

All these elements, non-Chinese but Chinese in appearance to one familiar with Animal Style, had been known for quite some time. It is also remarkable that on a saddlecloth in the same kurgan (and in the fighting scene between the sphinx and the fantastic bird) the medallions with "Greek crosses," one of the main simbols of the Cimmerians appear (Jettmar 1964: 112; Terenozkin 1976). But the consequence was not seen: The "local culture in the Chinese periphery" (where the style originated) was that of the Yueh-chih. Perhaps one of the pair buried in kurgan no. 5 belonged to this people.

I do not think them to be remnants of the legendary proto-Indo-Europeans of Central Asia, but they may have been relatively late immigrants, that is, the very element among the Western barbarians who had wondered the farthest. Their flight to Middle Asia after the victory of the Hsiung-nu was a sort of return.

Perhaps belt-plaques showing landscapes originated in the same milieu. This would explain a Western element of the costume, the fez-like wooden helmet with staff in order to bind up the pigtail of the woman, which is also know from Pazyryk kurgan no. 5 (Artanomov 1971b; Grjaznov 1961). But as plaques of this kind were found over a vast area, such a thesis is difficult to prove. Many of these things found over a vast area, such a thesis is difficult to prove. Many of these things found from Shensi to as far as Liaoning in the East may have been exports or booty (Cheng-Te-k'un 1963).

Few objects decorated in the Animal Style occur in assemblages found in northern Bactria. They belong to the time when nomadic tribes settled there after the conquest and combined local ceramics with weapons and implements they had brought with them (Mandel'stam 1966; 1975a). The small ornaments show intensive affinities to finds from Ch'en-pa-erh-hu-Ch'i, in the region of Hailar near the border of Transbaikalia (KK 1965 [6]). Perhaps one tribe of the Yueh-chih was displaced to the borders of Transbaikalia during the great exodus of their Hsiung-nu overlords to the north.


The Burials of T'ao-hung-pa-la

T'ien Kuang-chin (Kaogu Xuebao 1976 [1]) has assigned the burials of T'ao-hung-pa-la (south of the region) where the dead lie with their heads to the north and similar finds in the same region (e.g., Fan-chia-yao-Tzu, Holingol; WW 1959 [6]) to the Hsiung-nu. At the same time he protested strongly against the tendency not to differentiate the cultural groups in Manchuria, and against the merging of all complexes of this kind into a universal "Scythian Culture," being merely an unscientific device, useful only for the imperialistic theories of aggressors.

My impression is that T'ao-hung-pa-la may really represent an early stage of Hsiung-nu-Culture. The predominance of Western (from Altai and Tuva) and northern elements (from the Slab Grave culture mentioned earlier) is striking, and may have produced conditions where the Hsiung-nu were able to make the transition from the tribe to state earlier than their neighbors in regions farther to the east, such as the Hsien-pi. In addition, as is known from the historical sources, the Hsiung-nu were strongly influenced by the Yueh-chih, who lived at their southwestern borders. Embroideries on a tapestry found in Noin-Ula but only preserved in fragments (Rudenko 1962/1969, pl LX-LXIX) show male heads similar to those of the Heraos family. One recognizes horse gear, clothing, part of a weapon, all of which belong to the same complex. Such embroideries were considered to be imports from Bactria.

On the other hand, the immense cemetery of Hsi-ch'a-kou (comprising 500 burials, but only partly excavated) was assigned to the Hsiung-nu (WW 1957 [I] and WW 1960 [8-9]). On the basis of Chinese coins found there, the graves can be dated to a rather short span during the first half of the first century B.C. According to the prevailing opinion among Soviet scholars, the Hsiung-nu had by then already transferred their main camps to the lands north of the Gobi (Minjaev 1975). The question is whether the ethnic identification is correct or not. The differences are indeed essential, not only with regard to older Hsiung-nu monuments of the region, but also in relation to finds belonging to relatively well-known settlers in Transbaikalia (Konovalov 1975; 1976). Many of the warriors from Hsi-ch'a-kou were buried with sword-blades of Chinese manufacture, but some of them had hilts made according to their own taste, perhaps strictly following the model of the previous period. There are also affinities (cheekpieces with a knob at the end) leading to toe complex whcih we assign to the Yueh-chih. But in any case, neither weapons nor ornaments are identical to those found in Transbaikalia.

Alternatively, were the people of this burial site barbarian allies of the Chinese, at whose disposal they placed superior weapons? How can one explain this difference? Were the warriors of Hsi-ch'a-kou members of a sort of task force much better equipped than the steelers in Transbaikalia--and perhaps recruited from various tribes? A comparison with the results of the export of arms (as practiced by the modern superpowers) comes readily to mind. In any case, the one-edged sword brought by the Huns to Europe may be derived from Chinese blades, such as those found here (Werner 1956).

On Soviet territory west of China, finds were made that may have belonged to garrisons of the Hsiung-nu. To these may be assigned graves near Mount Bajdag in Tuva (Mandel'stam 1975b) and others in Mongolia (Ulangom; see my German translation of Novgorodova et al. 1982).

The Tastyk Culture

As yet unvolved remains the puzzle of the Tastyk Culture in the Minusinsk Basin. Kyzlasov (1960) explained the apparent differences between Tastyk and the preceding Tagar Culture as pertaining to the heritage of immigrants from northwesetern Mongolia expelled by the Hsiung-nu at the beginning of the second century B.C. But at least some of the newcomers much have previously settled in the immediate borderlands of Han China. In the graves were found wooden animal figures, imitating the well-known grave figures and ceremonial umbrellas. The graphic art of the Tastyk people was perhaps imitating the "narrative" Han reliefs (Grjaznov 1971).


Enormous discoveries have been made in the field of East Asian archaeology in recent years. It has been confirmed that a major group of hightly active cultures existed there for a very long time, and that their interaction gave rise to the Chinese civilization.

Concertration on these problems has meant that in the new edition of Kwang-chih Chang's book the paragraphs dealing with the steppe peoples now seem very conservative. But in fact there are too many new materials and perceptions in this field. My bibligraphy of 1964, reprinted in 1967 (cf. Kwang-chih Chang 1977: 396) is out of date and does not deserve further consideration.

The rapidly expanding and ever-growing China had not only to deal with its periphery as before (e.g. the Ti, cf. Prusek 1971: 209-228), but also to interact with immigrants from the "Far West". New syntheses cannot overlook this fact, which is intimately connected with the question: Which waves of influence brought Western culrual goods such as wheat and barley to East Asia? The unexpected finds made in the soil of northern Afghanistan reveal the lack of information on many parts of Inner Asia and the continuing instability of our present state of knowledge.


When intense studies backed by fresh radiocarbon and thermoluminescent dates (Barnard 1975) indicated that the ancient civilizations of China-- growing up in different ecogeographical zones but united by frequent interactions--had deep and strong roots in the local earth, many competent scholars shared the tendency to consider China an almost independent theater of culture. Communications with rival theaters were not ruled out but were estimated to be rather unimportant during crucial periods. This attitude became obvious at the most interesting Conference on the Origins of Chinese Civilization, held at Berkeley, 26-30 June 1978, and in the resulting volume edited by Keightley (1983).

Due to the normal mode of scientific progress, after such a peak of general agreement a backlash will sooner or later follow. The tide must change, for only then shall we get a well-balanced result. In fact the backlash, or let us say the necessary complement, came almost immediately. This article was one of the conference papers not included in Keightley's volume. (I myself propsed that Keightley publish instead my review of a work by Krjukov, Sofronov, and Ceboksarov [1978], which presents the Soviet approach to a closely related problem--the origin of the Chinese people. In the meantime, two more volumes of this kind have appeared [Krjukov, Majavin, and Sofronov 1979 ; Krjukov, Perelomov, Sofronov, and Ceboksarov 1983].) Today this paper seems almost prophetic, However, it remains for me to point out certain aspects that are corroborated by new findings.

Evidence of Early Diffusion

One of the essential ideas of my article is that not the steppe belt itself but the areas such of it--northern Afghanistan and parts of Soviet and Chinese Central Asia--have to be studied to find evidence for migrations and early diffusion. This is substantiated now by a considerable number of observations. The Western scholar will find a most useful synopsis in the volume The Bronze Age Civilization of Central Asia edited by Philip L. Kohl (1981). Only one sentence from the introduction may be quoted (Kohl 1981: xxiii): "Present evidence is far from conclusive but suggests that food-producing cultures of the 'Middle Eastern type' may have spread from southern Turkmenia to the borders of Xinjiang, China, as early as the third millennium B.C. and may have provided the backdrop for the exchange of materials, such as silk and metals, between two major culture areas." (In fact, silk was found near Sapallitepa, north of the river Oxus, in graves belonging to the second millennium B.C.--Askarov 1977:726.).

I am convined that it is only due to the political situation that evidence is still "far from conclusive." 1. Regular digging in northern Afghanistan came to an end, and many promising sites and graveyards have since been plundered. (Jettmar 1981).

2. The site of Sarazm, situated on a flood terrace of the Zeravshan River and evidently revealing something like a Bronze Age metropolis, ought to be studied by means of large-scale excavations, with specialists for related complexses joinging the team. international coolaboration had been planned, but there is no hope that is can be implemented in the near future.

3. Excavation of prehistoric sites in the Tarim Basin would be most promising. The material should be studied by specialists in Bronze Age civilizations in order parts of Central Asia. Routes to the Far East must have crossed this region.

4. Observations indicating metallurgical centers of Western type in Gansu fit this picture, according to a short but highly interesting article by S. Kuczera. He stated in his concolusion that "some of those cultures are so close, chronologically and geographically, to the early Chinese dynasties that it is necessary to raise (and eventually resolve) the question of their mutual contacts and the possible influence of the Gansu centre of metal-making on the Shang-Yin civilisation. The most general conclusion that can be drawn from everything discussed above is that the correct understanding of the early periods of human history in the middle reaches of the Huanhe River, on the one hand, and Central Asia (outside China), on the other, is hadly possible without the knowledge and study of Gansu material" (Kuczera 1982:66)

W.T. Chase (1983:106) mentioned that some of the "contenders for the earliest Chinese metal objects" were produced not only by casting but also by additional hammering-definitely a Western trait. Moreover we are told: "There is something about all of these finds, and especially those from Gansu, that reminds us of the hoards and grave goods found on the steppes of central Asia. In fact, some of the later Gansu material has a definite stylistic affinity to central Asian decorative bosses and belt plaques. The shapes of the early knives, especially that from the Majiayao Culture (3000 B.C.), are very close to the Shang knives found at Anyang which show Siberian influence, and the Gansu macehead is closely paralleled by stone maceheads from South Russian and the Caucasus, suggesting a central Asian affinity." (cf. Jettmar 1980.)

The Importance of the Okunev Culture

In the northern fringes of the steppe belt, I stressed the extreme importance of the Okunev Culture which had on the one hand connections with the Far East and, on the other, definite links with the south of Central Asia.

Meanwhile I discovered a group of petroglyphs in the Indus Valley, near Chilas, that is connected with the engravings of the Okunev Culture by the main motifs and stylistic peculiarities. In addition to one report on my findings (Jettmar 1982: 298-302), others are forthcoming. It is not improbable that during the third and early second millemmia B.C. there were relations over thousands of kilometers, perhaps due to migrations of cattle-keeping Early Nomads. Other connections leading in the same direction were observed by Stacul (1977:251-252) and the Allchins (1982:111-116).

The Karasuk Culture

For the time that saw the emergence of the warlike Northern Nomads, the Karasuk period, Huttel took up my hint that a restudy of the development of the bridling system in China from Shang to Chou would be most rewarding. His paper (1979) comes to exactly the same conclusions as I expressed at the Berkeley conference.

No comments are needed on the later parts of my study. The discovery of a royal necropolis in northern Afghanistan (Sarianidi 1983) representing a dynasty rivaling the preimperial Kushans is of extreme importance. But these rulers were perhaps Saka, not Yueh-chih. Other Soviet studies are more pertinent to the problem I dealt with, but their discussion would need a separate article.

   Contact us  | © 1997-2000