This article looks at the revolutionary developments in government and warfare in medieval China. It was in this period of history that that crucial figure, the scholar-official, fully emerged as the characteristic government servant in imperial China.
The centuries between the end of the Han dynasty in 220 AD and the founding of the Sui dynasty in 589 AD saw China weak, divided, invaded and partially occupied by tribal peoples from beyond its borders. Nevertheless, throughout these chaotic and violent times, the tradition which four centuries of stable Han government had laid down, of centralized, bureaucratic government by educated officials immersed in Confucian ideology, never completely vanished.
In the final stages of the Han dynasty a succession of weak emperors, increasingly isolated from senior officials within their private apartments (the “Inner Court”, as it would become known under later dynasties). They surrounding themselves with their wives and concubines, and the eunuchs who attend them. Power struggles between the families of imperial wives and mothers flared up, in which eunuchs, untrained and inexperienced in affairs of state, grabbed power and influence. Their intimacy with – and power over- the emperors caused them to be intensely disliked by the officials, who were kept at arms length.
From 180 CE a succession of child emperors meant that violent factionalism went unchecked at court. Sound government was utterly undermined, and the imperial administration became increasingly corrupt and oppressive, falling into the hands of a small group of immensely wealthy landowning families.
In the hands of the landowners
Since the time of the Wang Mang usurpation (CE 9 to 23), the later Han dynasty had owed its throne to the support of the landed elite, which had, as a result, gown in wealth and power. In the last years of Han rule, a change to recruitment methods for government officials further strengthened the landowning families’ hold over the bureaucracy, to the point where their members practically monopolized the senior positions.
The change in question gave local worthies the task of rating every eligible man in their area (i.e. those with the requisite education) on a one-to-nine grading system according to their suitability for official appointment. These worthies, being usually great landowners themselves or closely connected to them, naturally gave the highest rankings to men of their own class. Those chosen ones would then naturally be fast-tracked into the highest posts.
The basis of the late Han provincial administration was the 100 or so commanderies into which the empire was divided. These in turn were divided into counties. Both commanderies and counties were under officials appointed by the emperor and his ministers.
The commanderies were subject to inspection by regional commissioners who had acquired more and more authority, so that by the mid-2nd century CE they were in effect regional governors, with widespread powers over their territories.
One change which occurred at the very end of the Han dynasty, and would continue in practice after the fall of the dynasty, was the acquisition of military powers by these regional governors. The troubles of the period made this change hardly avoidable, with senior provincial officials needing provincial military force at their disposal to put down large-scale brigandage and low-level disorder.
This situation – factionalism at court, eunuch power and landowner monopolization of high office – continued under the Han empire’s successor states. Their courts continued to be wracked by violent faction-fighting, and eventually fell under the complete dominance of rival landowning families (from whom their royal families had sprung).
The “barbarian” kingdoms of northern China
With effective government thus absent, several nomadic groups were able to move into northern China in the late 3rd century and early 4th century. South China remained united under a succession of native Chinese dynasties; their regimes were faction-ridden and unstable, dominated by small numbers of wealthy landowning families. The north became divided between several kingdoms under barbarian rulers.
Despite the sacking of cities and the widespread destruction by the barbarian invaders, ordered government did not altogether disappear in the north. As the tribal groups moved in, the local Chinese looked to their officials for leadership and protection; the barbarians needed the co-operation of the same officials if they were to exercise any control over the native population.
The Chinese bureaucratic system therefore continued to function after a fashion in the north. Gradually, as the barbarian regimes became more and more “Sinified” (i.e. more Chinese in their culture and way of life), and their kingdoms more centralized along traditional Chinese lines, these bureaucracies were restored to something like their previous condition. The barbarian tribal leaderships and the Chinese landowning families merged into a single ruling class, which filled the highest offices in the states and provided a much more solid power base from which rulers could suppress factionalism and win back effective control over their states. This enabled the Northern Wei dynasty (386-534 AD), for example, to enact the “equal-fields” laws.
In both north and south China, the weakness of the royal courts in the early period of division and invasion ensured that provincial governors became even more powerful than before. Under the late Han they had already gained military authority over troops within their provinces; at some point now they asserted the right to appoint their own subordinates. This gave them huge powers of patronage within their territories. Governorships were usually held by senior aristocrats and even royal princes. These posts often provided the power base from which bids from the throne would be launched.
Later, rulers were able to reimpose their authority on the governors, at least in the north. Under the Northern Wei kingdom (386-534 AD) a major administrative reform greatly multiplied the number of provinces, so that each covered a much smaller area. They thus posed no threat to the central government. Henceforth in Chinese imperial history these officials are known as “prefects”. They continued to form a second tier of provincial government, above the county magistrates.
The emperor Wendi, The first emperor of the Sui dynasty, who reunified China under his sole rule, moved to establish firm government over the whole of China, which meant tightening his hold over the administrative apparatus of the country.
He tightened his control over the administration by making sure that all appointments were made by the central administration, answerable directly to him (in the previous centuries provincial governors had gained the power to appoint their subordinates themselves).
The examination system
The emperor Wendi also abolished the “nine-grade-system” of recruitment to the civil service (see above). In its place he adapted the old Han system for recruiting new officials. He ordered provincial governors to send three candidates annually to the capital, where they were examined for their suitability for an official career. Although each year only a few officials were recruited in this way (the rest were appointed through personal recommendation by senior courtiers and officials), officials from a broader section of society and from different parts of China began to enter the imperial bureaucracy.
Throughout the centuries of division Confucianism had retained its status as the medium of education and the ideology of government, and officials were expected to have a good grounding in Confucian teachings. They remained broadly Confucian in their public lives even while practicing Buddhism in the privacy of their homes. Knowledge of Confucian writings were tested in the new examinations.
The next dynasty, the Tang, expanded this rudimentary examination system inherited from the Sui. Candidates were first screened at county level, and at then at prefectural level, before being sent to the capital for the annual examinations.
Civil service officials
Successful candidates who obtained a degree were now qualified for official posts. As there were only about 25 such men each year under the early Tang, this was nowhere near enough to fill all the civil service vacancies to be filled. The majority of posts were filled by sons of senior officials, who automatically qualified for an official career, plus lower-ranking officials who were eligible for promotion.
Once appointed, every official was evaluated by his superior on an annual basis. Dossiers were kept on their performance at the ministry of personnel. A consistently good performing official could expect to rise steadily through the civil service. Both the Sui and Tang imperial families were drawn from the landowning aristocracy, and members of this class would therefore continue to predominate in high office for a long time to come. Nevertheless, those putting themselves forward for an official career came increasingly from a broader class of lesser landowners, and some of these succeeded in reaching high office, especially under the empress Wu.
The local gentry
Candidates for the examinations at the capital came to considerably outnumber successful examinees, but even those who sat the exams and did not pass could return to their home regions with their prestige enhanced, and able to take up unofficial (but paid) local government work, or work as tutors or school teachers. From this time a class of highly educated small-scale landowners came to prominence in the localities of China, a gentry elite who would form the natural leaders of Chinese rural society under future dynasties.
Just as under the Han, the Tang civil service included a department called the Censorate. The job of this office was to ensure that government functioned as intended. When the censors uncovered incompetence or corruption, they had the right to submit their reports directly to the throne.
The later Tang
Under the early Tang, the functioning of government was restored to at least the level of the Han dynasty, and probably attained a higher level. Despite the disasters of the An Lushan revolt, under the later Tang the civil service continued to function more or less as it had done before. The examination system continued to expand, with the 25 degree holders produced annually under the early Tang ballooning to about 100 per year. This meant that perhaps half of all officials were “scholar-officials” of this type. Such was the increase in numbers of young men studying for the examinations that a major reason for the development of block printing at this time is credited to the demand for text books. Whether true or not, the publishing of text books was certainly an early use of this technology.
Under the circumstances of the later Tang period it is hardly surprising that the standards of Tang administration eventually began to decline. In a repeat of late Han times, a series of weak emperors surrounded themselves with palace eunuchs who attended to their personal needs within the their private chambers (the “Inner Court”). This group came to dominate government, and a weakened court could not prevent corruption and incompetence from spreading throughout the civil service.
By the time the founder of the Sui dynasty had reunited China under his sole rule the provinces – now known as prefectures – had been severely cut down in size and made much more numerous. The Wendi emperor strengthened his central rule by ensuring that all officials in the provinces, from county magistrates upwards, were appointed directly by – and were answerable directly to – himself and his ministers, not senior provincial officials.
The early Tang applied very little change to the provincial system. The lowest level of regular government in the provinces were the counties, of which there were more than a 1000, and above them were the prefectures, of which there were some 360. There was no tier of provincial officialdom between prefectures and the central government.
In a replay of developments under the Han, in 711, to make the government of the provinces more efficient, the prefectures were grouped under the supervision of 10 regional commissioners. As time went by, these commissioners gained more and more authority over their regions, and at some point gained control of the military forces within their territories.
During the chaos of An Lushan’s rebellion and the years that followed, the regional commissioners, who had already acquired wide powers, including command over military forces, came to represent the authority of government in their territories. They raised and maintained mercenary armies of their own. They acquired enormous authority, becoming all-powerful regional governors.
Some of these governors managed to convert their office into a hereditary rulership of their provinces, while others stopped forwarding taxes to the center. Gradually, however, the central government was able to bring them back under a measure of central authority, but was never again able to control the provinces in the way that the earlier Tang had done.
At the end of the Tang dynasty, the weakness of the central government led to these governors grabbing more power for themselves. When the last Tang emperor was deposed by his chief minister, the governors of the southern provinces simply converted their offices into hereditary monarchies of independent kingdoms; and while the northern kingdom retained a semblance of unity, the situation was not much better here (see below).
The troubles of the half-century of division after the fall of the Tang dynasty destroyed the power (in many cases the very existence) of the old landowning aristocracy. This left the founder of the Song dynasty free to expand the examination system so that the majority of the civil service was recruited through it.
From top to bottom, educational achievement and administrative merit were what counted, rather than personal connections. The families of imperial wives and concubines were prohibited from holding high office at court, and the palace eunuchs were prevented from gaining too much influence.
The Song dynasty is particularly noted for its highly capable chief ministers. These were men who had been recruited to the civil service through educational achievement and had risen through its ranks by merit, in the process acquiring long experience of government. The Song emperors (most of whom were of mediocre ability) delegated most of the running of government to these figures, and the other ministers came to report to them on a daily basis. The chief ministers controlled what matters came before the emperor for his consideration.
Under the Southern Song, the reins of government remained with strong chief ministers, and the structures of central and local government remained very much as they had been under the Northern Song.
A wider pool of candidates
The government ensured that the examinations had a wide pool of candidates to draw on by encouraging the establishment of state-supported schools in the prefectures. Promising sons from poorer families could now be prepared for the exams. Also, the Song instituted preliminary exams at the prefectural level, the passing of which qualified a person to attend the central examinations in the capital. Candidates no longer needed sponsorship by a senior provincial official. This reduced opportunities for favoritism by these officials, and made it easier for candidates who were less well connected to embark on a good career. Thousands of candidates, still mostly from gentry backgrounds, produced about 200 degree holders a year.
The ratio of degree holders to the general population was around 1: 10,000, which helps to explain the respect in which they were held by the population at large. It was in Song times that the examination system came to be regarded as playing a key role in legitimating government in the eyes of the people, who knew that (1) they were being governed by men who merited their authority by their education and ability, and (2) that even people from poorer backgrounds – maybe even themselves or their children – had a chance of attaining power and status under this system. Though most officials came from the landowning gentry, there were plenty of examples of such men.
The examinations narrowed in scope under the Song, a reflection of the renewed interest in Confucianism in these years. No longer were there questions of current affairs and practical government, as there had been under the Sui and Tang.
In the period of division which followed the end of the Tang dynasty, the provincial governors in the north became even more powerful then before (those in the south had converted their provinces into independent kingdoms). These powerful regional governors had virtually private armies at their beck and call. The rivalries between them repeatedly broke out into open warfare, and one or more of their number could always be counted on to lend support to opponents of a reigning ruler.
Given this experience, it is no surprise that the early Song emperors did away with this tier of administration altogether. The more than 300 prefects reported directly to central government. Just as in Han and Tang times, however, this was an unworkable situation – there were simply too many prefects to be supervised properly from the center. It was inevitable that co-ordinating officials should be dispatched to the provinces sooner or later, but when this did occur, these senior regional commissioners had specifically defined roles (fiscal, judicial, transport, military affairs and so on), with their responsibilities overlapping geographically with one another, so that no one official had responsibility for all government matters within one particular area.
The number of more local officials – county magistrates – remained at roughly the same level as in Tang times, at over 1000. As the population more than doubled under the Song, this meant that people saw less and less of the representatives of central government. Local administration fell increasingly to the local gentry, who fulfilled this role on an irregular, mostly unpaid basis as a matter of public duty.
Throughout the Song and Southern Song periods, barbarian regimes from the central and northern Asia ruled much of northern China. The westernmost of these, the Western Xia kingdom, was hardly more than a loose confederation of tribes, with a Chinese institutional veneer on the surface. To its east, the Liao state, ruled by the Khitan tribe, had a dual government, with “north-facing” institutions to rule the Khitan people by tribal custom, and “south-facing” institutions modeled on Tang precedents, to rule the Chinese. This latter included an examination system for recruiting Chinese officials. Real power, however, remained very much in hands of the Khitan tribal leadership.
When the Jurchen people had conquered the Khitan to form the Jin empire, they initially followed the Liao by establishing a dual government, one for the Jurchen and one for the Chinese. They soon abandoned this in favor of a single system based entirely along Chinese line. This included an examination system for the recruitment of civil service officials, to which indeed many Jurchen aristocrats submitted themselves.
With the Mongol conquest of northern China, the examination system was abolished, and although the civil service remained in place, it had a subservient position in the state; Mongols and allied central Asian peoples filled all the positions of power within the Mongol empire.
China under the Yuan
Having conquered the south of China and founded the Yuan dynasty, Kublai Khan deliberately moved the Mongol regime towards the by-now traditional Chinese model of government. A major part of this was to reinstate the examination system for the whole of China. However, though some Chinese officials did attain high office, the power-players at the Yuan court came from many races: Mongols, Uighurs and Turks from central Asia, Persians, even some Europeans (most famously, Marco Polo).
Muslims in particular were valued for their financial expertise, and were given key roles in the finance departments of the central administration, and as tax farmers in the provinces.
The Yuan dynasty’s provincial government formed a complex hierarchy, with prefectures grouped into supervisory circuits, circuits into supervisory regions, and regions into provinces. This system was designed to concentrate as much power as possible at the center, but unsurprisingly it blurred lines of responsibility, leaving officials with a sense of powerlessness, especially in the face of the mounting disorder in the later Mongol period.
In the later 2nd century CE, the defenses of the Han empire, based on long-service professional troops, began to break down. Weaknesses in the state meant that complex and costly governmental operations which had to be undertaken on a regular basis, such as providing the frontier garrisons with supplies and troops, began to fail. With troops not arriving to keep frontier garrisons up to strength, defense fell increasingly into the hands of “loyal” tribesmen from across the border.
This development strengthened under the successor dynasties, and under the barbarian regimes in the north military responsibilities fell entirely to members of the conquering tribes. The armies of the northern states were in fact the tribal cavalry, and cavalry became the primary arm of all Chinese states at this time.
One innovation which strengthened this trend was the invention of the stirrup, which probably happened in China about now. The first firm evidence for a stirrup is in a stone carving of the late 3rd /early 4th century.
Having to deal frequently with horse-riding steppe nomads, often on a hostile basis, the Chinese equipped their cavalry with metal stirrups. It is possible that they were copying an innovation already used by the steppe nomads; or they may have come up with the innovation themselves, to enable their horsemen to match the nomads’ cavalry techniques on a more equal footing.
In any case, the stirrup caught on, and within a century was being used throughout China. Further afield it did not take very long for stirrups to cross central Europe and appear in the west. Many scholars think that the stirrup would be key to the rise of the knight as the dominant figure in medieval European warfare.
Ironically, back in China, cavalry did not dominate military tactics for long. In the 6th century the Northern Wei kingdom re-instituted the militia system of the early Han, with all males being made liable for military service for a short period of time, as infantry troops. This enabled the rulers to reduce the size and importance of the tribal cavalry forces which had brought them to power but which were always a potential power base for rebellious tribal nobles.
The emperor Wendi, the reunify of China and founder of the Sui dynasty, built on this militia system, but instead of all males being liable for short periods of military service, selected men were chosen to serve for long periods. He established several hundred garrisons around China to bolster his regime, manned by the militiamen who served as farmer-soldiers, cultivating lands around the garrisons to grow their own food. These troops were sent on rotation to serve in the capital, on the frontiers or on campaign beyond the empire’s borders.
This system provided a huge reserve of well trained military manpower. In addition, the old tribal army inherited from the preceding northern dynasties was maintained, now slimmed down and transformed into an imperial guard.
This military establishment enabled the emperor Wendi to expand the empire’s borders considerably, especially into central Asia.
The early Tang emperors inherited this system from the Sui. However, as time went by the militia proved less and less popular to join, and conscription had to be imposed, causing much ill-feeling. Increasingly, the poorest men in society were induced to sign up as long-term soldiers, and sent to serve as frontier troops.
At the same time the garrisons ceased to be the self-sufficient forces they had been intended to be (if they ever had been that), and in fact proved more and more expensive to maintain, in particular as a way of manning the frontier garrisons. During the 8th century, therefore, frontier units came to be largely filled by tribesmen from beyond the frontier. This was a comparatively inexpensive way of keeping troop strengths up, but turned frontier armies into virtually private mercenary forces at the beck and call of their commanders. The garrisons scattered throughout the empire were still maintained, but now used mostly for local policing duties.
The weaponry of the Tang armies had hardly changed from pre-Han days. The crossbow was the main weapon. The only major innovation was that the cavalry now used stirrups, allowing them to fire more accurately.
With the complete eclipse of central government in the years of An Lushan’s rebellion (755-61), especially in the north, commanders and governors loyal to the Tang were required to raise their own armies in order to restore order. From now on, provincial military forces were effectively the private armies of these commanders. When the central government had to raise an army for a specific campaign, it had to negotiate with them for the use of these troops, but also employed “friendly” tribal groups as mercenaries under their own leaders.
The Song empire was surrounded on its northern and western sides by powerful and hostile neighbors, and was therefore compelled to maintain a huge army at enormous expense. It remained fully manned and equipped at all times.
Given the trouble that powerful frontier generals had caused the Tang dynasty, the Song naturally concentrated the bulk of their forces around the capital, divided amongst several different commands. The army commanders reported directly to the emperor and his ministers – indeed, the overall direction of military affairs was exclusively in the hands of ministers who had an entirely civilian career behind them.
This reflected the fact that, in Song times, civilian and military careers had become entirely separate, which had not been the case earlier in Chinese history. The army was now officered by career soldiers. Although there was a military version of the civil service examination system, it were not deemed nearly so rigorous, and did not have the same prestige. Moreover, it was civil servants who had the opportunity to reach the very top of the tree, not army officers: as we have seen, even the most senior generals reported to high civilian officials, not directly to the emperor (as had been the case in previous dynasties).
This led to the Chinese ruling classes developing an ethos wherein army officers were not held in high regard, which of course meant that the brightest and best avoided a military career. It was in Song times that civilians came to predominate over military men in terms of status and power in the Chinese state system, and that the China ruling class developed its distinctively civilian ethic.
The lower ranks of ordinary soldiers in the Song army were also filled by full-time professionals.
A mixed record
The actual military record of the Song army was patchy. The Song dynasty was far less successful than the Tang had been in its dealings with foreign peoples. They not only failed to conquer the Khitan (the Liao kingdom), but, following a defeat in 1004, the Song were forced to pay the Khitan a heavy annual tribute to keep the peace. However, the Song army performed very capably on several key occasions, and kept the Khitan threat more or less at bay throughout the 11th and early 12th centuries; and they were able to repel the Jurchen invasion of the south in the mid-12th century. The Southern Song army fought with great courage against the Mongol armies of Kublai Khan.
A distinctive feature of Southern Song warfare was the emphasis on naval warfare, particularly in the Yangtze river system. In operations against the Jurchen navy in 1161, the Song fleets deployed fast paddle-wheeled warships armed with catapults which fired gunpowder bombs – the first recorded use of gunpowder in warfare. These new weapons won them two famous battles in that year. Other military innovations were bazooka-like weapons made from bamboo for firing explosives, and iron-clad carts.
The Yuan military establishment was composed of garrisons scattered throughout China, with land attached to them for their upkeep, worked by enserfed Chinese peasants. The largest concentration of forces was stationed in and around the capital, with the elite palace guard, staffed by the sons of Mongol nobles, at the center.
The Yuan army was made up of troops with different levels of status and pay. These reflected the Mongol notion of ethnic ranking, and in descending order were Mongol tribesmen, allied central Asian tribesmen (mostly Uighurs and Turks), North Chinese, and at the bottom, South Chinese.
Military service under the Mongols was hereditary; and indeed ranks within it became more or less hereditary as well. The civilian population provided a militia for local policing and anti-brigandage duties. Otherwise the native population was strictly forbidden to have weapons.