CC BY
92
Mongols in World-Systems History1
Thomas D. Hall
DePauw University, Greencastle
ABSTRACT
This paper is a summary of my thinking about the roles of Mongols in world-systems history, and draws heavily on my work with Christopher Chase-Dunn (1997), and summarizes some later work. In the interests of brevity I present as assertions a number of claims that are argued theoretically and empirically elsewhere in considerable depth. I further note that my primary interest in this work is not Mongols, per se, but understanding how world-systems form, transform, merge, contract, are absorbed by other world systems, and how they incorporate other peoples, territories, and world-systems. My fundamental claim is that the Mongols, and steppe pastoralists in general, have played pivotal roles in the growth of world-systems ringing the Central Asian steppes. Furthermore, examination and study of these roles offers a special opportunity to understand how world-systems change, and to see the often vital, if invisible, roles nonstate peoples have played throughout human history in processes of social change.
I begin my account with brief a recapitulation of world-system analysis, emphasizing recent attempts to stretch the initial paradigm developed by Immanuel Wallerstein to precapitalist, that is to say, pre ca. 1500 CE times. I then turn to a summary of the analyses of the roles of Central Asian nomads, concentrating on the Mongols, in the world-system history of Afroeurasia. I will conclude with some speculations about where further research might take us, and a discussion of some of the pressing theoretical and empirical issues.
Social Evolution & History, Vol. 4 No. 2, September 2005 89-118 © 2005 'Uchitel' Publishing House
WORLD-SYSTEMS ANALYSIS, A BRIEF SUMMARY
I refer to world-system analysis as a perspective or paradigm in Thomas Kuhn's sense rather than as a theory (1970, 1977). Briefly, a paradigm is more general than a theory. It is a set of assumptions that guide questions and the development of many related, yet competing theories. Mistaking world-systems analysis for a theory, rather than a paradigm has led many scholars to assume that Immanuel Wallerstein's early works (1974a, 1974b) encompass the whole of the 'theory'. While world-systems thinking has moved far beyond Wallerstein's original formulation, most of the basic assumptions derive from his early work2. The use of 'world' in world-systems has become somewhat confusing in recent decades. It is often seen as a synonym for 'global'. Rather, if refers to a self-contained, somewhat coherent internally, unit of social organization: 'my "world-system" is not a system "in the world" or "of the world". It is a system "that is a world"' (Wallerstein 1993b: 294).
Thus, a world-system is the fundamental unit of analysis within which all other social processes and structures should be analyzed. This, however, is not a claim that world-systems analysis supplants and replaces all other social analyses. Rather is a claim of a necessary, but far from sufficient, requirement for any social analysis. Thus, all social processes occur at a specific time and place and position within the world-system. That context is a crucial part of understanding any social process.
For Wallerstein there are three fundamental types of world-systems: world-economies, world-empires, and mini-systems. A world-economy consists of states that trade, compete, and make war with each other. A world-empire is the result of one state politically uniting the others, typically through conquest. World-economies and world-empires often are different historical phases of the same overall system. A world-empire may fragment due to any number of factors, such as a crisis of succession or an ecological crisis. Typically, another state, often a semiperipheral marcher state (Chase-Dunn and Hall 1997: Ch. 5), will conquer the others and build a new world-empire. Wallerstein does not discuss minisystems in detail. He also argues that the modern world-system is unique because it is based on capitalism, has not become a world-empire, and is the first such system to become truly global.
This 'modern world-system' arose in western Europe during 'the long sixteenth century' (1450-1640 C.E.). Early capitalist merchants needed labor, raw materials, and markets. These needs fueled the expansion of trade networks and colonization of many areas of the world. Expansion was continual, but cyclical, a fundamental quality of the world-system. The modern world-system had a division of labor consisting of: (1) core states that employ advanced industrial production and distribution systems, have strong states, a strong bourgeoisie, and a large working class; (2) peripheral areas and states that specialize in raw materials production and have weak states, a small bourgeoisie, and many peasants; and (3) semiperipheral states that are intermediate between core and periphery, economically, socially, and politically, and whose social structures are intermediate between, or a mixture of, core and peripheral states.
Core capitalists use coercion to force peripheral producers to accept lower prices and lower wages. Such unequal exchange promotes core development and peripheral impoverishment simultaneously. The spatial implications of this tripartite division are not always obvious and remain problematic. Further, these are 'ideal typical' descriptions of relative relationships. What was high tech in one era - textiles in the late eighteenth or early nineteenth century - can be low tech in another - textiles in the late twentieth century.
World-systems analyses have a dual research agenda: (1) how do system processes shape the internal dynamics and social structures of its components [states, regions, peoples]?; and (2) how do changes within components produce change in the system? World-systems analysis is often criticized for being overly determinative, and ignoring human agency. While sometimes justified, this critique is overstated. In part this is because analyses of the second type are reported in books and rely on thick historical description; whereas briefer articles, such as this one, present summaries that often read like the first type of analysis. Indeed, one of the promises of further study of the roles in of steppe pastoralists, including the Mongols, is to provide richer empirical and theoretical insights into world-systemic processes.
Brief structural accounts tend to underemphasize the dynamic aspects of world-systems. The dynamic dialectics between local and global is the heart of world-systems analysis. Systems exhibit
several trends with embedded cycles; producing a spiral of change (Boswell and Chase-Dunn 2000). The trends include: commodifi-cation, proletarianization, state-formation, increasing size of enterprises, and capital intensification. Two major cyclical processes are the Kondratieff wave and the hegemonic cycle.
The Kondratieff cycle (K-wave) is approximately 50 year cycles in prices. The upward part is called the A-phase, the downward part the B-phase. K-waves are difficult to date precisely because they must be measured indirectly (Grimes 2000). Basically development of a new technology allows economic expansion. However, the market saturates, competition increases, and expansion slows, until a new or renewed technology starts a new cycle.
Hegemony, in a non-Gramscian sense, is a condition in which one state in the core dominates the world-system through its sheer economic and political power. When a hegemon's power peaks, and hegemony is lost or abates, the core experiences more intense inter-state rivalry. Hegemons often achieve power through a war that involves all or most of the system. The combination of K-waves and the hegemonic cycle promote cycles of colonization, decolonization, war, state-formation, and social movements. Yet, they do not cause these other cycles, but rather create conditions that are more, or less, conducive to them.
FROM WORLD-SYSTEM TO WORLD-SYSTEMS
Initially, anthropological and archaeological research and world-systems analysis might seem to have little relevance to each other. Yet if we are to understand how the 'modern world-system' arose and to avoid reading contemporary processes into a false universal history we must critically examine its antecedents. Simultaneously archaeologists saw some potential in world-system analysis for understanding regional systems. While world-system analysis seemed to offer a way to integrate local studies with larger processes, it also seemed too programmatic, too structural, and gave too little attention to agency. These early attempts led some world-systems analysts to reexamine their basic assumptions and transform many of them into empirical questions (Hall and Chase-Dunn 1993). As an example, how many semiperipheral layers exist between core and periphery, and what are their roles?
Other issues are the roles of exchanges of various types in world-system dynamics and evolution. Jane Schneider (1977) questioned Wallerstein's emphasis on bulk goods exchange and his neglect of trade in luxury goods. Several writers analyzed how leaders used the exchange of luxury or prestige goods to enhance their standing and to consolidate political power (Peregrine 1992, 1995, 2000; Peregrine and Feinman 1996; Kardulias 1999). These writers produced studies that examined the roles of exchanges in luxury goods, military alliances, and ideas in the operation of world-systems.
Christopher Chase-Dunn and Thomas Hall (1991, 1997, 2000) pursued this line and argued that the extension of world-systems analysis to precapitalist settings requires that many of its assumptions must be transformed into empirical questions. They argued that there have been four broad types of world-sytems: kin-ordered, tributary, capitalist, and a possible future socialist3. Each type has many variations or subtypes. Still, each has a dominant mode of accumulation, (a more-or-less routinized way of accumulating wealth or capital), as opposed to how it is produced (mode of production).
Kin-ordered systems were the earliest, inchoate form of world-systems. They consisted of small, stateless groups of sedentary foragers. These systems had very little differentiation or hierarchy. Because they are stateless, their politics must be reconceptualized to include the politics of kinship, marriage, and gender relations. While such systems are far removed from the 'modern world-system', 'kin-ordered systems' are the base from which all subsequent world-systems evolved.
Some seven thousand years ago, chiefdom based world-systems started to develop. These systems had sharper hierarchy and some degree of differentiation into core and periphery. The tensions and dynamics of these systems (Hall 2001) gave rise to the first states and tributary world-systems which emerged some five thousand years ago. Very quickly tributary systems dominated the globe. They incorporated kin-ordered and chiefdom based systems until the Dutch developed the first state capitalist elite in seventeenth century. Chase-Dunn and Hall (1997) argue that this marked the emergence of the modern, capitalist world-system, not the long sixteenth century as Wallerstein argues4.
Studies of precapitalist world-systems have led to several working hypotheses. First, one mode of accumulation may encompass more than one type of mode of production.
Tributary world-systems often include some kin-ordered subsections, typically but not exclusively, in peripheral areas. They may contain pockets where capitalist relations exist. However, most wealth is wealth amassed through tribute paid to a central ruler. In this distinction Chase-Dunn and Hall differ with most other writers (Denemark et al. 2000; Denemark 2000; Thompson 2000). Second, in addition to bulk goods trade networks, world-systems often have networks of political/military exchanges, luxury or prestige goods exchanges, and information exchanges. For Chase-Dunn and Hall, information includes all sorts of nonmaterial, cultural content. Each network can define a set of limits or boundaries for a world-system. Each network delimits a successively larger system. The four seldom coincide, except on small islands or in the modern world-system. Relationships among these networks through time are far from clear.
Third, all world-systems 'pulsate', 'that is, expand and contract, or expand rapidly then more slowly. Pulsation is why sporadic and cyclical, not linear, world-system expansions occur. All kinds of world-systems pulsate, thus pulsations can not originate in a specific mode of production or mode of accumulation. Rather, such cycles are prima facie evidence of a system' (Straussfogel 1998, 2000). Fourth, Afroeurasia (in conventional terms Asia, Europe, northern Africa) has been linked, at least at the information and luxury goods exchange levels for at least two and half millennia. Hence, events and processes in Europe cannot be explained solely by European processes. This makes all the more puzzling why empire size and city size distributions at the western and eastern ends of Afroeurasia have been linked for at least two millennia (Teggart 1918, 1925, 1939; Chase-Dunn, Manning and Hall 2000). Turchin and Hall (2003) suggest that ecological cycles and other mechanisms may account for this synchronization of cycles across systems, even systems isolated by long distances.
Chase-Dunn and Hall's (1997, 2000) analysis differs from the rest in other ways. First, they assert that there have been many world-systems, in four broad categories, each with many subtypes. Second, they argue that there are distinct, complex processes driv-
ing world-system evolution (Chase-Dunn and Hall 2000)5. Third, the semiperiphery is a major locus of change. Fourth, while they disagree with Frank and with Wallerstein about the historical depth of the 'modern world-system', they recognize that its historical roots go back five millennia. Finally, Chase-Dunn and Hall see the origin of the state as part of world-system evolution, rather than its starting point, as most other analysts do after states have already been formed.
Concern with this latter issue is why they draw a theoretical and empirical distinction between core-periphery differentiation and core-periphery hierarchy. They define core-periphery differentiation as 'societies at different levels of complexity and population density in interaction with each other'; whereas core-periphery hierarchy is 'intersocietal domination or exploitation' (1997: 36, 272). This conceptualization makes it easier to investigate (empirically and theoretically), first how societies become differentiated, and second, how and when differences in social organization and interactions give rise to hierarchical relations, rather than assuming that they always do so.
INCORPORATION AND MEGERS
States, civilizations, and world-systems have incessantly encountered, confronted, warred with, conquered, destroyed, and occasionally been destroyed by various non-state groups. States, have often tried to absorb, or incorporate non-state or indigenous peoples in many ways. Incorporation produces profound effects even when it limited in degree. Incorporation is a two-way, interactive process that ranges from mild to extreme (Hall 1989a). Labeling this entire range 'incorporation' masks important variations and makes it difficult to understand the wide variety of consequences for and reactions to incorporation that occur on the frontiers of world-systems (Hall 2000). Some changes resulting from incorporation are reversible, others are not.
Incorporation creates many kinds of frontier zones (Hall 2000). Many of the Native American groups we know today were built from an aboriginal base of loosely connected living groups during the process of incorporation (Hall 1989b). Indigenous resistance to expanding world-systems, empires, or states, has been universal. Consequently, many putative evolutionary sequences are suspect
because they are products of extensive interactions with states and/or world-systems. Because such effects are ancient, it is clear that not all deleterious effects are due to capitalism, but due to states. Often, too, histories of these encounters have reflected the worldview of expanding states, which takes as axiomatic that states are inherently superior to non-state societies (Wolf 1982).
Occasionally incorporation pushes societies toward a more complex form of organization: bands become tribes, tribes become chiefdoms, and chiefdoms become states. This kind of reactive evolution, as result of incorporation can backfire on the incorporating state or world-system. The Mongol conquest of China demonstrates how forced centralization can prove harmful to the state that caused it.
The Mongol example also illustrates how incorporation-induced centralization can lead to transformation of the incorporated group, in this case into a secondary state, and even an empire can come into being. Often these secondary states are inherently unstable, as was generally the case for steppe confederations. They usually continued only as long as trade goods were delivered from the sedentary core state. Even the Mongols remained dependent on Chinese goods. As the Chinese said, 'an empire can be conquered from horseback, but cannot be ruled from there'. The steppe empires could only persist by extracting surplus from the core regions.
Studies of state - non-state relations in precapitalist settings form a basis for comparing the interactions of non-state with states in the modern world-system with such interactions in tributary world-systems. These comparisons help us understand what is unique about these processes in the modern world-system. One lesson is that oppression and cultural domination are not unique to capitalism or Europeans, but are common whenever and wherever states interact with non-state societies. Yet, the types, intensities, and consequences of such interactions vary considerably between tributary, early capitalist, and late capitalist world-systems (Hall 1998a). Furthermore, in ancient or precapitalist world-systems zones of incorporation or frontiers, occur along all four types of boundaries: bulk goods, political-military, prestige goods, and information. Too little work has been done in comparing these to draw any sorts of conclusions.
A closely related problem is the merger of world-systems. What is distinctive about merger situations is that one world-system in not truly absorbing or incorporating another, but they are joining together, that is, merging. One would expect that such mergers would begin along the widest boundaries, information and prestige goods, later to include political-military networks, and finally to include bulk goods networks. Indeed, that is what Chase-Dunn and Hall (1997) argue in their reconstruction of more than two millennia of Afroeurasian world-system history.
AFROEURASIA AND STEPPE PASTORALISTS 6
Chase-Dunn and Hall argue that there are at least three world-systems in Afroeurasia: one centered in west Asia, growing out of the first states in Mesopotamia, then Egypt, then Rome; a second set of world-systems and states that emerged in south Asia, which have been poorly studied from a world-system perspective; and a third Chinese world-system. The agents that made these unifications possible were a series of steppe confederations that facilitated, even while preying on, trade across Asia.
In Rome and China (1939) Frederick J. Teggart argued that the warfare on the eastern borders of the Roman empire and the western border of China were correlated. He further argued that the mechanism of correlation was the movements of various Central Asian steppe peoples, yet he did not specify the mechanisms of this unification.
In general, Central Asian pastoral groups have a segmentary lineage structure (Barfield 1989, 1993; Sahlins 1961). The salient feature of segmentary lineage structure is that it allows easy formation of alliances, and facilitates easy conversion of a former enemy to an ally, often rationalized via a kinship metaphor. This was one way steppe leaders could manipulate the flow of prestige goods economy to promote wider alliances, and thus amassing larger armies.
The Cimmerians and Scythians who may well have been the first mounted pastoralists, appeared sometime in the ninth century B.C.E. Chinese accounts of horse-riding nomads show up sometime in the fourth century B.C.E. The development of the Hsiung-nu, the first large steppe confederacy, coincided with formation of the Ch'in dynasty in China in the late third century B.C.E. This was
not an accident, but rather the first cycle of the simultaneous rise and fall of Chinese dynasties and steppe confederations as Barfield shows (1989: 13; 1991: 23).
Pastoral - sedentary relations exhibit several recurrent themes in Chinese history. First, gradual expansion of sedentary agriculture displaced nomads further into the steppe. However, a combination of agricultural technology and local ecology limited this expansion. Along the border zone agriculturists sometimes adopted a pastoral ways; at other times pastoralists became sedentary. This created a mixed zone along some parts of the Chinese frontier. Where the vegetative zones were sharp, so was the frontier. In short, the pastoral - sedentary distinction was fundamentally one of livelihood, not race or ethnicity, although it was often expressed that way. This is an early example of the persistence of a cultural boundary despite continual movement of individuals, families, or groups across it (Barth 1969). Second, geographic and political factors drove waves of migrations or invasions from east to west across the Eurasian land mass. There was a gradient in temperature and precipitation from east to west over the Eurasia steppe with the west having better pasture conditions (McNeill 1982: 17; 1987: 265ff, 323). However, there is a counter attraction in the east: the Chinese empire.
Third, because nomads produced little of interest or value to agricultural Chinese, they often used threats to induce trade: raiding and trading were different means to the same ends (Jagchid and Symons 1989: Ch. 1). Khazanov (1983: 202ff) points out that there was a distinct asymmetry in the demand for this trade. Nomads, due their specialization, had a much stronger need for sedentary goods - both agricultural and handicrafts - than sedentary peoples had for nomad products. The strongest demand for exchanges with nomads came from a second type of trade, one in which nomads played an intermediary role among sedentary peoples. Still, this asymmetry in demand for trade explains why nomads often forced the issue via raiding.
Christopher Beckwith (1991) documents an important contrary example. Uighurs supplied horses, vital to internal commerce and regulation, to T'ang China (618-906). Thus, at least for early middle ages some pastoralists did supply important, even vital, goods. In this case the asymmetry favored pastoralists. Di Cosmo (1994)
also seems to argue against this asymmetry on the basis of Hsiung-nu relations, noting that nomads occasionally mixed farming with pastoralism. While Di Cosmo does add several fine nuances to the analysis of pastoral-sedentary relations, nothing he reports vitiates Barfield's analysis. Beckwith's account, however, seems to support both Barfield's analysis and to strengthen the claim that pastoral confederacies at times took on semiperipheral role in trade. Clearly, more data on such trade is needed and could require rethinking of this analysis.
Eventually, the Chinese developed a cavalry to fight nomads. Raiding correlated with changing conditions of trade and changing state stability (Szynkiewicz 1989: 154; Barfield 1989). Chinese officials saw this trade as a way to control nomads. Pastoral leaders used prestige goods in their political economy to shore up and symbolize their power. These interactions drove changes in China and among pastoralists. When the state was in decline, pastoral leaders sometimes helped protect beleaguered areas. When the state was ascendant, unified Chinese response promoted wider unity among pastoralists. Thus, steppe confederacies were as often a source of change as receivers.
Barfield (1989) documents the intimate connection between Chinese empire and steppe political organization. He draws a clear distinction between inner and outer frontier strategies. The outer frontier strategy consisted of a dominant steppe leader using violence to terrify Chinese officials, alternating between war and peace to raise tribute payments (which Chinese usually called 'gifts') and improve terms of trade. The leader would assiduously avoid taking over Chinese territory thereby avoiding the entanglements in Chinese politics that conquest would entail.
The inner frontier strategy developed when a pastoralist confederation began to disintegrate. One or another pastoral faction leader, typically of a weaker faction, would seek alliance with some Chinese official against his steppe rivals. These Chinese officials acquiesced, intending to use 'barbarians against barbarians'7. The pastoral faction leader sometimes allied with some part of the Chinese military to aid in the defeat of his rivals. He would also try to use his good standing with Chinese leaders to sever the flow of Chinese goods to his rivals. This near-monopoly of access to Chinese goods became a potent tool to gain supporters. This worked
especially well in a segmentary lineage system in which exotic goods were used enhance the status and reputation of leaders, i.e., in a prestige goods economy. Once such a leader gained ascendance he could either use his power to unify pastoral groups and return to an outer frontier strategy or leave them politically fragmented and consolidate domination in a limited region.
The cycling between inner and outer frontier strategies was the mechanism that synchronized strong steppe polities with strong Chinese empires and fragmentation of pastoralist confederacies with fragmentation of the Chinese empire. Only when the empire was strong could it be 'milked' continuously via an outer frontier strategy. When the empire was weak, pastoral leaders tended to favor an inner frontier strategy, making alliances with local 'war lords'.
This cycling explains why this 'perilous frontier' (Barfield 1989) remained relatively permanent. Pastoralists could not rule a sedentary population without giving up their pastoralism. Yet, sedentary states could not conquer pastoralists - except by sedenta-rizing them. They could try to control them by a combination of barriers and highly mobile troops, who could essentially beat the pastoralists at their own game (Lattimore 1962a: 485). Thus, Central Asian pastoralists, especially the Mongols, could build huge empires, but could not maintain them. Conversely, the Chinese could manipulate, but never conquer, their pastoralist adversaries. The sinicization of pastoralists has been noted often, but the 'Mon-golization' of frontier Chinese has been noted rarely (except by Lattimore 1940, 1962a).
The Afroeurasian-wide effects of the cycling between inner and outer frontier strategies rippled along the steppe gradient, and sometimes caused pastoralists to move west. Whenever the Chinese world-system became multicentric the associated steppe confederacies would also fragment. Conflict among pastoral groups was often settled by migration of the weaker group into new territory. Because a confederacy allied with the Chinese empire was typically larger and stronger than its rivals located further out on the steppe, it could conquer or displace them. Hence, a net displacement westward was common. Thus, west Asia was subjected to repeated invasions by the losers in a long chain of displacements. This is one factor in the synchronization of the rise and fall of empires and the growth and decline of cities in the East Asian
and West Asian world-systems (Chase-Dunn, Manning, and Hall 2000)8. This is not an argument that raids by pastoralists were the main cause of collapses of East and West Asian empires. Imperial collapses were to a large extent due to overextension and diminishing returns to scale of the costs of empire (Tainter 1988; Hopkins 1978a). Still, these dynamics initiated migrations and intensified nomad raiding that accelerated imperial collapse.
This analysis of steppe confederations has implications for debates about the role of class relations in state formation. In traditional Marxist accounts, states only arise after classes are formed. While there were differences in wealth among Central Asian nomads, to call these classes is an overstatement. Furthermore, there was the recurrent problem that when these 'states' collapsed, the nomads reverted to 'tribal' organization. The problem here is the assumption that state formation is entirely endogenous. Steppe confederations were only purely 'tributary' states in the sense that the surplus that they extracted was from other states, and very little from their own members. The formation of steppe confederacies can only be explained by taking world-system relations into account. This was an unusual case in which the less complex and less hierarchical periphery or semiperiphery (in terms of core-periphery differentiation) exploited a more complex core region. This example illustrates why Chase-Dunn and Hall (1997) insist that the relationship between core-periphery differentiation and core-periphery hierarchy needs to be studied empirically, rather than assumed on theoretical grounds.
Rather than peripheral, steppe confederacies can be understood as semiperipheral. If the Chinese empire is the core, remote pas-toralists the periphery, and nearer steppe confederations are the semiperiphery, then this situation presents an intriguing world-system puzzle. Here the semiperiphery grows because of unequal exchange in its favor with the core and uses its power to block the formation of rival semiperipheries. Still, it remains dependent on the core in the sense that this relationship is contingent on core prosperity.
The seeming contradiction of the semiperiphery exploiting the core is resolved partially by the consideration of relative sizes. Chinese states were one hundred or more times more populous than steppe confederacies. Hence what they gave to steppe pas-
toralists was relatively minor compared to other exchanges, but was vital to the nomads. That is why the same exchange was seen by steppe pastoralists as a matter of 'tribute', while Chinese always thought of it as 'gifts' or 'bribes' (Jagchid and Symons 1989: Chs 2 and 4).
At the other end of Afroeurasia, the non-state peoples that Rome faced were more varied and presented different problems (Dyson 1985). Along the eastern and southern (that is, North African) limits the non-state peoples were broadly similar, more localize transhumant pastoralists. On the European borders peoples varied from shifting cultivators to complex chiefdoms (Wells 1999). Rome exerted considerable, if uneven, pressure toward political centralization on its non-state neighbors. At times for Rome, too, such pressure led to undesirable events, as when Arminius defeated several Roman Legions in the battle of Teutoberg Forest in 9 C.E. (Wells 2003).
Unlike steppe pastoralists, west Asian and north African pastoralists had kinship structures that were less amenable to expansion by 'fictive' kin incorporation. This may have reflected important geographical differences (Barfield 1990). These nomads shifted more readily and more often between sedentary and nomadic life styles (Cribb 1991). This may have been because these pastoralists, especially where they occupied highly variegated terrain, were transhumant. Consequently, they often had a symbiotic relationship with the sedentary agriculturalists in the core regions (Barfield 1993: 94).
Finally, Roman strategy toward non-state societies differed from that of the Chinese, and that of the Parthians, Seleucids, Sas-sanians, and other West Asian states. Continuing its early federation strategy, Rome often sought to enlist erstwhile non-state foes as allies. Their strategy alternated between territorial or direct and hegemonic or indirect control.
Mattingly (1992) notes that Roman conflicts with North African pastoralists, contrary to Edward Luttwak's (1976) famous analysis of Roman strategy, did not always try to keep nomads out, even in the late empire: 'Roman frontiers of whatever type (walls, earthworks, rivers, mountains, deserts, or road and fort networks) were not intended to be lines of blockade or first defenses against invading forces. Rather they were filters, designed to facilitate ob-
servation and supervision of movement between the territorial and hegemonic zones' (Mattingly 1992: 56). That is, Rome seduced pastoralists into an inner frontier strategy and blocked any shift to an outer frontier strategy by a judicious combination of tribute, alliances, and divide-and-conquer techniques (Mattingly 1992: 54). These techniques were aided by the geographic and kinship differences already noted.
Thus, Rome seldom faced concerted pastoralist confederations of the type that challenged China. This did not eliminate threats and impacts of pastoralists on Rome or other empires in west Asia. The Central Asian cycle of pastoral confederacies caused the syn-chronicity of border warfare in China and Rome, drove the unification and dissolution of the Afroeurasian-wide prestige goods and information networks.
THE MONGOLS
Much has been written about the Mongols and Chinggis Khan. Here I will only stress the insights that world-systems analysis can add to the discussion and not recapitulate their rich and complex history. The Mongol conquest was one of the most important processes in the world-system history of Afroeurasia. Barfield says: 'The exceptional nature of the Mongol Empire has been largely misunderstood because, as the most powerful nomadic state that ever existed, it was presumed to be the culmination of political evolution on the steppe rather than the exception that it was' (Barfield 1991: 48). It was exceptional in several ways: in the process of Chinggis's ascendancy, in the Jurchen reaction him, in the scope of the conquest, and in the building of city-based states, including the Yuan dynasty.
Chinggis rose from a marginal position, and was often opposed by his own relatives. Thus, he did not rely on kinship for organization of followers, but on loyalty and autocratic control. He drew a multi-tribal elite from friends and retainers:
Chinggis's political organization was not, therefore, the culmination of a long evolving steppe tradition because it rejected the imperial confederacy model. Instead, the Mongol state was based on the principles of centralized administration, the destruction of tribal patterns of leadership, and a rigid discipline to a degree previously unknown among nomads. It was a unique creation.
After the fall of the Mongol Empire the nomads reverted to the older and more traditional imperial confederacy model of organization (Barfield 1991: 49).
This new style state confronted the Jurchen semiperipheral marcher state in Manchuria that refused to follow the time-honored Chinese path of appeasement. Chinggis overplayed his hand in pursuing the outer frontier strategy of destructive raids to inspire appeasement, and somewhat reluctantly conquered north China. When combined with Chinggis's low tolerance for resistance, this led to widespread destruction of cities and agricultural infrastructure. It was only with Khubilai that the Mongols took up the responsibility of ruling northern China rather than despoiling it.
The Mongol Empire united most of Eurasia into a single empire for the first time. Yet the merger was far from complete. The Mamlukes prevented military incorporation of Egypt and North Africa. The inhospitality of forest zones to mounted archers (Lindner 1981, 1983) in combination with the success of knights in the forest zone (McNeill 1963, 1964, 1982) kept Western Europe outside the Mongol Empire.
The explanation of Mongol success lies in factors and processes occurring at different levels simultaneously (Saunders 1971; Morgan 1986; Lindner 1981, 1983; Barfield 1989). First, western Asian states were weak compared to China and vulnerable to conquest. Mongols were unaware that their ecological adaptations were more fragile than was so for China where destroyed towns were often rebuilt and repopulated. However, in southwest Asia such destruction often became permanent, especially when they destroyed irrigation systems. Because Mongols did not often bother with local administration, frequently no one remained who was capable of rebuilding ruined tows and irrigation systems (Barfield 1989: 201-202).
Second, a client relationship with the Mongols was an attractive 'bargain' under conditions of incessant warfare. Polities that resisted Mongol rule or repudiated treaties (Chin China, western Turkestan, and the Tangut kingdom) were often destroyed.
Polities that acquiesced to Mongol rule (in Manchuria, Korea, Uighar oases) survived and often kept their own leaders.
Chinggis was particularly intolerant of disloyalty. Punitive wars 'were so devastating that they led to the overthrow of the
ruling dynasties and, by default, their direct incorporation into the Mongol Empire' (Barfield 1989: 200). This was something new for Central Asian pastoral conquests. In the west they were forced to incorporate and administer states they did not destroy. In the east their own vigor ultimately trapped them into founding a new dynasty in China.
The persistence of pastoral groups on the steppe was a third factor in Mongol success. These pastoralists maintained a flexible kin-ordered social structure that allowed them to recruit other nomads, and even occasionally sedentary groups, into a larger and larger machine for conquest. Chinggis built a command structure that would not easily segment along tribal lines by choosing leaders for loyalty and performance, rather than for kinship.
The presence of several leaders who astutely balanced the drive for conquest and plunder with the needs of administration was a fourth factor. Mongke (Khubilai's brother) implemented administrative innovations that kept conservatives occupied in (successful) battles, which gave him a free hand in the center (Allsen 1987).
This is not a reversion to a 'great man' theory of history. Rather, it is recognition that Mongol leadership was form of big man leadership (Sahlins 1961, 1963, 1968) that is especially sensitive to the competence of a leader. Several Mongol leaders maintained a sufficient volume of plunder and tribute to insure the loyalty of tribes that might otherwise have left the confederation. That is, internally they perfected the outer frontier strategy of 'milking' sedentary states, even as externally, they overplayed it. Furthermore, Mongol leaders had extensive experience and knowledge of how states operated and no doubt used that knowledge to selectively adopt and adapt state techniques for their own administration.
A fifth, and key, factor in Mongol success was superior logistic ability in military communication, transportation, and movement. The pastoral way of life is the root of this ability: ready availability of horses, intimate knowledge of geography, and ability to move their entire means of production (families and herds) with them.
Through these advantages all the great Khans capitalized on the inclusive kinship structure and permeability of group boundaries to build large confederacies. Conquered groups had the option of changing allegiance rather than die in a futile battle. While this worked well with pastoralists, it did not work as well with seden-
tary peoples. The problem of revenue was alleviated by continued expansion and by a constant inflow of booty. This continual flow helped to minimize factional rivalry and quiet objections to changes that came with empire. Superior mobility and communications made the formation of large armies relatively easy. Still, these advantages were temporary, because they were inherently unstable.
Technological and political problems were the source of this instability. Orderly succession of rulers was a political problem. Personal skills, not the least of which are alliance-building and warfare, are basis of big man power. Thus, fighting was a vital part of succession processes. Principles of lateral and lineal succession within Mongol society generated and exacerbated internal com-petion and conflict9. This is why succession by arms became inevitable due to the lack of clear priorities for succession. Yet, in-stitutionalization of succession would have undermined the very basis of leadership. The problem is not simply that Mongols failed to institutionalize political succession [as Eisenstadt (1963) argued], but that they could not do so and remain Mongols. Similar problems were inherent in political control and revenue garnering.
Superiority in communication and mobility contributed to instability because these capabilities were integral to pastoral life and hence skills of all pastoral groups and leaders. Thus it was nearly impossible for any one leader to monopolize control of strategic resources in order to coerce compliance. Dissenters always had the option to fold their tents and leave with their herds.
This situation was rooted in the material adaptation to the volatile and uncertain steppe environment. The pastoral economy was highly adaptive culture for living in a volatile steppe environment, utilizing a flexible kinship system and a fluid form of leadership that could quickly respond to changing circumstances. This adaptation made it possible to organize mobile armies, but simultaneously imposed limits to expansion. Furthermore, it was most assuredly not well adapted to the administration of sedentary agriculture. This is why the edge of the steppe remained a permanent frontier (Lattimore 1940; Lindner 1983; McNeill 1964; Whittaker 1994).
Still, Mongol unity, despite brief existence of little more than a century, brought major changes to Afroeurasia. The Mongol conquest opened a northern connection between China and Europe
by-passing connections through what is now southern Iran and Iraq or through the Indian Ocean. The resultant steady traffic east-west across the steppes opened other circuits of trade: 'Gradually a north-south exchange of slaves and furs for the goods of civilization supplemented the east-west flow of goods that initially sustained the caravans' (Bentley 1993: 56).
Another important consequence was the transmission of the Black Death or bubonic plague to Europe and to China (McNeill 1976)10. The Black Death first swept through China in 1331 causing immense population losses, and through Europe in 1348, ultimately killing one third to one half of its population, which fundamentally changed relations between lords and peasants.
The spread of the plague may have also been an important cause of Mongol undoing (McNeill 1976: 132-175). The bacillus Pasteurella pestis probably traveled with Mongols from Manchuria to the Central Asian steppes. There it became endemic among native rodents. It then spread to both China and Europe, and probably infected many Central Asian nomads. If so, it too, would have slowed Mongol expansion. The spread to China (1331) contributed to undermining the strength of Yuan dynasty (which fell in 1368). The presence of the plague on the steppe also explains why net migration was onto the steppe by sedentary peoples, rather than movement from the steppe by pastoralists, after the collapse of the Mongol empire.
CONCLUSIONS, QUESTIONS, SPECULATIONS
The net effect of all of these events and processes, in world-system historical terms was a pulsating unification and separation of world-systems in west Asia, East Asia, and to a much lesser extent South Asia. Indeed, the relatively continuous isolation of south Asia - except critically at the level of information and the transmission of religious ideas - may explain why the synchrony of change between east and west Asia did not affect south Asia. Clearly, Central Asian steppe pastoralists played crucial roles, albeit changing, conflicting, and volatile in the merger of these world-systems at the information and prestige or luxury good levels. The Mongols, however, were the only group to achieve unification at the level of political-military interaction.
The Mongol unification also contributed to the rise of Europe in many ways. New ideas, new luxuries, new pathogens, and new enemies (or the fear of them) led to many changes in extreme west Afroeurasia (conventionally known as Europe). Not the least of these was a drive to explore and find shorter, or less contested routes, to Asia, to develop new kinds of shipping. The edge of the steppe frontier (McNeill 1964) also helped protect Europe from military conquest. All these factors helped set western Europe on a trajectory of change that gave rise to colonialism, mercantilism, the rise of capitalist states, and the industrial revolution. Clearly, the 'rise of Europe' can not be explained solely by internal factors. Those who claim to do so ignore the Afroeurasian wide processes that created a context within which features in local European social organization could have the effects that they did. As Ken Pom-eranz (2000) has argued so eloquently and forcefully, much of what happened in Europe can only make sense in a larger con-text11. What I have argued here is that Central Asian pastoralists, including especially the Mongols, were a key part of that context.
There are other conclusions, or at least speculations and questions that may be drawn from this account. First, events on the Central Asian steppe cannot be explained by reference solely to local factors. As Barfield (1989) has shown, the pastoralists can only be understood by looking in all directions, east, west, north, and south. Khazanov (1983) has long since advanced a similar argument. I would add to this that their complex, if shifting and volatile, roles within various world-systems are crucial to understanding their social structures and evolution. Steppe confederations could not exist without the presence of strong core states. World-system analysis at the minimum offers a way to sort out and study these complex relations with increased precision. But the consequences of the study of steppe pastoralists flow both ways.
Second, the study of Central Asian pastoralists suggests ways to examine both synchrony and merger of world-systems. At many times the steppe confederacies played the role of a contested periphery, and certainly with the Mongols a contested semiperiphery. A contested periphery or semiperiphery is a social group that exists between world-systems, and may simultaneously be a periphery, or more rarely a semiperiphery in each (Allen 1996; Berquist 1995; Cline 2000). Thus these contested positions become links and con-
duits for exchanges between world-systems. Yet at times they may be barriers, albeit differentially permeable with respect to what may pass through and the direction in which they pass. Thus, the study of steppe pastoralists offers abundant opportunities to study this complex, yet important role within world-systems. No doubt similar or analogous processes occurred elsewhere, but until we have more knowledge of such cases we can not make reasonable claims for either the uniqueness or generalness of these steppe processes.
Third, by using this sort of approach we can re-examine evolutionary sequences, and understand how pastoral societies change, and how those changes are shaped by their world-systemic contexts. Nikolai Kradin argues that steppe confederacies are a form of super complex chieftaincies and a form of xenocratic pastoral polities, and not really states (2002). I have treated them as a type of state. Yet there is less difference in our approaches than these differences in terminology might suggest. If states, they are proto, nascent, or inchoate states, not fully developed states. Just how and when does a chiefdom become a state remains a thorny problem. Closer examination of steppe confederacies offers an opportunity for researchers examine this crucial transition in human social evolution. A fourth conclusion or speculation grows from this: namely that the study of steppe pastoralists offers an opportunity to add more threads to a multilinear study of evolution. Indeed, future studies may force us to reconsider our understanding of the processes and sequences of human social evolution.
Finally, the entire subject is a cautionary tale about assumptions regarding the roles of class, the forms of states and chiefdoms, the roles of kinship and political succession, and many other social processes. It is often useful to re-examine basic assumptions and theoretical premises and subject them to empirically driven criticism. Cleary there is much more to be learned. It is also clear, I hope, the study of Central Asian pastoralists is not an arcane exercise, but intrinsic to our understanding of the origins of our globalizing world-system.
NOTES
1 I am indebted to my colleague Nikolai Kradin for suggesting that I summarize my thoughts on the Mongols, drawing on my work with Christopher Chase-Dunn
in our book, Rise and Demise: Comparing World-Systems (1997, Boulder, CO: Westview Press). I see this paper as a contribution to a dialog among scholars, who due to various historical problems [e.g. the 'cold war'] and differences in language have not known of each other's work. I suspect, informed in part with many email discussions with Prof. Kradin, and from his English-Language paper (2002) published in the Journal of World Systems Research, that many Russian readers may see considerable 're-inventing of the wheel' as we say in colloquial English in the U.S. In this essay I cite mostly literature available in English.
2 See Wallerstein 2004 for a brief, but elegant summary of his views. This book also contains a very useful appendix in which Wallerstein lists key works on world-systems analysis.
3 See Boswell and Chase-Dunn 2000 for an elaborate discussion of what they mean by this. Briefly, a socialist world-system, would be global, and run by strong democratic means for pursuing collective rationality, as opposed to private gain. Under this definition there has never been socialism on the modern world, at best only localized state socialism.
4 They do not disagree with Wallerstein's argument about the long sixteenth century origins of the modern world-system, and indeed, concur with Janet Abu-Lughod (1989) that it roots go back two or three centuries earlier. The disagreement is when the term capitalist becomes appropriate. For other arguments about his see Frank and Gills (1993).
5 Chase-Dunn and Hall (2000) revise somewhat the explanations they presented in 1997, and is the one that should be consulted.
6 This section draws very heavily on Chapter 8 or Rise and Demise (Chase-Dunn and Hall 1997) and Barfield (1989, 1990, 1991, 1993). Other important sources are Lattimore (1940, 1962a, 1962b, 1980), Khazanov (1983), Irons (1979), Burn-ham (1979), Morgan (1986), Seaman (1989, 1992), Seaman and Marks (1991), Szynkiewicz (1989), Jagchid and Symons (1989) and Beckwith (1987a, 1987b, 1987c, 1991). Jachid and Symons (1989, Introduction) review the debates in this literature. Hall (1991) and Frank (1992) give world-systems interpretations.
7. The phrase, 'barbarians against barbarians', is found often in translations of Chinese accounts of frontier policy. A specific citation is found in Ying-shih Yu (1967: 15). See also Kwanten (1979: 12ff) and Beckwith (1987d).
8. Shifts in climate that cut the productivity of local grazing lands could produce identical effects. In fact, following the arguments of Turchin and Hall (2003) a large variety of factors could produce such synchronization. Much more research is needed to understand this synchronization. Whatever the mechanism, however, steppe pastoralists clearly played an important role in it.
9. Lateral descent or succession refers to inheritance that passes from older to younger sons, whereas lineal descent refers to inheritance that passes from father to son. Lineal descent can be by primogeniture, that is to first (legitimately) born son or ultimogeniture that is to last (legitimately) born son. The practice of polygyny (multiple wives) complicates all these forms of descent. Barfield (1989) has many detailed accounts of how this worked, and did not work, among Central Asian nomads.
10 Recently there has been some debate as to just which disease spread. Yet whatever the pathogen, and whatever the vectors of its spread, many of these processes would have occurred. Still, detailed understanding of the processes must await certain identification of the pathogen(s) and its vectors.
11 Since first drafting this paper several important books addressing the exogenous origins of the European 'miracle' have appeared. Hobson (2004) provides extensive, detailed evidence on these external origins of European development. Good (2004) provides cogent summary of the 'great debate' about the origins of modern Europe.
REFERENCES
Abu-Lughod, J.
1989. Before European Hegemony: The World System A. D. 12501350. New York: Oxford University Press.
Allen, M.
1996. Contested Peripheries: Philistia in the Neo-Assyrian World-System. Unpublished Ph.D. dissertation, Interdepartmental archaeology program, UCLA.
Allsen, Th. T.
1987. Mongol Imperialism: The Policies of the Grand Qan Mongke in China, Russia, and the Islamic Lands, 1251-1259. University of California Press.
Barfield, Th. J.
1989. The Perilous Frontier. London: Blackwell.
1990. Tribe and State Relations: The Inner Asian Perspective. In Khoury, Ph. S., and Kostiner, J. (eds.), Tribe and State Formation in the Middle East (pp. 153-182). Berkeley: University of California Press.
1991. Inner Asia and Cycles of Power in China's Imperial Dynastic History. In Seaman, G., and Marks, D. (eds.), Rulers from the Steppe: State Formation on the Eurasian Periphery (pp. 21-62). Los Angeles, CA: Ethnographics Press, Center for Visual Anthropology, University of Southern California.
1993. The Nomadic Alternative. Englewood Cliffs: Prentice-Hall.
Beckwith, Ch. I.
1987a. The Tibetans in the Ordos and North China: Considerations on the Role of the Tibetan Empire in World History. In Beckwith, Ch. I. (ed.), Silver on Lapis: Tibetan Literary Culture and History. Vol. 20 (pp. 3-11). Bloomington, IN: The Tibet Society.
1987b. The Tibetan Empire in Central Asia. Bloomington: Indiana University Press.
1987c. Tibetan Science at the Court of the Great Khans. The Journal of the Tibet Society 7: 5-11.
1987d. The Concept of 'Barbarian' in Chinese Historiography and Western Sinology: Rehetoric and the Creation of Fourth World Nations in Inner Asia. Papers presented at the Association of Asian Studies meeting, Boston.
1991. The Impact of the Horse and Silk Trade on the Economies of T'ang China and the Uighur Empire. Journal of the Economic and Social History of the Orient 34 (2): 183-198.
Bentley, J. H.
1993. Old World Encounters: Cross-Cultural Contacts and Exchanges in Pre-Modern Times. Oxford: Oxford University Press.
Berquist, J. L.
1995. The Shifting Frontier: the Achaemenid Empire's Treatment of Western Colonies. Journal of World-Systems Research 1: 17 [ejournal: http://jwsr.ucr.edu/index.php].
Boswell, T., and Chase-Dunn, Ch.
2000. The Spiral of Capitalism and Socialism: The Decline of State Socialism and the Future of the World-System. Boulder, CO: Lynne-Rienner.
Burnham, Ph.
1979. Spatial Mobility and Political Centralization in Pastoral Societies. In la direction de l'Equipe écologie et anthropologie de sociétés pastorales (ed.), Pastoral Production and Society / Production Pastorale et Société (pp. 349-360). Cambridge: Cambridge University Press.
Chase-Dunn, Ch.
1998. Global Formation: Structures of the World-Economy, 2nd ed. Boulder, CO: Rowman and Littlefield (Originally 1989, London: Blackwell).
2001. World-Systems Theorizing. In Turner, J. H. (ed.), in Handbook of Sociological Theory (pp. 589-612). New York: Kluwer/Plenum.
Chase-Dunn, Ch., and Hall, Th. D. (eds.)
1991. Core/Periphery Relations in Precapitalist Worlds. Boulder, CO: Westview Press.
1997. Rise and Demise: Comparing World-Systems. Boulder: Westview Press.
2000. Comparing World-systems to Explain Social Evolution. In Denemark, Friedman, Gills and Modelski 2000: 86-111.
Chase-Dunn, Ch., and Manning, S.
2000. Rise and Fall: East-West Synchronicity and Indic Exceptionalism Reexamined. Social Science History 24 (4): 727-754.
Cline, E.
2000. 'Contested Peripheries' in World Systems Theory: Megiddo and Jezreel Valley as a Test Case. Journal of World-Systems Research 6 (1): 8-17. [ejournal: http://jwsr.ucr.edu/index.php]. Cribb, R.
1991. Nomads in Archaeology. Cambridge: Cambridge University Press.
Denemark, R. A.
2000. Cumulation and Direction in World System History. In Denemark, Friedman, Gills and Modelski 2000: 299-312.
Denemark, R. A., Friedman, J., Gills, B. K., and Modelski, G. (eds.)
2000. World System History: The Social Science of Long-Term Change. London: Routledge.
di Cosmo, N.
1994. Ancient Inner Asian Nomads: Their Economic Basis and Its Significance in Chinese History. Journal of Asian Studies 53(4): 1092-1126.
Dyson, S. L.
1985. The Creation of the Roman Frontier. Princeton: Princeton University Press.
Eisenstadt, S. N.
1963. The Decline of Empires. Englewood Cliffs, N. J.: Prentice-Hall.
Frank, A. G.
1992. The Centrality of Central Asia. Amsterdam: VU University Press for Center for Asian Studies Amsterdam (CASA), Comparative Asian Studies No. 8.
Frank, A. G., and Gills, B. K. (eds.)
1993. The World System: Five Hundred Years of Five Thousand? London: Routledge.
Goody, J.
2004. Capitalism and Modernity: The Great Debate. Cambridge, UK: Polity Press.
Grimes, P.
2000. Recent Research on World-Systems. In Hall (ed.) 2000: 29-55.
Hall, Th. D.
1989a. Social Change in the Southwest, 1350-1880. Lawrence, KS: University Press of Kansas.
1989b. Is Historical Sociology of Peripheral Regions Peripheral? In Martin, M. T., and Kandal, T. R. (eds.), Studies of Development and Change in the Modern World (pp. 349-372). New York: Oxford University Press.
1991. The Role of Nomads in Core/Periphery Relations. In Chase-Dunn, Ch., and Hall, Th. D. (eds.), Core/Periphery Relations in Precapitalist Worlds (pp. 212-239). Boulder, CO: Westview Press.
1998. The Effects of Incorporation into World-Systems on Ethnic Processes: Lessons from the Ancient World for the Contemporary World. International Political Science Review 19 (3): 251-267.
2000. Frontiers, and Ethnogenesis, and World-Systems: Rethinking the Theories. In Hall (ed.) 2000: 237-270.
2001. Chiefdoms, States, Cycling, and World-Systems Evolution: A Review Essay. Journal of World-Systems Research 7 (1): 91-100 [ejour-nal: http://jwsr.ucr.edu/index.php].
2002. World-Systems Analysis and Globalization: Directions for the Twenty-First Century. In Dobratz, B. A., Buzzell, T., and Waldner, L. K. (eds.), Theoretical Directions in Political Sociology for the 21st Century, Research in Political Sociology. Vol. 11 (pp. 81-122). Oxford: Elsevier Science Ltd.
Hall, Th. D. (ed.)
2000. A World-Systems Reader: New Perspectives on Gender, Ur-banism, Cultures, Indigenous Peoples, and Ecology. Lanham, MD: Rowman & Littlefield Press.
Hall, Th. D., and Chase-Dunn, Ch.
1993. The World-Systems Perspective and Archaeology: Forward into the Past. Journal of Archaeological Research 1 (2): 121-143.
Hobson, J. M.
2004. The Eastern Origins of Western Civilisation. Cambridge: Cambridge University Press.
Irons, W.
1979. Political Stratifcation among Pastoral Nomads. In la direction de L'Equipe écologie et anthropologie de sociétés pastorales (ed.), Pastoral Production and Society / Production Pastorale et Société (pp. 361-374). Cambridge: Cambridge University Press.
Jagchid, S., and Van Jay Symons.
1989. Peace, War, and trade Along the Great Wall: Nomadic-Chinese Interaction through Two Millennia. Bloomington, IN: Indiana University Press.
Kardulias, P. N. (ed.)
1999. Leadership, Production, and Exchange: World-Systems Theory in Practice. CO: Rowman and Littlefield.
Khazanov, A. M.
1983. Nomads and the Outside World. Cambridge: Cambridge University Press. (2nd ed. 1994, Madison, WI: University of Wisconsin Press.)
Kradin, N. N.
2002. Nomadism, Evolution and World-Systems: Pastoral Societies in Theories of Historical Development. Journal of World-Systems Research 8 (3): 368-388 [ejournal: http://jwsr.ucr.edu/index.php] Kuhn, Th. S. nd
1970. The Structure of Scientific Revolutions, 2nd ed. Chicago: University of Chicago Press.
1977. Second Thoughts on Paradigms. In Kuhn, Th. S. (ed.), The Essential Tension: Selected Studies in Scientific Traditions and Change (pp. 293-319). Chicago: University of Chicago Press.
Kwanten, L.
1979. Imperial Nomads: A History of Central Asia, 500-1500. Philadelphia: University of Pennsylvania Press.
Lattimore, O.
1951. Inner Asian Frontiers, 2nd ed. Boston: Beacon Press. (Originally [1940]. New York: American Geographical Society).
1962a. The Frontier in History. In Lattimore 1962c: 469-491.
1962b. Inner Asian Frontiers: Defensive Empires and Conquest Empires. In Lattimore 1962c: 501-513.
1962c. Studies in Frontier History: Collected Papers, 1928-1958. London: Oxford University Press.
1980. The Periphery as Locus of Innovation. In Gottman, J. (ed.), Centre and Periphery: Spatial Variation in Politics (pp. 205-208). Beverly Hills: Sage.
Lindner, R. P.
1981. Nomadism, Horses and Huns. Past & Present 92: 3-19.
1983. Nomads and Ottomans in Medieval Anatolia. Vol. 144. Indiana University Uralic and Altaic Series. Bloomington, IN: Research Institute for Inner Asian Studies.
Luttwak, E. N.
1976. The Grand Strategy of the Roman Empire: From the First Century A.D. to the Third. Baltimore: Johns Hopkins University Press.
Martin, W.
1994. The World-Systems Perspective in Perspective: Assessing the Attempt to Move Beyond Nineteenth Century Eurocentric Conceptions. Review 17 (2): 145-185.
2000. Still Partners and Still Dissident After All these Years? Wallerstein, World Revolutions and the World-Systems Perspective. Journal of World-Systems Research 6 (2): 234-263. [ejournal: http://jwsr.ucr.edu/index.php]
Mattingly, D. J.
1992. War and Peace in Roman North Africa: Observations and Models of State-tribe Interaction. In Ferguson, R. B., and Whitehead, N. L. (eds.), War in the Tribal Zone (pp. 31-60). Santa Fe, New Mexico: School of American Research Press.
McNeill, W.
1963. The Rise of the West: A History of the Human Community. Chicago: University of Chicago Press.
McNeill, W. H.
1964. Europe's Steppe Frontier, 1500-1800. Chicago: University of Chicago Press.
1976. Plagues and Peoples. Garden City, New York: Doubleday.
1982. The Pursuit of Power: Technology, Armed Force, and Society since A.D. 1000. Chicago: University of Chicago Press.
1987. A History of the Human Community. Vol. I. Prehistory to 1500, 2nd ed. Englewood Cliffs, NJ: Prentice Hall.
Morgan, D.
1986. The Mongols. London: Blackwell.
Peregrine, P. N.
1992. Mississippian Evolution: A World-System Perspective. Monographs in World Archaeology No. 9. Madison, WI: Prehistory Press.
1995. Networks of Power: The Mississippian World-System. In Nas-saney, M., and Sassaman, K. (ed.), Native American Interactions (pp 132-143). Knoxville, TN: University of Tennessee Press.
2000. Archaeology and World-Systems Theory. In Hall (ed.) 2000: 59-68.
Peregrine, P. N., and Feinman, G. M. (eds.)
1996. Pre-Columbian World-Systems. Monographs in World Archaeology No. 26. Madison, WI: Prehistory Press.
Pomeranz, K.
2000. The Great Divergence: China, Europe, and the Making of the Modern World Economy. Princeton: Princeton University Press.
Sahlins, M. D.
1961. The Segmentary Lineage: An Organization of Predatory Expansion. American Anthropologist 63 (2): 322-345.
1963. Poor Man, Rich Man, Big-Man, Chief: Political Types in Melanesia and Polynesia. Comparative Studies in Society and History 5 (3): 285-303.
1968. Tribesmen. Englewood Cliffs, NJ: Prentice-Hall.
Saunders, J. J.
1971. The History of the Mongol Conquests. New York: Barnes & Noble.
Schneider, J.
1977. Was There a Pre-Capitalist World-System? Peasant Studies 6 (1): 20-29.
Seaman, G. (ed.)
1989. Ecology and Empire: Nomads in the Cultural Evolution of the Old World. Los Angeles, CA: Ethnographics Press, Center for Visual Anthropology, University of Southern California.
1992. Foundations of Empire: Archaeology and Art of the Eurasian Steppes. Los Angeles, CA: Ethnographics Press, Center for Visual Anthropology, University of Southern California.
Seaman, G., and Marks, D. (eds.)
1991. Rulers from the Steppe: State Formation on the Eurasian Periphery. Los Angeles, CA: Ethnographics Press, Center for Visual Anthropology, University of Southern California.
Shannon, Th. R.
1996. An Introduction to the World-System Perspective, 2nd ed. Boulder, CO: Westview Press.
So, A. Y.
1990. Social Change and Development: Modernization, Dependency, and World-system Theory. Newbury Park, CA: Sage.
Straussfogel, D.
1998. How Many World-Systems? A Contribution to the Continua-tionist/Transformationist Debate. Review 21 (1): 1-28.
2000. World-Systems Theory in the Context of Systems Theory: An Overview. In Hall (ed.) 2000: 169-180.
Szynkiewicz, S.
1989. Interactions between the Nomadic Cultures of Central Asia and China in the Middle Ages. In Champion, T. C. (ed.), Centre and Periphery: Comparative Studies in Archaeology (pp. 151-158). London: Unwin Hyman.
Teggart, F. J.
1918. The Processes of History. New Haven: Yale University Press.
1925. Theory of History. New Haven: Yale University Press. (Both reprinted University of California Press 1942, Peter Smith 1972.)
1939. Rome and China: A Study of Correlations in Historical Events. Berkeley: University of California Press.
Thompson, W. R.
2000. Comparing Approaches to the Social Science History of the World System. In Denemark, Friedman, Gills and Modelski 2000: 287-298.
Turchin, P., and Hall, Th. D.
2003. Spatial Synchrony among and within World-Systems: Insights from Theoretical Ecology. Journal of World-Systems Research 9: 1 (Winter): 37-64 [ejournal: http://jwsr.ucr.edu/index.php].
Wallerstein, I.
1974a. The Rise and Future Demise of the World Capitalist System: Concepts for Comparative Analysis. Comparative Studies in Society and History 16 (4): 387-415.
1974b. The Modern World-System: Capitalist Agriculture and the Origins of European World-Economy in the Sixteenth Century. New York: Academic Press.
1993. World System vs. World-Systems. In Frank, A. G., and Gills, B. K. (eds.), The World System: Five Hundred Years of Five Thousand? (pp. 291-296). London: Routledge.
2004. World-Systems Analysis: An Introduction. Durham, NC: Duke University Press.
Wells, P. S.
1999. The Barbarians Speak: How the Conquered Peoples Shaped Roman Europe. Princeton: Princeton University Press.
2003. The Battle that Stopped Rome: Emperor Augustus, and the Slaughter of the Legions in the Teutoburg Forest. New York: W.W. Norton.
Whittaker, C. R.
1994. Frontiers of the Roman Empire: A Social and Economic Study. Baltimore: Johns Hopkins University.
Wolf, E. R.
1982. Europe and the People Without History. Berkeley: University of California Press.
Yü, Ying-shih.
1967. Trade and Expansion in Han China: A Study in the Structure of Sino-Barbarian Economic Relations. Berkeley: University of California Press.
