Extensive mobile pastoralism, including semi-nomadism and pure pastoral nomadism in its most extreme forms, is, or rather was, represented by but a limited number of types that reflect both its geographic diversity and economic similarity1. In addition, their formation was also not infrequently linked to historical circumstances, such as diffusion, borrowing, migration, etc. Intermediate and marginal forms excluded, the main types of mobile pastoralism are as follows: North Eurasian type (reindeer pastoralism of the tundra zone); Eurasian steppe type, which occupied the temperate zone of the steppes, deserts, and semi-deserts, from the Danube to North China, and sometimes also the wooded steppe to the north; the Near Eastern type, including Northeast Africa; the Middle Eastern type, which in some respects is intermediate between the Eurasian steppe and Near Eastern types and embraces the territory of contemporary Turkey, Iran, and Afghanistan; the East African type of predominantly cattle-breeders, who do not use transport animals such as the horse and camel; and the High Inner Asian type with the pastoralists of Tibet as its principal representatives.
These main types can easily be divided into sub-types, sub-sub-types, etc., but typologies and classifications are not the subject of this paper, which is mainly devoted to the pastoralism of the Eurasian steppes, semi-deserts and deserts. The latter is fairly homogeneous, although it is possible to single out its several sub-types: The Inner Asian (Mongol), the Kazakstan, the Central Asian, the East European (in the ancient and medieval periods), and the South Central Asian (Turkmen). The differences between these sub-types, however, are not only economic, but cultural as well. Only the South Central Asian type stands out on its own. On the territory of Turkmenistan, the deserts of the temperate zone become the deserts of
Khazanov / Nomads of the Eurasian Steppes in Historical Retrospective, pp. 476–500
the sub-tropical zone (the so-called Iranian-Turanian or South Turanian deserts); correspondingly, the pastoralism there acquired some characteristics similar to the Middle Eastern or even Near Eastern varieties.
Inasmuch as pastoral nomadism still lacks a generally accepted definition, I have to start with arguing my own understanding of this phenomenon. Some scholars pay particular attention to mobility and use the term ‘nomadism’ very broadly. They consider such different groups as wandering hunters and gatherers, mounted hunters (the Great Plains Indians of North America), all kinds of pastoralists, some ethno-professional groups like Roma (Gypsies), the ‘sea nomads’ of Southeast Asia, and even certain categories of workers in contemporary societies (the so-called industrial mobility) to be nomads. Others perceive nomadism as a sociocultural system or primarily in cultural terms of a specific way of living, lifestyle, world view, value system, etc. These definitions, however, neglect the economic side of nomadism, which, in my opinion, is the most important criterion. Above all other characteristics, extensive mobile pastoralism is a specific type of food-producing economy that implies two opposites: between animal husbandry and cultivation, and between mobility and sedentism. The size and importance of cultivation in pastoralist societies, along with ecological factors, determines the degree of their mobility and may serve as a criterion for different varieties of pastoralism.
In this case, pastoral nomadism in its most specialized variety is based on the following characteristics: (1) Pastoralism is the predominant form of economic activity; cultivation is either absent altogether or plays a very insignificant role. (2) Pastoralism has an extensive character connected with the maintenance of herds all year round on free-range grazing without stables and without laying in fodder for livestock. (3) The pastoralist economy requires mobility within the boundaries of specific grazing territories, or else between such territories. (4) All, or at least the majority of the population, participates in this mobility. (5) Pastoralist production is aimed at the requirements of subsistence. The traditional pastoral nomadic economy was never profit-oriented, although it was often considerably exchange-oriented. (6) Social organization of pastoral nomads is based on kinship, and, in the case of the nomads of the Eurasian steppes, and of the Near and Middle East, also on various segmentary systems and genealogies, whether real or spurious. (7) Pastoral nomadism implies certain cultural characteristics connected with its mobile way of life, sociopolitical peculiarities, and some other factors.
Any specialization implies dependency, and pastoral nomadism is no exception. It was an innovative solution for assimilating certain, previously underexploited ecological zones. The emergence of pastoralism, and later of pastoral nomadism, was a crucial moment in the spreading of food-producing economies in the arid, semi-arid and tundra zones of the oikumene, because for a very long time they had an advantage there over all other types of economic activity.
However, the shortcomings of pastoral nomadism are also quite evident. First, its specialization was principally different from that in industrial and even in traditional farming and urban societies. This was appreciable already in the early stages of Near Eastern history (Nissen 1988: 43ff.). Specialization in pastoral nomadism implied the division of labor between societies with different economies. The internal division of labor within nomadic societies was very undeveloped.
Second, unlike many types of farming which had the potential for diachronic technological development, in pastoral nomadism, once its formation was complete, the reproduction of similar and highly specialized forms prevailed. Its ecological parameters significantly limited the capabilities for economic growth through technological innovation; they also placed very serious obstacles to the intensification of production.
Thus, an increase in productivity of natural pastures requires extensive development projects that only industrial societies are capable of carrying out. Even temporary maximization of the number of livestock could be achieved mainly by increasing the production base through territorial expansion. As a rule, this was done by military means and/or by turning the sown into pasturelands. Both ancient and medieval histories are abundant with examples of such events. However, this extensive way of increasing production could be neither permanent nor stable. It was too much at the mercy of the balance of power between nomadic and sedentary societies. Besides, sooner or later even the enlarged ecological zone of pastoral nomadism would be completely filled out, which would make further growth in stock numbers impossible.
There was another reason for the dependence of nomads on the outside world. Pastoral nomadism as an economic system is characterized by constant instability. It was based on a balance between three variables: the availability of natural resources (such as vegetation and water), the number of livestock, and the size of the population, all of which were constantly oscillating. The situation was further complicated because these oscillations were not synchronic, as each of the variables was determined by many factors, both temporary and permanent, regular and irregular. The simplest and best known case of temporary imbalance was periodic mass loss of livestock and consequent famine due to various natural calamities and epizootic diseases. In other cases, stock numbers sometimes outgrew the carrying capacities of available pastures. It was just such cyclical fluctuations that maintained the long-term balance in the pastoral nomadic economy, however ruinous they might be in the short run. In other words, the balance was not static but dynamic.
One of the means of overcoming the deficiencies of the pastoral nomadic economy was the creation of the farming sector. Actually, during long historical periods, many if not most of those who roamed the Eurasian steppes were not pure nomads but rather semi-nomads, practicing cultivation as a supplementary and secondary form of subsistence. However, as the Soviet ‘virgin lands’ campaign has proven, even in the twentieth century, cultivation without irrigation is a risky endeavor in the dry zones and often results in overexploitation of productive ecosystems. In pre-modern times, it was even less stable and reliable. Semi-nomadism was unable to solve the problem of non-autarchy of the pastoral economy. Under this situation, the nomads needed sedentary societies as a kind of external fund vital to their survival. They invested little in this fund, but it was indispensable to them when they got an access to its interest, and sometimes even to its principal. But in order to get an access to this fund, pastoral nomadic societies had to adapt to external sociopolitical and cultural environments.
An integral part of the nomadic ideologies was the antithesis between nomadic and sedentary ways of life, which to some extent reflected the differences in actual conditions of existence. As a manifestation of the universal ‘we – they’ opposition on a symbolic level, this antithesis played an integrating function within nomadic societies and a differentiating one regarding the sedentary world. Moreover, it created a negative view of the sedentary way of life. Nevertheless, historical sources since the very first mention of nomads make it clear that grains and other farm products formed an important part of their dietary systems. These sources, as well as numerous archaeological data, also demonstrate beyond doubt that the nomads procured a substantial part of their material culture from sedentary territories. The economic dependence of nomads on sedentary societies, and their various modes of adaptation to them, carried corresponding cultural implications. As the nomadic economy had to be supplemented with products of cultivation and crafts from external sources, so too did nomadic culture need sedentary culture as a source, a component, and a model for comparison, borrowing, imitation, or rejection. Even ideological opposition was relative. Suffice it to say that nomads never created any world and universal religion, but made significant contributions to the dissemination of these religions around the world (Khazanov 1993, 1994a). The nomads understood very well certain social and military advantages of their way of life. At the same time, they also comprehended that their culture was less complex, rich and refined than that of their sedentary counterparts. Their attitudes towards the latter had some similarities with the attitude towards Western culture of many in the Third World. Experiencing its irresistible glamour but being outside its socioeconomic sphere they reject it in principle, but strive to borrow some of its achievements. However, nomads did not suffer from an inferiority complex and did not resort to terrorism. Borrowings always underwent selection and filtration with regard to their correspondence to nomadic culture as well as to their utilitarian value (Allsen 2001).
This is quite evident with regard to those nomadic states in which new cultures emerged. Although the nomads, or, to be more precise, their elites, initiated the formation of these cultures and were their main patrons and consumers, they were created mainly by specialists from various sedentary countries: artists, craftsmen, traders, religious preachers, intellectuals, literati. This is why these cultures were eclectic more than synthetic. Perhaps they should be called state cultures because they were created to provide comfort and luxury to ruling nomadic elites and, even more importantly, to facilitate state management. Since these cultures were by no means ethnic ones and were quite different from the synchronous cultures of ordinary nomads, their fate was directly connected with that of the states which engendered them. There was no Golden Horde people, but there was the distinctive culture of the Golden Horde (Kramarovsky 1991: 256–257). The cultures of the Saljuq sultanates, of some Mongol states simultaneous with the Golden Horde, to a lesser degree of the Turkic qaghanates, especially that of the Uighur, the still obscure culture of the Khazar qaghanate, and, perhaps, even the state culture of the Scythian kingdom may serve as examples.
There is a peculiar tendency in several Central Asian countries, and even in some republics of the Russian Federation, that has become especially noticeable in the post-Soviet period. It is connected with the specifics of their nationalist mythologies, which, like in other countries, spare no effort in attempting to glorify real or imaginary ancestors (for analysis and criticism of nationalist mythologies in the former Soviet countries, see, for example, Shnirelman 1996; Eimermacher and Bordugov 1999; Olkott and Malashenko 2000). Since these ancestors, or at least some of them, not infrequently were nomads, a number of scholars and pseudo-scholars either tend to overstate their development and achievements, or, on the contrary, strive to prove that they were no nomads at all, but practiced instead a mixed economy. To some extent, such attitudes are an overreaction to Soviet concepts of historical processes that considered pastoral nomadism a blind alley and praised the forced sedentarization and collectivization of the nomads in the late 1920s and early 1930s as the only way of their economic development.
In genuine scholarship there is no need, or reason, for any kind of ideological speculation and unbridled fantasy. The search for direct ancestors prior to the early modern period is a hopeless endeavor considering the specificity of ethnic history in the Eurasian steppes and in the adjacent regions of Central Asia. However, if the glorious ancestors are really indispensable for nation-state building, the nomadic ones are nothing to be ashamed of.
The importance of pastoralism in general, and of pastoral nomadism in particular, far exceeds their successful response to the challenge of climatic and geographic conditions. In the political and ethno-linguistic history of the Old World their impact is hard to overestimate (Khazanov and Wink 2001). Nomads played an enormous role in radical border changes, the destruction of some states and empires and the emergence of new ones. While it is still unclear whether the original Indo-Europeans were nomads or incipient agriculturalists, the spread of Semitic languages, of the languages of the Iranian branch of the Indo-European linguistic family, of many Altaic languages, especially the Turkic ones, and, apparently, of some African languages, i.e. Nilotic, was certainly connected with the migrations, conquests, and/or political dominance of the pastoralists and nomads. In some periods, the nomads served as organizers of and intermediaries in cultural exchange between different sedentary societies. Their contribution to the transcontinental circulation and transmission of goods and ideas was quite significant. In this regard, polyethnic and polycultural empires created by the nomads played a certain positive role (Bentley 1993).
Not only did sedentary peoples influence the cultures of nomads; nomadic cultures in turn influenced those of their sedentary counterparts. The invention of, or even more so the spread, of some cultural traits was their indisputable achievement. Nomadic arms, ornaments, modes of fashion, and traditions were often imitated in sedentary countries. Among other things, this was reflected in the phenomenon of post-nomadism. The nomadic system of values, the way of life, the rules of social behavior and political traditions were considered prestigious and were imitated in certain strata of sedentary societies long after the nomads themselves had sedentarized or had ceased to be politically dominating.
Still, the indisputable fact is that in economic respects, the pastoral nomads depended on sedentary farming and urban societies much more than the latter on the nomads. Pastoral nomadic economies were never autarkic and could never be so.
The nomads always strove for the acquisition of products from sedentary societies by all means possible. It was a matter of sheer survival for them. This was noticed already by the great medieval sociologist, Ibn Khaldun (1967: 122), who wrote:
…The desert civilization is inferior to urban civilization, because not all the necessities of civilization are to be found among the people of the desert… While they [the Bedouins] need cities for their necessities of life, the urban population needs [the Bedouins] for convenience and luxuries. Thus, as long as they live in the desert and have not acquired royal authority and control of the cities, the Bedouins need the inhabitants of the latter.
Likewise, the nomads of the Eurasian steppes had to adapt not only to a specific natural environment but also to external sociopolitical and cultural environment. Their interrelations with sedentary societies varied from direct exchange, trade, trade mediation and other related services, or mercenarism, to raids, looting, blackmailing, occasional payments, more or less institutionalized subsidies, regular tribute extraction, and direct conquests and subjugations.
Nomadic conquests and their consequences always attracted great attention. However, a related question, why the nomads with their limited human and economic resources were, for centuries and even millennia, so strong in military respects, has not been sufficiently addressed. Each individual case certainly depended on many circumstances and deserves a special study, but in general terms the answer seems to be connected with the undeveloped division of labor and wide social participation, which provided the nomads the edge in the military realm. With but few exceptions, in sedentary states, the war was a specialized and professionalized sphere of activities. On the contrary, in nomadic societies, every male commoner (and in some of their pre-Islamic societies, even some female when necessary) was a warrior, most of them mounted ones. Only this allowed the nomads, despite their relatively small number, to mobilize sufficiently large armies. Moreover, their specific way of life, among other things, implied an availability of a large number of horses and almost natural military training. In terms of individual skills, only the medieval European knights and the Middle Eastern mamluks were a match to nomadic warriors; but the training and military equipment of the latter sometimes reflected nomadic military traditions (Nicolle 1976; Amitai-Preiss 1995: 214ff.; Smith and Masson 1997: 255ff.).
Thus, the economic and social backwardness of the nomads turned out to confer military advantages in the interrelations with their sedentary counterparts. Up to modern times, this advantage often allowed them to transfer these interrelations from the purely economic, or cultural, onto a political plane. Their military superiority gave them the leverage for political domination. This was particularly true for the great nomads of the Near and Middle East and the Eurasian steppes capable of the large-scale intrusions and conquests so numerous in ancient and medieval history.
There is still one more and very important factor that should be taken into account in order to understand the functioning of pastoral nomadic societies. The necessary prerequisites for pastoral nomadism had first been created by the transition from food-extracting to food-producing economies usually labeled the Neolithic revolution (Shnirelman 1980; Clutton-Brock 1987, 1989). Both cultivation and animal husbandry contained the potential for the dissemination with further specialization in different ecological zones. However, for many millennia after the Neolithic revolution, food-producing economies in the Old World were usually based on a combination, although in different proportions, of cultivation and animal breeding. In my previous publications (i.e. Khazanov 1994: 85ff.; 1990: 86ff.), I argued that pure pastoral nomadism with complete separation from cultivation was a rather late development, although its many important technological preconditions, such as horse riding (Anthony and Brown 1991), had appeared much earlier. Even in the main indigenous areas of its dissemination, being the Eurasian steppes and the Near East, it emerged only around the turn of the second millennium B.C. (I mean the forms that without drastic modifications continued then to function for three thousand years). In both areas, it developed from a mixed economy through intermediate forms of extensive and mobile pastoralism with cultivation as supplementary activity. In other areas, pastoral nomadism was formed and spread even later, under the direct or indirect influence of the already existing forms.
This hypothesis still seems to me the most plausible, but not necessarily exclusive. Inconclusive as most of the archaeological data are despite the growing sophistication of their analysis (Bar-Iosef and Khazanov 1992), they provide some materials for suggestion that primitive forms of pastoral nomadism could from time to time appear in a few regions of the Eurasian steppes as early as the Bronze Age, or even earlier (for example, Shishlina 2000). However, these were hardly more than a temporal adaptation to specific local conditions. Given its disadvantages, pastoral nomadism required not only a trigger, a special motivating stimulus, to emerge, but also a favorable external sociopolitical environment. Only in these conditions could it become a viable economic alternative for extended historical periods. It seems that these requirements, which made possible a long-term break of pastoralism from other forms of food-producing economies, were met only by the end of the second millennium and start of the first millennium B.C.
On the one hand, the desiccation of the climate noticeable at that period had apparently modified the natural environment and required further specialization on the part of the pastoral economies. On the other, only from the first millennium B.C. did the pastoral nomads form the periphery of sedentary states emerging on the southern borders with the steppe zone. Only at that time did they acquire their optimal milieu. My point here is that for the most efficient and long-term historical functioning, pastoral nomads not only needed simply sedentary societies, but those of a certain developmental level – not primitive and not industrial, but traditional (i.e., pre-modern or pre-industrial) societies that had already achieved the level of statehood, such as the ancient and medieval states of the Near and Middle East, Central Asia, or China.
Primitive societies were unable to supply all the products needed by the nomads and even less to supply them in sufficient quantity. Considering their lack of centralized political organization, the collection of whatever surplus was produced would constitute a serious problem. These societies could easily be robbed, but not systematically exploited. In the pre-colonial period, such a situation existed in East Africa. It is hardly accidental that pastoral nomads there lacked even stable and institutionalized leadership. Horses and camels might have begun to be used for riding earlier than had been assumed by scholars twenty years ago or so, but there is no evidence of mass cavalries before the beginning of the first millennium B.C. After all, in order to mobilize and maintain them, one should have adversaries against whom they could efficiently be employed. It is worth remembering that at the dawn of their history, the Cimmerians and the Scythians, in a search for such adversaries, had to cross the Caucasus and invade the ancient Near East. On the contrary, modern and contemporary states have in abundance everything that the nomads may dream of. However, they are much stronger and almost always are on the offensive in their interrelations with the latter.
Unlike primitive and modern societies, pre-modern states with stratified social systems provided the optimal environment for pastoral nomads. Not only did they produce a regular surplus product; they possessed mechanisms for its extraction, distribution, and redistribution. In addition, they had a fairly developed division of labor, exchange and trade, and a culture, including the ‘Great Tradition’, on which nomadic cultures might depend. The gap between these societies and their nomadic counterparts in the ancient and medieval periods was not very deep. After all, with the exception of the reindeer pastoralists of the North and the cattle-breeders of East Africa, pastoral nomadic societies were but varieties of the traditional ones.
This brings me to the specifics of sociopolitical development of the pastoral nomads. The Soviet studies of pastoral nomadism, however serious and important they were in many respects, suffered from one important deficiency. Their fundamental premise was the Marxist concept of universal and progressive socioeconomic formations. In accordance with this, every society had to develop in a similar way and in the same direction, and the nomads were in no way considered an exception. Thus, ideology forced Soviet scholars to deal with an unsolvable problem: how to prove that the nomads were developing towards higher socioeconomic systems. This was difficult to do with regard to many sedentary societies and even whole historical regions; with regard to the nomads this was simply a hopeless endeavor, involving a long but fruitless discussion (for one of its last surveys see Kradin 1992). Few scholars tried to avoid the Marxist dogmas (Khazanov 1975; Markov 1976); many more wrote about nomadic feudalism; others about the ‘military democracy’ (using Engels's term) as the developmental level occupied by the nomads; still others imagined a specific ‘nomadic formation’. There are also scholars who construct a more sophisticated evolutionary sequence: the archaic empires – the barbarian states – and the early feudal states; apparently in an attempt to prove that the historical development of the Eurasian steppes followed the West European model (Kljashtorny and Sultanov 2000: 82). The circular pattern for the nomads' development was sometimes admitted, since it was too striking to be ignored. But this was often explained by the fact that the nomadic polities which succeeded each other in the steppe had a somewhat different ethno-tribal composition – thus, each time starting the development anew. In spite of these differences of opinion, the emergence of nomadic statehood was perceived as a spontaneous process, just as had been prescribed by Marxist theory. External factors were either played down or completely ignored.
The whole problem, however, is farfetched. It exists only to the unilinear evolutionists and Marxists, whose understanding of historical process is essentially teleological. At present, only by a great stretch of the imagination and at the great expense of factual data can one adhere to the antediluvian theory of universal socioeconomic formations. So far, the historical process has never been universal and unidirectional. In different regions and in different societies, it took on different patterns, forms, directions, tempos, etc. As for similarities, they were at least connected as much with movements of ideas and populations, cultural and technological borrowings, or with forced imposition of external patterns, as with parallel indigenous developments. Still, even imitation or imposition of forms and patterns alien to recipient societies have many limitations and are far from always successful. This is demonstrated by the difficulties that the process of modernization, and globalization as its latest stage, are facing in many Third World and some other countries. There is no reason to assume that in other periods the situation was any different.
In my opinion, the major units of the historical process were not universal formations, but cultural regions. These are often called civilizations, but I am reluctant to use this term because of its semantic ambiguity. It involves a never-ending discussion about a number of civilizations and the criteria for their definition. In other words, it involves many speculative and subjective taxonomic and culturological questions that I prefer to eschew. In any case, serious long-term regional differences existed and were connected with many factors of geographic, economic, social, political, cultural, and many other orders. All major breakthroughs in human history were results of unique combinations of many and various factors, some of which were almost accidental. History is not a deterministic project. Actually, one can observe only a few, if any, laws and regularities in history, and they are mainly limited to a sequential order (Gellner 1988: 15ff.).
Be that as it may, for almost three thousand years the political development in the Steppe region oscillated between similar forms of statelessness and statehood; this very oscillation was connected not so much with spontaneous development, but mainly with the specifics of interrelations between nomadic and sedentary societies. For these same reasons, the development was reversible but not completely circular, since the very character and peculiarities of the nomadic polities to a large extent depended on the character of their sedentary counterparts, which were quite different. The pattern was more or less the same but in the course of history some modifications and innovations can be traced in the forms of nomadic statehood, which reflected changes in the sedentary world.
The nomads of the Eurasian steppes were the most successful of all nomadic conquerors. Their horses trampled the fields of France and Italy, Syria and Palestine, India and Indochina. Noteworthy but not accidentally, the level of their social organization was also higher than that of the nomads in other regions (one should also take into account that warriors on horseback are much stronger than those on camelback – Sinor 1972, 1981).
Beginning from the first millennium B.C., fairly big polities appeared to become quite common in the Eurasian steppes. They may be considered as a functional, although not structural, analogue of chiefdoms well described on the example of many sedentary societies. These nomadic polities are alternatively called ‘tribal associations’, ‘tribal federations’, ‘tribal confederations’, etc. I do not consider these terms sufficiently felicitous, because most of these ‘federations’ and ‘confederations’ were created by force. But after all, any terminology is conditional and, therefore, should be a matter of consent, not debate. Much more important is to understand the nature of the political organization of these formations.
Archaeological indicators of the emergence of these polities are the very rich burial mounds which are not infrequently (and not always substantially) called ‘royal’, by analogy with the Scythian ones. In many cases, archaeological materials alone are insufficient in determining the exact level of political development of nomadic polities. As a rule, this can be done only when they are complemented by literary sources, especially if the latter contain sufficient information on the interrelations between nomadic polities and sedentary societies. Thus, we may assume with a high degree of confidence that the Scythians and Hsiung-nu reached the state level, because we have sufficient knowledge about the economic bases of their states. But one should be much more cautious with regard to those who left such burial mounds as the Arzhan, Issyk, Pazaryk mounds, and several others.
Far from all segmentary forms of social organization are egalitarian. Some of them are asymmetric. Structural relativity and balanced opposition in them are upset, and a dominant lineage group acquired greater control of resources and greater power than the others. I call such systems differential or even stratified, and the latter two were the most characteristic of the nomads of the Eurasian steppes. Segmentary systems do not totally preclude the possibility of the emergence of a central governance agency, but they place serious obstacles to its functioning and limit the sphere of its activities. To a large extent, power in nomadic polities was diffused and was mainly connected with military and managerial-regulatory functions. Correspondingly, their composition was fluid; they were loose and short-lived except in the cases when they underwent transformation as a result of their specific relations with the outside world. In other words, internal requirements for political integration in nomadic polities were too weak to result in irreversible structural change. Their leaders seldom acquired the monopoly of legitimate violence, which, according to Max Weber, is the most important characteristic feature of the state. There is even less ground to assume that these polities were based on class divisions (an opposite opinion was recently put forth by E. I. Kychanov [1997: 5]). Only in a few, rather rare cases did internal development in nomadic societies lead to the emergence of hereditary (although still reversible) social stratification. By and large, ‘managers’ in nomadic societies were less expansive than ‘managers’ in their sedentary counterparts. Ordinary nomads might respect their authority, high status, and even hereditary rank, but they were less inclined to pay for this, especially when regular payments were required. For all these reasons, nomadic aristocracies were not able to create an autonomous power base within their own societies, which would provide them with sufficient freedom of action.
To some extent, social stratification in nomadic societies increased when their aristocracy succeeded in the subjugation of other nomadic groups. However, such subjugations were seldom stable and lasting. The history of various Turkic qaghanates proves this unambiguously. Inherent deficiencies of the pastoral nomadic economy made the production of regular and fairly large surplus by the commoners very problematic. In cultural respects, the problem consisted of the same way of life, which implied mobility and, therefore, the possibility of break away and migration to new territories of dissatisfied groups of nomads. In socio-political respects, the problem was connected with a lack of strong machinery of coercion. The example of the early medieval Turks, Mongols, and many others demonstrates that the mobilization of many nomadic formations was indispensable for the creation of a large nomadic state. However, only the anticipation of benefits from joint exploitation of sedentary societies might for a while reconcile subjugated nomadic groups with their dependence on other groups.
In the final analysis, the emergence of nomadic states and their characteristic features, as well as their fate, were connected with specific forms of exploitation of sedentary societies. As a matter of fact, the very term ‘nomadic states’ is to some extent tentative. They were nomadic insomuch as they were founded by the nomads and/or the nomads occupied the dominant positions in them. However, in one way or another, all of them were based on asymmetric relations with sedentary societies. Otherwise, nomadic states were but rare and transitory exceptions, if such exceptions existed at all. The history of nomadic statehood in Central and Inner Asia, from the Hsiung-nu to the Mongols, and even to the Manchus (the latter were never pure nomads, but in many respects followed the Mongol political tradition) illustrates this point very well.
Following Lattimore (1940), Barfield (1989, 1991) has suggested an interesting model for the explanation of cycles of Chinese dynastic history and nomadic statehood in Inner Asia. He claims that all nomadic empires in the Mongolian steppes and the Chinese dynasties that managed to unite the whole country rose and fell together. By contrast, Manchurian states could develop only in times of anarchy on the northern frontier, when central authority in both China and in the steppe had collapsed. In my opinion, this model has its weak points along with the strong, and, therefore, needs further elaboration. For example, the Hsiung-nu state was founded in 206 B.C., when China was on the brink of civil war; it flourished in the early Han period, when central power in the country was not yet completely consolidated (Yamada 1982). The Turkic qaghanate emerged in the middle of the sixth century, at least thirty years before China, divided into local regimes, was united again. Barfield's claim that the nomads never played an important role in the collapse of unified Chinese empires is an overstatement, although their impact was often indirect. Suffice it to mention the Uighurs' assistance in the suppression of the An Lu-shan rebellion, which for a time saved the Tang, but certainly ruined the country by turning it into a huge hunting ground and, thus, contributed to the eventual collapse of the dynasty (Pulleyblank 1955). It is also hard to agree with Barfield that the Mongol conquest of China was an aberration of the steppe conquest patterns and was almost accidental. The term ‘aberration’ does not deliver the causes of the conquest and its success from explanation (on this see Franke and Twitchett 1994). Still, the historical process of long duration in China was much more connected with internal rather than external factors. Even China's repeated failures to deal with nomadic threats should be attributed to specifics of her socio-political system and political philosophy at least as much as to the strength of the nomads. On the contrary, the character of nomadic statehood in Inner Asia and its very existence to a large extent depended on the vicissitudes of the political situation in China.
With a great degree of simplification and schematization one may single out the two main types of nomadic states according to their relations with sedentary societies. In the nomadic states of the first type, these relations were mainly confined to vassal-tribute and other undeveloped and not always completely institutionalized forms of collective dependence. Sometimes sedentary states continued to exist2, in other cases, nomads and peasants-townsfolk were joined within one and the same state. But even in the last case their limited integration took place primarily in the political sphere, without affecting the social and economic foundation of sedentary societies.
Ordinary nomads in such states remained their main military and social support. With regard to their own society, or sub-society, the nomadic aristocracy was positioned as the leading estate rather than as the dominating class. Correspondingly, redistributive mechanisms continued to function among the nomads. It is true that in almost all nomadic states of the first type their own farming and urban sector emerged – in the main, owing to voluntary or involuntary migrants from the sedentary territories. However, as a rule, it was too weak to provide for all of their economic requirements. Thus, these states could not do without towns, which were centers of political power and to a lesser degree of handicrafts and trade. Admittedly, their emergence looks somewhat artificial. It was not so much the state which existed on their account as that they existed on account of the state. They perished with the downfall of the states that had brought them into existence. The fate of Itil, Karabalaghasun, Sarai-Batu, and Sarai-Berke is very indicative indeed.
The Scythian and Hsiung-nu states, the Turkic qaghanates, the state of Khitans (Qara-Khitay) in Central Asia, the Mongol Empire under the first great khans, later the Golden Horde3, and some others, may serve as examples of nomadic states of the first type. Eventually, they either underwent a transformation connected with their growing complexity or ceased to exist when opportunities for the primitive exploitation of sedentary societies diminished and vassal nomadic formations broke away. Usually this was followed by primitivization of the socio-political order. In more rare cases the process of sedentarization took the upper hand, especially when the nomads migrated to other ecological zones. As a result, the society ceased to be nomadic and for the most part became a farming-urban society. It still could preserve a significant pastoral sector, but its general development took quite different directions. The Uighurs ousted by the Qirghiz to East Turkistan may serve as an example.
The nomadic states of the second type not infrequently were the outcome of a certain transformation of those of the first type. Among other developments, it was connected with the integration of nomads, peasants, and townsmen into a single political system. Most often this happened when the nomads, after conquering sedentary states, moved onto their territories. In ancient times, nomadic states of the second type were represented by the Kushans Empire; in the early medieval period by a number of states created by nomads in North China in the fourth to sixth centuries; later by the Khitan and Jürchen states in China, the Qarakhanid state in Central Asia, and by the Saljuk state in the Middle East; in the Mongol period by the Yüan in China and the Hülegüid state in Iran; and in the late medieval period by the Shaybani state in Central Asia.
These states were created by the nomads and were ruled by dynasties of nomadic origin. Social stratification in them to a certain extent coincided with economic specialization and ethnic division. These characteristic features allow them to be called nomadic. In such states, nomads and the sedentary population could even belong to separate sub-societies, but only socially, not politically. The integration process usually started with the dynasty and its immediate entourage; then affected all or part of the nomadic aristocracy, which became the upper class of the sedentary population, or one of its upper classes. Political synthesis in these states was somewhat accompanied by a social synthesis, although in practice the latter was nowhere near always fully realized. Nevertheless, socioeconomic and even political relations and characteristics of subjugated sedentary societies, which were more developed than those of their nomadic conquerors, demonstrated a remarkable resilience. In sedentary sub-societies the consequences of nomadic conquest affected mainly the privileged strata. Not infrequently, the nomadic aristocracy, or the ruling strata of nomadic origin, became a landed estate. But more often than not, even the turnover of ruling elites was not complete, and a certain institutional continuity can be traced in many cases. In the Muslim countries, it was much easier for victorious nomads to replace ‘people of the sword’, the military estate of subjugated countries, than ‘people of the pen’, their bureaucracy. Besides, those nomads who converted to Islam, such as the Saljuqs, Qarakhanids, and later the Shaybanids, never considered, or dared, to encroach upon another group of sedentary Muslim society: the religious nobility, the ulama, and the Sufi shaykhs (Bartold 1968). In China, the literati officials survived all nomadic conquests because they were indispensable for ruling the country. In this regard, it is worth repeating once more the old and much-quoted aphorism of the ancient Chinese orator, Lu Tsia, taught to the Great Khan Ögödei, son and successor of Chinggis Khan, by his Chinese counselor, Yehlu Ch'uts'ai: ‘Although you inherited the Chinese Empire on horseback, you cannot rule it from that position’. Ögödei got the message, and allowed Confucian scholars to be drawn into the civil administration (Munkuev 1965).
It seems that the change in the social and economic relations in the sedentary societies caused by nomadic conquests was often less drastic than it is sometimes assumed. A permutation within the existing socio-economic order was their more frequent consequence than radical transformation. There are many examples of nomadic conquerors that willingly adopted institutions of subjugated societies when they were considered expedient. Thus, the Saljuqs adopted and extended the iqtā system because it facilitated their rule over a conquered sedentary population.
The main forms of dependence and exploitation in the nomadic states of the second type were connected with the relations between the ruling class, in which the nomadic aristocracy occupied the dominant position, and the conquered sedentary population, in particular the peasantry. However, the positions of the rest of the nomads did not remain immutable either, although they never constituted the main class or even a single class or estate. The nomadic sub-society was becoming more differentiated. The nomads divided into the privileged, less privileged, and the non-privileged, depending on their ties with the dynasty and the nomadic aristocracy, their ethnic and tribal membership, etc. Usually, they formed several intermediary estates and strata, some of which were closer to the ruling class, and others to the dependent ones.
As the integrative process developed, the nomadic aristocracy, and in particular the dynasty, had to decide whether they should identify their interests with those of the state as a whole and with the sedentary sub-society, or to preserve the loyalty of nomadic sub-society and in doing do sometimes act against the interest of the state.
The dilemma was never an easy one, and a consensus was far from always taken on it, even by various groups of the nomadic aristocracy. As a result, new conflicts often emerged in states of the second type, for example, between the dynasty of nomadic origin and its supporters, on the one hand, and the traditional nomadic aristocracy, on the other; between different groups of nomads; between the dynasty and ordinary nomads, etc.
One of the widespread further developments of these states was their eventual transformation into sedentary states, in which some nomads gradually became sedentary, while others little by little turned into a backward social and sometimes ethnic minority. They became encapsulated in the more developed socioeconomic and political systems. Ottoman Turkey may serve as an example of such development, but this is already another story.
In all, the decisive factor in the emergence and functioning of nomadic statehood was the specific relations between nomadic and sedentary societies. Still, the internal factor should not be dismissed completely. It was, however, connected with the remarkable stability and continuity of the political culture in the Eurasian steppes rather than with the very dubious evolutionary development. This culture was polyethnic and was by no means confined to individual nomadic polities and states (Golden 1982, 2001; Trepavlov 1993; Kljashtorny and Savinov 1994; Allsen 1996).
Ancient and medieval contemporaries of the nomads, as well as many modern scholars, were often astonished by the swift rise of strong nomadic polities, which seemingly sprung from out of nowhere and almost immediately initiated military campaigns against their neighbors, both nomadic and sedentary. Perhaps this would be less surprising if one would take into account that already in the medieval period, most if not all of the nomads of the Eurasian steppes were well acquainted with the idea and practices of statehood. Knowledge, models, symbols, even some traditions of higher forms of political organization continued to exist, though in their latent or semi-latent forms, even in those polities that can hardly be characterized as nomadic states.
Apparently, the original political culture had emerged in the Eurasian steppes already in the first millennium B.C.4 This polyethnic culture was represented by different synchronic and diachronic variants, but nevertheless bore many similar characteristics across the whole region. Despite modifications, it also demonstrated remarkable stability. This should not be too surprising, since the main characteristics of the sociopolitical organization of the nomads also had many common and stable features. The political culture of the nomads underwent substantial changes only after the cultural space in the Eurasian steppes was fragmented by dissemination of different world religions, especially after most of the nomads converted to Islam (Cohen 1968), and the Mongols converted to Buddhism. Still, some of its traits were noticeable even much later (Manz 1989).
The sources of this political culture are far from being completely understood, and the search for them constitutes an intriguing problem. At the moment, one may tentatively assume that it was based on the original nomadic traditions in conjunction with borrowings from political traditions of various sedentary states and societies which were adapted to nomadic conditions. Likewise, mechanisms of transmission of this culture in different ethnic and linguistic milieus are not yet sufficiently researched (see, however, Trepavlov 1993: 31ff.). The culture, however, was quite indigenous. The sedentary contemporaries of the nomads, and their distant descendants, might consider them barbarians, but they were rather sophisticate ‘barbarians’. To prove this one may refer to several concepts and practices which for many centuries have been widespread in the Eurasian steppes. They included: the notion of the divine mandate to rule bestowed upon a chosen clan, or even of the divine origin of this clan (Golden 1982, 2001)5, and translatio imperii – the possibility of transfer of this mandate and, correspondingly, of the legitimate supreme authority from one polity to another; the notion of charisma – the Iranian farnah, the Turkic qut, the heavenly ordained good fortune and the aura connected with this fortune (Bombaci 1956, 1966; Frye 1989; Gnoli 1990); a quite developed system of imperial (royal), noble, and administrative titles6; imperial symbolism, including color; elaborate status and rank traditions and practices associated with crowning, dressing, belting, robbing and headdress; special investiture ceremonies (Allsen 1997: 85ff.); refugia, sacred territories and cult centers; the notion of collective or joint sovereignty, according to which a state and its populace belong not to an individual ruler but to all members of the ruling clan or extended family as their corporate property, and a corresponding appanage (ulus) system; specific succession patterns based on different variations of the collateral or scaled rotating system and seniority within a ruling clan (Fletcher 1979–1980); diets or convocations composed of members of the ruling clan, nobles, and worthies, such as the Mongol quriltais; a partial overlapping of administrative systems with the military organization (bipartite or tripartite organization of polities, left-right military-political division, decimal systems); a patrimonial mode of governance that implied a redistribution of various kinds of wealth and goods among vassals, followers, and even commoners; and several other concepts and institutions.
Such was the state of affairs in the Eurasian steppes for approximately two and a half thousand years. Everything has changed only since the onset of modern times. The ‘European miracle’, the transition to civilization based on technological innovations, gradually began to influence the development of the sedentary countries of Asia. The nomads, however, remained the same. The great geographic discoveries and improvements in seafaring sharply diminished the importance of transcontinental overland trade, as well as the role of nomads as intermediaries in this trade (Steensgaard 1973; Rossabi 1989). In Eurasia, caravels, and later steamboats, defeated caravans. The centralized colonial empires of Russia, Ottoman Turkey, and China created massive regular armies. They were increasingly employing firearms with ever-growing lethal power (Headrick 1981). Against such armies, the irregular cavalries of the nomads were ineffective. Bows and spears were as toys compared to guns and cannons.
The consequences soon followed. Nomads were losing their independence and had to adjust to new situations beyond their control. Their growing dependence on colonial powers, and later on national governments, indeed on the outside world in general, all of which remained alien to the pastoral nomads, had many detrimental effects. It decreased their territories, disturbed their migratory routes, overstressed their subsistence-oriented economies, and undermined their sociopolitical organization, ideology, and political culture. As a result, traditional pastoralism in the Eurasian steppes, just as everywhere else on earth, cease to exist (Naumkin, Shapiro, and Khazanov 1997; Naumkin, Tomas, Khazanov, and Shapiro 1999; Humphrey and Sneath 1996, 1999; Khazanov 1998; Sneath 2000).
Still, climate and environment are not subject to even our post-industrial civilization. It is worth keeping in mind that pastoralism was originally developed as an alternative to cultivation in the very regions where the latter was impossible or economically less profitable. In many of these areas the situation remains basically the same. In several arid ecological zones, mobile pastoralism, if sufficiently modernized, may retain some advantages in comparison with other forms of agricultural activity. Soviet communists, and to a lesser degree their Mongol vassals, have already tried to modernize it, but in their own characteristic fashion – i.e., by the worst and most inefficient means possible. The results of this are well known. Now the trauma of the past should be overcome. Now a great deal, if not everything, must be rebuilt from the bottom up. The time has come to perceive that not only the politics but also the economics of development are the art of the possible.
Modernization is a beneficial but cruel process. It has its own winners and losers, but it does not allow anyone to simply sit on the fence. So far, the post-communist period has not been marked by significant achievements in the (re)modernization of mobile pastoralism. On the contrary, at present many tendencies can best be characterized as anti-modern. At the same time, the traditional pastoralist way of life was destroyed in most countries of the region, and many characteristic features of the traditional pastoral culture were probably irreversibly lost. Not ecological and economic factors, and not modernization per se, but abusive, corrupt, mismanaging and exploitative state powers, since the nineteenth century almost always alien to the pastoralists and detrimental to their interests, have ruined extensive and mobile pastoralism in a region where it thrived for millennia, all without replacing it with any viable modern type.
At least since the Bronze Age onward, people have tried to predict the future, but for better or for worse (personally, I think for the better), they have never succeeded. It is impossible to know what will happen to mobile pastoralism in the twenty-first century, but at present there appears to be little cause for optimism.