Introduction
The Roman and the Han empires have been compared in recent publications with the purpose of understanding the overall conditions of their emergence and sustainability over centuries. They were remarkably similar in size, organization, and in their history, emerging as vast unified territorial states from a mix of competing city-states, kingdoms, and stateless societies, and succumbing to the pressure of warlords and external military forces who undermined the bargains struck between central government, local government, and local elites.Reference Scheidel1 The administrative and military structures of the Roman Empire as well as its artistic, literary, and religious practices have provided a template for Christian empires both east and west throughout European history albeit in very different ways.Reference Bang2 Similarly, not only Chinese but also East Asian and Central Asian states adopted and adapted Han models of imperial rule during the last two millennia. The memory of classical empires has been prominent in both Chinese and European history but it is obvious that it has had a different imprint.
The Han territories were reunified in part and periodically at first (Eastern Jin, Sui, Tang, Song dynasties) and were more consistently ruled as unified empires from the 13th century onwards.
The Eastern Roman or Byzantine Empire lost its control of Northern Africa, Syria, and Palestine to the Arabs in the 630s, and could no longer resist Seljuk expansion in Anatolia and the Balkans after 1071. Its capital Constantinople finally succumbed to the military superiority of the Ottoman Empire in 1453. In medieval Western Europe, the Carolingian and the Holy Roman empires boasted of being renewals of the glorious ancient models, but they developed in a different environment, were no longer built on the Roman scale and only borrowed selectively from the Roman repertoire.
In Europe, the memory of the Roman Empire has been cherished until deep into the 20th century. It remained an ideal to be imitated in its imaginary and in many of its structural aspects, such as its system of law. Its language is still the standard in the Catholic Church, and – until the 18th century – also in university teaching and science. Architectural models from Roman times were time and again imitated to demonstrate the age-old solidity of the institutions they housed. The huge cupola church that Emperor Justinian commissioned in Constantinople in the 530s became the model not only for churches in the Orthodox tradition, but also for mosques in large parts of the Islamic world. The seven hills of Rome found their replicas in Constantinople, Lisbon, Bergen, Moscow, and many other cities. The Catholic and Orthodox Churches are the largest institutions directly continuing the traditions of the Late Roman Empire until the present day. However, beyond the cultural sphere, imperial power had only a limited impact in Western and Central Europe, while the Ottoman and Russian empires were harsh realities in the East.
Empires survived in Central and Eastern Europe until the 20th century, while in the West they were impotent or short-lived. In the Early Modern period, the Ottoman, the Holy Roman/Habsburg/Austro-Hungarian, and the Russian Empires competed for centuries, until their paroxysmal role in the Great War accelerated their demise. Charles Tilly observed that empires could only grow and remain consolidated in the least populated, least urbanized, least commercialized and latest industrialized parts of the continent. They are characterized by the lowest per capita GDP. Resources were scarce and taxes hard to extract from subjects scattered across wide spaces.Reference Tilly3 The contrast with the Chinese empire seems total.
In this essay we examine how differences in power relationships, fiscal regimes, and territoriality help explain both the peripheral impact of the classical model in the European context and the enhanced prospects for it in Chinese history from the 12th century onwards. We will therefore not only compare the contemporary Han and Roman empires, but also their successors in both continents through the 18th century.
The Classical Model
Before addressing the tensions that impacted on the maintenance of empire and the solutions that were developed in postclassical Chinese and European history, let us briefly set out the main features of the classical model.Reference Burbank and Cooper4 In both Rome and Han China vast territories were brought under the control of one person, the emperor, who oversaw from a large capital the administration of over 1000 districts. The centralization of power in the person of the emperor and in the capital was made possible through the cooperation of bureaucratic and aristocratic elites and military governors who managed the mobilization of large armies and the extraction of resources from the predominantly farming population. With the manpower and resources thus mobilized, classical imperial governments invested heavily in the development of infrastructure including transportation (waterways and highways), coinage, legal systems, and common languages – some of it still visible to this day. Markets and towns grew in number and size and the state sought to interfere in their operation. Classical empires made claims to universal power and expressed such claims in monumental art and architecture, in new literary and historiographical genres, and in religious discourse. Much of this kind of cultural production was controlled by a small elite of cultured men who contributed to the formation of classical canons. Classical empires faced threats from other military powers on their peripheries, differentiated themselves from non-civilized others, but also created ways for incorporating different ethnicities and social statuses in legal systems, civilizational discourses, and legitimating ideologies.
There were convergent trends in the development of imperial government in Han China and Rome but there were also significant differences between the Roman and Han models. Walter Scheidel and Jane Burbank point to more local autonomy and more power to the military in the Roman case and to greater central control over cities and state agents and the emergence of elites more adept at and interested in imperial administration in the Han Chinese case.Reference Scheidel5 The fact that the first Chinese empire was based on the unification of well-organized competing states, while the Roman empire developed out of a republic that had gradually extended its territorial domination, may well explain these differences.
As early as the fifth century BCE, states located along the Yellow River launched programmes to improve agricultural productivity without destroying the livelihood of family-based peasant freehold. Large-scale irrigation and land reclamation projects, the facilitation of iron implements, encouragement of multiple crops, combined with the setting-up of granaries protecting peasants against harvest fluctuations, all contributed to the high land-productivity, which allowed the states to extract ever more services and taxes from the population. The mobilization of hundreds of thousands of workers for large-scale construction activities such as the Grand Canal, the new capital cities, and the Great Wall under the first empires would have been impossible without this foundation.Reference Hui6 In the second century CE, the capital city Luoyang had an estimated population of half a million, which is comparable size to Rome in the same period. Even if a sizeable proportion of the Chinese lived in poor conditions and had to perform conscript military service and corvée, they were not slaves. Instead, they were encouraged to improve productivity. This contrasts with the Roman economy in which massive slave labour did not provide any incentive in that direction.
There is no reason to assume that the above-mentioned differences and other differences in scale can explain why the legacies of the Han Chinese and Roman empires would turn out to be so different in the ensuing centuries. Hindsight should not blind us to the challenges that aspiring empire builders faced on both ends of the Eurasian continent. In the 1000 years following the fall of the Han, its former territories were under unified rule for about 300 years. This trend coincides with the findings of the much acclaimed 11th-century historian Sima Guang who wrote in 1061: ‘In these 1700 or so years [from the move of the Eastern Zhou capital in the eighth century BCE until the foundation of the Song] there have only been 500 or so in which the realm was united.’7 Sima Guang’s words, addressed to the then ruling emperor, Renzong, suggest that imperial reunification remained a powerful and deeply rooted ideal among top-ranking politicians in the 11th century, but these words equally show that up until the 10th century this ideal had seldom been realized and in any case when it had it was for a period no longer than one third of recorded history.
Yet, in the centuries that followed, Sima Guang’s observation about the longue durée of imperial Chinese history would no longer hold. During these centuries a number of factors contributed to a trend in which reunification of the territories ruled by the classical empires became a more attainable and a more desirable goal for Chinese than for European military and political elites. The latter had never shared any other common ideology than their Christian belief, and even that motivated them to bloody internal persecutions and wars against so-called heretics, especially in the 16th and 17th centuries.
Central Control and Local Power
With regard to the sustainability of central control, post-Han imperial regimes confronted the same underlying tensions as their Qin and Han predecessors. The main challenges to the continuity of imperial power came from both internal and external sources. Internally, the history of the Qin and Han empires had demonstrated that tensions between the inner court (imperial family and in-laws, eunuchs, and other personal attendants) and the outer court (central government officials), between the central and local bureaucracies, and between the government and local power holders and communities could escalate and lead to an irreversible breakdown of unified central control. Measures to manage and reduce those tensions were recalled from the dynastic histories and new ones implemented under the Sui, Tang, Song, and other succeeding empires. Measures included the incorporation of princely domains over time, the stationing of imperial family members away from the centre of power, prohibitions on the serving of imperial family members and members of the inner court in office, an emphasis on the open and impartial consultation of officialdom by the sovereign, short tenures and regular rotation of bureaucrats, avoidance of tenure in officials’ places of origin, the regular monitoring of officials’ performance and the institutionalization of petitioning mechanisms to counter official corruption. The incorporation of these and other measures in administrative manuals addressed to crown princes and officials alike and their standardization over time solidified the foundations laid down for the centralization of power in classical times.
From the history of the Tang dynasty, policy advisors and Chinese readers of history drew the conclusion that administrative centralization could work to great effect and yield the best model for imperial government but the fragmentation of power and the territorial disintegration between the mid-eighth and mid-tenth century had shown the continuing threat of military power in the periphery. In the early empires the mass conscript armies that had been waging wars across the interior of the Chinese territories in the centuries preceding unification had been demobilized and deployed along the border regions. Massive conscript and mercenary armies similarly defended and expanded Tang borders deep into Central Asian territory but the power and discretion accorded to Tang generals, many of whom were of foreign origin, was later seen as a principal cause for the mid-eighth century crisis of power and the gradual decentralization and fragmentation that ensued. Even though those implementing centralizing measures under the Song Dynasty remained wary of military autonomy and were less prone to court displays of martial power, border policy and imperial self-representation under the last three dynasties suggest that late imperial courts accepted that their capacity to project power from the centre depended on their ability to keep out external forces (especially along the vast northern and north-western borders) or to neutralize the threat they posed by integrating them in a larger multi-ethnic empire.8
The tensions discussed so far were not only common to the first Chinese empires but arguably to all pre-industrial complex states; their negotiation was, in the case of Chinese history, significantly impacted by one of the major social developments in imperial Chinese history: the expansion and enduring localization of political elites. Up until the 11th century, political elites tended to be concentrated in the metropolises that served as imperial capitals. Conquerors moved rival families into their capitals either forcibly or by luring them. During the Tang Dynasty, aristocratic families were ranked and a relatively small number of them dominated politics by virtue of their ancestry. The regional warlordism that started off in the mid-eighth century led to the decline of the old aristocratic families, and the centralization policies devised by the Northern Zhou and Song emperors set off a chain of transformations that, in large part through their unforeseen consequences, led to a new and far more stable relationship between imperial court and political elites.
The institutionalization of the civil service examinations and the growing importance accorded to them led to a major increase in the number of political elites. The major increase in numbers (20,000 to 30,000 prefectural examination candidates in the early 11th century and 400,000 or more by the mid-13th)Reference Chaffee9 suggests that these increases outpaced demographic trends. Examination candidates were political elites in the sense that their preparation involved them in debates about current affairs, that they constituted a pool from which officials at all levels were drawn, and that participation and success in the examinations also translated in their involvement in the government of local communities as middlemen between local government and the population.
While the localization of political elites was most likely not unprecedented and may to some extent be part of the development of imperial powers (Han historians have also noted localizing tendencies),Reference Brown10 a consensus has emerged amongst Chinese historians in recent decades that political elites settled down in the provinces far more visibly than before and with lasting effect from the 12th century onwards.Reference Hartwell11 Apart from the growing numbers of political elites (and the lack of corresponding growth in the numbers of positions in officialdom), the crisis of the 1120s also played a critical role in consolidating the trend towards localization. The Song court lost the northern half of its territories, bringing about the relocation of the court and tens of thousands of subjects to various places in the south.
Did the adoption of localist strategies and the celebration of local elite involvement in local gazetteers, biographies, collected writings, or Neo-Confucian texts, which started circulating in ever-growing numbers from the12th century onwards lead to de-imperialization (i.e. a process of decline in the concentration and centralization of power in the imperial capital)? Some have suggested that such is the case.Reference Tackett12 Yet the rest of Chinese history suggests that the continuing expansion and localization of political elites in the Ming and Qing dynasties did not coincide with a decline in the ability of courts to maintain large empires. Rather the opposite, localization and the consolidation of unified imperial rule appear to be positively correlated. As political elites settled down locally they contributed towards the further extension of state power in the provinces. Local governments and local elites relied upon each other and collaborated in the development of local infrastructure. Political elites remained, moreover, tied to broader networks of peers and continued to cultivate an interest in matters of empire-wide significance. Their local as well as supra-local ties and interests posed threats as well as opportunities to late imperial courts. Song, Ming, and Qing courts oscillated between moments of relaxation and moments of restriction of the parameters of elite political participation and debate, as authoritarian governments past and present have tended to do. As the history of elite cultural production suggests, successive courts tended to accept that elite information needs would be accommodated by private and commercial printers and that, with persistent central monitoring and targeted prosecutions, elite participation could bolster rather than undermine the stability of the center.Reference De Weerdt13 The high degree of centralization in late imperial times was recognized by some outstanding critics who advocated a return to a feudal system rather than a centralized administrative hierarchy under the emperor.Reference Tsung-hsi and de Bary14
The cultural (literati) elite constituted the most influential subgroup among the political elite but other groups, including religious leaders, merchants, or the so-called local bullies were also powerful stakeholders at the local level, and – to a lesser extent – at the supra-local level. On the whole, the literati community was a closer structural equivalent to the church in medieval Europe than the Buddhist, Daoist, or other religious institutions and communities in imperial China. The literati community bore little resemblance to the hierarchical and cross-cultural organization that was the church. It did not have a head who could disagree openly with monarchs, threaten them, and thus reveal that the tension between church and state was always real. Yet, as an amalgam of cross-regional networks of men inside and outside government, the literati community ensured that actors outside the court and the bureaucracy could exert political influence at different levels. There were few formal channels through which literati opinion could be collected. The role of schools herein remained uncertain and in any case insufficient. In the eyes of Philip Kuhn it is this lack of a constitutional agenda for the incorporation of the political elite that has remained a persistent tension in imperial and modern Chinese history.Reference Kuhn15
In Europe, several successful rulers conquering a number of peoples fashioned themselves as emperors. Cnut, king of Denmark, England, and Norway, and overlord of the Scots (1016–1035) labelled himself emperor, as did King Alfonso VII of Castile one century later, without either of them being able to consolidate this claim. The most famous case, with the longest lasting effects, is Charlemagne’s coronation by Pope Leo III in Rome on Christmas Day 800. The acclamation by the people of Rome confirmed him in his new dignity. The empire was named ‘Roman’ or ‘of the Romans’, although its territorial basis showed little congruity with its ancient model, and its core lay in Northern Gaul. The pope had implored Charlemagne’s support against rivalling powers in Italy. He was then king of the Franks and the Langobards, and the strongest ruler in the West. Therefore, he and his successors were entitled to the particular position of guardian of the Church. As the anointed, he became a sacral person.
The penetration of the Carolingian and Holy Roman Empires into the societies they pretended to dominate was very weak as a consequence of the very low level of surplus accumulation and concentration. By bare necessity, the exercise of power was directly personal, physical, and local, its source was the possession of landed domains from which surpluses were extracted through coercion. Domains could be inherited, conquered by military means, or granted in fief as the material counter-value in a bilateral personal relationship, in return for loyal services and advice, possibly related to an office. Given the difficulty of bulk transport, the low levels of commercialization and monetization of the economy in the German part of the empire, the military aristocracy (Ritterschaft) needed to live on their domains and to consume its returns locally. Even the highest-ranking nobility and the emperors themselves had to travel from one domain to the other, manifesting the reality of their rank by the number of followers to be fed and entertained, and trying to make their authority felt by holding courts of justice and mediating conflicts, of which the outcome might be consolidated in ceremonies of swearing oaths, and written down in charters. This form of traveling court (Reisekönigtum) did hardly lead to the building of palaces and even less to capital cities, especially since dynasties were relatively short-lived and held their domains in different regions. The most impressive architectural remains left by the Roman emperors were the metropolitan domes built in various cities along the Rhine, where the Salian dynasty (1024–1125) settled their mausoleums.
The exercise of power was thus founded on the domain economy, which provided the income required to uphold the elements of public authority held by the lords: the right to command the people who were bound to live and work on the domain, to appropriate their customary services and deliveries in kind by which the lord’s military life style had to be supported, and to administer the law on the basis of local customs. Great lords such as dukes may have held large patrimonies and commanded great numbers of vassals and serfs, but none of them ever concentrated a patrimony allowing them to outclass all or most of their competitors. In other words, the Holy Roman Empire had no stable centre of power, and the largest landowners ran the risk of losing control over their vassals as well as their domains, since landed property was by definition localised and could not be concentrated. The natural tendency of the lords was to try by all means to expand their domains, to consider the fief as heritable private property, and to care as little as possible about their distant suzerain(s).
The sheer extension of the Holy Roman Empire, the low level of capital accumulation, the lack of concentration and the elective character of German kingship led inevitably to centrifugal tendencies. These were aggravated by the ambition to rule over rapidly urbanizing Italy as well, which detracted resources form Germany without any other lasting effect than the weakening of the emperor’s position, as he had to bargain for support by making concessions to the various local and regional power holders. There was no concept of public authority, hardly any tangible means connected to the empire as such, and no officialdom committed to the empire: all these features made the emperors depend entirely on their dynasty’s resources (Hausmacht). Power was personal, the abstract notion of empire hardly more than a symbol.
The only professionally trained officials were clerics, from the archbishops down the whole hierarchy, but their training was aimed at targets other than administrative ones. Their loyalty to the empire was seriously challenged during the investiture contest (1075–1122) by which the popes tried to get a full grasp on church personnel and to impose a hierocratic ideology to which no adequate secular response could be formulated. During this conflict, and until well into the 14th century, successive popes intentionally undermined the emperor’s position by calling on the vassals to break their oaths of loyalty towards him, and by deposing and excommunicating some of them. In sum, the comparison between the Tang and Song empires, on the one hand, and the Carolingian and Holy Roman empires, on the other, leads to a clear conclusion: the latter lacked all the means of consolidated power beyond temporary military mobilizations.
The Fiscal Basis
Chinese imperial states tried and implemented different fiscal models. New models were developed in response to military threats, economic developments, and the need to balance the cost-effectiveness of taxation against the negative impact of middlemen. Judging from scanty records, per capita revenue appears to have been at about the same levels during the Han and Roman empires, even though it is likely that Han land and poll taxes were spent more on maintaining a larger bureaucracy and sub-bureaucracy and less on subsidizing aristocrats and the army than was the case in Rome.Reference Scheidel16 The Sui and Tang governments continued to rely upon poll taxes. Military costs were initially kept down through the adoption of the equal field system inherited from the Northern Wei regime, which had ruled large parts of the north between 386 and 534. Under this system, families provided military service on a rotating basis and continued to provide for themselves by farming government land. In the area of fiscal organization major changes took place in central government penetration and in the relationship between central and local power during the period covering the eighth and the eleventh centuries.
Tang rulers and military governors supplemented the regular armies with mercenaries. Increasing military costs and, following the court’s loss of control over several provinces from the mid-eighth century onwards, difficulty in registering the population and collecting poll taxes were some of the reasons behind the implementation of a new tax system in the late eighth century. In this new policy, called the two-tax method, taxes were assessed on the amount and quality of land owned and were collected twice a year following harvests in the summer and fall. Even though implementation took time and discrepancies in tax collection amongst localities always remained, the two-tax policy set off a long-term trend away from poll taxes and towards a property-based system of tax assessment. This trend persisted through the end of the imperial period. The property-based system meant that the state had given up on the ideal of an egalitarian system based on family size and equitable land distribution and aimed to raise revenue by taxing wealth.17
This rethinking of the foundation of state revenue also translated in the more aggressive collection of commercial taxes. Commercial taxes included both state monopolies in basic necessities (especially salt but also tea and liquor) and transit and sales taxes on goods traded by merchants. Monopolies became a major source of income for the late Tang government and became the ground of operation of fiscal experts rather than the regular bureaucracy. The technique was also exploited by forces competing with the Tang court. In the ninth and tenth centuries, the wealth generated by monopolies administered by fiscal experts formed the basis of several regional governments. From the tenth century onwards, the Song government fully exploited the potential of commercial taxes in the creation of a strongly centralized empire. It did so to such an extent that some economic historians perceive in the Song model an early articulation of the mercantilist state.Reference Guanglin18
A comparison between the revenue of the Tang state in the 740s and the Song state in 1077 demonstrates the extent to which the fiscal basis of the Song Empire differed from its predecessors. Land taxes were nearly exclusively collected in grain (25,000,000 shi) and in textiles (27,000,000 bolts) in the mid-eighth century. Even though payment in kind remained standard in 1077 (17,887,000 shi and 267,000 bolts) a substantial part of the land taxes were collected in cash (5,646,000 strings). The monetization of the economy was even more pronounced in the predominance of indirect taxes (2/3) collected in cash in the Song revenue structure (42,484,000 strings of cash in 1077).19
Song expenditures suggest that a motivating force behind the state’s strategy to tap commercial taxes were high military costs. These costs had risen sharply since the mid-eighth century given the Song decision to rely on mercenary forces rather than conscripts and to vastly increase the size of the army from around 0.5 million in the mid-eighth century to 1.2 million three centuries later. Comparing these figures with the size of Roman armies reveals the unprecedented scale of the growing capacity to extract resources on the part of Chinese regimes: 375,000 around 200 and 0.5 million c. 350.Reference Mitchell20 The size and cost of maintaining this army was a matter of concern to many Song political elites, not surprisingly given that military costs occupied an estimated 70–80% of the total budget.21 The Song government was able to sustain these costs (as well as substantial payments to competing states in the north) for extended periods of time. Its ability to do so depended on its capitalization on commercial and urban income. Commercial taxes were disproportionately raised from major urban centres. A large number of tax stations were set up across the empire to enforce transit and sales taxes. The ability of local governments to retain resources at the local level and the impact of mercantilist policies on commerce are matters of debate and were subject to variation over time and space but for now it appears safe to conclude that, overall, Song fiscal policies did not lead to a stifling of commercialization (given the government’s continued ability to tax trade) or to a decline in local government resources (given local government involvement in local infrastructure).
Subsequent imperial regimes abandoned the mercantilist model of the Song and returned to a physiocratic model. During the Ming and Qing dynasties land taxes once more became the mainstay of state revenue. The share of indirect taxes declined precipitously and per capita land and indirect taxes also went down.Reference Guanglin22 This long-term shift can be partly explained by the loss of confidence in the Song model, which resulted in hyperinflation as soon as tax revenue did not suffice to cover military expenses and the court printed paper money that could no longer be backed up in cash. It also resulted from the early Ming state’s imposition of an autarkic model of governance in which local communities were organized to be self-reliant and in which subjects were locked into social status categories by the state. Even though the second wave of commercialization and urbanization in the 16th century rendered these categories out of date, the physiocratic model of tax collection remained by and large in place.Reference Huang23
The difficulties faced by successive imperial governments in collecting land taxes are well known. The relatively small size of the bureaucracy meant that tax collection relied on local intermediaries. Even during the Song, when increased tax revenues could theoretically have been used to extend the size of the bureaucracy and sub-bureaucracy, local governments in rural areas continued to rely on middlemen, which resulted in far lower tax yields and efficiency in these areas than in the more urbanized areas where government-operated tax stations were active. In order to cope with relatively low rural tax rates and the underfunding of the bureaucracy, local state agents designed various kinds of surcharges. The excessive collection of such surcharges led to repeated tax revolts and threatened the stability of the imperial bureaucracy. Despite repeated attempts to grapple with the problems inherent in the physiocratic model (most famously the Yongzheng reforms in the 1720s, which sought to regularize some of the surcharges to place local government on a more secure footing)Reference Zelin24 the tension between a central government desirous of maintaining the faith in benevolent and frugal government and control malfeasance and local governments struggling to keep up with the business of growing populations lasted into the 20th century. As Philip Kuhn wrote, by the 18th century (if not a few centuries earlier) ‘society had outgrown the political system that sought to govern it’.Reference Kuhn25 The fiscal basis of successive Chinese empires thus remained precarious. Land taxes may have been sufficient to support a minimal bureaucracy but they proved inefficient when courts were faced with major internal and external threats.
Despite these weaknesses, the contrast with the Holy Roman Empire underscores the superiority of the Chinese administrative capacity. The empire hardly had any central institution other than the imperial court and a small chancery. Occasionally, the emperors invited the German princes of the empire (Reichsfürsten) for consultation in a Hoftag. After the extinction of the Hohenstaufen dynasty in 1254, only Louis of Bavaria (1314–1347) still tried to play a role in Italy, without great success. From 1274 onwards Roman Kings and emperors levied extraordinary taxes on the imperial cities only, in addition to their annual contributions, but, overall, this practice remained highly irregular and the returns modest. Most inhabitants were subjects of a territorial prince or lord, and therefore not liable to taxation by the empire. Fiscal and other state institutions had developed earlier in the territorial states, and in England on the scale of the kingdom since the 12th century. On the occasion of the Hussite war, 1420–1434, a first attempt was made to levy a repartition tax on all classes and entities in the empire, but the assemblies of the estates’ major territorial principalities, such as those of Bavaria, simply refused to transfer the collected income to the imperial treasury. From the 1480s onwards, another attempt was launched to install a general tax for the defence against the Ottoman expansion. In 1495, a ‘common penny’ was agreed upon in the Imperial Diet – an assembly that was just developing – but some territorial estates again refused to contribute. Instead of the planned four years’ levy, the collection was dropped after the first instalment. The revenue was only half of the expected amount, but the idea of a regular imperial fiscality had finally been introduced and could be implemented on several occasions in the 16th century.Reference Isenmann26 Compared with its competitors, the equivalent of 5 tonnes of pure silver collected by the Empire in 1500 was a meagre result; Austria had 4.5 tonnes of silver to spend yearly, Bavaria, the Palatinate, and Tyrol about 3 tonnes each and Württemberg 2 tonnes. Combining the revenues of the multiple territories under the House of Habsburg, from the Low Countries to Bohemia, Roman King Maximilian might have reached a potential expenditure of 35 tonnes or more, comparable to the income of the Republic of Venice and the kingdoms of France, England, and Castile.Reference Körner27 However, the Roman King had no means to coerce the territories. Using other sources of income, Emperor Charles V tried to do it in his war against the Schmalkaldic League of Protestant princes, which united about half of the German territories. He won a battle at Mühlberg in 1547 but lost the war. His successors, nearly all of them belonging to the House of Habsburg, effectively ruled on the basis of their dynasty’s substantial domain income.Reference Prinz28
Territoriality
From what has been said so far about long-term trends in relations between centre and local stakeholders and in fiscal structure, we may conclude that, over time, negotiations with, and the cooperation of, local stakeholders and the political elite in particular were crucial to the formation and maintenance of imperial government over the Chinese territories. During the last millennium, the literati, who formed the core of the political elite, proved to be firmly in support of the imperial model of government and fiscal management. They stood to benefit from being associated with government and having their status officially recognized. On the other hand, the decreasing odds of obtaining a degree and official rank led to much social stress, and the dissatisfaction of a politically informed population posed risks for a small imperial government. The continuation (at least in form) of the bureaucratic models of classical and later empires during times of multi-state rule suggests that bureaucratic organizations, which had penetrated deeply during the Warring States and early imperial period, had staying power for political and military elites. The history of multi-state rule further suggests that bureaucratic techniques and the accompanying Confucian and Legalist ideologies of a well-ordered society could work at various scales and that in and of themselves these need not lead to the realization of large unified empires and exclusive loyalty to one ruler or one state. The weight that should be given to religious and ethical ideals in the realization of large empires may be legitimately questioned given the comparative lack of impact that Christianitas or the unity of the umma have had in West Asian and European history.Reference Scheidel29 As noted above, since the late 11th century, the papacy even actively tried to subdue emperors to its authority and therefore fostered divisions in the empire. But what about attachments to territory? In combination with deep-rooted traditions of bureaucratic rule, the adoption of a territorial imagination that posited the indivisibility of the Chinese territories on universal (cosmological, topographic, and classical) grounds as well as historical precedent may have been an important factor in cementing elite support for governments that set out to incorporate all Chinese territories under unified imperial rule.
The ideal of the unification of different regions was represented in both texts and maps during the Qin and Han dynasties. ‘The Tribute of Yu’ (‘Yugong’ 禹貢) described the natural features and products of the areas covered in Yu’s mythical tour of the Chinese commonwealth and became the paradigm for representing the empire by means of its major administrative subdivisions (jiuzhou 九州). ‘The Tribute of Yu’ was most likely incorporated as a chapter in Shujing 書經 (The Book of Documents), one of the Five Classics, around the time of unification under the Qin and Han dynasties. The first emperor’s tomb reportedly included a monumental map of the empire he had created, made up of natural materials and metals. His Han successors were depicted with maps of the empire in hand, suggesting that by this point empire maps had become metaphors for imperial sovereignty and territorial unification.30
In the 12th century, the social uses and political meaning of empire maps changed significantly. Until then, empire maps had been the prerogative of emperors, court advisors, and generals. Following the loss of the north to the Jurchens, empire maps were also produced by and for the expanding community of Song cultural elites. They appeared in commercially printed atlases and were also disseminated as large-size single sheet rubbings taken from map stelae. These maps, the earliest extant maps of the Chinese territories as a whole, show all territories from where the great walls were imagined to have been in the north, down to where the most southward of the sacred mountains was located, east to the coastline, and west to where the course of the major east-west rivers ended (Figure 1). In this way and by overlaying past and present place names these maps helped foster a sense of belonging to a transhistorical empire at a time when the reality of Song territorial control looked rather different. Cultural elites from this time onwards gave voice to an imperial mission and called for the restoration of the empire based on deeply held beliefs about its territorial scope.
Figure 1 ‘Huayitu’ (‘Map of Chinese and Non-Chinese Territories’), 1136 stele. (Source: Ref. Reference Wanru31)
The weakness of the Holy Roman Empire directly followed from its creation, in 800, by the conquests of a heavily armed cavalry led by a dynasty of four generations of charismatic leaders, from 719 to 814. They met no countervailing powers when they incorporated defeated peoples including Muslims, Saxons, and Slavs. Similarly, Otto I, Duke of Saxony, elected Roman King in 936, took the title of King of Italy in 951 and was crowned emperor in 962 thanks to his stunning victory over the Magyars in 955. He restored imperial power due to his ability to unite the Saxon forces with those of Franks, Alamans, Swabians, Bavarians, Lotharingians, and Bohemians who shortly before this challenge were fully occupied in fighting among themselves. The electoral system for selecting the Roman King, the lack of institutionalization for succession on the level of the Empire, the diversion of scarce resources to hopeless attempts to curb the power of Italian cities, and papal interventions constrained centralization. State formation occurred at the level of the principalities and a stalemate amongst those prevented change. Some of the territories detached themselves more or less formally from the empire: the Low Countries effectively since the 13th century and the Swiss Confederation in 1499. The French kingdom gradually incorporated the principalities east of the Rhône and Saône. The Habsburg dynasty operated in their own interest and successfully expanded their own realm, to which the imperial crown had become just an ornament.
It is illustrative of the Empire’s lack of conceptual reflection and administrative organization that no visual representations have come to us from the medieval period. It was not the German princes, Roman Kings, or emperors who designed a visualization of the Empire. The oldest relatively detailed map of Europe, in which the Empire’s boundaries were delineated, was created between 1106 and 1121, not by its intellectual and political centre, but in the chapter and Saint-Bertin’s abbey at Saint-Omer, then in the county of Flanders and belonging to the kingdom of France. It represents the continent as a quarter of the circular world with Jerusalem in the middle (Figure 2). The map thus shows a cosmological order, not an administrative structure. Notwithstanding this traditional worldview, some details are strikingly concrete. A red line precisely delineates the Carolingian empire, consisting of Italy (which is not named, only Rome, with a marked church, and eight regions), separated by the Alps from a northern section. That northern section encompasses the West- and East-Frankish kingdoms, separated by a meandering red line identifiable as the Rhine. Left appear the names Alemania and Germania, Bavaria, Suevia, Histria, Saxonia. Franconia, the territory of the reigning Emperor Henry V, isn’t mentioned on the map. To the right of the Rhine, the following names are given, from the Alps down to the Northern Seas and the Pyrenees: Burgundia, Aquitania, Gallia (six times), Narbona, Colonia (with a marked church), Neustria, Morini, Flandria. The British Isles are just mentioned in the margin. The references to the Carolingian and ancient traditions are obvious.Reference De Coene, De Reu and De Maeyer32 The encyclopaedic compilation Liber Floridus which contains this and many other originally designed maps of the world, was created in the lively intellectual environment of the chapter and abbey at Saint-Omer, the latter of which was closely connected with the counts of Flanders.Reference Deploige and Verommeslaeghe33 Successive counts had been involved in the First Crusade and remained committed to the crusade and to a wide-ranging worldview.Reference Rider34 The ‘Deeds of the Franks Conquering Jerusalem’ occupy a central place in the voluminous encyclopaedia, a text composed by its compilator canon Lambert in collaboration with monks, at the instigation of Bohemond of Antioch, one of the crusade’s heroes. In combination with the special and well-informed interest in the ongoing Investiture contest, the Liber Floridus and especially its maps, reflect an apocalyptic worldview rather than a vision of Empire.Reference Derolez35
Figure 2 The oldest independent map of Europe, drawn around 1121 by Lambert, canon in Saint-Omer, in his richly illustrated encyclopedia Liber Floridus. Europe is represented as a quarter of the circular world with Jerusalem in the centre. The Italian peninsula, with Rome marked prominently, Gaul and Germany are outlined in red (to view this figure in colour please see the online version of this journal), marking the kingdoms of the Carolingian empire. (Ghent, University Library, ms. 92, f°241 r°.)
Conclusions
The legacies of the Roman and Qin-Han empires have been profound and long-lasting in cultural, legal, administrative, and other respects. In both European and Chinese history the restoration of the classical empires remained an appealing prospect to rulers but throughout the first millennium after the fall of the classical empires this remained an ideal seldom realized. As the odds declined for territorial reunification in Western Europe, the Chinese territories were more or less continuously integrated under imperial rule from the 13th century. The continuation of bureaucratic structures of government during times of multi-state rule made reunification under a larger bureaucratic empire more feasible. Moreover, the bargains struck between centralized power and the expansion of local power (e.g. in the area of elite political participation and local fiscal organization) from the 12th century onwards created a more robust foundation for imperial governance. Finally, the emergence of a commitment to territorial unification facilitated elite cooperation with the late imperial regimes; these regimes’ adoption of Eurasian techniques of imperial rule could be accommodated given the existence of overlapping political imaginaries.
The comparison between the Roman empire and the Han empire has revealed similarities and variation, especially with regard to local autonomy, which we related to the different paths along which the empires developed, by gradual conquest respectively from a republican basis and from well-organized kingdoms.
The fiscal basis of the Han and Song empires allowed for an impressive growth of cities, administrative structures, and military organization that outgrew by far the highest level reached in ancient Europe. The Holy Roman Empire was created thanks to the temporary military superiority of the Carolingians in the eighth century, but it failed to build administrative and ideological structures by which the conquest might have been consolidated. Moreover, agricultural yields did not result in the accumulation of surpluses, nor was it possible to create a unified economic system. Surplus extraction remained essentially limited to the domains, which severely constrained the centralization of power. Local and territorial entities were able to expand slowly and to resist centralizing monarchs. In rapidly urbanizing Northern Italy, this led to the total collapse of imperial authority. While military and literati classes supported imperial authority in Chinese empires, these as well as the merchant and peasant classes organized themselves as corporate estates opposing the monarchy and the Holy Roman Empire. This effectively escalated during the Thirty Years’ War (1618–1648), after which the Habsburg dynasty carved out its own kingdoms, which constituted an empire in name only.
Wim Blockmans is professor emeritus in medieval history at Leiden University and has served as rector of the Netherlands Institute for Advanced Study (2002–2010). His research focuses on the 12th- to 16th-century Low Countries, with a special interest in comparative approaches. Popular representation and political systems have been his main focus, including studies on state formation as well as on power structures created by commercial elites. He is particularly interested in the relations between politics, economic systems and cultural expressions. He is member of four academies, including the Academia Europaea and the British Academy.
Hilde De Weerdt is Professor of Chinese History at the Leiden Institute for Area Studies. Prior to this she taught at King’s College London (Reader in Chinese History, 2012–2013), Oxford University (University Lecturer/Associate Professor in Chinese History, 2007–2012) and Pembroke College (Fellow, 2007–2012), and the University of Tennessee at Knoxville (Assistant Professor of Chinese History, 2002–2007). She obtained her doctorate in East Asian Languages and Civilizations from Harvard University with a dissertation on the intellectual history of the civil service examinations. This work, later published as Competition over Content: Negotiating Standards for the Civil Service Examinations in Imperial China (1127–1276) (Harvard University Asia Center, 2007), was awarded the Achilles Fang Prize for the best dissertation in the East Asian Humanities (1999). Her research focuses on the question of how social networks shaped Chinese politics. Her interests in intellectual and political history, information technologies, social networks, and digital research methods have also led to her involvement in several comparative research projects including ‘Communication and Empire: Chinese Empires in Comparative Perspective’ (funded by the European Research Council, 2012–2017). She is the co-editor of Knowledge and Text Production in an Age of Print–China, Tenth–Fourteenth Centuries (Brill, 2011). Her most recent book, Information, Territory, and Networks: The Crisis and Maintenance of Empire in Song China (Harvard University Asia Center, 2015), takes a fresh look at the question of how the ideal of the unified territorial state took hold in Chinese society.
