Tusi and the Chinese Administration Chinese influence in western Yunnan became increasingly important through much of the time period of interest here by means of extension and penetration of Imperial administration through transformation of the role of indigenous rulers, Han dominance of trade, and the massive influx of Han migrants. Qing penetration was not consistently successful, however; it weakened with periods of unrest and administrative decentralization. The role of the tusi was hollowed out and transformed but not eliminated and the tusi remained significant middlemen in Chinese administration. Full administrative control was not implemented throughout southwest China until the Communist government took control. The system of ruler through native rulers or tusi was established in Yunnan in the Yuan Dynasty (Mongols, 1279-1368) as a temporary necessity, transitional to the incorporation of the frontiers of Yunnan and Sichuan into the Chinese state (Lee 1982a, b; Mote 1999:703). The system continued through the Ming Dynasty, in which the main activities were military campaigns and the establishment of military garrisons. Some colonization occurred in the vicinity of the garrisons and soldiers both worked as farmers and were furloughed from service to settle nearby (Mote 1999:710). The main means of Chinese control of the frontier was the appointment of indigenous rulers to represent the Chinese state, but these rulers functioned within the pre-existing Tibetan conceptual framework of power and Tibetan influence and local autonomy reasserted itself whenever Chinese power weakened. The rise of the Qing Dynasty changed the role of the tusi. Before the mid-Qing, most of the southwest frontier of what is now China was administered under this tusi system. It was not until the late Qing Dynasty that the Sino-Tibetan frontier became a site for more extensive Chinese imperial expansion. The Qing reformed the native chieftain inheritance process and tied it to the Chinese public education system in hopes of taking control of the selection process and to increase their “political legitimacy and cultural prestige” (Herman 1997:48). Tusi were “hereditary” indigenous rulers invested by the Chinese government to represent it in the border regions, a form of indirect rule. A whole parallel administration was established in this way. Native people were not registered as Chinese subjects; rather their own chiefs were confirmed to rule their own people in the” traditional” ways. Tusi had offices analogous to those in the Chinese system of ranked civil service posts, tu being the prefix for “local” or “native.” Appointees received formal attire, official seals, and patents of their office (Mote 1999:703). While the position was formally hereditary, it was not in fact (Chao 1995), and the Chinese government could remove the tusi if needed. The status was considered a reward for voluntarily submitting to Chinese authority or for military service to imperial armies (Lee 1979:54). Leaders of rebellions, for instance, could be incorporated into the Chinese military administration, an example of transformation from heterodox to orthodox military leadership (Feuerwerker 1975); Wiens 1967 [1954]:222). Native chiefs were required to maintain peaceful relations among themselves and with neighboring populations of Chinese. Tusi were nominally under the control of Chinese magistrates, but in most cases the Chinese rulers were too far distant (or, in later times, too weak or corrupt) to have great impact on the dynamics of tusi relations with their subordinates. In time, requirements for tusi rank were established, such as a certain degree of literacy in Chinese and knowledge of Chinese literature, so that it served as an avenue for Sinification. A tusi could also prove his worthiness to the Chinese through military service or a career with the imperial guards and battalions garrisoned in their territory (Mote 1999:704, 706). The position of tusi was one of cultural duality (Mote 1999:706). Even while adhering to Chinese requirements of the role, a tusi‘s authority occurred within indigenous and Tibetan political and ritual frames and depended on his ability to validate that power in culturally appropriate ways, such as displays of generosity. This cultural duality created a “soft border” of interaction (Mote 1999:706). In this role, tusi served as buffers against foreign polities and between ‘barbarian’ peoples and the Chinese state. Chinese rulers had their strategies in the tusi system and the local rulers theirs. Tusi did not take the interests of the Chinese state; they sought their own power, autonomy, wealth and control, and they sought these in their own terms regardless of the goals of the Chinese state. Their terms were more congruent with Tibetan than with Chinese culture. The tusi relationship did not mean that the territory was under Imperial control, even if Chinese courts believed so (Sun 2000). And they were considered troublesome; from the Imperial point of view, tusi were easily seduced by external forces or prone to internal disorders; indigenous rulers were unreliable, instigators of strife (Wiens 1967 [1954]:240). Native chiefs worked to their own interests and used the ‘validation’ of the Chinese state toward their own ends vis-à-vis their followers. In fact, Herman points out that the Qing reforms were in response to the instability of inheritance, but what started out as reform policies inadvertently resulted in an increase in violence from new social tensions that arose as tusi used Chinese legitimation to strengthen their own positions at the expense of other indigenous elite (Herman 1997). Administrative problems with tusi are often discussed in terms of integration of a ‘frontier’ area of China. Yet whatever its political and administrative difficulties, the tusi system had deep social significance for segments of indigenous southwestern societies. It further legitimized the power of native chieftains. Tusi were important in the cultural and political fabric of southwest China, the point men for assimilation into Han culture and incorporation into the Chinese empire, and simultaneously local actors strategizing in the context of the intertwined indigenous and Tibetan cultures. The Qing saw the tusi as the origin of lack of control. Chinese fugitives and rebels could trade on their literacy in Chinese to collude with native chieftains and challenge Qing control (Chao 1995:45; cf. Giersch 2001:67-68), thus providing legitimation for indigenous polities and haven for people considered dangerous to the state. The Qing also held that native leaders were to blame for the prolonged state of disorder in the southwest as they competed for inheritance, control of legal and illegal trade, and benefits to be gained from collusion with Chinese officials in the region. The Qing also faced the problem of “upstart” indigenous rulers who contested Chinese rule (Mote 1999:902; Chao 1995:45). An illustrative case is of the tusi of Lijiang. The Naxi of Lijiang, while heavily Tibetanized, had also been under Chinese suzerainty for nearly the longest period of time among the peoples of the northwestern Yunnan Sino-Tibetan region. The tusi family had made itself rich through this status (Mote 1999:711). At the time of Qing penetration into western Yunnan, the power of the tusi of Lijiang was waning and their domain shrinking. The tusi family resisted Wu Sangui’s rebellion (Lee 1979:47), likely in response to Wu’s strategies in relation to them, an attempt to court one patron (the Qing) over another (Wu). Wu was pursuing a policy of consolidating direct control over tribal chiefs, winning the friendship of the Dalai Lama, and ensuring open trade routes between Kham and Yunnan in order to trade Yunnanese tea for Tibetan horses for Wu’s army, and twice ceded Lijiang tusi territory to Lhasa (Lee 1979:53). Before the Qing had firmly established their government in Yunnan, tusi power had grown and peaked, especially during the Wu Sangui feudatory. The Qing concluded that the tusi had to be curbed (Lee 1979:51). Starting in the 1720s, the Qing eliminated a number of the most powerful tusi in favor of direct government administration; in 1723 the Qing overthrew the tusi family of Lijiang and asserted direct administration over Lijiang. Tusi strategizing to resist Wu and support the Qing had made no difference in the light of Qing administrative needs. The ruling family was allowed to keep its rank as tusi (Lee 1979:47) but without the power to rule; the family continued to petition for restoration of their powers for many years (Rock 1947:296, 310n, 320). This hollowing out of the tusi role continued through the Qing and into the Republican eras in many parts of the southwest. Profitable trade, ‘open’ lands, and expansion of direct administration facilitated a huge influx of Han into Yunnan, which in turn necessitated even better administration (Mote 1999:953). Beginning with the early eighteenth century there was both increased immigration of Han, often in the wake of military campaigns, and a more energetic policy of Sinicization of the non-Han indigenous peoples (Chao 1995). This gave rise to conflicts between indigenous and Han peoples – often defined as ‘banditry’ perpetrated by indigenous peoples against Han – that the Qing government sought to quell. In response, the Qing established a stronger organizational structure in Yunnan in the early eighteenth century. A special administrator was appointed to the southwestern provinces (E’ertai, administered 1726-31); his goal was to transform the tusi native chieftainships into appointed officials. In much of Yunnan, but not the far northwest, the process was largely effective. E’ertai conducted campaigns to pacify warring tribes; abolished half the native chieftainships; imposed regular Chinese local government, especially where increasing numbers of Han civilians were settling; extended taxation to large sectors in the provinces of the southwest; brought about reforms in taxation and governance; supervised improvements in irrigation; established Confucian schools; and encouraged the copper mining industry in Yunnan (Mote 1999:902-903). By the late eighteenth century, the territory under tusi semi-autonomous jurisdiction in Yunnan had been reduced from half of the provincial area to one-quarter. This increased the tax base of the government. Under the tusi system, non-Han had not been registered in census records, nor had they been taxed as Chinese citizens. Under the standard provincial administrative structure, even non-Han peoples became taxable, recorded residents of the Chinese state (Lee 1982a:728-30). And so, imperial control was extended. The form of the role of the tusi changed through this history. The nature of their position in the Chinese state and their strength in their regions varied considerably. The administrative reforms of the 1720s were most effective in central and along the southwestern Sino/Burmese border. They were not complete or permanent in northwest Yunnan (Gros 1996). In some cases, as in Kham, tusi were for the first time appointed as the “international” frontier expanded westward. But control was incomplete. On the Sino-Tibetan border, indigenous chiefs resisted Qing attempts to regularize administration through setting of tax rates and ending debt slavery. The chiefs continued to collect a variety of taxes and rents and it was not until eight years later, in 1803, that the chiefs were dismissed when they were suspected of supporting a Lisu uprising (Gros 1996). Where tusi power was entrenched or the region very remote, administrative reforms proved to be short-lived. Because the later Qing and Republican periods in Yunnan were marked by administrative weakness, official corruption, banditry, uprisings, and rebellion, the tusi system remained in some form through Republic and up to the beginning of Communism, particularly in the Ta-liang Shan, and southern and western Yunnan frontier areas. Many tusi maintained or regained elements of local power and certain positions remained up until the 1940s. Indigenous rulers remained important in part because the weak central Chinese government made them necessary. The government did not have the resources to maintain full government on this periphery, but had to maintain some presence to prevent European powers from expanding their sphere of influence and snatching Yunnanese territory (Wiens 1967 [1954]:240). But the tusi were only minimally able to protect the frontier. Aside from their own interests, there were no systematic instructions from the imperial government (Wiens 1967 [1954]:240). As a result, tusi domains were lost: Meng-mi, Mu-pang, Meng-yang, Man-mu, and Meng-lang were lost to British Burma; and Meng-so, Meng-la, Meng-wu, and Wu-te were lost to French Indochina (Wiens 1967 [1954]:240). One large patch of territory claimed by both Burma and China was demarcated only after WWII. The establishment of the Republic (1911) further weakened central administrative control, thus strengthening the position of local rulers and warlords. There was no national policy toward the tusi system in the Republican period; each province was allowed to make their own policies (Wiens 1967 [1954]:240). Between 1911 and 1923, six tusi were abolished in Yunnan, but all districts reverted back to tusi rule in 1929. There were 20 tusi in power in Yunnan in 1994 (Wiens 1967 [1954]:245; Gros 1996, 2001b). Even when Han civil servants were appointed, the tusi could and did retain their old power. Some tusi retained their power; some their title; some retained both (Wiens 1967 [1954]:240, 245). Many lost their political juridical power, but retained the power to collect rents. In effect, many tusi became glorified landlords (Wiens 1967 [1954]:246), but landlords with guns, local respect, and control of significant trade. Local militarization in the opium wars had allowed local gentry – tusi and Han – to gain considerable influence over local government as a landlord class. Personal authority replaced imperial command, and after the 1911 Revolution the armies of the local gentry became a destructive force as autonomous warlords (Walker 1991: 9). Of great significance, especially in the warlord period, was that the tusi not only had deeply entrenched political control over other indigenous peoples, they were often well armed with large quantities of regular army rifles. The ‘customary’ law of draft by the tusi, made it possible for them to call up large military forces when threatened. Tusi understood indigenous law and ritual, so that local people had more confidence in them than in the Qing and later the Republican states. Furthermore, the tusi had made alliances among themselves. It would have required large numbers of military force to suppress the tusi in their own terrain (Wiens 1967 [1954]:249-250). Local Chinese officials who attempted to extend administrative control or enforce taxation were resented not only by local rulers but by the merchant class, as well. Kingdon-Ward writes that the merchants of A-tun-tzu (north of Lijiang, culturally Tibetan and an important city in the Sino-Tibetan trade) hated the Chinese civil officer because he fought opium smoking and introduced a tax on musk passing through the village. The musk trade was the most important on the border (Kingdon-Ward 1986 [1923]). Thus, were the Chinese to attempt to abolish the system, the tusi were likely to have brought about civil war, which would have invited British intervention. Tusi were retained because of Chinese fear of British control in Burma; the Qing and the Republicans had to do the best they could to retain control in Yunnan (Wiens 1967 [1954]:254-5). Early twentieth century Chinese political scientists noted how effective the British system of indirect rule was and argued for maintenance of the tusi system because to destroy native rulers’ power at this time and remove local institutions such as the Tibetan Buddhist monasteries would drive people into the arms of the British colonial government (Wiens 1967 [1954]:254-5). In the end, Han civil administrators were forced to depend upon tusi to enforce order, and the central Chinese state was even more dependent upon them to protect the southwestern borders (Gros 1996, 2001b; Wiens 1967 [1954]:245). The tusi system had lasted longer than planned when it was first set up, and it may have been the reason for many of the problems of Chinese governance in the southwest (Mote 1999:902), but it continued to exist in large part due to the internal conditions of Yunnan – economic, political cultural, and geographical. In short, tusi were the point men for administration and cultural assimilation but working within the indigenous/Tibetan conceptual frameworks of power. Shifting alliances were the norm. In this, the southwestern frontiers were administratively unique from both Tibetan regions to the west and Chinese areas to the east. Tusi were the main medium through which local people had contact with Chinese government until the nineteenth century. With administrative changes from the 1720s to the 1940s, the role of the tusi was transformed as well. Historical changes also brought about a “hardening” of a formerly soft border subject to frequent change when French and British colonial interests began surveying the southwestern border in the late nineteenth century (Mote 1999:709). As will be seen below, removal of local leaders and remaining leaders’ collusion with corrupt Chinese administration brought about profound social dislocation for local people, such as banditry and alienation of land.
Warfare Between China and Tibet When Ming China had extended administration in Yunnan, it focused on the more congenial environment of the southwest borders than on the austere northwest where environmental conditions were inhospitable to Chinese methods of agriculture (Lee 1979:35; Wiens 1967 [1954]:8) Tibetan influence was strong and not challenged until later under the Qing. Even then, Tibetan and Chinese influence coexisted in uneasy equilibrium while both attempted to extend their dominance (Lee 1979:36) By the mid-nineteenth century relations between Tibet and China were unstable due to the expansionist policies of both (Gros 1996). Relations between China and Tibet were very complex. At times, Tibetan rulers called on the Chinese Court for support in dynastic disputes or for aid in fighting off other powers. When the Seventh Dalai Lama was found in Kham, the Qing emperor sponsored him and claimed control of Tibet at his accession. But Tibet remained administratively independent; while the Qing considered the Dalai Lama a subordinate ruler, the Tibetans regarded the relationship as one between priest and patron (Samuel 1993:50-1; Mote 1999:876-878) The uncertain relations between these two powers, as well as local polities’ attempts to maintain autonomy or choose one of these powers as a patron, contributed to conditions of recurring violence and warfare that profoundly affected the people of northwest Yunnan. In the early eighteenth century, Qing administrative reforms and consolidation of control brought some measure of peace to Yunnan. The boundary between Kham and Lhasa was fixed at this time and the Tibetan chiefs of the Kham states and tribes were made semi-independent feudatories of China. However, the polities of the northwest of Yunnan, particularly Kham, remained autonomous and a source of instability for the Sino-Tibetan frontier (Teichman 1922:2). As Qing control over Lhasa waxed and waned through the eighteenth century, so too did the degree of control exercised by China over these semi-independent feudatories of Kham (Samuel 1993:69). In the far west, peace was achieved not through imposition of direct administration but a negotiated peace with Tibet and Tibetan rulers in Kham, and through investing tusi in 1728 (Lee 1979:60). Relations with the people of Kham had been problematic for some time (Samuel 1993:69) In 1718, the Khang-hsi emperor (Qing Dynasty) was determined to regain control over Tibet, despite opposition from his court. He maintained that Tibet was a shield for Sichuan and Yunnan and had to be controlled (Lee 1979:55) The southern army of the Tibetan expedition had to pass through southern Kham (in the northwest of Yunnan), which was not yet under Qing control. The conquering Qing armies accepted the surrender of a number of chiefs. To facilitate the movement of troops, the emperor set up 87 garrison depots along the route of march. Even after imperial troops withdrew from Lhasa, garrison forces remained in Kham. This was the beginning of the Qing control of southern Kham. When the rulers of Tibet again asserted their independence from Qing rule, Chinese troops were again sent and many native rulers of Yunnan who had professed loyalty to Tibetan rulers asserted their loyalty to the Qing in 1724 (Lee 1979:56) Still later, the Qing carried out two huge campaigns against one tribal group in Kham until the mid-eighteenth century; these people were eventually annihilated and replaced with a more peaceful group (Mote 1999:936-7). The Chinese administration attempted to win the loyalty of indigenous people and their rulers by maintaining the jurisdiction of indigenous rulers; setting an imperial tax at a lower rate than they had paid to their rulers in Tibet; and eliminating the tax that trading caravans of the Dalai Lama had been exacting from the local monasteries and the tariff to the Chinese paid by trading caravans at a key way station. Military garrisons were to be established at all strategic points in Kham. Chung-tien and Weixi districts were returned to Yunnan from Tibetan authority, but lands west of the Nu River were given to the Dalai Lama in 1725 (Lee 1979:56-57). Lisu resided west of the Nu (Salween) River. Elsewhere in Yunnan through the eighteenth century, there was a border war against the Burmese in 1766-1770 and another against the Annamese in 1788-1789 (Mote 1999:936-9). The Qing expelled the Gurkhas from Tibet in the 1790s (Mote 1999:936-9), all of which doubtless affected the local populations through demands for troops, corvee, and taxes. Despite a period of peace in the early nineteenth century, the Tibetans, both as a population and in terms of their religious-based political influence, continued to be a problem to the Qing Court in terms of establishing both borders and hegemony. Dergé, in Kham, was the third largest independent Tibetan polity after Lhasa and Bhutan. It reached the height of its power in the eighteenth and nineteenth centuries. Also significant were the Hor or Trehor states (Samuel 1993:320-22). The Rimed movement, a synthesis of clerical and shamanic elements of Tibetan Buddhism that retained a more central place for shamanic elements, arose out of monasteries in Dergé and Kham; violent factional disputes among monasteries adhering to Gelugpa or Rimed tenets sometimes arose and the consequences spread (Samuel 1993:533-37). The limited and varying nature of Chinese and Tibetan control can be illustrated by the following incident. In 1848, a Qing imperial expeditionary force was dispatched against one of the Nyarong tusi for robbing people in a neighboring region and killing a Chinese officer. The chief was captured, killed, and his territory divided among the other tusi of Nyarong (Lee 1979:60). A little over a decade later, when a Nyarong chief arose in Dergé who engulfed eastern Tibet in the 1860s, it was the Tibetans who intervened rather than the Chinese, despite the fact that the area had previously been independent of Lhasa (Samuel 1993:69, 77, 542). Nyarong was off the main road to Lhasa, thus shielded from the eyes of the Chinese officials, but incorporated on a tusi basis into the Qing empire in 1728 (Lee 1979:60). This chief united Nyarong (Chan-tui) by conquering his smaller neighbors and all of the other Nyarong tusi. By doing this, they could block the caravan route to Lhasa. In 1862, after a tea caravan had been robbed on its way to Lhasa, the Tibetan government petitioned the Qing imperial court for a combined Sino-Tibetan campaign against the chief. The petition was granted but the provincial government of Sichuan was too involved with the Taiping and Muslim Rebellions to spare troops. Finally, in 1865, the Qing court agreed to a joint campaign just as Tibetan troops were approaching Nyarong with the support of many Kham chiefs. The governor-general of Sichuan asked them to stop their advance. The Tibetans refused. When the Tibetans won without Chinese support, they demanded payment from the Chinese for their withdrawal. Unable to meet this demand, the governor-general suggested that Nyarong be given to the Tibetans (Lee 1979:60). From that time on, a series of Tibetan governors of the region sought to expand their influence, which was easy to do in the face of the weakness of the Qing court (Lee 1979:61). Western observers who were beginning to explore the territory and proselytize the people of western Yunnan noted that the Chinese did not seem able to assimilate the Tibetans, either culturally or politically (Davies 1909:386). Tibetans and related people did not accept Chinese rule. The pattern was described as one of Chinese subjugation of Tibetan and other local populations; local garrison joins with natives in a revolt; lamas and local people expel the Chinese; the Chinese re-subjugate them; and the cycle starts again (Teichman 1922:197). Following the Muslim Rebellions, the vice-governor of Yunnan published a circular against the Tibetan Buddhist lamas in 1874, which led to local unrest in northwestern Yunnan (Gros 1996). Tensions heightened in 1904 with a series of ‘rebellions’ and increased intervention by the Qing. An incident in 1904 in which Chinese officials and French missionaries were killed and a series of skirmishes with Tibet and the Tibetan rulers of Kham prompted the Chinese state to inaugurate a new interventionist policy in Tibet and tighten control over “parts of Tibet in China,” i.e., in Sichuan and northwestern Yunnan, through 1905-1911 (Davies 1909:387; Teichman 1922:195). Chinese interests were also heightened by concerns over British and Russian competition for control of Lhasa, instigated by the 1903-1904 British Expedition to Central Tibet and the Anglo-Tibetan Convention that resulted from it (Lee 1979:35, Samuel 1993:70-1); the prospect of Tibet opening up independent foreign relations with the British and Russians worried the Chinese imperial government (Wiens 1967 [1954]:237). Yet the extension of Qing administration further exacerbated unrest and resulted in uprisings among the Tibetan population throughout Kham. Qing troops sent to Kham killed several thousand people, destroyed lamaseries, and forcibly removed the rulers of the several small states of Kham. In the end, there was not a single Tibetan ruler left in Eastern Tibet – from Tachienlu up to the Mekong, Chinese magistrates administered the country (Samuel 1993:70-1). West of the Mekong and Salween, several districts were planned but not actually established. The consolidation was successful, but only temporary. When the Qing Dynasty fell, there was a general uprising against the Chinese in Central Tibet and Kham, resulting in Lhasa establishing control over Kham. In the words of the British consular official who negotiated a border agreement between Tibet and China in 1918, “the whole edifice of Chinese control was ... but a hastily constructed framework, imposed on an unwilling people taken by surprise, and the greater part of it collapsed completely when put to the test at the time of the [Republican] revolution ...” (Teichman 1922:33). Multiple, non-border-bound allegiances remained the pattern. The territorial border was established at the Yangtze River, but Lhasa retained rights over lamaseries in territories to the east of the Yangtze, which was under Chinese control. Chinese control remained slight (Samuel 1993:70). It was only nominal at the time of Teichman’s travels and he encountered constant rumors of attacks and roving bands in China who claimed the authority of Lhasa, which Lhasa denied (Teichman 1922:198). In 1930-1933, the Chinese took control of the lamaseries following sectional conflict among them. Still, as in the twentieth as in the nineteenth century, Chinese authority rarely extended past the small garrisons on two main trade roads for the Chinese-Tibetan trade, and only nine of the thirty-one Chinese magistracies established following the 1904 rebellions remained in Chinese hands. In areas where local chiefs and rulers had been removed, banditry was common (Samuel 1993:71). Peace was best obtained by leaving them alone (Teichman 1922:197). Even earlier, about 1922, ‘Tibetan’ outlaws often marauded on farmers in western Yunnan. Rock commented: “This state of affairs is only attributable to the short-sightedness of certain Chinese militarists who personally kill or order the killing of inoffensive native rulers who alone are able to control these wild border tribes” (Rock 1947:417). French travelers in Kham in the late 1930s and 1940s recorded the local autonomy despite a veneer of Chinese control. They noted that Kham possessed a Chinese name only and did not differ in any profound way from the other districts of Tibet; Chinese administration was largely ignored by the Tibetan population, who obeyed only their own chiefs; and Chinese officials had to engage in barter to subsist as Chinese currency was not accepted by the local population (Samuel 1993:72, quoting Migot and Guibaut). Thus substantial Lisu populations in the north were more closely under Tibetan rule or influence than Chinese, especially in the far northwest. Lisu living further south were far more exposed to the state penetration and administrative reforms of the Chinese state. When the Communist government took over and established direct administration of Eastern Tibet and extended that into Tibet proper, the rebellion against Chinese government that resulted in the Dalai Lama’s flight to India in 1959 started in Eastern Tibet (Samuel 1993:544). It was a region that remained largely independent from governmental control until the 1950s.
Shifting Borders in Northwestern Yunnan Tusi, as indigenous leaders, worked within indigenous frameworks that were understood across the region by many different indigenous peoples and the Tibetan polities. People continued to follow the political lead of their tusi (Wiens 1967 [1954]:249). In this context, the tribal peoples tended to choose the governance of whomsoever they found to their advantage. Chinese authorities feared that either villages or tusi would side with the British with good reason. In one case (the P’ien-ma Frontier Incident), two headmen in a tusi domain who were both Han but had been brought up in Kachin society complained of the tusi to Chinese authorities. Chinese officials were swayed, however, by bribery from the tusi and found against the headmen. The headmen then told their followers that to escape this tyranny they needed to petition the British authorities to recognize P’ien-ma as being under British jurisdiction. The British accepted and sent troops to protect the village. In winter of 1910, 2000 British troops occupied the territory and stationed garrisons in all the important villages (Wiens 1967 [1954]:256). Nine years earlier, a Lisu rebellion in Yunnan had also brought in British troops (Hertz 1912, see below). In another case, a tusi refused to punish Kachin bandits of a Chinese opium caravan because the Kachin were their neighbors and their interests lay with them (Davies 1909:38-39). This does not reflect the objective superiority of one form of rule over another. In an earlier time, Chinese rule had presented an attractive alternative to Tibetan rule. In 1720 when a Chinese garrison conquered the salt marshes of Yangjin, south of Tibetan Tsarong, local Nu (Lutzi) paid tribute of animal skins, medicinal plants, and hemp cloth to the Chinese to ensure their protection against Tibetans and Lisu, and to receive land tenure rights (Gros 1996). Such strategies were possible in the context of the multiple spheres of authority in the region, and indigenes’ strategies continued within that framework even in the face of European colonialism and expanding Chinese hegemony. An example of how the system worked ‘on the ground’ can be found in Kham. Here, tusi by and large remained powerful throughout the Qing and Republican periods, rooted as they were in indigenous power. The political system was a complex one, in which Chinese and Tibetan power were encased locally. While the Chinese were replacing indigenous rulers with Chinese administrators in much of the rest of Yunnan, in the northwest frontier area they were just beginning to attempt control. They started by appointing tusi. In the Kham region at the end of the Ming Dynasty, the indigenous rulers in the east submitted their allegiance to the Qing court (Lee 1979:55). Many tribal chiefs submitted to the Qing authorities in Yunnan during the seventeenth century because as Qing armies entered Yunnan in pursuit of the Ming remnants, tribal peoples along the route of the march submitted to the Qing, their leaders being rewarded with formal positions (Lee 1979:54). After a round of warfare between Tibet and China from 1718 to 1724, the Chinese state established the minor prefecture of Weixi in the Mekong Valley in 1729 as a means of ending continuous conflicts between the Naxi and Tibetans. The prefecture controlled villages in both the Mekong and Salween Valleys, which are very close to each other at that point. There were Chinese garrisons at various locations, but indigenous “hereditary” Naxi chiefs were appointed to administer the territory on the behalf of the Chinese state. The Naxi tusi were confirmed in their offices by the Chinese, granted fiefs, and bestowed military titles in exchange for collecting taxes and quelling ‘rebellions’ (Gros 1996). By the mid-nineteenth century, two main Naxi chiefs ruled the Salween Valley under Chinese support – the tusi of Kangpu and Yezhi. The latter collected taxes in the southern part of the Salween Valley, and his territory adjoined that of a population of independent Lisu; the former was closely involved with Tibetan polities to the north. In addition, certain Tibetan Buddhist temples were in charge of collecting tribute in the name of the Chinese Prefect of Weixi (Gros 1996). Further northwest in Kham, the tusi of one prefecture were eliminated, opening the way for the establishment of military and civilian colonies in 1733. However, a traveler in 1744 noted that only the road and the towns itself belonged to the Chinese; the rest was tribal territory (Lee 1979:51). Agricultural conditions were so poor that few Chinese settled outside of the main towns. The establishment of tusi rulerships in Weixi Prefecture appears to have followed a pattern set up centuries before, similar to when Lijiang was established as a tusi kingdom. Lijiang had been an indigenous kingdom with a king recognized in Tibetan chronicles. The Chinese government recognized it as important in their battle with Tibet for control of the region, and confirmed the ruler of what became Lijiang (Gros 1996). The Chinese courts used indigenous kingdoms such as Lijiang as proxies in holding off or usurping other powers in the region, particularly Tibetan states. They used traditionally hostile native groups to rule and suppress each other (Mote 1999:707). Despite their at least nominal allegiance to the Chinese state, the tusi advanced their own interests, as well as those of Tibetan polities and monasteries, and the Chinese were often rather more dependent on the Naxi chiefs than otherwise. In practice, the tusi enjoyed great independence; confirmed in their offices and granted fiefs and titles, they were very influential locally and quite independent due to their geographical distance from Chinese administration. The fiefs in fact resembled principalities based on the indigenous Naxi model of rulership, and the subject population called them king (wang). The tusi also delegated a certain part of their authority to other, smaller chiefs to represent them in regions distant from their fiefs. These included influential Lisu in certain areas (Gros 1996, 2001b). Tusi derived important private benefits from taxes, commercial activities, and revenues from the estates granted them (see Daniels 1994; Mote 1999:711). In the north of the Salween Valley, they had to reconcile with the Tibetan authorities as well; the Tibetans, as mentioned previously, were collecting different taxes on the same population ruled by an important local temple (Gros 1996, 2001a). As was previously discussed, these tusi also built and supported a number of Tibetan Buddhist temples and monasteries (Gros 1996). The statuses within the tusi system among the Naxi rulers and elements such as the use of surnames for hereditary (patrilineal) positions were “markers of social standing that cemented the authority of local ... rulers over their subordinate subjects” and implied a specific relationship of entitlement on the part of the tusi based on ties with the Chinese empire (Chao 1990:36-7). As in Southeast Asian galactic polities, this likely created chieftains and rulers where none had permanently existed before (Jonsson 1996). In form, these political relations seem far more similar to galactic polities in that they are relational rather than absolute. Leaders allied themselves with greater powers to consolidate their positions; local people switched ethnic identity depending on political and cultural power. The interest of the imperial government in the tusi system was not political effectiveness but control of ‘barbarian’ populations. Their chiefs, whose tenure in office was contingent upon how few problems they caused the Chinese government, and probably on how well they were able to enrich local officials, imposed much of the ruthless exploitation of native peoples. By emphasizing control over governance, the Chinese encouraged the degeneration of native leadership institutions (Mote 1999:716-717; Wiens 1967 [1954]:232-233). Indigenous leaders and Han Chinese are recorded as allying in ‘rebellion’ against Chinese government; we cannot see such political relations as a strictly Han vs. indigenes phenomenon (Giersch 2001). And the imperial administration was not always able to withstand the demands of tusi and other native officials. One example of this comes from Ming Dynasty Sichuan, but is instructive. The family of hereditary Pacification Commissioners in southeastern Sichuan was Han but had intermarried with local indigenous people and by the late sixteenth century had built a strong base of widespread support among Miao people in three adjoining provinces. The leader at that time was reported as being degenerate, cruel, guilty of many murders, and a political trickster. His own associates in the tusi government complained about his behavior to Chinese provincial authorities. However, they were reluctant to act because of his family’s record of loyal service to the throne. This leader had personally contributed valuable goods and Miao labor to the construction of an imperial palace and also dispensed large quantities of gold in bribes to local officials. Most importantly, he was seen as a supplier of military forces to be used against other uprisings. Thus, the court sought to negotiate a resolution. This leader bided his time, making himself indispensable to the imperial government, and doing as he wished. It was not until nearly 20 years later that he was suppressed (Mote 1999:715-6). By playing this game of subservience and tribute to the imperial court, a native leader could maintain a high degree of internal autonomy for long periods of time providing he had the loyalty of his subordinates and control of valued natural resources. To look at Yunnan as a ‘frontier’ of China does little to contribute to our understanding of these places, particularly their internal dynamics. The period of the mid-eighteenth century marked the extension of Chinese administrative control over the whole of the Sino-Tibetan border, as China continually fought the power of Tibet and its rulers, whether the Dalai Lama or other Mongols. Instead of being an external frontier demarcating China from foreign rulers, the Sino-Tibetan frontier had “become an internal frontier separating Chinese culture from tribal cultures” (Lee 1979:59). Geography, religion, politics, and trade oriented this region toward Burma and Thailand to the southwest of Yunnan and Tibet to the northwest (Giersch 2001:73; Gros 1996; Sun 2001). Another model of Yunnan and the southwest is of a “soft border” (Mote 1999:706), in which Chinese and indigenous cultures and forms of rulership intertwined in a kind of cultural duality. This may perhaps be a better description of the situation, but it is most appropriate in discussing the ‘point men’ of this system, the tusi and other local rulers through whom the Chinese and Tibetans acted. What borders there were overlapped – frontier Yunnan in the eighteenth and nineteenth centuries was more than a place between or a middle ground where the standard Qing bureaucracy met the native officials of the ‘frontier’ (cf. Giersch 2001). Few demarcated political boundaries existed until the 1890s (Giersch 2001:71), when western colonial powers delineated borders through the complex array of small, relatively autonomous polities (cf. (Thongchai Winichakul 1994). This was the “hardening [of] a line that until that time had remained soft and subject to frequent change” (Mote 1999:709). To view Yunnan’s society from the point of view of its eventual incorporation into the Chinese state is to fall to presentism. The place of indigenous concepts of rulership and of Tibetan cultural, religious, and political influence cannot be ignored. This is not to negate the important role of Chinese political institutions in the constitution of these societies. China was a powerful empire; under certain emperors, it conquered significant lowland parts of the southwest; the Chinese state had strong interests in Yunnan and carried out significant trade with Yunnan and other polities further southwest such as Burma. And, the southwest served as a locus for rebellions, haven for failed usurpers or overthrown emperors – although curiously they may not have been so powerful had it not been for the partial political assimilation introduced Chinese titles and literacy to the region. The institution of tusi added to the complexity of the political conditions by contributing another factor to be reckoned with, the validation of Chinese recognition and confirmation of positions, thus providing a counterpoint to Tibetan and other influences. Yet until the nineteenth and twentieth centuries, the impression is that Chinese influence was a veneer upon the complexities of Yunnan’s culture and society. Sinicized tusi and Chinese garrisons were far more common along major trade routes or in larger, lowland cities. Their influence had not reached into more distant areas, those more isolated and inaccessible due to rough travel conditions or simply being in the uplands. It was not until the later Qing and the Republican governments, with the influence of British and French colonial powers attempting to grasp these territories and with a huge influx of Han migrants into Yunnan, that Chinese government penetrated the furthest reaches of Yunnan, and even then not all (note Kham’s relative independence). This influx was based in part on state policy and Qing penetration into Yunnan, but was also carried forward by the spontaneous migration of thousands of Han traders and miners seeking economic opportunity in Yunnan and trade with Burmese and Tai polities. I will consider the dynamics and consequences of this great migration in a later section. This political context bespeaks the turmoil in the daily life of local people of Yunnan, particularly peasants and upland farmers who were subject to the various polities that arose and receded. In the northern area of Lisu settlement, Chinese settlement and administrative control came very late and Lisu societies in those regions were often involved in either alliance or patronage relationships with Tibetanized or partially-Tibetanized polities such as those of the Naxi. Heads of independent Lisu led within a Tibetan cultural framework of power. Nevertheless, Lisu, like other peripheral peoples of western Yunnan, were subject to the vicissitudes of warfare among Tibetans, Naxi, and Chinese. Particularly in the case of Lisu further south, they were subject to the multiple spheres of influence that burdened others. As we have seen, one survival strategy was to seek the protection of whichever polity appeared more fair or less demanding. Leaders and communities played these powers against each other. But, as we shall see, if Lisu responded then as they have in the past 175 years, one response to the dangers of war, the burdens of dependency, not to mention the endemic natural disasters, was to move away. While there are no early references to migration on as great a scale as in the late nineteenth and early twentieth centuries, Gros states that Lisu settled in the upper Salween as a result of fighting between Tibetans and Chinese (2001b). In short, Lisu people were not isolated and separate, but part of a cultural continuum of Tibetanization and extension of Chinese administration.
Trade and Economy Overland trade in northern mainland Southeast Asia (cf. Sun 2001) was of long-standing significance. It was an important element of the ecological adaptation of Tibetans and others in western Yunnan such as the Naxi, if not as traders then as protection for or scavengers on caravans and trade routes. The trade networks that passed through this region were vast, of a ‘global’ nature long before the era of Western-dominated trade, extending from the Middle East to China, north into Mongol regions and south into Burma, Thailand, and Vietnam. Chinese trade was but a part of this, but became increasingly important in the Ming and expanded throughout much of the Qing. This included trade with Yunnan and with places south and west through Yunnan. Trade helped to create societies that were far from isolated; being, rather, heterogeneous and cosmopolitan. Researchers have discounted overland trade in this region for a century. Research focus has been almost exclusively on the maritime trade. This ignores the significance of overland trade through the southwestern route during the Ming; the wealth it brought into the court, in the form of jewels and metals; the silver it sometimes drained out of the center; and the superior military weaponry that the Ming sold to its southern neighbors. Rather, researchers assume the trade to have been small-scale, unchanged for centuries except for the addition of opium as a trade item, and controlled by Yunnanese Muslims (Hui) and Han (Sun 2000). In fact, trade ebbed and flowed with political conditions, conquest, and market forces. There were three periods of particularly active trade between Burma and China: after the Yuan conquest in the late thirteenth century; in the late fourteenth and early fifteenth centuries during the Ming (Sun 2000); and at the end of the eighteenth century (Frank 1998), in the period of relative peace as the Qing expanded its administration in the early eighteenth century, which then ebbed again dramatically with the Muslim uprisings in 1856-1873. Local indigenous people of Yunnan and Burma were also involved in this trade, albeit on a smaller scale. After the Uprisings, the nature of overland trade changed significantly due to the influence of colonial powers, particularly the French in Indochina, who built a railroad from northern Vietnam to Yunnan in the early twentieth century (Baber 1882), with British control of Burma, and with a weak administrative center in the Republican period. Trade was also reoriented from overland routes and the sorts of commodities available through them to coastal cities through which western goods came. Finally, the inclusion of opium as a commodity in trade networks was highly significant, not perhaps for the trade itself (except insofar as it was legal or illegal) as for the social and economic consequences for the people who produced it. Significantly, the chronicles of local rulers and administrative history leave out trade and economy, although it was likely important in Chinese and Tibetan competition for control of this region and related to the nature of upland/tribal relations with the lowlands. Aside from such regional considerations, China was an important player in the world economic system at the end of the Ming Dynasty in 1644 and remained an important regional player at least up until 1800. Silver flowed into China through Europe from South America and since most people did not care to buy western goods the silver did not flow out again. Chinese society was wealthy and eager to consume luxury items (Mote 1999:953), many of which originated in Yunnan and to its south and west. A regional consideration for China was its tea trade with Tibet, important both economically and militarily.
History and Items of Trade Our picture of Asian intraregional trade is inadequate (Frank 1998) for two main reasons. One is that the overland trade, undertaken by a myriad of small-scale traders from many different states and ethnic groups, was simply poorly documented compared to sea routes where inventories (often in western languages) exist. Another is the Eurocentric view of western hegemony in trade. In fact, Europe was not the core of a world embracing economy from 1500-1800. “If any regions were predominant in the world economy before 1800, they were in Asia. If any economy had a ‘central’ position and role in the world economy and its possible hierarchy of ‘centers,’ it was China” (Frank 1998). Among the important Asian economies, we must include those of what are now Burma, Cambodia, Laos, Thailand, and Vietnam (Frank 1998). Burma and Siam maintained overland caravan trade relations with each other, other kingdoms in Southeast Asia, and China and India (Frank 1998). Western views of trade with China have also been biased by the view that it was all in the form of tribute; but even from that point of view, the importance of overland routes cannot be ignored. Burma’s and Laos’s periodic tribute to China occurred through overland routes in Yunnan (see Fairbank and Teng 1941:176). The India-China trade across Nepal and Tibet had probably been going on for more than a millennium when westerners came on the scene (Frank 1998). There was significant trade between China and Southeast Asia documented in the second century A.D., and even recorded in the second century B.C. This was actual trade that affected local production, not simply a case of Southeast Asia providing entrepôts or way stations for trade (Frank 1998). From the 1400s to the 1800s, the overland items traded were varied. From Burma and the Tai polities to China were traded rice, sugar, cotton, rubies, amber, jade, deer and tiger skins, timber, ships, woods, jaggery, paper, cutch, betelnut, birdnests, shark fins, tobacco, pepper, sappan wood, tin, lead, saltpeter, and silver. From China came ceramics, lacquerware, silks and textiles, clothing, arms and powder, copper cash, copper wares, quicksilver, iron products, lead, zinc, cupro-nickel, salt, fruit, rhubarb, tea, satin, velvet, brocades, thread, paper dyes, carpets, shoes, stockings, house wares, labor, and shipping services (Frank 1998). Burma imported silk, salt, iron and copper utensils, arms, powder, cloth, satin, velvets, brocades, thread, carpets, paper, fruits, tea, and copper cash from China and exported amber, rubies and other precious stones, jade, ivory, fish, birds’ nests, shark fins, jaggery, jasper, catechu, betel nut, tobacco, and raw cotton (Frank 1998:102, citing Sun 1994). In Yunnan itself, silver and gems, copper and copper ware, salt, horses, cotton, and grain were important trade items to China and the polities to the southwest (for instance, see Giersch 2001:82-3), aside from Yunnan’s important role as a crossroads of trade. Sun’s sources record substantial caravans, and while this trade was less substantial than the overseas trade, it was significant. Its value in part lay in it sometimes carrying contraband trade of prohibited metals and arms from China (Frank 1998). In addition, Burmese mines attracted Chinese entrepreneurs, merchants, and labor (Frank 1998). Trade had important effects on the societies and polities of Yunnan and its neighbors. For instance, Burmese trading fairs were tied into the overland trade and people were said to run out of daily necessities if trade were interrupted (Frank 1998). An interregional, even global, trade expressed itself in market forces that formed local incentives and exigencies. The results were land clearing, crop choices, and deforestation. When China suffered deforestation in key areas, timber was imported from Southeast Asia (Chew 2000). Ecological and demographic consequences also resulted from the introduction of American crops such as maize and potatoes (Frank 1998). Ming (1391-1580s) traders bought goods to central China originating in the southwest frontier, Burma, Thailand, and India. Laichen Sun has documented the overland gem trade with Yunnan and Burma. They traded for silver and firearms from the Chinese center (Sun 2001). After Ming conquest of the region, tens of thousands of Chinese civilians followed the soldiers and set up as merchants trading with Tai and Burmese markets or mining silver (Giersch 2001:74). Starting with the early Qing there was increasingly heavy exploitation of Yunnan’s natural resources by Han Chinese. Trade underlay the powerful influence of Tibet in the societies of western Yunnan. Tibetan social and political structures were deeply influenced by trade. Samuel argues that Tibet was built on long-distance trade and Tibetan society was one based on traveling for both religious pilgrimages and trade. The large religious orders were also heavily involved in trade, which financed the spectacular expenditures on monasteries. Travel kept Tibetans in close contact with the various other ethnic groups to its east, south, and west. As a result, their cities, including Kham trading centers, had a heterogeneous character due to their extensive trade routes to and from China (often via Chinese Muslims), India, Nepal, Kashmir, and Iran. Trading was an intrinsic part of the Tibetan ecological adaptation and trade was what made Tibetan states viable. The rise and fall of Tibetan states was probably linked to the degree of prosperity of trade routes (Samuel 1993:42-3, 145-6). Thus Tibetan trade was a vehicle for religious missionization, increased contact and influence with surrounding peoples, and the incorporation of surrounding groups into the Tibetan cultural and political sphere. As Tibetan trade went, so went the fortunes of many upland peoples in the Sino-Tibetan borderlands.