That was also the time when Chinese civilization and politcs began to develop a distinct attitude of closedness. Partial responsibility for that lies in relations of conflict with the outer world. This attitude did not prevent territorial expansion towards the central Asian steppes, whose inhabitants in their turn considered China a periphery to their own economic and political centre. In this region the process of absorbion was seriously hampered by the basic difference between the respective economies.

The foregoing is a most simplified piece of history, picking up just selected trends deemed to be of great consequence. The transitive character of frontiers in the northern regions, which allowed the extension of economic activities and political control in both directions, has been well established by Lattimore (1940).

The diffusionist legacy in our thinking, especially pronounced in orientalist writing, would attach the notion of a 'centre' to great societies and cultures, that is cases where a high complexity of culture correlates with a similar complexity of society. Much hesitation would, however, follow if one applied the term 'centre' to the nomadic culture of the east Asian steppes, since, notwithstanding its attainments, it was not correlated with a fixed social substrate with permanently developed social institutions. Neverthless, any political organization emerging from that substrate would consider itself a centre and let itself be known as such to its neighbours, often including China itself. Thus we should agree that 'centre' is a relative concept asserted by a culture in consequence of its prevailing world-view and sytem of values. It is then a claimed value on the level of social practice.

Any objectivezation of the concept is precarious, since it is not properly defined. If complexity of culture were a defining component, we would not be able to measure it. On the other hand, we can measure the intensity of cultural borrowings and the results of acculturation, but it is rather dubious whether this can be made an index of the attribute of being a 'centre'; the more so as China would show varying indices in relation to her various neighbours. Besides there is a notable contrast between the minimal disposition of Chinese culture to assimilate alien cultures (recorded as a feature of long duration) and its great effectiveness in the adaptation of those cultures to Chinese political and economic organization (Pershits & Khazanov 1978, pp 112-118).

A stereotyped view would reduce centre-periphery relations to a onesided flow of cultural values and artefacts from a developed region to an underdeveloped one, and would treat partners in the process solely as emitting or receiving agencies. The great renown won by Chinese culture seems to exemplify without question such a stereotype, if China is regarded as a centre. It could then easily be accepted that China was a patron and a source of innovation for her neighbours of dissimilar cultures. It is often overlooked, however, that in the Shang and Zhou (Chou) dynasties at the turn of the 2nd millennium BC the dissimilarity was not so distinctly pronounced, and hence relations were more balanced. In later times too the situation was not clear in this respect.

With the growing advancement of agricultural and pastoral economies, the respective societies became more differentiated. This included a difference in complexity of social organization. Chinese society consisted of various groups composing a vertical stratification and a horizontal network, mutually interdependent and controlled with varying intensity by combined means of political, social and economic dependence. The interests of a particular group were liable to suppression by a central authority without endangering the interests of the whole society, provided that it was functioning normally. Even the removal of some groups or the disruption of society would not necessarily destroy the organization of the remaining part.

The opposite was the case with nomadic societies, which were much simpler in their organization. Intra-tribal institutions and authority were derived from kinship principles, while inter-tribal ones were based on political and military dependence. Both combined into a horizontal network of comparable units. Political structures emerged often as purposeful organizations and could be easily split and transformed, depending on the capabilities of the individual leaders and on external pressures. As a rule, nomadic polities were established in conditions of prolonged contact with sedentary societies for the purposes of securing compatibility with the latters' complexity, hence their aim was to regulate the contacts. If these polities acquired a degree of permanence, the political organization of the nomads would become stabilized over a period, thus achieving some complexity of social structure and of authority. These were, however, still transitory unless a state combined both nomadic and agricultural populations (Khazanov 1984, pp 228)

The history of contacts between the Chinese and the nomadic pastoralists has been divided by Eberhard (1965, pp. 119-34) into three stages: (a) neighbourly relations, (b) Chinese rule over some nomads, (c) nomadic rule over China. These stages cannot be precisely fixed in time, and particular circumstances changed independently of this general progression. As a classification of contacts, however, the suggestion is still valid.

Eberhard's neighbourly relations cover the early period of east Asian nedieval history, traditionally extending from the 2nd century BC to about the 3rd century AD. They included political relationships based on 'brotherhood', as well as military clashes, and attempts at increasing each side's influence. Contacts during the period from the Zhou (Chou) dynasty until the erection of defensive walls including the Great Wall were probably not very different from those of ancient times.

The nomadic contribution to the building of Chinese statehood, ethnicity and nation has been recognized but not adequately studied. Quasi- nomadic elements of old Chinese culture are inferred rather than established, since even a vague method of diachronic analogy has not been developed in this case, while written sources remain unreliable for a critical historian.

At the end of the 2nd millennium BC a two-way diffusion of artefacts from southern Siberia and from further west to China is reported (Watson 1971, p. 51). The go-between were then semi-nomads since at that time a full pastoral nomadism had probably not yet developed. These intermediary peoples also left some imprint on Chinese culture, we evidenced by chariots, some types of swords and knives, and in ritual as well.

It is known that the steppes of east central Asia yield agricultural produce from neolithic times until the last millennium BC, and exceptionally even later. The reasons for the discontinuation of agriculture are uncertain; most probably, however, they were of internal origin pertaining to ecological circumstances. Thus Eberhard's interpretation (1965, pp. 11, 120), that the pastoral peoples gave up agriculture when they got interested in robbing Chinese peasants of their produce, seems very doubtful. It would appear that they ceased cultivation because they were in want of foodstuffs and decided to acquire them by a new, uncertain, and politically dangerous strategy. The first conflicts of this sort on the Chinese border are recorded with certainty in the 2nd century BC, although they could well have a longer tradition.

Few accounts of relations between nomadic and sedentary populations overlook the use of force, or raids for booty. This is a predictable chapter of most writing on nomads, as poosed to works on settled peoples. The literature, including that of specialist anthropology, is full of accounts and instances of the pillage-oriented economy of the nomads. Little care has been taken to verify such cases in order to esclude biased interpretations arising from an unconscious prejudice against strange nomads by authors rooted in sedentary cultures.

As important attempt at re-evaluation of this archetype of nomadic behaviour in their contacts with the sedentary Chinese has been undertaken by Lattimore (1940). He has described the nomadic turbulence in frontier areas as a reaction to Chinese pressure exercised through expanding agricultural settlement and the tax-collecting vocation of regional officials. Subsequently some authors have begun to correlate raids by pastoralists with the changing conditions of their external trad (Khazonov 1984, p. 202). The theory has been most positively formulated by Zlatkin (1974, p. 141). In his opinion, raids by the nomads on their agricultural peripheries constituted a primitive form of normal trade, and when trade had been established, raids occurred as a substitute for broken trade relations. The following analysis draws on the second part of Zlatkin's theory, not excluding, however, the possibility that other conditions may also have been responsible for some of the cases under consideration.

Raids are certainly part of the nomadic way of like, although we have to be careful to distinguish between unprovoked ones, intended simply for plunder, and those which are punitive expeditions of some kind. The former are s simple auxiliary to the pastoral economy. The latter exert power and thus belong, not to the sphere of economic practice, but to the sphere of administrative means (the concept of administration, however, has to be related to a particular cultural specificity). Of course, both may bring gain to those who carry them out, but analogous results do not necessarily imply a similarity of purpose.

It is possible to suggest a correlation between the two types of raids and the level of social complexity. Robbery raids seem to be characteristic of simple, usually small, groups of pastoral societies, lacking a central authority with an established hierarchy of leaders and complexity of aims. Their system of values and political ambitions do not favour constant relations with agricultural neighbours nor any responsible planning for the benefit of the whole group in the present or future.

Notwithstanding raids, almost all nomads are eager to trade with agriculturalists in order to supplement their simple diet and limited range of handicrafts. Moreover, the concept of the sedentary economy as complementary to that of the nomads has long been part and parcel of their world-view. Central Asiatic nomads, as opposed to their western counterparts, have not been able, however, to dictate the terms and conditions of the trade since their main partners in that part of the world were militarily strong and politically bold. Efforts to continue the trade and to set favourable terms have been an essential features in relations between the pastoral and agricultural peoples for centuries.

Trade contacts between China and her northern steppe neighbours were probably established in the 6th century BC, and some of this trade was surprisingly not on a simple barter basis, as wvidenced by coin finds (Watson 1972, p. 143). Written sources refer only to barter exchange, conducted at selected points on the border where, according to mutual agreement, fairs were held annually or even more frequently. The first report of the existence of such fairs relates to the year 129 BC (Bichurin 1950, p. 63).

Chinese exports included cereals, tea, sugar, spirits, silk and cotton cloth, metals, luxury goods for display of wealth and status, pottery, and paper. Among imports the dominant items were animals (notably horses), husbandry produce (hides, white felt, wool), salt, antlers and sometimes slaves.

The partners in this exchange usually displayed varying interests as to their essential goals. The pastoral populations wanted goods necessary for their existence, mostly consumable ones. Capital investment was provided inside their own economy, although some technical investment in weapons was required from outside. The sedentary populations were theoretically independent of nomadic produce, though this was a political theory rather than a practical reality. These two contradictory attitudes, of explicit interest and of manifest disinterest, have defined the mutual relations of the east Asian peoples in question.

This allowed the Chinese to aim at regulating the trade according to their wishes. Applying political pressure to the nomads by denying them rights to trade with sedentary societies was not an infrequent ploy throughout central Asia on the part of the rulers of agricultural states (Khazanov 1984, pp. 206-7). The motives behind such an attitude were sometimes as simple as prejudice, but particular considerations concerning the possible rise in the power of the nomads with the help of imported strategic goods weran also involved. The Tanguts even restricted access to such strategic items as dishes for the Mongols shortly before their country was totally subdued by Chinggis-Khan (Khazanov 1984, p. 207).

Periodically providing, and then refusing, goods in exchange as a stratagem in relations with the nomads is said to have been invented by Chia I, a Chinese expert on the nomadic Hsiung-nu of the 2nd century BC (Eberhard 1065, p. 121). This trick was continued until modern times.

Trading with the nomads was synonymous with political control in official Chinese thinking. A chronicle Ming Shi, describing a system of purchasing horses from the Tibetans in the 15th century, concludes in this way: 'and thus Tai-Zu administered the Fans', i.e. the Tibetans. Another chronicle explains the aims of exchange: 'not only that China is short of good horses ...but also that the Fan barbarians living in frontier areas are thus getting attracted [to China]' (quoted from Martynow 1970, pp.. 82, 84).

Foreign trade was a state monopoly in China and any private entrpreneur could face the death penalty (Kychanov 1968, p. 95). Refusal of barter might have been detrimental to the local Chinese of the frontier provinces, but not to the state as a whole. The Chinese state was interested in few items from outside, its only constant need being for horses for the army. Since the northern neighboours and the Tibetans to the west were horse exporters all eager to sell, there was practically no danger of severing this strategic import link.

Trade restrictions as a means of foreign policy were practised towards the neighbouring states of the Tanguts and, to a lesser extent, the Kitans, Their dual economy provided much agricultural produce for internal markets but some imports were necessary. Besides, raw materials for handicrafts were needed, as well as luxury goods, Their relatively developed economy produced a surplus which had to be traded abroad, and the Chinese market was the most promising. At the same time it was the most difficult because, being under government control, it had to follow Frontier fairs were therefore often closed at will, and it was always the Chinese who acted in this way, even contrary to the interests of the Chinese in the region (Kychanov 1968, pp. 92-9). Except for some local troubles, mainly over cultivated frontier lands, no wars between the two states broke out solely for reasons of trade. The determining factor here was the centralized authority of the Tanguts and Kitans, which restrained their own citizens and their ambitions.

Such a restraint was weak or absent in fully nomadic societies organized on tribal principles. This was true for at least 19 centuries, starting from a peace treaty with the Hsiung-nu at the beginning of the 2nd century BC and closing with the Mongol expeditions led by Altan-khan, who demanded fairness in trade. Annoying pressure exerted by the Chinese who were the stronger party at barter fairs was resisted by the nomads seeking their own advantage here.

Reactions to the Chinese manupulation of trade are indicated by a report in a Chinese chronicle that Tanguts, arriving at a closed fair-place, shouted at the guards in anger: 'unless you let us barter whenever we come, we will start killing again'(Kychanon 1968, p. 95n.) THis should serve as a motto for any research on the relations with the nomads of easter Asia.

Any early form of acquiring merchandise from China by nomad societies was a 'tribute' given to their leaders by the Chinese court. The 'tribute' is an imprecise term for ceremonial gifts presented reciprocally, although those bestowed by the Chinese side often exceeded the balance. Already the peace treaty with the Hsiung-nu had established that the Chinese would provide the Hsiung-nu leader with some wine, 50,000 sacks of rice, and 10,000 pieces of silk annually. It was reaffirmed in AD 50 after a long period of not infrequently interrupted practice (Bichurin 1950, pp. 76, 119). The Chinese contribution to the welfare of the northern stats increased greatly with time, and some ten centuries later the Kitan emperor received as much as 300,000 pieces of silk per year in addition to an amount of silver (Tashkin 1979, pp. 283, 290).

Distribution of these 'tribute' goods was certainly unequal, and limited only to groups important to the continued exercise of the Khan's power. The amount of goods acquired was obviously too small to apportion an adequate share to each family, but there was enough of them for selected recipients among close associates and allies from more distant tribes. Ordinary members of the tribe were still in need of imported goods and on occasion conducted petty raids on the march-lands. The Hsiung-nu leaders constrained them, but at the same time tolerated them (Bichurin 1950, pp 110-11). It seems that irregular border fairs allowing for barter, a new institution at that time, were not sufficient to meet the nomads' various demands.

The tribute system of procuring gains was rather free of conflict, though it depended on the changing nature of political relationships. The reason why cessation of tribute alone did not provoke the use of force lay in its content: luxury goods prevailed over those of practical use.

Nomadic leaders understood that in order to consolidate their power they would have to draw on a more stable resource than tribute. The pastoral economy did not provide them with this opportunity, since even the exchange of surplus produce was not extensive not directed for consumption. Subsidising a state organization, in particular a complex one after the Chinese model, was an expensive business. A ruler had to rely on internal income for that purpose, and only a diversified economy could support such amitions. For this reason nomadic aristocracy tended to build a state which would include some agricultural peripheries. This policy, however, though well conceived, was not well carried out.

A great number of local dynasties organized by Turkic, Mongol and Tibetan leaders on China's periphery in the first millennium were short-lived and of limited territorial extent. Their most important feature was the underlying model of a state organized by the nomads: a combination of two types of economy, modes of occupying space and interacting cultures. Such a diversified state, approaching the nomadic ideal of economic autonomy,could not last for long in territories with a tradition of Chinese statehood, since the latter tended to revert to its uniform character. The reason behind this reversion was the difficulty which the predominantly Chinese officials faced in administering the state structure combining the two varying forms of social organization: tribal among the nomads and communal among the agricultural villagers.

To summarize this review of relations between the pastoral nomadic and the sedentary agricultural systems of east Asia, it seems essential to abandon the commonly-held view which sees them simply in terms of invasion and resistance. They are rather relationships of exchange, combined with policies of conflict, which preceded relations based on force. It seems unwarranted to regard the nomads either as just destructive or as a destablizing agent in the regions they occupied. They were rather an active factor in intercultural contacts and communication. and a dynamic one. The underlying motive was that the nomads were handicapped in the division of labour, and hence the outside world was complementary to theirs. It was also inherent in this position of the nomads as structurally related to sedentary societies that they could in certain circumstances be transformed from suppliers of products of their pastoral economy into a major military threat to the agricultural 'centre', with the potential for politcal domination from the periphery.