Hostname: page-component-7b7567c8c-tnzcf Total loading time: 0 Render date: 2026-11-12T20:59:36.199Z Has data issue: false hasContentIssue false
  • English
  • Français

The nomadic other: Ontological security and the Inner Asian steppe in historical East Asian international politics

Published online by Cambridge University Press:  21 September 2015

Joseph MacKay
Rights & Permissions [Opens in a new window]

Abstract

A growing literature in IR addresses the historical international politics of East Asia prior to Western influence. However, this literature has taken little note of the role of Eurasian steppe societies and empires in these dynamics. This article offers a corrective, showing that relations between China and the steppe played an important role in regional politics. I argue that Chinese elite conceptions of the steppe as other played an important role in maintaining China’s ontological security. Imperial Chinese elites pursued a deliberate strategy of ‘othering’ steppe societies, presenting them as China’s political-cultural opposite. Doing so both provided a source of stable identity to China and justified their exclusion from the Chinese ‘world order’. Empirically, I proceed in three sections. First, I consider Chinese identity building, framed in terms of ontological security, both under the founding Qin and Han dynasties, and under the later Ming dynasty. Second, I address recent historiography of the steppe, showing Chinese conceptions of it were inaccurate. Third, I address the long history of hybridity between the two regions.

Information

Type
Articles
Information
Review of International Studies , Volume 42 , Issue 3 , July 2016, pp. 471 - 491
Copyright
© 2015 British International Studies Association 

Introduction

Recent research on historical East Asian international politics has addressed the region’s Sinocentric hierarchical structure, and the ‘long peace’ that accompanied it.Footnote 1 Such accounts focus largely on Imperial Chinese Confucian political identity, as an explanation for the region’s international politics, treating identity as an explanation for other phenomena. Less attention has been paid to explaining the sources and persistence of those ideas, as puzzling unto themselves. I link the Sinocentric East Asian long peace to China’s very different relations with peoples to the north and west: the nomads of the Inner Eurasian steppe.Footnote 2 China’s disposition toward the steppe was persistently bellicose. While it viewed its East Asian neighbors – Korea, Vietnam, occasionally Japan, and many smaller polities – as participants in a Confucian social order, official Chinese accounts described the peoples of the steppe as ‘barbarians’,Footnote 3 nomadic groups that did not and perhaps could not participate in Confucian society.Footnote 4

I argue China’s treatment of the steppe constituted a deliberate identity-securing strategy. I build an explanation of China’s self-understanding as zhongguo (the ‘Middle Kingdom’, a uniquely powerful polity at the center of world order), tasked with governing, directly or indirectly, tianxia (‘all under heaven’, the sum total of governable political space). China claimed uniquely powerful legitimate authority, to which the steppe was intractably resistant. From its founding at the end of the Warring States period, Imperial China could neither conquer the steppe nor impose its ideology on it. The steppe thus challenged China’s identity as the paramount actor in a Sinocentric ‘world order’.Footnote 5 China responded with a conscious strategy of dialectical identity building, designating the steppe as an ‘other’: an identity foil that was different in almost every way: migratory rather than sedentary, diffuse rather than hierarchical, violent rather than harmonious, and natural rather than historical. Chinese elites portrayed it as a social order not just outside the Confucian world, but as one inherently unable to participate in it. This image of the steppe as a constitutive other afforded China a source of consistent identity – one that recurred across dynasties.

To develop this account theoretically, I turn to two related conceptual approaches. The first concerns ontological security. Over the last decade, ontological security arguments have become prominent tools of constructivist explanation in IR.Footnote 6 Such accounts posit that actors seek not just physical, military security, but also security of identity. Recurring cognitive scripts stabilise actors’ interactions with their environments, permitting stable self-conceptions. I argue Chinese imperial elites leveraged an account of the steppe as other to construct such a stable, relational identity for the Sinocentric world, becoming bound to it. In so doing, I put ontological security to relatively novel theoretical use. While more often used to explain how identity drives putatively counterintuitive actions, here it explains the creation and persistence of identity itself, on a macro-historical scale. I emphasise the processes by which identity is made secure through periods of geopolitical change. The second approach, related to the first, concerns dialectical identity construction of self and other.Footnote 7 On such accounts, identities emerge in part from contrast. As scholars have recently argued, ‘othering’ is not a necessary result of interaction, so much as an overt and deliberate strategy of identity building.Footnote 8 In Imperial China, it persisted across multiple imperial dynasties.

As I show below, China’s image of the steppe as ‘natural’, or uncivilised, was not especially accurate. The historical steppe was in fact a complex and diverse place, the site of dynamic and historically varied institutional innovation. Nor was it inherently violent – war came and went, as we might expect in any region. The traditional Chinese conception of the steppe did not reflect an attempt to describe it accurately. Instead, its purpose was to secure identity: it provided China with an inversion of its self-image. While China understood itself as irreducibly different from the steppe, I show the two engaged in a long and varied history of cultural and political hybridisation. The othering of the steppe both consolidated Chinese (and eventually East Asian) identity, and also concealed or downplayed recurring process of interaction and exchange.

This account is important for at least three reasons. First, it improves historical- IR-theoretic understanding of the region itself. In so doing, it may offer a basis for adjudicating between competing accounts of the region’s international politics, some of which emphasise China’s historical bellicosity.Footnote 9 Distinguishing more closely between targets of Chinese policy may help explain why policy was sometimes harmonious and sometimes belligerent. Second, it expands the study of East Asian historical IR beyond the region itself, taking into account inter-regional politics. Third, theoretically, it provides a novel application of ontological security to explaining not specific events or processes, so much as macro-historical persistence of identity itself.

Below, I proceed in four sections. The first frames the puzzle of Chinese identity persistence, and develops the theoretical account. The second deals with China’s place in the East Asian hierarchy, and its conception of the steppe as ‘other’. The third section deals with the steppe itself, showing China’s historical conception of it was inaccurate. The fourth concerns cultural hybridisation between the two regions. I conclude by summarising and considering implications for theory and for historical East Asia.

Self, other, and ontological security in Chinese frontier policy

This section sets out the puzzle of Chinese identity formation, linking it to the steppe. In so doing, it shows extant accounts elide the steppe and its explanatory value. It then develops a theoretical account based on ontological security and self/other relations.

The Puzzle: Chinese Identity Consolidation and the Steppe

IR theorists have largely followed historians in understanding premodern East Asian international politics as hierarchical and pacific, relative to the anarchy and bellicosity of its modern, Eurocentric equivalent. Imperial China was located at the apex of an international hierarchy ordered by Confucian values of harmony.Footnote 10 For more than two millennia – from the founding of the Qin dynasty in 221 BCE, to the formal end of the Qing dynasty, in 1921 – a relatively consistent idea persisted of what Imperial China was or should be.Footnote 11 When China was ascendant, as during the Han and Ming dynasties, this identity justified Chinese regional dominance. When China was in decline, it provided a source of aspiration. When foreigners occupied the country, as did the Mongols under the Yuan dynasty and the Manchus under the Qing dynasty, they justified their rule by claiming the Mandate of Heaven (tianming) for themselves. In short, a relatively constant identity persisted across a wide variety of political conditions.

The persistence of a single imperial Chinese identity or self-understanding over such an extended period presents a striking social-scientific puzzle. As Pines has noted, what is puzzling is ‘not the empire’s indestructibility – for it witnessed several spectacular collapses – but is almost miraculous resurrection after years of disorder’.Footnote 12 What needs explaining is not persistent order, so much as persistent identity. What endured for more than two thousand years was not the specific institution of empire, but the ideas underpinning it. Indeed, given that they were practiced only inconsistently, their recurrance seems all the more striking.

Since what is at stake is identity persistence across differing historical conditions, physical security seeking alone cannot offer an explanation. We are not strictly in the realm of power politics, at least in traditional, realist, or neorealist formulations, which cannot explain varied imperial Chinese behaviour toward its neighbours. Thus, Kang notes that China’s foreign policy was not reducible to realist unipolarity: ‘China’s relations with the nomads were characterized by war and instability, whereas relations with the Sinicized states were characterized by peace and stability. Unipolarity cannot account for both of these outcomes.’Footnote 13 An appeal to identity is necessary.

Nonetheless, constructivist accounts of the region treat identity as an explanation for other phenomena, rather than as a puzzle unto itself. Identity maybe made persistent by clearly circumscribing it. Kang goes on to note that: ‘Creating civilization also entailed creating the idea of an other, the “barbarian”, in contrast to “civilization”.’Footnote 14 The role of this ‘othering’ has received little systematic attention. Kang and others who focus on ideas treat the steppe as exogenous to the region, regarding sinocentric East Asia as a predominantly closed system. Thus, Kang claims that ‘the nomads were generally more of a nuisance than a threat to China’.Footnote 15 Similarly, Kelly treats East Asian peace as puzzling, and bellicose relations with the steppe as unexceptional.Footnote 16 Such accounts identify a real cultural difference between ‘the steppe and the sown’, but do not link it to Chinese identity formation.Footnote 17 Below, I link Chinese identity formation to China-steppe relations. Since that identity went on to shape East Asian international politics, the production and reproduction of the one likely shaped the distinctive form of the other.

Ontological security and the steppe as other

I argue the point in terms of ontological security.Footnote 18 On such accounts, actors seek not just physical security, but also security of identity: ‘Ontological security-seeking is the drive to minimize hard uncertainty by imposing cognitive order on the environment.’Footnote 19 Actors do so by establishing stable and recurring relationships with their surroundings. On such accounts, identities are defined in terms of cognitive scripts. Recurring behaviours and interactions grant actors persistent conceptions of themselves.

Despite focusing on stasis or stability, previous applications of ontological security have emphasised explaining putatively counterproductive or self-destructive behaviours, usually over the short to medium term. Mitzen emphasises the power of ontological security in explaining actions that undermine actors’ presumed interests, chiefly the persistence of avoidable armed conflict.Footnote 20 Steele emphasises a wider range of activities, but nonetheless emphasises specific policy choices.Footnote 21 More recently however, Subotic has emphasised the role of elites in constructing state ‘autobiographies’ that provide a fixed identity, and thus ontological security.Footnote 22

I pursue a similar approach, over a longer historical period. I emphasise the capacity of ontological security to generate ideational and identity stability on a macro-historical scale, through long periods of repeated political change. Where it persists through long periods of recurring transformation, stability of identity itself becomes puzzling: how, that is, are we to explain how a consistent identity persisted alongside significant political and institutional change? Since ontological security theory understands identity maintanence as a product of routines, often framed in terms of interactions with actors’ environments, it offers an opportunity to explain how identities are made persistent. Here, it explains the persistence of China’s identity as the regionally dominant and politically Confucian Middle Kingdom through multiple dynastic collapses and foreign occupations, and other political transitions.Footnote 23

To make the argument, I turn to conceptions of self and other. Dialectical identity formation has a long history, both in social theory more broadly and in critical and constructivist IR. On such accounts, identities are generated and reinforced through interaction. However, as Neumann noted two decades ago, we should not take the rhetoric of self and other wholly at its word, assuming that ‘human collective identities are unequivocally bounded actors’. That is, social groups are prone to shifting identities and permeable memberships.Footnote 24 Collective actors may deploy accounts of others strategically, in order to stabilise their own boundaries and identities.Footnote 25 Actors are not merely passive products of interaction; they shape those interactions to their own ends. As Steele notes, ‘Identity construction is a political project, where states distinguish the “we” as a basis for social action.’Footnote 26 Locating the self/other dichotomy within the framework of ontological security allows us to track these processes.Footnote 27

The resulting representations of the other need not be accurate, in the sense of representing others’ cultures and political orders as they understand themselves. Rather, actors generate representations of others in ways that stabilise their own identities. In East Asia, traditional Chinese representations of the steppe were sometimes strikingly at odds with the reality north of the frontier. They nonetheless served a purpose, providing an opposed category useful in building a stable and circumscribed identity.

The task confronting early Chinese elites was that of constituting a clear collective identity for their newly consolidated empire – the production of a collective self, ‘a narrative process of identification whereby a number of identities that have been negotiated in specific contexts are strung together into one overarching story’.Footnote 28 Doing so was predicated upon the availability of another identity group, one counter-posed in important respects to the peoples the new empire comprised. Strategically emphasising them in elite rhetoric aided in constituting a collective identity – one in which the sedentary, hierarchical, Confucian social order of the Chinese heartland could be contrasted with a nomadic, horizontally structured, and culturally non-Confucian other. That identity proved useful not just for the early imperial dynasties, but also for later dynasties, following periods of collapse or foreign occupation.

To demonstrate this empirically, I proceed below in three stages. First, I detail Imperial Chinese identity, and link it to China’s conception of its northern neighbors. Second, I develop a historical account of the steppe itself. Doing so serves two purposes. First, it shows the traditional Chinese view was not especially accurate. Second, it shows us what the best available scholarship indicates steppe societies actually to have been.Footnote 29 Third, I turn to the long history of cultural and political hybridity between the Chinese heartland and the steppe, showing the deep and frequent pattern of interaction and mutual influence that China’s self/other identity framework allowed imperial elites to elide.

Chinese identity construction and conceptions of the steppe

This section details historical China’s conception of the steppe. I begin with the period when Chinese elites began identifying the steppe as other, during the founding Qin and Han dynasties. I then turn to the more recent Ming dynasty, which reasserted China’s canonical relationship with the steppe after the Mongol/Yuan period, and which was located at the heart of a Confucian East Asian regional order. The former captures the identity’s emergence, the latter its endurance.

The classical dynasties

Imperial Chinese identity emerged during the founding Qin and Han dynasties. ‘China owes its ability to endure across time, and to re-form itself again and again after periods of disunity, to a fundamental reshaping of Chinese culture by the earliest dynasties.’Footnote 30 At that time, imperial integration brought diverse cultural groups under a single Chinese identity:

Such regional variations were not an inconvenient fact of life but, rather, became essential to an empire that justified itself by making just this kind of hierarchical distinction – between the universal, superior culture of the imperial center and the limited, particular cultures of regions and localities.Footnote 31

Imperial unification was hard won – no previous polity in the region had wholly achieved it. Creating a unifying political culture required an identity under which more localised identities could be subsumed.

The geographical limit of Qin and Han power was the steppe frontier. It is difficult to overstate the impact the steppe had on China:

The encounter with the nomads became the single most significant event in the political, cultural, and ethnic history of the Chinese. Not a single ethnic group on China’s frontiers had a comparable impact on China’s life, or was able to challenge Chinese political culture as the nomads did … [T]he arid steppe zone was basically unconquerable and ungovernable … [T]he nomads swiftly established their independent tribal confederations, the very existence of which undermined the notion of the singularity and universality of the Chinese emperor’s rule.Footnote 32

However, the impact of the nomadic encounter was dual. On the one hand, it undermined the universality of China’s claims to political legitimacy. On the other, in circumscribing China, it provided it with a clearer and sharper self-definition. The conceptual distinction between the steppe and the sedentary heartland clarified Chinese thinking about China’s own unity. Elite conceptions of Chinese collective identity emerged from it. Mark Edward Lewis summarises:

In contrast with the Warring States period, when regional cultures constituted the primary divisions in the Chinese sphere, the imagining of a world divided between nomads and Chinese marked a major step. It posited the fundamental unity of a single Chinese civilization defined by what was not nomadic, and it reduced regional divisions to secondary status. China first emerged as a unity through the invention of a Chinese/nomad dichotomy, and this bipolar concept remained central to Chinese civilization in later periods.Footnote 33

Thus, Chinese identity had at its center a contrast with an other: an opposed party that could be usefully described as opposite in every way. By instrumentally deploying an account of the steppe, Chinese elites articulated a systematic and expansive conception of a collective Chinese self, one that took in all the previous collective selves, now construed as imperial subordinates.

This effect is visible in the writings of Qin and Han elites. The period coincided with the consolidation of the first large steppe empire, that of the Xiongnu. The Han statesman Chao Cuo described it, focusing on its political cultural inversion of civilised Chinese life:

The Xiongnu do not depend upon the soil for their food and clothing, and due to this they easily cause disturbances on the frontier. How can one explain this? The Xiongnu eat meat and drink milk; they wear skins and furs. They have no walled towns, fields, or houses to serve as bases. They are like flying birds, or running beasts in the vast, open fields. When they find good grass and fresh water they halt, and when the grass and water are used up they move on. Coming and going, wandering about, now arriving and now departing, this is the livelihood of the Xiongnu.Footnote 34

This at least partially captures the mobile lifeways and military strategy the Xiongnu adopted, but in so doing casts them as the inverse of the sedentary, agricultural, hierarchical, and culturally sophisticated civilisation that classical China saw in itself. Similarly, the Han historian Sima Qian wrote that:

It is the custom of the Xiongnu to support themselves in ordinary times by following their flocks and hunting, but in times of hardship they take up arms to raid. This would appear to be their nature … When it is to their advantage, they advance; when not they retreat, as they see no shame in retreat.Footnote 35

Defining the nomads by their differences from the settled, agrarian, hierarchical Chinese, helped to define China as civilisationally superior. Having adopted this attitude, military victories over them become occasions for civilisational aggrandisement. Thus, he credits the first Qin emperor with sending a general to ‘build the Great Wall and defend the borders, driving back the Xiongnu … so that the barbarians no longer ventured to come south to pasture their horses and their men dared not take up their bows to vent their hatred’.Footnote 36 The Xiongnu, and later steppe empires and confederacies, nonetheless persisted. At some points the Xiongnu extracted tribute payments from Han authorities. When Imperial China was founded, the region had not one power concentration, but two: one sedentary, and the other nomadic.Footnote 37 It was this tension that fed Chinese official attitudes: the duality of Chinese-led tianxia and the steppe meant that the latter had to be imagined as external to the civilisational space of ‘all under heaven’.Footnote 38

Chinese identity had other sources, of course. Qin and Han elites understood themselves as the descendants of the Zhou polity – the long dynasty that preceded the warring states period – from whom they claimed the Mandate of Heaven. But the status of pre-imperial dynasties like the Zhou and Shang as Chinese in the full sense is uncertain, as is much of the historical record concerning them. Zhou rhetoric deployed a language of tianxia, but the underlying meaning of their usage described a culture rather than a political unit, and was geographically relatively restricted. Later usages referred explicitly to an expansive and culturally diverse political space, under imperial authority.Footnote 39 Only after the Qin consolidation did the tianxia category arise in its final, political form:

Qin territorial expansion and the concomitant establishment of the Xiongnu empire changed the nature of Sino-alien intercourse. The nomadic way of life placed formidable obstacles in front of possible Sinification, while nomadic grasslands were of little value for the Chinese. The establishment of the Qin Great Wall, later expanded and restored by the Han dynasty, marked the separation of the two worlds. It also marked, even if unintentionally, the end of Chinese pretensions to build a truly universal empire.Footnote 40

The rhetoric of tianxia became resolutely political and universal only when Sinicised consolidation reached a relatively hard limit. It did not describe a situation that applied in practice, since it did not accurately account for political practice. Instead, its purpose was to describe a rhetorically universal political sphere from which the nomads could be plausibly excluded – not from just the Chinese polity, but from civilisation as such.

Contrasting the Confucian tianxia framework against the less hierarchical and rigid social order of the steppe clarified Chinese imperial identity by reference to its other. Doing so was not necessary or unavoidable. It was a deliberate strategy for managing the nomadic challenge. As we will see below, the underlying political reality was different. The Xiongnu were neither marginal nor weak, and eventually extracted significant political concessions from their Han interlocutors. Nor was China an undifferentiated whole: internal cultural differences persisted after unification.Footnote 41 Official rhetoric nonetheless remained universalist and exclusionary. Its political value lay in providing a consistent political identity, in contrast to an other. It was, in short, a source of ontological security.

The Ming dynasty

Under subsequent dynasties, Chinese rhetoric of universal empire recurred and grew, becoming more elaborate and providing ideological materials for Confucian regional order. Official discourse came to divide subject peoples into the ‘raw’ and the ‘cooked’ (sheng and shu), where the former remained rebellious and the latter had formally submitted to imperial authority, recognising the cultural superiority of Chinese civilisation.Footnote 42 A Song dynasty official account illustrates the distinction:

Those who have submitted and are attached to our county and township administration are the Cooked Li. Those who live in mountain caves and are not punished by us or supply corvée labour are the Raw Li. These sometimes come out to engage in barter with the administered population.Footnote 43

Ethnically non-Chinese populations were viewed as civilised to the extent they adopted Chinese ways of life, and recognised the empire as the seat of power.Footnote 44 Those that did not remained ‘raw’: an uncivilised other.Footnote 45

This distinction was dramatically upset by the Mongol conquest, and declaration of the Yuan dynasty, in 1271. While previous steppe empires had made incursions against the Chinese heartland, the Mongols were the first to conquer it outright. The Yuan attempted to combine the tenets of steppe and Chinese governance. When they were overthrown in 1368, the proclaimed purpose of the new Ming dynasty was to restore more purely Chinese political order. Official rhetoric aimed to reassert the traditional distinction between Chinese self and steppe other. By this time, however, China was at the center of a Confucian regional order in East Asia.Footnote 46

The dynasty’s founding Hongwu emperor forbade his successors from invading most neighbouring countries – he named Korea, Vietnam, and Japan, and several smaller tribute-paying states – noting that they would be too much trouble to control directly. In total, fifteen states were designated as ‘not to be invaded’.Footnote 47 Most of these participated in the formal institutions of tribute. However, he singled out the steppe: ‘As for the hu and jung barbarians who threaten China in the north and west, they are always a danger along our frontiers. Good generals must be picked and soldiers trained to prepare carefully against them.’Footnote 48 His concern was practical – he had, after all, founded his dynasty by expelling such foreigners from the Chinese imperial throne – and he proclaimed elsewhere that ‘his mission was to rid China of its Mongol influences’. Nonetheless, practical political exigencies pushed in another direction: ‘his new regime tended to reproduce the Yuan practices with which he was familiar’, and he privately revered Kubilai Khan, China’s first Mongol emperor as a ‘great conqueror’ – a figure to be celebrated for his achievements.Footnote 49 The externalising of the nomadic other was, thus, at least as much a matter of rhetoric – of identity – as of military security seeking.

The Ming dynasty was rigorously Sinocentric in its rhetoric. Rossabi notes that Ming official sources on Inner Asian societies were

to say the least, biased. The scholars who compiled these records professed lack of concern for foreigners and foreign relations. They depicted foreigners as barbarians and scarcely accorded them much attention, offering scant information on the commercial and economic relations with the peoples of Central and Inner Asia.Footnote 50

Such biases served the restorationist project of reclaiming Chinese civilisation from Mongol rule. As one Ming historian wrote, ‘When our imperial Ming received Heaven’s mandate, it unified Chinese and aliens. The breadth of the territory that submitted extended all the way east of the Liao River, west as far as the desert, south beyond the maritime coast, and north into the steppes.’Footnote 51 The account is inaccurate – the new dynasty was more cohesive than its precursor, but geographically smaller and less diverse. However, the claim served the purpose of elevating Ming achievements above those of neighbouring peoples. The Ming rearticulated a unified and independent China, one that had recurred for centuries before the Mongol invasion.

The Ming dynasty also saw the return of neo-Confucian thought, a movement that had originated under the Tang and Song dynasties. In its earliest form, neo-Confucianism had been in part a reaction to outside influences. These were both spiritual, including Buddhism, and geopolitical, chiefly invasions from the north.Footnote 52 Under Ming rule, it became a source of renewed orthodoxy, of doctrinaire values, after the comparatively ecumenical period of Mongol rule.Footnote 53 It was, in short, a way to reassert Chineseness, after foreign occupation. Moreover, by this time Chinese political culture was not limited to China itself. A network of tribute-paying states (the ‘tribute system’ or ‘Chinese world order’)Footnote 54 now constituted a relatively stable hierarchy around China. The rearticulation of a strong and cohesive identity served to reinforce China’s role as regional hub.

In sum, the Ming period was typified by political cultural reconsolidation, and an increasingly homogenous identity. That identity construction was carried out by contrast to the steppe peoples just expelled from imperial leadership. Doing so involved constituting a consistent environment for itself, by imposing consistent relations with its neighbours – that is, the pursuit of ontological security. China did so by making the steppe its constitutive other, treating its erstwhile conquerors as a persistently uncivilised inversion of itself.

The new steppe history

How did the Chinese image of the steppe compare to the steppe itself? We can infer that Chinese conceptions of the steppe were motivated by identity building – by othering, and ontological security seeking – not just because they aligned with the motivations of imperial elites, but also because they presented an inaccurate image of steppe culture. The region was a more dynamic and varied political and cultural space than the received Chinese view suggests. The implied purpose of China’s images of the steppe was not to document the region as it was, but to provide an account useful for identity building.

Historians, archaeologists, and geographers of Inner Asia once shared similar misperceptions. Canonical accounts often deprived nomadic peoples of agency, reducing their history to exogenous factors, including pressures from the environment or from neighbouring societies. Ecological accounts traced patterns of migration, social aggregation, and war to climatic cycles.Footnote 55 ‘Dependency’ accounts traced nomadic social hierarchy and complexity to settled societies, from whom nomads borrowed institutions and ideologies, and upon whom they preyed.Footnote 56 Such accounts now seem, if not wholly wrong, then at least incomplete. While environmental pressures and relations with China did influence pastoral steppe societies, other factors were at work as well. Recent research suggests the premodern steppe was a more diverse and politically innovative place than traditional accounts suggest.Footnote 57

Alternative origins

Written records of early Inner Asian social aggregation are limited to a few Qin and Han sources, including those discussed above.Footnote 58 However, recent archaeological evidence suggests it occurred much earlier, independent of China.Footnote 59 Jean-Luc Houle has shown that, on at least one occasion, primary emergence of a complex society occurred prior to contact with China, as early as the Late Bronze Age (late second or early first millennium BCE). The evidence ‘suggests substantial efforts at maintaining an integrated community that went beyond the immediate locale and existed in the absence of everyday face-to-face interaction’.Footnote 60 Pre-imperial political order thus extended beyond the local kinship ties usually associated with pastoral life.

Similarly, Gideon Shelach identifies a range of non-Chinese artifacts across the region, indicating distinct social practices and power structures, including a distinctive militarist ideology (evidenced in part by weapons in grave sites).Footnote 61 These groups were not alone in forming indigenous social orders: early societies across the region emerged through varied trajectories and took diverse forms.Footnote 62 Additional evidence indicates they consumed grains, suggesting steppe societies were partially sedentary, or traded with sedentary peoples, as early as the Iron Age.Footnote 63

The rise of steppe confederation

Beginning in the late third century BCE, steppe societies formed much larger polities. These have been called empires, but are now also termed confederations. These confederation- or empire-building projects represented periods of relative unity over large parts of the steppe, and on some occasions over virtually all of it. Over the last twenty-three centuries, Central and Inner Asian societies have formed such polities on at least a dozen occasions.Footnote 64 Nicola Di Cosmo attributes steppe empire building to ‘a social response to a state of crisis’.Footnote 65 While such crises might be environmental or geopolitical, the varied ways societies responded – developing new political structures, or expanding into new territory and new economic activities – were matters of internal political innovation.

The first of these, the Xiongnu (Hsiung-nu, Hunnu) Empire, existed from 209 BCE to 150 CE, occupying a large portion of steppe to China’s immediate northwest.Footnote 66 The Xiongnu empire-builder Modun proclaimed he had unified ‘all the people who live by drawing the bow’.Footnote 67 Accounts of the empire’s construction differ – while some describe a process of conflict, conquest, and imperial consolidation, William Honeychurch argues the empire may have been assembled non-violently, as a loose-knit confederation.Footnote 68 The Xiongnu posed a significant threat to China, one that persisted across four Chinese polities – Qin, Western Han, Xin, and Eastern Han.Footnote 69

While some empires were short-lived, others persisted for two or more centuries, suggesting a robustness of social order not explained by temporary tribal alliances of convenience, or occasional charismatic leaders such as Chinggis (‘Genghis’) Khan.Footnote 70 Nomadic tribes have sometimes been thought prone to fission, or fragmentation into smaller, more manageable units.Footnote 71 On the steppe, such accounts portray confederation as temporary alliances of convenience.Footnote 72 However, recent research suggests political order extended beyond alliance, to a consolidated and persistent aristocratic class.Footnote 73 Fission through migration was not likely a reliable option, since leaving one area meant finding a new one, and contending with its population.Footnote 74 Since even small pastoral groups required large swathes of semi-arid grazing land, territory easily became scarce. Confederation or empire building thus likely occurred with some regularity.

New political structures

These large polities were structured in varied and flexible ways, prone to institutional innovation. Such innovation responded to the distinct and diverse living conditions of Inner Asian geography. Environmental accounts were right to emphasise geographical context, but wrong to conclude steppe societies were unsophisticated or uniform. Steppe environmental conditions drove wide-ranging adaptations: ‘pastoral nomadic polities represent alternative forms of complex organization that were different from [sedentary counterparts] … but still quite complex in unexpected ways’.Footnote 75

Mobility imposed distinctive problems of political order, to which steppe peoples responded with distinctive solutions. Steppe societies adapted to varied geographical and climatic conditions in ways ranging from pastoral herding, to sedentary agriculture,Footnote 76 mobile predation,Footnote 77 and proto-urban settlement.Footnote 78 Still, a largely mobile population made for an inherent problem of power projection. The Xiongnu solution was a mobile court, allowing ‘central’ authority to migrate across multiple nomadic subgroups.Footnote 79 Its authority was nonetheless diffuse: steppe rulers led by negotiation as much as by decree. The Xiongnu ‘state’ was an emerging process as much as a hierarchical certainty, constantly under renegotiation, typified not only by kinship relations, but by loose aristocratic leadership, coupled with a ‘decimalised’ military structure. Leaders oversaw ten, a hundred, a thousand, or ten thousand men, in nested units – a structure of military organisation that would persist through future large steppe polities. It was an early variation on what David Sneath has called a ‘headless state’.Footnote 80 Steppe ‘state’ formation was thus an especially gradual and intermittent process. Throughout, mobility was not simply a hindrance to political integration. Horses allowed elites to move at speed over great distances. Mobility reduced risk of political breakdown through isolation. Negotiation and renegotiation of political order occurred over vast territories with relative frequency.

Such negotiations were coupled with economic innovations concerning the convertible valuation of food, resources, crafts, livestock, animal products, and other goods, which could be exchanged both within and beyond the polity, giving rise to the Silk Roads.Footnote 81 Inner Eurasian peoples have often been portrayed as passive conveyances for goods they circulated across Eurasia. It now appears these long-distance markets were created effortfully and innovatively by steppe peoples who enriched themselves through trade. Nomads likely became expert at these conversions of value, requiring sophisticated knowledge of what was worth how much, to whom, over time. Economic development reinforced political institutions, creating ties of interdependence that bound these loose-knit polities together. Foreign luxury goods also served as status objects, reinforcing social hierarchies.Footnote 82

Later empires

Early empires like that of the Xiongnu established a pattern, as later large polities consolidated, fragmented, and re-emerged. By this time, the steppe was no longer in any sense isolated from China – contact was frequent and extensive. Nonetheless, large steppe polities remained distinctive and innovative. The obvious example is the Chinggisid Mongol Empire, one of the largest land empires in history.Footnote 83 The empire’s heterogeneity and rapid expansion – it absorbed much of Eurasia in a matter of decades – created a need for novel methods of rule, as Mongol imperial rule was grafted onto existing local political networks, many themselves foregoing empires.Footnote 84

Out of necessity, the Mongols employed those available in the aftermath of conquest to govern, producing regional variance in structures of government. Thus, ‘the basic building blocks of the Mongols’ system of rule over the peoples south of the steppe were the remnants of multi-ethnic regimes they destroyed: diverse communities of post-nomads, and international merchants who, be the very nature of their profession, were specialists in inter-ethnic relations’.Footnote 85 In some places, they imposed bureaucratised rule, in others they ruled indirectly through traditional local elites, in others still they combined the two. The empire’s size and pluralism allowed them to move administrative officials around, cross-pollinating institutions that became influential beyond imperial borders: ‘The success of the Mongol empire created the conditions for the export of techniques of governance on an enormous scale.’Footnote 86

The last steppe polity to attempt such widespread conquest was the Timurid Empire, in the late fourteenth and early fifteenth centuries.Footnote 87 The last pure steppe confederacy of any kind was that of the Zunghars, one of the three dominant empires of seventeenth- and eighteenth-century Eurasia (the other two being Chinese and Tsarist Russian).Footnote 88 The Zunghar polity was structurally more diffuse than its Chinggisid precursor, typified by ‘a confederated pattern of several ruling lineages competing for domination, and linked by marriage alliance’.Footnote 89 It persisted for about eight decades, before being conquered by the expanding Manchu empire.Footnote 90

Analysis

While the political history of the steppe overlaps in some way with the traditional Chinese view, in important ways it does not. Substantial political order originated independently on the steppe. When confederation or imperialism later appeared, the distinctive geographical conditions of the region gave rise to innovative political and economic structures, adapted to mobility. The Mongols went on to borrow the political structures and legitimacy claims of conquered peoples, including the Chinese, combining them with their own. Large-scale steppe political orders proliferated for almost two millennia. This history is not wholly inconsistent with the traditional Chinese account or Western accounts.Footnote 91 However, steppe political history is not neatly reducible to an inversion of the Chinese way of life – the region has a political history of its own.Footnote 92

Concomitantly, the traditional Chinese view should be read not as an account of the region, so much as a set of active interventions in Chinese political life.Footnote 93 Chinese elites are best understood as offering not to much an account of what the steppe was, in it’s various interactions with China – for their descriptions were strikingly inaccurate – but instead a defence of the ontological security of Chinese civilisation. They other they constructed in so doing in some respects resembled the historical range of societies found on the steppe, but in at least as many ways did not. The steppe presented itself not simply as a challenge to China’s physical security, but also as an idea – one created and perpetuated by Chinese elites that could be leveraged to consolidate and stabilise Chinese imperial identity. That the idea bore only a limited resemblance to the neighbouring region itself helps us infer their intentions in so doing.

Steppe-sedentary hybridisation

The standard Chinese account of the nomads also somewhat inaccurately reflected relations between the steppe and China itself. While the division is absolute in Chinese identity discourse, historical practice was much more ambiguous and varied. The frontier was a locus of recurring interaction and hybridisation. In the case of Eurasian steppe polities and European peoples, Iver B. Neumann and Einar Wigen see both persistent difference and a centuries-long history of hybridity between the two.Footnote 94 A similar set of processes recurred on the East Asian-Inner Asian frontier. This section surveys recurring interaction, influence, and hybridity along it. I aim to show that Chinese insistence, in diplomatic or identity-constituting discourses, on a distinction between the steppe and the sown’ did not reliably describe frontier relations. Indeed, if anything, an insistence on the distinction served to conceal ongoing hybridisation on the ground. This insistence on a persistent distinction from the nomadic other constituted ontological security seeking.

Cultural and political mixing and interaction dates from the beginning of steppe-sedentary relations. The Han and Xiongnu paid tribute to one another by turns, matching periods in which one would attempt to conquer the other with periods of trade and elite inter-marriage.Footnote 95 The Xiongnu had a longstanding dependence on China for grains, and other fruits of sedentary life, but also had goods the Chinese wanted – not least horses, which were key to defending the Chinese heartland. Thus, even early Chinese relations with the steppe were varied, and sometimes interdependent.

The Tang dynasty was in many respects defined by close links to the steppe. Its ruling family had a long history of inter-marriage with steppe peoples, and adopted some steppe military practices. A combined effort at conquest and cultural suasion, facilitated by familiarity with nomadic culture, allowed the second Tang emperor to expand Chinese influence over the eastern steppe. Thus, hybridisation with steppe cultures facilitated Chinese aggrandisement during this period.Footnote 96

The most striking periods of hybridity between East and Inner Asian social orders occurred during periods of foreign rule in China. The Mongol Yuan adopted the official imprimatur of a Chinese imperial dynasty, thus claiming the legitimacy of the Chinese imperial mandate to govern ‘all under heaven’. However, Yuan emperors also claimed the Mongol title of Great Khan, thus asserting formal (if not actual) authority over the whole of the steppe. Yuan emperors thus asserted a dual mandate, to rule both the Middle Kingdom and the notionally confederated Mongol hordes.Footnote 97 Such claims were rarely recognised by other steppe hordes, but nonetheless indicated an assertion of legitimacy under two distinct political systems.Footnote 98 Moreover, Mongol rule over China itself was predicated on having fulfilled a Chinese ambition to restored unity. The Mongols took up the political mandate to govern tianxia, claiming to have unified China, which had been fragmented at the time of their rise, thereby justifying their conquest.Footnote 99 China’s ideological emphasis on geopolitical unity and rule over ‘all under heaven’ could be played both ways.

Even the Ming, whose political project was largely one of ethnic chauvinism, found they could not wholly efface the diverse hybridity of Chinese culture. The Ming founders necessarily inherited some of the political practices and administrative trappings of the Yuan polity.Footnote 100 They claimed (although they never controlled) territories extending far into the steppe itself. They permitted Mongols to reside within China’s frontiers, as settled, Sinicised imperial subjects.Footnote 101 They also moved the capital north to Beijing, closer to the Inner Asian frontier. The exacting purism of the dominant neo-Confucian political ideological was ‘at odds with the world in which Neo-Confucians lived’.Footnote 102 Once hybridity with the steppe had set in, it was not easily undone.

No Imperial Chinese dynasty is more emblematic of this hybridising trend than the Qing. Traditional accounts of Qing political culture view it as highly Sinicised: upon conquering China, the argument goes, the Manchus essentially became Chinese. However, the recent ‘new Qing history’ suggests this was not the case: the Manchus remained culturally distinct.Footnote 103 Instead of adopting whole the political structures of the Ming state they conquered, they adopted what might be called a ‘multi-vocal’ approach to imperial rule, representing themselves and their authority in different ways to different audiences.Footnote 104 On this account,

[A]t the height of their power, the Qing regarded China not so much as the center of their empire, as only a part, albeit a very important part, of a much wider dominion that extended far into the Inner Asian territories of Mongolia, Tibet, the Northeast (today sometimes called Manchuria) and Xinjiang, or Chinese (Eastern) Turkestan.Footnote 105

Manchu authorities wrote Qing court documents in Chinese, deploying the language of Confucian hierarchy and harmony. However, elsewhere they addressed their imperial possessions quite differently. To Mongol or other Inner Asian polities, they spoke the traditional steppe language of decimalisation and kinship. In Tibet, they deployed claims of allegiance to, and privileged cosmological status in, Tibetan Buddhism – a practice they adapted from the Yuan dynasty.Footnote 106 In Xinjiang (East Turkestan), they left varying traditional Islamic legal and governance structures in place, alongside Manchu military government.Footnote 107 A Qing emperor’s written instructions to Kyrgyz nomads are indicative:

[If] you are drawn into the civilized ways of the empire and submit … then We shall ensure that you continue to live peacefully. There is no need for you to adopt our ways. We shall not give you official rank and titles, no burden you with taxes. You have only to send an envoy to pay your respects and We shall extend Our favor with rewards.Footnote 108

In expanding westward, the Qing established a range of new and expressly colonial administrative structures.Footnote 109 They took in vast new ethnically diverse territories, and governed them, directly or indirectly, in ways variously adapted to each context.Footnote 110 The Manchu empire was an assemblage of distinct ethno-national silos, existing in parallel with one another, each governed by distinct claims to political legitimacy.

The Manchus expanded more successfully than the Ming, assembling the largest empire ever governed under the Mandate of Heaven. It was also the last Chinese polity to grapple with the ideological problem steppe societies presented. Once Manchu forces conquered the neighbouring portions of Inner Asia, independent steppe empires never reappeared.Footnote 111 By then however, the steppe’s influence on Chinese civilisation was extensive. The Qing period represents as much a merger with the steppe as conquest of it.

China’s successive ruling dynasties had experiences of the steppe as varied as was the steppe itself. However, they had in common being substantially influenced by it, having faced it as a threat, been ruled by it, or experienced prolonged cultural influence, trade, and inter-marriage. The same is true of the eastern steppe itself, which bore the influence of centuries of trade and cultural exchange with the Chinese heartland.

A useful example of this variation in Chinese policy toward the steppe is the history of the Great Wall of China. As Arthur Waldron has shown, the idea of a massive and persist barrier on the frontier is ‘a historical myth’.Footnote 112 There was and is a wall, of course – there have been many – but not one single wall consistently, nor one unchanging motivation for building it, nor a transhistorical Chinese elite consensus about wall building as a policy. Instead, wall construction and maintanence as a practice came and went and came again as relations with the steppe shifted. In other words, the wall itself varied historically, but the idea of it did not – much same as did the practice of China-steppe relations, and Chinese official beliefs on the subject. Eliding this variation made denying the steppe’s influence possible, stabilising China’s ontological security in the face of changes on the steppe and in its own policy alike.

Chinese elites were nonetheless at pains to elide this influence, proclaiming in its place a recurring Chinese Confucian ideology. Asserting consistent political and ideological scripts across multiple dynasties and transformations served a purpose, making a persistent political identity possible in the face of profound and often painful recurring change. The assertion of those scripts amounted to ontological security seeking. They were made possible precisely by a consistent account of the other.

Conclusion

China is sometimes described as a single polity more than two millennia old. This is true not in the sense of continuous institutions – Imperial China’s political structures transformed, collapsed, were conquered from without, and were rebuilt repeatedly – so much as in the sense of a persistent idea of proper Chinese political order. The underlying ideological materials of Chinese imperialism endured, with occasional adjustments, throughout that long period. Eventually, the spread of those ideas to neighbouring East Asian states led to a prolonged period of peace among them.

The limit point of that shared political space – the frontier with the Eurasian steppe – became a negative conceptual focal point for Chinese identity construction. The identity that persisted through the wracking changes described above was made possible in large part by a deliberate elite project of identity construction and maintenance. That project was one of ontological security seeking. By imposing consistent ideas or scripts upon what were in practice quite varied relations with the steppe, Chinese elites took on the steppe as foil, an other, portraying it as different from them in almost every way. They were wrong, factually speaking, on two counts: the steppe was not as they described it, and was not wholly different from China itself. Steppe societies were adaptive and institutionally innovative in ways Chinese sources elide. Moreover, China and the steppe experienced prolonged periods of cultural and political hybridisation. But the inaccuracy of the story of steppe life Chinese elites told was, in important respects, beside the point. China’s carefully constructed self-other conception of the steppe provided a stable source of ontological security – in effect, a cover story to excuse the very real political and social change that occurred beneath.

I conclude with a few words on consequences for the East Asian international hierarchy, and the late imperial long peace of the Ming and Qing dynasties. Historical accounts of the region in IR disagree on how pacific it really was. This account suggests that neither a strictly pacific nor bellicose view is appropriate. Instead, harmonious and warlike Chinese tendencies were interdependent, insofar as both bore on Chinese identity. The same value system that called for harmony within tianxia – the Confucian Chinese world order – also called for a defensive and even belligerent attitude toward that which was excluded from it. The maintenance of consistent identity within that cultural space required that whatever was without it be treated differently.

Consequently, the steppe should not be understood as wholly exogenous to the hierarchical political dynamics of East Asia proper. If the legitimacy of regional hierarchy was based on a shared Sinocentric identity and value system, and if that identity was predicated on a particular relationship to the steppe, then the steppe needs to be thought of as central to the persistence of the East Asian system itself. The argument may thus have consequences for how we think about the borders or limits of an ‘international system’. Premodern international systems, being non-global in scope, necessarily had peripheral points of contact with other systems. Such points of contact or seams between international orders may be more important than is commonly recognised.

More basically though, I have tried to show that a certain conceptualisation of the steppe played a key role in Imperial Chinese identity formation and foreign or frontier relations. Across multiple dynastic transitions, and varied practical relationships with the steppe and its other neighbours, China cemented a stable identity, both for itself and before the world, by contrasting itself with something it believed it persistently was not. Indeed, that consistency served not just to stabilise Imperial China’s ontological security – it also permitted a variety of hybridisation to go largely ignored by official accounts, providing an alibi of sorts for a volume of interaction and exchange across the frontier. Since that interaction was likely inevitable over the long haul of Chinese history, the role of this self-other construction seems especially important.

Footnotes

*

I would like to thank Simon Pratt, Christopher David LaRoche, Lincoln Rathnam, three extremely helpful reviewers, and the editors of the RIS for help in preparing this article.

1 Zhang, Yongjin, ‘System, empire and state in Chinese International Relations’, Review of International Studies, 27:5 (2001), pp. 43–63; Kang, David C., East Asia Before the West: Five Centuries of Trade and Tribute (New York, NY: Columbia University Press, 2010); Zhou, Fangyin, ‘Equilibrium analysis of the tributary system’, The Chinese Journal of International Politics, 4:2 (2011), pp. 147–178; Kelly, Robert E., ‘A “Confucian long peace” in pre-Western East Asia?’, European Journal of International Relations, 18:3 (2012), pp. 407–430; Zhang, Yongjin and Buzan, Barry, ‘The tributary system as international society in theory and practice’, The Chinese Journal of International Politics, 5:1 (2012), pp. 3–36; Ringmar, Erik, ‘Performing international systems: Two East-Asian alternatives to the Westphalian order’, International Organization, 66:1 (2012), pp. 1–25; Joseph MacKay, ‘Rethinking the IR theory of empire in late Imperial China’, International Relations of the Asia-Pacific (forthcoming).

2 Nomadic peoples generally have recently received attention in IR. Neumann, Iver B. and Wigen, Einar, ‘The importance of the Eurasian steppe to the study of International Relations’, Journal of International Relations and Development, 16 (2013); MacKay, Joseph et al., ‘Before and after borders: The nomadic challenge to sovereign territoriality’, International Politics, 51:1 (2014), pp. 101–123.

3 The word is, of course, contentious. ‘Barbarian’ has been use to translate multiple Chinese terms that may usefully parallel the equivalent Greek root, a practice I follow here, for simplicity’s sake. An alternative usage is simply ‘foreigner’. For a detailed discussion of these translation issues, see Nicola Cosmo, Di, Ancient China and Its Enemies: The Rise of Nomadic Power in East Asian History (Cambridge: Cambridge University Press, 2002), pp. 95–96, fn. 7.

4 Kang, , East Asia Before the West, p. 140.

5 The term is due to Fairbank, John K. (ed.), The Chinese World Order: Traditional China’s Foreign Relations (Cambridge, MA: Harvard University Press, 1968).

6 Mitzen, Jennifer, ‘Ontological security in world politics: State identity and the security dilemma’, European Journal of International Relations, 12:3 (2006), p. 341; Steele, Brent J., Ontological Security in International Relations: Self-Identity and the IR State (London: Routledge, 2007); Rumelili, Bahar (ed.), Conflict Resolution and Ontological Security: Peace Anxieties (London: Routledge, 2014).

7 Neumann, Iver B., ‘Self and other in International Relations’, European Journal of International Relations, 2:2 (1996), pp. 139–174. ‘Others’ have been associated with identity formation by philosophers and social scientists at least since Hegel. Georg Wilhelm Friedrich Hegel, Phenomenology of Spirit, ed. John Niemeyer Findlay, trans. Arnold V. Miller (Oxford: Clarendon Press, 1977).

8 Rumelili, Bahar, Constructing Regional Community and Order in Europe and Southeast Asia (Basingstoke: Palgrave, 2007). Nor, as Abizadeh has recently argued, does identity building require an other. Abizadeh, Arash, ‘Does collective identity presuppose an other? On the alleged incoherence of global solidarity’, American Political Science Review, 99:1 (2005), pp. 45–60.

9 Johnston, Alastair Iain, Cultural Realism: Strategic Culture and Grand Strategy in Ming China (Princeton, NJ: Princeton University Press, 1995); Hui, Victoria Tin-bor, War and State Formation in Ancient China and Early Modern Europe (Cambridge: Cambridge University Press, 2005); Wang, Yuan-Kang, Harmony and War: Confucian Culture and Chinese Power Politics (New York, NY: Columbia University Press, 2011).

10 The canonical account is Fairbank, The Chinese World Order. While historians increasingly question Fairbank’s framework, few doubt the system was usually hierarchical, and no alternative account has been widely accepted. For a survey of criticisms, see Hevia, James L., ‘Tribute, asymmetry, and imperial formations: Rethinking relations of power in East Asia’, Journal of American-East Asian Relations, 16:1–2 (2009), pp. 69–83.

11 This is not to say the regional order persisted for the entire imperial era – rather, it accrued gradually, as neighbouring states joined its Confucian value system. Kang describes it peaking during the Ming and Qing dynasties. Kang, East Asia Before the West.

12 Pines, Yuri, Envisioning Eternal Empire: Chinese Political Thought of the Warring States Era (Honolulu, HI: University of Hawaii Press, 2009), p. 1.

13 Kang, David C., “Hierarchy and legitimacy in international systems: The tribute system in early modern East Asia’, Security Studies, 19:4 (2010), pp. 593, 595. For a realist account, see Wang, Harmony and War.

14 East Asia Before the West, p. 140.

15 Ibid., p. 10.

16 Kelly, ‘A “Confucian long peace” in pre-Western East Asia?’ The peace was nonetheless puzzling. In contrast, the same period in Europe (roughly the fourteenth to nineteenth centuries) included all the Reformationary wars, including the Thirty Years War, as well as the French Revolutionary Wars and the Napoleonic Wars. It was an extraordinarily bloody period. Alternately, the East Asian peace was more than ten times as long as, and considerably less violent than, the original ‘long peace’ – the Cold War.

17 This literature belongs to a broader trend toward conceptualising non-Western historical international systems. See especially English School accounts, including Watson, , The Evolution of International Society: A Comparative Historical Analysis (London: Routledge, 1992), pp. 85–93; Buzan, Barry and Little, Richard, International Systems in World History: Remaking the Study of International Relations (Oxford: Oxford University Press, 2000), pp. 57–60. The subject has often arisen in the context of comparing international systems: Donnelly, Jack, ‘Rethinking political structures: From “ordering principles” to “vertical differentiation” – and beyond’, International Theory, 1:1 (2009), pp. 49–86, Donnelly, Jack, ‘The elements of the structures of international systems’, International Organization, 66:4 (2012), pp. 609–643; Ringmar, ‘Performing international systems’. See also Hobson, John M., The Eastern Origins of Western Civilisation (Cambridge: Cambridge University Press, 2004).

18 Mitzen, ‘Ontological security in world politics’; Steele, Ontological Security in International Relations.

19 Mitzen, , ‘Ontological security in world politics’, p. 346.

20 Ibid., pp. 353–63.

21 Steele, Brent J., ‘Ontological security and the power of self-identity: British neutrality and the American Civil War’, Review of International Studies, 31:3 (2005), pp. 519–540.

22 Jelena Subotić, ‘Narrative, ontological security, and foreign policy change’, Foreign Policy Analysis (2015). See also research by social psychologists – for example, Kinnvall, Catarina, ‘Globalization and religious nationalism: Self, identity, and the search for ontological security’, Political Psychology, 25:5 (2004), pp. 741–767.

23 Mitzen treats ontological (in)security as a property of states as unitary actors. Others, such a Krolikowski, are sceptical. I focus on Chinese elite individual efforts to construct a collective identity, eliding the methodological problem of treating states as unitary, but nonetheless focusing on the collective identity of the state. Subotic adopts a similar approach. Mitzen, ‘Ontological security in world politics’; Krolikowski, Alanna, ‘State personhood in ontological security theories of International Relations and Chinese nationalism: A sceptical view’, The Chinese Journal of International Politics, 2:1 (2008), pp. 109–133; Subotić, ‘Narrative, ontological security, and foreign policy change’.

24 Neumann, ‘Self and other in International Relations’, p. 165, emphasis in original. The article also reviews self/other logics in IR and social theory. See also Inayatullah, Naeem and Blaney, David L., International Relations and the Problem of Difference (London: Routledge, 2004).

25 See, for example, Rumelili, Constructing Regional Community and Order in Europe and Southeast Asia.

26 Steele, , Ontological Security in International Relations, p. 40.

27 Not all ontological security frameworks in IR emphasise self/other relations. While Mitzen makes no direct reference to it, the concept is important to Steele’s account. Ibid., pp. 26–7, 30–2.

28 Neumann, Iver B., Uses of the Other: The ‘East’ in European Identity Formation (Minneapolis, MN: University of Minnesota Press, 1999), pp. 218–219.

29 As Neumann and Wigen have recently argued, steppe societies have been ‘roundly neglected’ in IR, in terms of understanding relations both between steppe polities, and between steppe and settled peoples. Neumann and Wigen, ‘The importance of the Eurasian steppe to the study of International Relations’.

30 Lewis, Mark Edward, The Early Chinese Empires: Qin and Han (Cambridge, MA: Belknap Press, 2010), p. 1.

31 Ibid., p. 2.

32 Pines, Yuri, The Everlasting Empire: The Political Culture of Ancient China and Its Imperial Legacy (Princeton, NJ: Princeton University Press, 2012), p. 34.

33 Lewis, , The Early Chinese Empires, pp. 135–136, emphasis added.

34 Quoted in Lewis, Mark, ‘The Han abolition of universal military service’, in Hans van de Ven, Warfare in Chinese History (Leiden: E. J. Brill, 2000), pp. 45–46. The passage is from a biography of Chao in the Han shu or Book of Han, by the historian and court official Ban Gu.

35 Qian, Sima, ‘The world beyond China’, in Patricia Buckley Ebrey (ed.), Chinese Civilization: A Sourcebook (2nd edn, New York, NY: Free Press, 1993), p. 55.

36 Qian, Sima, Records of the Grand Historian of China, trans. Burton Watson (Hong Kong: Columbia University Press, 1993), p. 12. Sima Qian’s account of the nomadic other had historical parallels, including the Greek view of the Scythians, found in Herodotus. Stuurman, Siep, ‘Herodotus and Sima Qian: History and the anthropological turn in Ancient Greece and Han China’, Journal of World History, 19:1 (2008), pp. 1–40.

37 Cosmo, Di, Ancient China and Its Enemies, pp. 196–199.

38 The tian-xia concept has been revived as a normative political-theoretic programme by the Chinese philosopher Zhao Tingyang (‘Rethinking empire from a Chinese concept “all-under-heaven” (tian-xia),” Social Identities, 12:1 (2006), pp. 29–41; ‘A political world philosophy in terms of all-under-heaven (tian-xia)’, Diogenes, 56:1 (2009), pp. 5–18.) For critique, see Callahan, William A., ‘Chinese visions of world order: Post-hegemonic or a new hegemony?’, International Studies Review, 10:4 (2008), pp. 749–761. For a defence, see Ren, Xiao, ‘Traditional Chinese theory and practice of foreign relations’, in Yongnian Zheng (ed.), China and International Relations: The Chinese View and the Contribution of Wang Gungwu (London: Routledge, 2010), pp. 102–116.

39 Pines, Yuri, ‘Changing views of Tianxia in pre-imperial discourse’, Oriens Extremus, 43 (2002), pp. 101–116.

40 Pines, Yuri, ‘Beasts or humans: Pre-imperial origins of sino-barbarian dichotomy’, in Reuven Amitai and Michal Biran (eds), Mongols, Turks and Others: Eurasian Nomads and the Sedentary World (Leiden: Brill, 2004), pp. 90–91.

41 Lewis, , The Early Chinese Empires, p. 2.

42 Fiskesjö, Magnus, ‘On the “raw” and the “cooked” barbarians of Imperial China’, Inner Asia, 1:2 (1999), pp. 139–168.

43 Quoted in ibid., p. 143. ‘Li’ refers in general to uncivilised peoples of non-Chinese origin. Ibid., p. 158.

44 A Song dynasty historian took an analogous view of steppe peoples: ‘Let us not seek victory over them … They are like unto all manner of insects, reptiles, snakes, and lizards. How could we “receive them with courtesy and deference”?’ quoted in Wright, David C., ‘The northern frontier’, in David Andrew Graff and Robin D. S. Higham (eds), A Military History of China (Lexington, KY: University Press of Kentucky, 2012), p. 57.

45 This view was not limited to the steppe – it applied to multiple peripheral peoples. See Scott’s account of imperial relations with Southeast Asian hill tribes. Scott, James C., The Art of Not Being Governed: An Anarchist History of Upland Southeast Asia (New Haven, CT: Yale University Press, 2009). Tellingly, Scott references Barfield’s ‘shadow empires’ in framing his account. Barfield, Thomas J., ‘The shadow empires: Imperial state formation along the Chinese-Nomad frontier’, in Susan E. Alcock et al. (eds), Empires: Perspectives from Archaeology and History (Cambridge: Cambridge University Press, 2001). Only the steppe both received this rhetorical treatment and presented a military threat. Exceptionally, Ming and Qing authorities had difficulty dealing with pirate polities on the Chinese coast. Being relatively short-lived, institutionally adaptable, and opposed to Chinese rule, they had striking parallels with the steppe. MacKay, Joseph, ‘Pirate nations: Maritime pirates as escape societies in late Imperial China’, Social Science History, 37:4 (2013), pp. 551–573.

46 Even during the Yuan and later Qing nomadic conquest dynasties, the bureaucratic elites of the empire remained ethnically Chinese, passing through the meritocratic selection process of civil service exams. Nomads joined the imperial elite only through conquest. While nomads occasionally served in China’s northern armies, they did not lead them.

47 By this time, the states were states, in the sense of being centralised and bureaucratised. For example, civil service exams dated from the Han dynasty, and were formalised during the Tang dynasty (618–907). Woodside, Alexander, Lost Modernities: China, Vietnam, Korea, and the Hazards of World History (Cambridge, MA: Harvard University Press, 2009), pp. 1–2. East Asian state consolidation thus predates equivalent processes in Europe. It occurred in China, Korea, and Vietnam, and to a lesser extent in Japan, where a feudal warrior aristocracy persisted. Kang, , East Asia Before the West, pp. 25–53.

48 Quoted in Wang, Gungwu, ‘Ming foreign relations: Southeast Asia’, in Denis C. Twitchett and Frederick W. Mote (eds), The Cambridge History of China: Volume 8, The Ming Dynasty (Cambridge: Cambridge University Press, 1998), p. 12.

49 Brook, Timothy, The Troubled Empire: China in the Yuan and Ming Dynasties (Cambridge, MA: Harvard University Press, 2010), pp. 86, 8.

50 Rossabi, Morris, ‘The Ming and Inner Asia’, in Denis C. Twitchett and Frederick W. Mote (eds), The Cambridge History of China Volume 8: The Ming Dynasty 1368–1644 Part 2 (Cambridge: Cambridge University Press, 1998), pp. 222–223.

51 Quoted in Brook, , The Troubled Empire, p. 29.

52 Huang, Siu-chi, Essentials of Neo-Confucianism: Eight Major Philosophers of the Song and Ming Periods (Westport, CT: Greenwood Press, 1999), p. 4.

53 Brook, , The Troubled Empire, pp. 182–185.

54 Fairbank, The Chinese World Order.

55 Mackinder viewed nomads as akin to natural disasters occasionally visited upon sedentary civilisations. Mackinder, H. J., ‘The geographical pivot of history’, The Geographical Journal, 23:4 (1904), pp. 421–437. Lattimore, while more nuanced, sharply distinguished between settled and steppe peoples, emphasising environmental factors. Lattimore, Owen, Inner Asian Frontiers of China (New York, NY: American Geographical Society, 1940). For recent discussion, see Rogers, J. Daniel, ‘Inner Asian states and empires: Theories and synthesis’, Journal of Archaeological Research, 20:3 (2012), pp. 209, 215–16; Christopher I. Beckwith, Empires of the Silk Road: A History of Central Eurasia from the Bronze Age to the Present (Princeton, NJ: Princeton University Press, 2009), pp. xxi–xxv.

56 Barfield, Thomas J., The Perilous Frontier: Nomadic Empires and China 221 BC to AD 1757 (Cambridge, MA: Blackwell, 1989); Barfield, ‘The shadow empires: Imperial state formation along the Chinese-Nomad frontier’.

57 For a critical review of dependency, see Rogers, , ‘Inner Asian states and empires’, pp. 214–217.

58 Chiefly Sima Qian, Records of the Grand Historian of China.

59 Rogers, , ‘Inner Asian states and empires’. For a useful review of archaeological literature on the region, see Philip L. Kohl, The Making of Bronze Age Eurasia (Cambridge: Cambridge University Press, 2007). For a systematic account of China’s interaction with these polities, see Di Cosmo, Ancient China and Its Enemies. For a review of recent archaeological evidence, see Hanks, Bryan K., ‘Archaeology of the Eurasian steppes and Mongolia’, Annual Review of Anthropology, 39:1 (2010), pp. 469–486. For additional synthesis on Mongol hierarchy during the period, see Kradin, Nicolay N., ‘Heterarchy and hierarchy among the ancient Mongolian nomads’, Social Evolution & History, 10:1 (2011), pp. 187–214.

60 Houle, Jean-Luc, ‘Socially integrative facilities and the emergence of societal complexity on the Mongolian steppe’, in Bryan K. Hanks and Katheryn M. Linduff (eds), Social Complexity in Prehistoric Eurasia: Monuments, Metals and Mobility (Cambridge: Cambridge University Press, 2009), p. 371.

61 Shelach, Gideon, ‘Violence on the frontiers? Sources of power and socio-political change at the easternmost parts of the Eurasian steppe during the late second and early first millennia BCE’, in Bryan K. Hanks and Katheryn M. Linduff (eds), Social Complexity in Prehistoric Eurasia: Monuments, Metals and Mobility (Cambridge: Cambridge University Press, 2009), pp. 241–271.

62 Frachetti, Michael D., ‘Multiregional emergence of mobile pastoralism and nonuniform institutional complexity across Eurasia’, Current Anthropology, 53:1 (2012), pp. 2–38.

63 Murphy, Eileen M. et al., ‘Iron Age pastoral nomadism and agriculture in the Eastern Eurasian steppe: Implications from dental palaeopathology and stable carbon and nitrogen isotopes’, Journal of Archaeological Science, 40:5 (2013), pp. 2547–2560.

64 Numbers vary with the nebulous borders of Inner Eurasia as a region, and with the definition of ‘empire’. Rogers finds 15 large-scale steppe polities. Turchin finds 13 Central and Inner Asian ‘“mega” empires’, although not all were consistently nomadic. Rogers, , ‘Inner Asian states and empires’, p. 208; Turchin, Peter, ‘A theory for formation of large empires’, Journal of Global History, 4:2 (2009), p. 202.

65 Cosmo, Nicola Di, ‘State formation and periodization in Inner Asian history’, Journal of World History, 10:1 (1999), p. 14; see also Di Cosmo, Ancient China and Its Enemies.

66 Honeychurch, William, ‘The nomad as state builder: Historical theory and material evidence from Mongolia’, Journal of World Prehistory, 26:4 (2013), p. 284.

67 Quoted in Golden, Peter B., Central Asia in World History (Oxford: Oxford University Press, 2011), p. 28.

68 Honeychurch, , ‘The nomad as state builder’, pp. 313–314.

69 Hansen, Valerie, The Open Empire: A History of China to 1600 (New York, NY: W. W. Norton & Co, 2000), pp. 5–6.

70 Rogers, , ‘Inner Asian states and empires’, p. 208.

71 Carneiro, Robert L., ‘A theory of the origin of the state’, Science, 169:3947 (1970), p. 735.

72 Barfield, , The Perilous Frontier, pp. 24–26.

73 Sneath, David, The Headless State: Aristocratic Orders, Kinship Society, & Misrepresentations of Nomadic Inner Asia (New York, NY: Columbia University Press, 2007).

74 Rogers, , ‘Inner Asian states and empires’, p. 213.

75 Honeychurch, William, ‘Alternative complexities: The archaeology of pastoral nomadic states’, Journal of Archaeological Research, 22:4 (2014), p. 277. Many innovations were technological, including horse riding and the chariot. Anthony, David W., The Horse, the Wheel and Language : How Bronze Age Riders from the Eurasian Steppes Shaped the Modern World (Princeton, NJ: Princeton University Press, 2007).

76 Murphy et al., ‘Iron Age pastoral nomadism and agriculture in the Eastern Eurasian steppe’; Honeychurch, , ‘Alternative complexities’, p. 304.

77 Barfield, The Perilous Frontier.

78 Honeychurch, , ‘Alternative complexities’, pp. 292, 294, 310.

79 Ibid., p. 306.

80 Sneath, The Headless State, pp. 23, 181.

81 Roads plural – multiple routes intersected, overlapped, and varied over time. Parzinger, Hermann, ‘The “Silk Roads” concept reconsidered: About transfers, transportation and transcontinental interactions in prehistory’, The Silk Road, 5:2 (2008), p. 7.

82 Honeychurch, , ‘Alternative complexities’, pp. 295, 307–8; Honeychurch, William, Inner Asia and the Spatial Politics of Empire: Archaeology, Mobility, and Culture Contact (New York, NY: Springer, 2015), pp. 277–278; see also Beckwith, , Empires of the Silk Road, pp. 26–28.

83 Morgan, David, The Mongols (2nd edn, London: Blackwell, 2007), p. 5. Mongol Imperial historiography presents unique research challenges, including ‘a bewildering array of languages, of which Chinese and Persian are undeniably the most important but which also include Latin, Greek, Russian, Georgian, Armenian, Arabic, Syriac, Tibetan, Korean and even (to some extent) Mongolian.’ Jackson, Peter, ‘The Mongol Empire, 1986–1999’, Journal of Medieval History, 26:2 (2000), p. 190. Jackson nonetheless usefully reviews recent research. The key primary source concerning the empire itself is Urgunge Onon, trans., The Secret History of the Mongols: The Life and Times of Chinggis Khan (London: Routledge, 2011). It’s historical accuracy remains disputed. Morgan, The Mongols.

84 Skrynnikova, Tatyana D., ‘Relations of domination and submission: Political practice in the Mongol Empire of Chinggis Khan’, in David Sneath (ed.), Imperial Statecraft: Political Forms and Techniques of Governance in Inner Asia, Sixth-Twentieth Centuries (Bellingham, WA: East Asian Studies Press, 2006), p. 86.

85 Allsen, Thomas T., ‘Technologies of governance in the Mongolian empire: A geographic overview’, in David Sneath (ed.), Imperial Statecraft: Political Forms and Techniques of Governance in Inner Asia, Sixth-Twentieth Centuries (Bellingham, WA: East Asian Studies Press, 2006), p. 123.

86 Sneath, David (ed.), ‘Imperial statecraft: Arts of power on the steppe’, Imperial Statecraft: Political Forms and Techniques of Governance in Inner Asia, Sixth-Twentieth Centuries (Bellingham, WA: East Asian Studies Press, 2006), p. 19; Allsen, , ‘Imperial statecraft’, pp. 37–38.

87 Darwin, John, After Tamerlane: The Rise and Fall of Global Empires, 1400–2000 (New York, NY: Bloomsbury, 2009), pp. 4–6.

88 Rogers, , ‘Inner Asian states and empires’, pp. 236–237; Perdue, Peter C., China Marches West: The Qing Conquest of Central Eurasia (Cambridge, MA: Harvard University Press, 2005).

89 Atwood, Christopher P., ‘Titles, appanages, marriages, and officials: A comparison of political forms in the Zünghar and thirteenth-century Mongol Empires’, in David Sneath (ed.), Imperial Statecraft: Political Forms and Techniques of Governance in Inner Asia, Sixth-Twentieth Centuries (Bellingham, WA: East Asian Studies Press, 2006), p. 209.

90 On Qing conquest, see Perdue, , China Marches West. On late steppe political structures, Lhamsuren Munkh-Erdene, ‘The 1640 Great Code: An Inner Asian parallel to the treaty of Westphalia’, Central Asian Survey, 29:3 (2010), pp. 269–288.

91 Thus, for example, Honeychurch cites positively elements of Lattimore’s and Barfield’s arguments.

92 These new accounts have not been accepted without debate. Golden, Peter B., ‘Review of the headless state: Aristocratic orders, kinship society, and misrepesentations of nomadic Inner Asia. By David Sneath’, The Journal of Asian Studies, 68:1 (2009), pp. 293–296.

93 A strong variant of the revisionist account appears in Beckwith, Empires of the Silk Road. Beckwith posits long historical cycles of conflict between Central Eurasia and polities at their continental peripheries, in Europe and East Asia. Inner Asian empires, he argues, created economic exchange networks that benefitted core and peripheries alike (the Silk Road being the most famous), but which were periodically destroyed by conquering peripheral empires. He thus offers not an account of Inner Asian history alone, so much as a geographical theory of world history – or, at least, a large part of it. It is thus not necessarily conducive to assessing the specifics of China-steppe relations. His synthesis has also been somewhat controversial – even proponents of the revisionist view do not wholly follow it. (See, for example, Cosmo, Nicola Di, ‘Empires of the Silk Road: A history of Central Eurasia from the Bronze Age to the present, by Christopher I. Beckwith’, Journal of Global History, 4:3 (2009), pp. 510–512.) I follow some of Beckwith’s specific observations, but not his general theory.

94 Neumann and Wigen, ‘The importance of the Eurasian steppe to the study of International Relations’; Neumann, Iver B., ‘The steppe and early European state formation’, Uluslararasi Iliskiler / International Relations, 8:30 (2011), pp. 3–12.

95 Wright, , ‘The northern frontier’, pp. 64–66.

96 Lewis, Mark Edward, China’s Cosmopolitan Empire: The Tang Dynasty (Cambridge, MA: Harvard University Press, 2009), pp. 147–148.

97 Brook, , The Troubled Empire, p. 27.

98 Allsen, Thomas T., ‘The rise of the Mongolian empire and Mongolian rule in North China’, in Denis C. Twitchett, Herbert Franke, and John King Fairbank (eds), The Cambridge History of China: Volume 6, Alien Regimes and Border States, 710–1368 (Cambridge: Cambridge University Press, 1994); Morgan, The Mongols.

99 Brook, , The Troubled Empire, p. 27.

100 Ibid., p. 81.

101 Rossabi, , ‘The Ming and Inner Asia’, p. 275.

102 Bol, Peter K., Neo-Confucianism in History (Cambridge, MA: Harvard University Asia Center, 2008), p. 216.

103 Rawski, Evelyn S., ‘Presidential Address: Reenvisioning the Qing: The significance of the Qing period in Chinese history’, The Journal of Asian Studies, 55:4 (1996), pp. 829–850. See also Crossley, Pamela Kyle, A Translucent Mirror: History and Identity in Qing Imperial Ideology (Berkeley, CA: University of California Press, 1999); Elliott, Mark C., The Manchu Way: The Eight Banners and Ethnic Identity in Late Imperial China (1st edn, Stanford, CA: Stanford University Press, 2001). For a sceptical view, see Ho, Ping-Ti, ‘In defense of sinicization: A rebuttal of Evelyn Rawski’s “Reenvisioning the Qing”’, The Journal of Asian Studies, 57:1 (1998), pp. 123–155.

104 Nexon, Daniel H. and Wright, Thomas, ‘What’s at stake in the American Empire debate’, American Political Science Review, 101:2 (2007), p. 264.

105 Waley-Cohen, Joanna, ‘The new Qing history’, Radical History Review, 88:1 (2004), pp. 194–195.

106 Ibid., p. 199.

107 Millward, James, Eurasian Crossroads: A History of Xinjiang (New York, NY: Columbia University Press, 2009), pp. 97–101.

108 Quoted in Newby, Laura, The Empire And the Khanate: A Political History of Qing Relations With Khoqand c1760–1860 (Leiden: Brill, 2005), p. 29.

109 Perdue, Peter C., ‘China and other colonial empires’, Journal of American-East Asian Relations, 16:1–2 (2009), p. 93.

110 Larsen, Kirk W., ‘The Qing empire (China), imperialism, and the modern world’, History Compass, 9:6 (2011), p. 499.

111 Perdue, China Marches West.

112 Waldron, Arthur, The Great Wall of China: From History to Myth (Cambridge: Cambridge University Press, 1990), p. 6.

* I would like to thank Simon Pratt, Christopher David LaRoche, Lincoln Rathnam, three extremely helpful reviewers, and the editors of the RIS for help in preparing this article.