THE CONSOLIDATION OF QING RULE
When Qing armies completed the military conquest of southwestern China in 1659, Guizhou was an unruly internal frontier. Although the Ming government had organized the region as a province in 1413, Guizhou still bore only the faintest imprints of imperial control. The province was an ethnic patchwork, populated by a variety of indigenous non-Han groups, each with its own languages and customs. Some of these groups showed varying degrees of assimilation with mainstream Chinese culture, but many others remained fiercely independent. Sub-provincial administrative units existed only where Han tax-paying populations were large enough to support them. Hereditary native officials (tusi) controlled most of the land within Guizhou and along its borders with other provinces. Although these native officials received their titles and legitimacy from the throne, most of them functioned beyond the reach of imperial law. The central government had little sway over such matters as criminal justice, or even over the rules governing inheritance of the titles it conferred upon the native officials.
It was not until the reign of the third Qing emperor, Yongzheng, that the imperial state took decisive steps to consolidate military, political, and cultural control over frontier communities throughout southwestern China. After consulting with officials in the region, Yongzheng endorsed a plan to “reform the native and return to the regular” (gaitu guiliu). Under this policy, native rulers were deposed and their domains were incorporated into regular administrative units. Indigenous populations that had traditionally enjoyed relative autonomy were thus brought under direct rule for the first time.1 Central and local authorities hoped that imperial law and Confucian ethics would eventually displace local customs, transforming Guizhou’s indigenous peoples into civilized Qing subjects.
Both the process and the short-term effects of the Yongzheng-era campaigns in central and southwestern Guizhou are worth examination. Relatively tolerant attitudes of the first two Qing emperors, Shunzhi (1644–1661) and Kangxi (1662–1722), gave way to more confrontational tactics during the Yongzheng period. These tactics unfolded in the Ding-fan-Guangshun region of central Guizhou, and Nanlong, in the province’s far southwest. Both areas were home to large communities of the Zhongjia ethnic group whose penchant for banditry, raiding, and feuding had long been a source of consternation for imperial authorities. In 1724, officials in Dingfan-Guangshun started a campaign to bring order and civility to the region. The enterprise met with fierce resistance from villagers who deeply resented the state’s intrusion. After two years of battling local vigilantes, provincial officials finally managed to lay the groundwork for a much larger military and administrative presence. Shortly after the campaign in Dingfan-Guangshun concluded, Qing authorities turned their attention to Guizhou’s ill-defined boundary with Guangxi. For centuries, this region had been a no-man’s-land rife with unresolved territorial disputes and blood feuds. In an effort to create a new administrative and social order, provincial officials deposed the inept native ruler whose family had controlled the region since the last decades of the Yuan period. After establishing a permanent boundary between Guizhou and Guangxi, the provincial officials cobbled together the recently annexed lands into a new prefecture, Nanlong.
This narrative covers both new and familiar terrain in the history of the Qing southwest, offering a reinterpretation of the events in Dingfan-Guangshun and providing the first English-language account of the campaign in Nanlong.2 The reforms in southwestern China are most often associated with the Yunnan-Guizhou governor-general (zongdu) Ortai. Only after Ortai assumed his post in 1725 did Qing troops begin to gain any traction against the local bandits who plagued the area. In a matter of months, Ortai secured the region and began taking steps to integrate it into the Chinese world.3 Ortai’s efficient handling of the Dingfan-Guangshun crisis convinced the emperor that the success here could be replicated in other parts of Yunnan and Guizhou.4 This proved a grave miscalculation, for subsequent campaigns proved far more complex and far bloodier than the engagement at Dingfan-Guangshun. Qing incursions into the Miao regions of southeastern Guizhou, the Lolo (Yi) heartland of northeastern Yunnan, and the Tai polities of southern Yunnan provoked violent opposition from local populations.5
Qing expansion in these regions was both an imperial and colonial phenomenon. The Yongzheng-era policies brought parts of southwestern China under direct imperial rule for the first time, and Han settlers soon displaced native populations. Some officials who advocated the overthrow of the native rulers voiced “overtly colonial” arguments, stating that “natural resources could be more efficiently tapped if the areas in question were brought under the direct control of the central government.”6 The colonial and imperial nature of the Qing enterprise in Dingfan-Guang-shun and Nanlong is most apparent in the state’s efforts to impose direct rule. “Overtly colonial” arguments were far less important in these regions than in other parts of the Qing southwest, particularly the copper-rich lands of northeastern Yunnan.7 Dingfan-Guangshun had few natural products to offer. Nanlong was more richly endowed, with reserves of silver and other metals and an abundance of forest products, but economic exploitation does not appear to have been a major impetus in determining policy for this region.8 Han settlement is a thornier issue. In theory at least, the policy of eliminating indigenous rulers was intended to protect Han Chinese living in or near native areas—not necessarily to attract new settlement.9 Subsequent Han settlement of Dingfan-Guangshun and Nanlong appears to have been more of a by-product—albeit one welcomed by imperial authorities—rather than an express aim of the Yongzheng-era program.10
The correspondence between the emperor and his officials in the field suggests that policy in Dingfan-Guangshun and Nanlong was motivated above all by the desire to create order where none had existed before—or at least no order existed that the Qing considered recognizable or usable. In view of this, Yongzheng’s plans for Dingfan-Guangshun and Nanlong appear to represent an exercise in what James Scott calls “seeing like a state.” Here and elsewhere in southwestern China, the state took steps to rationalize and standardize “a social hieroglyph into a legible and administratively more convenient format. The social simplifications introduced not only permitted a more finely tuned system of taxation and conscription but also greatly enhanced state capacity.”11 Before the Yongzheng-era reforms, large areas of Guizhou resembled social hieroglyphs, indecipherable to Qing authorities. Many indigenous communities lived beyond the reach of direct administration, isolated by geography as well as their own languages and customs. Officials of the Yongzheng era hoped to transform these hieroglyphs into a legible form, enabling the state to better understand and control the region’s population, land, and resources.
It may be useful to examine the matter in terms of state space and non-state space.12 State spaces describe lowland areas where the ecology was conducive to stable human settlements and surplus grain production, two prerequisites for administrative oversight and revenue extraction. Nonstate spaces, by contrast, were upland areas where the ecology did not support these preconditions, where inhabitants practiced slash-and-burn or shifting agriculture, or engaged in other economic activities that produced little surplus. In the absence of stable populations or steady sources of revenue, such areas remained largely impervious to state control and taxation. The central government generally regarded these non-state spaces as a source of violence and barbarism, and as places of sanctuary for bandits, rebels, and other dangerous characters who posed an immediate or potential threat to the state.13
With some refinement, these concepts carried deep resonance for Guizhou on the eve of the Yongzheng-era reforms. State spaces, such as they were, existed wherever the imperial government had managed to establish regular administrative units. Non-state spaces characterized areas without regular administrative units, native officials, or even centralized indigenous polities. Such regions were most commonly found in the Miao heartlands of southeastern Guizhou.14 Between the state and the non-state spaces lay the Zhongjia regions of central and southwestern Guizhou—areas perhaps best labeled “semi-state spaces.” These areas had both native officials and a smattering of regular administrative units, but neither functioned well enough to meet the government’s increasing demand for standardization and centralization.15 Yongzheng and his officials recognized that their capacity to “see like a state” would be severely constrained unless they reorganized space into a format that could accommodate their vision. One goal of gaitu guiliu, therefore, was to transform the semi-state spaces of central and southwestern Guizhou into full-fledged state spaces.
QING POLICY IN SOUTHWEST CHINA UNDER SHUNZHI AND KANGXI
During the Shunzhi and Kangxi reigns, the central government tried to operate within the existing framework of the native official (tusi) system. As noted in chapter 1, the military conquest of Guizhou in 1658 owed much to the cooperation of native rulers. When Qing troops marched across native domains, the local chieftains usually pledged loyalty to the new dynasty, exchanged their Ming seals for Qing ones, and then offered military assistance to the imperial armies. In this way, Beijing secured control over the region while leaving the institution of native rule largely intact.16
Although the native officials remained in power, the central government soon took steps to curb their traditional autonomy, beginning with new regulations on the inheritance of official titles. In 1659, the Shunzhi emperor moved to assert greater control over the succession process by stipulating that only direct, patrilineal descendants could inherit the Qing title. This marked a sharp departure from Ming precedents, which allowed native officials to designate heirs according to local custom or personal whim.17 Under the new Qing rules, the court refused to consider a deceased leader’s brothers, sons by concubines, or daughters for the post. In order to confirm the lines of descent, applicants for the titles were required to submit their genealogies to local officials. In addition, potential successors to the title were encouraged, but not required, to attend the nearest state-sponsored public school. These reforms affected the culture of native elites in two crucial ways. First, by assuming a direct supervisory role over the inheritance process, the Qing state greatly curtailed the political independence of indigenous frontier societies. Second, the opportunity to receive a Confucian education encouraged the native rulers to begin identifying more closely with Chinese culture.18
These reforms continued to evolve during the Kangxi reign as the court extended more educational opportunities to native populations. In 1682, the Guizhou public schools were opened to all male relatives of the native officials. Four years later, the schools were opened to all male children in the native domains. In 1703, the regular examination system was opened to students from native chieftain families, although their exams were graded separately from those written by Han Chinese candidates. Two years later, Beijing recommended that “charity schools” (yixue) be established throughout Guizhou for the designated heirs—that is, sons, and brothers—of the native officials. The court also stipulated that it would assign official titles only to candidates who could prove they had attended these state-sponsored schools.19
The purpose of these reforms was to tie the native rulers’ political fortunes even more firmly to the Qing court. Kangxi preferred this approach to the more aggressive policies advocated by civil and military officials who pressed for the abolition of native rule. The emperor’s attitude stemmed not from any particular affection for the tusi, but from profound mistrust of the provincial officials in southwest China, a product of his bitter experience with the traitor Wu Sangui.20 As explained in chapter 2, Wu won hereditary control over Yunnan and Guizhou for his role in the Qing conquest of these provinces and the capture of the last Ming pretender. Beijing had intended for him to govern on the court’s behalf, but he soon came to view the region as his personal empire. He amassed a small fortune by increasing taxes and establishing monopolies on salt wells, gold and copper mines, and trade in ginseng and rhubarb.21 Wu supplemented the ill-gotten gains from his local enterprises with generous subsidies from Beijing, ostensibly provided to quell local uprisings. His modus operandi was to antagonize native officials until they took up arms, and then petition the court for funds and munitions to put down the uprisings that he himself had fomented.22
By the 1670s, the Kangxi emperor was convinced that Wu Sangui posed a grave threat to the dynasty’s political and financial stability, but he was not sure how to deal with the powerful general.23 An opportunity for action arose in 1673, when Wu Sangui asked Beijing for permission to retire to Manchuria. Wu’s intentions were anything but sincere. He was simply trying to gauge the court’s reaction to his petition and fully expected Beijing to reaffirm his authority over southwestern China. Instead, the court turned the tables on Wu and ordered his immediate transfer to Manchuria. Outraged, Wu decided to rebel against the dynasty, sparking an eight-year war.24 In its aftermath, Kangxi remained wary of the provincial officials in Yunnan and Guizhou, finding echoes of Wu Sangui’s treachery in every request for military action against the tusi.25 Faced with the real or imagined threat of another war in the southwest, Kangxi preferred to forgive the native rulers’ occasional transgressions.
THE YONGZHENG REIGN: ABOLISHING NATIVE RULE AND ‘SEEING LIKE A STATE’
The Yongzheng reign saw a decisive shift in both central government attitude and policy toward southwest China. Whereas the Kangxi emperor had steered away from aggressive change in southwest China, his son was determined to transform the region’s political, administrative, and cultural landscapes. Yongzheng’s campaign in southwestern China represented an integral part of his mission to extend the imperial writ to the farthest-flung reaches of the Qing realm.26 Although some historians have portrayed Yongzheng as an autocrat or a despot, he was above all committed to the pursuit of effective government.27 He restructured state institutions and practices in ways that both increased his personal power and also promoted more honest, efficient administration at every level.28
Yongzheng’s Mission in Southwest China: Local Security and Broader Strategies
Soon after ascending the throne, Yongzheng turned his reformist zeal to the chronic unrest and maladministration in southwestern China. His initial attempts at reform were fairly conservative, demanding nothing more than a stricter application of the existing laws on native rulers. In a 1724 edict, Yongzheng leveled a series of accusations at the native officials, charging that they possessed “only scant knowledge of the laws of the Empire” and governed their domains with cruelty and avarice. According to the emperor, native officials thought nothing of stealing their subjects’ livestock, kidnapping women and children, and even murdering at will. Even worse, they colluded with fugitives from other provinces, the so-called “treacherous Han” (Han jian), to oppress and swindle the native commoners. After enumerating the native officials’ many crimes, the emperor ordered provincial authorities to keep a closer watch on the tusi under their jurisdiction and strictly punish those who broke the law. Native officials who failed to mend their ways even after a reprimand were subject to impeachment, dismissal, and punishment.29
After this tentative start, the emperor moved toward a more aggressive policy. He began staffing key provincial posts in Guizhou and Yunnan with “new men,” relatively young officials who shared his commitment to reform.30 As disorders persisted throughout the region, these activists became increasingly convinced that the native official system had ceased to function as a viable means of governing non-Han areas. The chieftains continued to flout imperial law in numerous ways. Some were personally involved in banditry, while others provided sanctuary to criminals, Han and non-Han alike. The activists worried that if such infractions went unpunished, then even the more law-abiding native rulers might be emboldened to engage in illegal activities.31 Even worse, as Ortai cautioned in a 1725 memorial, the continued presence of the tusi could undermine any chance for long-term stability in the region. “If we do not replace all the native rulers with regularly appointed officials . . . then even if the land tax, military affairs, and legal matters are managed with the utmost thoroughness, it will amount to nothing in the end. We dare not approach this matter with anything less than the greatest care and vigilance. . . . In my humble estimation, this is the first order of business in Yunnan and Guizhou.”32 Warnings like these persuaded the Yongzheng emperor of the need to overthrow native rulers and incorporate their domains into regular administrative units (gaitu guiliu).
It should be noted that broader strategic concerns also informed the decision to take action against the native officials. Southwestern China was vital to Qing interests in Inner Asia, and Yunnan occupied a key position as the gateway to Tibet, where the Qing had recently gained a foothold they could ill afford to lose. Tibet had become more and more important to the imperial state. During the late Kangxi period, imperial troops had engaged in fierce battles with the Western Mongols for dominance in Inner Asia. Qing authorities gradually came to realize that in order to contain this formidable adversary, it was essential to control the Mongols’ religious leaders (who were loyal to the Yellow Hat school of Buddhism in Tibet). In addition to its role as a gateway to Tibet, Yunnan was also important in its own right, as the empire’s chief source of copper, the primary ingredient in Qing coins. Qing policy in Guizhou was in some respects a by-product of Yunnan’s importance. Imperial authorities hoped that pacifying Guizhou would ensure easier access to its neighbor to the west.33
Imposing Order in Dingfan-Guangshun
Dingfan-Guangshun, often abbreviated Ding-Guang in Qing documents, served as the laboratory for Yongzheng’s activist policies in southwestern China. The transformation of this region occurred in four stages. The first stage involved the long and frustrating struggle to subdue Zhongjia bandits in the heart of Ding-Guang. The second stage entailed the abolition of a native chieftain whose family had controlled the lands just west of Ding-Guang. His surrender allowed provincial officials to consolidate and expand their control in central Guizhou. During the third stage, Qing authorities took armed action against local residents who obstructed the erection of military compounds in Ding-Guang. In the fourth and final stage, provincial officials enrolled residents of the newly annexed lands on tax rolls and issued last names to those who had none.
The name Dingfan-Guangshun refers to two small administrative centers in central Guizhou. The Ming government established Dingfan and Guangshun as departments (zhou) in 1587 and 1602, respectively. Both were subordinate to Guiyang prefecture, located 64 kilometers (40 miles) to the north.34 Until the reforms of the Yongzheng period, Dingfan and Guangshun stood as tiny outposts of imperial rule in the heart of native territory. The surrounding countryside remained in the hands of native chieftains. In Ming and Qing sources, these chieftains were referred to by their surnames followed by the suffix fan, or “barbarian.”35 Indeed, “Dingfan” means “pacified barbarians,” a name that reflected imperial ambitions rather than realities on the ground. Like much of Guizhou, the region was an ethnic mosaic. The Zhongjia were the predominant non-Han group, but Miao and other indigenous populations lived in the area. Small numbers of Han Chinese also began to settle in the region during the Ming period.36
As noted in chapter 2, imperial officials generally regarded the Zhongjia as the scourge of Guizhou. Ethnographic accounts from the late Ming and early Qing periods never failed to mention the ethnic group’s penchant for banditry and violence. Like all stereotypes, these unflattering portrayals carried a certain element of truth, but did not tell the entire story. The waning years of the Kangxi period did see an increasing number of Zhongjia raids on villages in the Ding-Guang region as more and more Han immigrants settled in their midst and seized what little good land existed.37 Although imperial officials could only see these predatory behaviors as threats to social stability, banditry and violence provided an important way for the impoverished Zhongjia to supplement their livelihoods. Not surprisingly, the Han residents of Ding-Guang did not view Zhongjia activities in such a compassionate light. Members of the local gentry lodged numerous complaints with provincial authorities during the last years of the Kangxi reign, but the apathetic officials usually ignored these pleas for help. The officials posted to Guizhou at this time were a sorry lot, so demoralized by the dank climate and lack of cultural amenities that they were unwilling to exert themselves even when crises arose.38
However, the rise of the Yongzheng emperor infused the provincial administration with new vigor. Soon after assuming the throne, the monarch made it clear to Guizhou officials that he would not tolerate the lassitude so prevalent during the last years of his father’s reign. In a 1724 edict to Mao Wenquan, his first appointee as Guizhou governor (xunfu), Yongzheng blamed the previous administration for the recent spate of raids in Ding-Guang and other areas. By failing to relieve the grinding poverty in the Guizhou countryside, local officials had effectively driven people to illegal pursuits like banditry and raiding. Yongzheng ordered Mao and his colleagues to govern with more compassion and rigor than their predecessors.39 The emperor’s decree showed that he understood why the Zhongjia behaved as they did, but his insight did not mean he would allow this behavior to continue. It was up to Governor Mao to alleviate the conditions that spawned criminal behavior, and to end any illegal activities already in progress.
Unbeknown to the emperor, Mao and his colleagues had already launched a military campaign in central Guizhou with humiliating results. Acting in concert with the Guizhou provincial commander-in-chief (tidu) Zhao Kun and Yunnan-Guizhou governor-general Gao Qizhuo, Governor Mao had targeted a gang of bandits who had been terrorizing villages throughout the Ding-Guang region. The ringleader was a Zhongjia strongman named A Jin who had proclaimed himself “king of the Miao” (Miao wang). From his headquarters deep in the mountains near Dingfan, he commanded a network of kidnappers and robbers who had forced many nearby villages into submission. In the spring of 1724, A Jin’s band raided several market towns. Without first seeking authorization from his superiors, Colonel Chen Yuanxun of the Ding-Guang garrison had ordered a force of two hundred troops to launch an attack on A Jin’s lair. Three Qing soldiers were injured in the ensuing skirmish. A Jin escaped with all of his followers, and the imperial troops were forced to retreat to their base without capturing a single bandit.40 To make matters worse, one of Colonel Chen’s sublieutenants had needlessly burned down a Zhongjia village that had never offered overt assistance to A Jin’s band. In his analysis of the events, written several months later, Governor-General Gao Qizhuo stated that Chen’s men should have asked the villagers to help capture A Jin, offering assurances of clemency to those who cooperated. Instead, the troops had resorted to extreme measures that only provoked the Zhongjia to turn against the Qing and join the bandits.41
Colonel Chen tried to redeem himself with another unauthorized attack on A Jin, but further disaster ensued. Early in the autumn of 1724, he set out with three hundred troops and marched toward A Jin’s base. The Zhongjia shot arrows at the Qing soldiers, wounding several. Once again, Colonel Chen was forced to beat a hasty retreat.42 His repeated debacles called for a more aggressive plan to control the Zhongjia. If the Qing were to have any credibility in the region, they would have to attack the Zhongjia with more than two or three hundred troops. Accordingly, Governor-general Gao authorized a 2,400–man expedition against A Jin’s stronghold in November of 1724. Instead of relying solely on men from the Ding-Guang garrison, the Qing force also included troops from units in northwestern Guizhou.43
This time, the Qing expedition succeeded. The sheer size of the Qing army made resistance impractical, and many Zhongjia collaborators served as guides and translators. The imperial troops easily made their way from Dingfan to A Jin’s base. A Jin set up a roadblock and fled into the nearby forest, but Qing troops quickly rounded up all of his subordinates. Within a week, they caught A Jin himself and delivered him to Guiyang for trial and subsequent execution. Eventually, Qing troops withdrew from the area, leaving behind small units to guard strategic points.44
Several months after capturing A Jin, Qing officials in Guizhou scored another modest victory. In early 1725, Governor Mao informed the emperor that he had decided to depose the Kangzuo native chieftain, Xue Shiqian. Xue’s family had controlled this small fiefdom just northeast of Dingfan since the early years of the Ming dynasty.45 According to Mao, Xue had long engaged in brigandage, eluding capture by slipping into his mountain lair. In Mao’s estimation, Xue was more evil than the other native chieftains. If the authorities allowed him to persist in his depredations, then other native chieftains in Guizhou might emulate his behavior.46 Implicit in Mao’s report was the concern that Xue might threaten the hard-won peace in nearby Ding-Guang.
Much wiser and warier after the long battle against A Jin, Mao felt it best to avoid another large-scale military expedition. In the hopes that friendly overtures would suffice, he sent a local magistrate to offer Xue life imprisonment in another province rather than capital punishment if he went to Guiyang without protest. Xue surrendered, and the government placed him in jail, garrisoned Bailaohu, and in 1727 placed that area under the direct jurisdiction of Zhenning department.47 Kent Smith suggests that Xue’s easy capitulation was a direct response to the show of force in Ding-Guang, which had begun to “lend new credibility to the power of provincial officialdom.”48 Xue probably also calculated that exile to a faraway region was preferable to certain death in battle or by execution if he resisted.
The Standoff at Changzhai
During the last year of his tenure as governor-general, Gao Qizhuo formulated a program for the final consolidation of Qing rule in central Guizhou. In the spring of 1725, he put forth a plan to greatly increase the military presence in the Ding-Guang region. Rather than increase the province’s troop quotas, Gao planned to redeploy some of the units already stationed elsewhere in Guizhou.49 The plan encountered numerous obstacles along the way, and it fell to Gao’s successor, Ortai, to bring it to completion. The official with immediate responsibility for executing Gao’s plan was Shi Liha, the newly appointed acting governor of Guizhou.50 After surveying the Changzhai region in the autumn of 1725, Shi Liha voiced his intention to establish a substantial military force in the area. He planned to increase the number of troops in Dingfan by 340 men, and to add 431 troops to the garrison in Guangshun. In addition, provincial authorities planned to build seventeen guardhouses and stations in each of these departments, and eventually to construct a grand total of 370 new offices and military compounds. Workers would first build guard posts and stations near the administrative centers at Guangshun and Dingfan and continue the construction on a line gradually moving toward the interior of Zhongjia territory. Shi Liha hoped that this gradual penetration into the Zhongjia heartland would forestall any resistance from the local population.51
In spite of Shi Liha’s precautions, Qing plans soon encountered a major stumbling block. When the line of new military outposts approached the Zhongjia stronghold near the hamlet of Changzhai, just south of Ding-Guang, local residents blocked all entrances to the construction site, and prevented workers from entering with building materials, thus halting construction.52
The standoff at Changzhai persisted for the better part of 1725. In the meantime, the Yongzheng emperor made a pivotal decision for the people of Ding-Guang, and indeed for the whole of southwestern China. He appointed Ortai as governor of Yunnan, charged concurrently with the responsibilities of the governor-general of Yunnan and Guizhou. Immediately after assuming office in February of 1726, Ortai turned his attention to the crisis in Changzhai. Other provincial officials were divided on how best to deal with the Zhongjia. Some advocated a military strike, while others were reluctant to pursue such an aggressive policy. Among those clamoring for a military solution was the recently appointed provincial commander-in-chief Ma Huipo, but the Yongzheng emperor had thus far refused permission for armed intervention. He urged Ma to consult with Ortai before taking any action. Accordingly, soon after Ortai assumed office, Ma Huipo reported to him that Changzhai residents were burning the few structures the Qing workers had managed to build and were frequently blocking access to the construction sites. He suggested that troops from northwestern Guizhou already designated for permanent transfer to Ding-Guang might be sent to guard the construction work near Changzhai.53 Ortai readily accepted Ma’s suggestion, adding that he was by no means averse to using military force to break the stalemate.54
Matters on the ground in Changzhai further convinced Ortai of the need for an armed solution. During the first week of May, local officials informed him that the local Zhongjia, having rejected numerous attempts at negotiation and reconciliation, had set fire to Qing military buildings. Adding insult to injury, the Zhongjia had also flogged and released naked a Qing sublieutenant sent to remonstrate with them. Local officials also reported that the Zhongjia had stationed 50 to 60 men at each of the heavily barricaded passes leading to the Changzhai area. In addition, the local rebels had also stockpiled a large cache of war materiel, including poison arrows, spears, armor, and helmets. Shi Liha and Ma Huipo dispatched 2,700 troops to support those in the Ding-Guang battalion. Ma also mobilized several units of non-Chinese auxiliary troops and ordered the combined forces to assemble at a strategic point just south of Chang-zhai. Shi Liha and Ma Huipo then asked Ortai for permission to attack the Zhongjia and pacify the region before the hot, rainy season set in at the end of May.55
In his reports to the emperor, Ortai made it clear that he wholeheartedly supported the military solution. He also hinted at his intention to use Ding-Guang as a testing ground for policy in other non-Han regions of the southwest. He declared that if the government did not suppress the Zhongjia, “all the Miao will notice and become even more intractable. This suppression is not just calculated to exterminate the [Zhongjia] rascals. It is designed to pacify all of the Miao.”56
Ortai won approval for a military engagement, and in June of 1726, the first skirmishes broke out between Qing troops and Zhongjia gangs. The Qing army enjoyed overwhelming logistical superiority. Their army numbered between four and five thousand regulars, along with more than a thousand native troops. The Zhongjia were at a distinct disadvantage. They had no firearms and relied on an arsenal consisting chiefly of knives, armor, helmets, and spears. Most villages fell within a few days, although some held out for up to two weeks. In due course, however, Qing troops were able to subdue the entire region.57
Consolidating the New Order
The military victory only marked the beginning of the government’s job. Now it remained for Ortai to secure the region’s long-term stability and prosperity. To begin, he arranged for a larger permanent military presence in the area, with new units stationed at both Changzhai and Zongjiao, another strategic town. An additional 350 troops would be deployed in the area, evenly distributed between the two new command posts.58
Ortai’s next order of business was to rebuild the local economy. Following a personal visit to the region in the autumn of 1726, he devised a program to encourage the Zhongjia to resume normal agricultural activities, and to secure orderly conditions by imposing Chinese social, economic, and cultural patterns. He extended tax amnesties and a monthly grain allowance as incentives for Zhongjia to return to the area. Ortai’s plan also called for the establishment of a new resident official, the subprefectural magistrate (tongban) of Guiyang.59
Kent Smith notes that the physical presence of this official signified the region’s formal entrance into the imperial orbit. The government took additional steps to affirm this transition. Indigenes were urged to shave their heads and wear the queue, the distinctive hairstyle that signaled loyalty to the Qing. Local officials also prepared population registers, issued surnames to those who had none, and arranged the villagers into public security (baojia) units.60 By assigning surnames to local residents, encouraging them to wear the queue, and enrolling them on tax registers, the Qing state furnished itself with a means to identify its subjects, or, to explain the project in Scott’s terms, “to create a legible people.”61
A NEW ORDER IN SOUTHWESTERN GUIZHOU: THE CREATION OF NANLONG PREFECTURE
The land lay athwart the boundaries of Qian [Guizhou] and Yue [Guangxi]. Disputes and murders were frequent, and they were most difficult to adjudicate. When there was a crisis in Qian, the criminals fled to Yue; when there was a crisis in Yue, the criminals fled to Qian. When a crisis arose in the territories that belonged to neither Qian nor Yue, this created profound difficulties.
—Xingyi fuzhi (Gazetteer of Xingyi prefecture), 1851, 46: 10b
The next major undertaking in Guizhou involved the annexation of the Sicheng native prefecture (Sicheng tufu).62 This area, which extended from the northern banks of the Hongshui River in southwestern Guizhou to the You River in central Guangxi, had long been ruled by native officials of the Cen, a lineage with deep but obscure roots in southwestern China.63 Ethnohistorians usually classify the Cens as Zhuang, but it is unclear when and how they became Zhuang, and whether or not they were ethnic Han at some point in the distant past. The matter is complicated by the need to rely on genealogies designed to enhance the clan’s prestige and legitimacy by creating ancestral ties to Han Chinese personages from the central and eastern provinces.64 Some genealogies establish descent from the Han dynasty (206 B.C.E.–220 C.E.) general Cen Peng.65 Genealogies from the Ming and Qing periods list the family’s founding ancestor as Cen Zhongshu, a Song dynasty hero from Zhejiang who supposedly helped the general Di Qing quell a mid-eleventh century rebellion in Guangxi.66 The Song court rewarded Cen Zhongshu for his services by granting his family hereditary control over several native districts in northern Guangxi. In the centuries that followed, a succession of native officials surnamed Cen—who may or may not have been Cen Zhongshu’s direct descendants—expanded the clan’s domains through military conquests and rewards from the Song and Yuan governments.67 In 1340, the Yuan granted an official named Cen Shumuhan hereditary control over the Sicheng region. The Ming government renewed his charter soon after the Ming assumed power in 1368.68 By the fifteenth century, Cen rulers had conquered lands north of the Hongshui River, including a number of communities in present-day Guizhou. Territories south of the river in present-day Guangxi were known as inner dependencies (nei shao), while those north of the river were called outer dependencies (wai shao). Wherever the Cen acquired new lands, the original rulers became their vassals. Local rulers who already held titles as native officials were required to pay tribute to both the Ming court and to the Cen ruler.69
The Cens maintained cordial relations with the imperial government throughout the Ming period. The government periodically asked Cen native officials for military support in campaigns against non-Han rebels in the southwest. In return, the Cen rulers could rely upon imperial troops for assistance when they were harassed by other native officials, or when renegade vassals within their own domain needed a firm hand.70 When the Qing troops entered southwestern China in pursuit of the Southern Ming, the Cen rulers transferred their loyalty to the new dynasty.71 The Qing throne subsequently reaffirmed the Cens’ hereditary claim to the title of Sicheng native prefect and allowed them to retain control over their territories on both sides of the Hongshui River.72
After this auspicious beginning, the Cens’ relations with the Qing court gradually soured. Eighteenth-century writers often depicted Sicheng as a bumptious region of fuzzy boundaries, ambiguous jurisdiction, and frequent bloodshed. Numerous chieftains (tingmu), the landowning elite described in chapter 2, controlled small fiefdoms within the Cen domain. They were notoriously warlike, frequently engaging in land disputes and blood feuds. To make matters worse, these chieftains operated in a jurisdictional no-man’s-land, answerable to neither the provincial officials in Guangxi nor those in Guizhou. The Sicheng native prefect was responsible for disciplining wayward vassals and sending the more recalcitrant ones to provincial authorities, but he seldom did so. However, even if he had been more proactive in handling these disputes, he would have been hard-pressed to find an imperial official to help resolve them. Government offices in and near the Cen domains were widely scattered and poorly staffed. Magistrates assigned to the region often chose to live elsewhere for at least six months out of the year because northwestern Guangxi was a breeding ground for malaria and other tropical maladies. Officials decamped to healthier climes during the warmer months and returned to their posts only when cooler temperatures reduced the possibility of disease.73
The administrative presence was equally thin in Guizhou. The Ming government had established a department (zhou) at Pu’an; and in 1687, the Qing court had created a subprefecture (ting) at Nanlong in southwestern Guizhou, near the northern edge of Cen territory. These administrative units had little influence over affairs within the native prefect’s domain, and they were too remote from Qing political centers to function effectively. Both were jurisdictional subunits of Anshun prefecture, which lay 120 kilometers (75 miles) east of Pu’an and 160 kilometers (100 miles) northeast of Nanlong. Poor roads and rough terrain made it difficult for the local officials in either town to communicate with their superiors in the prefectural seat, or in the provincial capital at Guiyang.74 If a real crisis erupted, local officials could not depend on military assistance. During the Kangxi period, only 1,500 troops were stationed at the garrison in Huangcaoba, a small town in the southwestern corner of Guizhou.75
In the mid-1720s, Yongzheng’s reform-minded officials in southwest China resolved to impose order on the chaotic landscape of the Guizhou-Guangxi borderlands. The catalyst for this decision was a longstanding feud between two chieftains, one in northern Guangxi and the other in Guizhou, with contesting claims to a group of hamlets on both banks of the Hongshui River.76 Cai Chenggui, the highest-ranking military official in the region, reported the matter to Gao Qizhuo, who was then governor-general of Yunnan and Guizhou. Gao ordered the magistrates of Pu’an and Sicheng to conduct a joint investigation to determine the rightful owner of the disputed hamlets. When more than a year passed with no action on the magistrates’ part, Cai sent the emperor a lengthy memorial describing the sorry state of affairs in southwestern Guizhou. He explained that although he had some troops at his disposal, he was reluctant to use military force until he determined whether the hamlets in question belonged to Guizhou or Guangxi. He suggested that provincial officials jointly investigate the region to ascertain which chieftain governed the villagers, and to which province the chieftains paid taxes. Cai went on to decry the lack of civilian administration in the region. (The highest-ranking civil officials were the subprefect of Nanlong and the district magistrate in Pu’an; and although there was a small military presence in the area, army figures like Cai himself generally preferred not to get involved in civilian matters.)77 Cai ended by imploring the emperor to appoint more regular officials in southwestern Guizhou. The Yongzheng emperor responded by ordering the other high-ranking officials in Guizhou and Guangxi to assess the situation.78
The first report on the matter came from Ortai, who had by now assumed office as governor-general of Yunnan and Guizhou. In his view, the solution to the problem lay in the establishment of a clear boundary between Guizhou and Guangxi. As he wrote:
The Miao fight and plunder endlessly, and the local officials are lax, letting matters drift until they ferment into severe upheaval. . . . Only because [the provinces] adjoin each other along a jagged line do these disputes reach such an extreme. Now if we use the river to create order, with territory south of the river belonging to Guangxi and that north of the river going to Guizhou, then the boundary will be orderly and natural, military defense and criminal investigations will be greatly expedited, and clashes . . . between the two provinces will be forever resolved.79
Several days later, Guangxi governor Han Liangfu weighed in on the crisis. He opined that the two warring chieftains were not the real source of trouble in the Guangxi-Guizhou borderlands. The worst culprit by far was the native prefect, Cen Yingchen, Governor Han charged that Cen had repeatedly failed to deliver criminals to local officials, and that he was generally incapable of governing his people. Despite Cen’s many transgressions, Governor Han was reluctant to depose him, stating, “Only if this native official [Cen] acts in an outrageous and brutal manner . . . should we consider enacting gaitu guiliu.” The governor contended that the changes wrought by the sudden introduction of direct imperial rule might be too extreme for indigenous peoples who could not read or speak Chinese. To ease the transition, he suggested that the Qing court establish a subprefect for managing the “Miao” (li Miao tongban) at Sicheng. With the guidance of this new official, Han argued, native populations would “have the imperial way bestowed upon them and slowly reform their hearts and minds.”80 However, Han’s proposal was completely out of step with the activist policies that now carried the day. In his responding edict, the Yongzheng emperor approved Ortai’s proposal to use the Hongshui River as a natural boundary between the two provinces and summarily rejected Governor Han’s proposal to install a new subprefect at Sicheng. He also warned Han that military action might well be necessary if Cen resorted to rebellion. He ordered Governor Han and Ortai to proceed to Yunnan for a meeting with other high-ranking officials in the region.81
The conference in Yunnan ended with a resolution that all parties found reasonable. Because Cen Yingchen had never shown any inclination to mount an armed insurrection, military action would be premature at this stage. Everyone agreed, however, that Cen could not remain in power because he controlled both the northern and southern banks of the Hongshui. Ortai’s plan to use the river as a hard boundary between Guizhou and Guangxi could be implemented only if Qing authorities deposed Cen and annexed his territories. The officials at the Yunnan conference decided to give Cen the opportunity to surrender peacefully. Governor Han agreed to visit him in Sicheng and to extend a personal invitation to a meeting with Ortai at the Huangcaoba garrison in the summer of 1727. Han would travel to Sicheng with an entourage of no more than fifty soldiers in order to avoid arousing Cen’s suspicions. If Cen showed signs of resistance, Han would summon troops from nearby battalions in Guangxi. Once Cen reached the garrison in Guizhou, Ortai would invite him to resign from office. If Cen accepted, the boundary setting would proceed smoothly; if he resisted, Ortai would order military action.82
Although Governor Han remained skeptical about gaitu guiliu, this was a plan he could live with. He also agreed wholeheartedly with the decision to establish a firm boundary between Guizhou and Guangxi. As he told the emperor in a memorial from the spring of 1727:
As for using the river to delineate the Qian-Yue border, I think this is not only natural for [Guizhou] but also what ought to be for [Guangxi]. When Sicheng’s land is registered [as Guizhou’s], it will create more tax revenue for Guizhou. . . . Once we set Guangxi’s border according to the natural boundaries . . . these regions will no longer be beyond the reach of Guangxi soldiers when disturbances occur. . . . No more will there be the calamity of the strong bullying the weak. The winds of enmity, murder, and feuding may cease. Great indeed will be the benefits to the two provinces.83
The plans eventually came to fruition, but only after a few twists and turns. Upon receiving Governor Han’s invitation to Huangcaoba, Cen Yingchen sent out spies to determine the provincial officials’ true intentions. His informants came back with the news that Ortai and his colleagues planned to attack Cen in the hopes of forcing him to surrender his official title. Upon learning this, Cen sent three thousand of his personal troops across the river to Guizhou in advance of his visit to Huangcaoba. After Cen himself had crossed the river into Guizhou, he once again dispatched spies to investigate the situation in Huangcaoba. This time, his spies returned with the news that the Qing officials had no intention of taking military action against Cen unless he attacked first. Cen hastened to send a petition to the officials assembled in Huangcaoba, begging for clemency. He then ordered his troops to return to Sicheng. Once the soldiers had retreated, he prostrated himself before the provincial officials and relinquished his seal of office. Cen and his family were subsequently exiled to Zhejiang.84 Although puzzled by Cen’s actions, Ortai and his cohorts were pleased with the results, for the Guizhou-Guangxi borderlands were now under direct imperial rule.85
Over the next several months, Ortai carried out his plan to establish a new administrative order in the region. He suggested that the court create a new prefecture to consolidate control over the lands newly annexed to Guizhou. On Ortai’s recommendation, the old subprefecture of Nanlong was elevated to a prefecture, with jurisdiction over the Guizhou territories of Pu’an department and Annan county (Annan xian). Nanlong also governed two additional subunits that consolidated the newly acquired lands from Guangxi: Yongfeng department (Yongfeng zhou), and Ceheng subprefecture (Ceheng).86 The localities south of the Hongshui River were placed under the jurisdiction of Sicheng prefecture, which was now fully incorporated into the regular imperial administrative system. For better or worse, the Qing had asserted sovereignty over a land that had been part of China’s territory for centuries, but where no previous dynasty had ever successfully asserted direct political control.
When Qing officials decided to depose Cen Yingchen, they risked provoking an armed conflict much larger than the one in Ding-Guang. The native prefect commanded a sizable personal army, and other tusi were sure to offer assistance in the event of a Qing attack. The worst-case scenario could have involved a coordinated revolt by all of Cen’s allies in Guangxi, plus, perhaps, a few in neighboring areas of Guizhou and Yunnan—in other words, this could have been a regional war only slightly smaller in scale than the Southern Ming insurrection or Wu Sangui’s rebellion. However, this did not come to pass. Neither the Qing troops nor the native officials’ soldiers fired a single shot. Sicheng stands out as one of the few regions in southwest China where gaitu guiliu was achieved through a bloodless coup d’état rather than a protracted military struggle. Kangzuo was another region where gaitu guiliu involved a peaceful transfer of power, but there the stakes were lower because Xue Shiqian was less powerful and his domain was much smaller.
It is also worth noting that resistance to gaitu guiliu came not from the native officials, but from the commoners in Ding-Guang. Xue Shiqian and Cen Yingchen both capitulated readily because they stood to lose too much if they did not. The terms of surrender were favorable; the Qing would resettle the native rulers elsewhere in the empire, in areas where they could continue to enjoy lives of privilege. Resistance, on the other hand, meant certain death for the native rulers, if not in battle, then by execution at the hands of imperial officials who would surely judge them guilty of treason. Villagers in Ding-Guang, however, saw no such benefits in surrendering to the Qing. The arrival of direct rule threatened to constrain their livelihood choices. The increased administrative oversight meant that the raiding and banditry they relied upon to supplement their family incomes would now be criminalized. Faced with this knowledge, many local residents recognized that they could either submit meekly to the new order, which guaranteed only a life of continued hardship, or they could die fighting the Qing authorities. Thus, when local officials began erecting administrative buildings and army garrisons in Dingfan-Guangshun, these physical symbols of the imperial presence became easy targets for the people’s outrage.
Commoners in the newly constituted Nanlong prefecture did not initially protest the arrival of direct imperial rule because it had little immediate impact on their daily lives. Indeed, the changes might have been too subtle for them to notice. Even after additional Qing functionaries were installed throughout the region, indigenous social and economic structures remained largely intact. Most Zhongjia villagers continued to live as tenant farmers on the estates of tingmu.87 The main difference was that the tingmu were now answerable to Qing authorities rather than the do-nothing Cen Yingchen. If disputes or feuds erupted, local officials could put a stop to them before they escalated into drawn-out conflicts. In theory at least, all of the mechanisms were in place for Nanlong to become a full-fledged state space. The Qing government had succeeded in increasing its capacity to “see like a state”—to carve out new administrative units and project legibility into them. As chapters 4 and 5 will illustrate, however, local residents soon exhibited an equal capacity for perseverance, adaptability, subversion, and sometimes outright resistance. Their livelihood choices transformed the Yongzheng-era templates for standardization and centralization into a shifting matrix of conflict and compromise.