Status: The native officials of Mahu and their titles

Since the earliest official records, Mahu was a liminal area in which the Shu-Han 屬漢 court gained a foothold by establishing it as a county (Mahu xian 馬湖縣). The toponym emerged in histories after Zhuge Liang’s 諸葛亮 southern expedition in the year 225, where he allegedly confronted a strong local chieftain called Moho ꂽꉼ (also known as Legge Ashy ꇍꈩꀋꏂ; Ch. Meng Huo 孟獲) and his tribe-clan designated by the name of their leader. “Mahu” in fact seems to be a Chinese transcription of “Moho,” as the local toponyms often derived from native clan designations. Tang and Song-Dynasty sources called the administrative unit the “Mahu tribal area” (Mahu bu 馬湖部). The Mongol rulers of the Yuan Dynasty restructured it as the Mahu Circuit (Mahu lu 馬湖路) ruled by a native chieftain, or nzymo, of the Hxala ꉐꇁ clan, under which six subordinate native offices were created in Nixi 泥溪, Pingyi 平夷, Manyi 蠻夷, Yidu 夷都, Muchuan 沐川, and Leibo, all lying westwards from Hxala nzymo’s seat. The westernmost Leibo office was held by the Azho ꀋꍐ clan. This whole setup created a buffer zone between the perpetually untamed anarchic area of Liangshan to the west, and the eastern empire-dominated region now known as Yibin, an important settlement on the axis of the local trading routes firmly under imperial control. Mahu Prefecture was thus a double periphery from the viewpoints of the court-dominated state and the Nuosu-dominated core Liangshan area, where everyone mingled, exchanged, traded, and fought with each other. At one point, the Yuan rulers moved the seat of the Mahu circuit to Yibu xikou 夷部溪口 and connected it to the southeastern bank of the Jinsha River (金沙江, Nuo. Ahuo Shyyy ꀉꉷꏁꒉ) that immediately bordered the Wumeng Circuit (烏蒙路). However, in 1305, the seat was moved back to the northwestern riverbank and shifted eastwards to the Nixi area,Footnote ²⁰ which was easier to defend. In official Chinese records, the Hxala appear under the locally ubiquitous surname An 安 (“Tranquil,” representing the court’s hope for an appeased borderland), while the Azho are surnamed Yang 楊 (another widespread surname pointing to several kinds of trees, the local commodity). After the Yuan, the Ming Dynasty turned Mahu into a “native prefecture” (Mahu tufu 馬湖土府; see Figures 2 and 3).Footnote ²¹

Figure 2. View of the Mahu area during the mid-Ming (Yu, Mahu fu zhi, “Xu” 7a–7b).

Figure 3. Detail of the Mahu Area (Courtesy of the author, Google and TerraMetrics, 2022).

Sitting with Hxala Shyge in a smoky and stuffy teahouse in Liangshan’s metropolis of Xichang 西昌, where we sipped green tea from Leibo, we tried to make sense of the history of Hxala/An in what used to be Mahu, and as seen through official and personal accounts. Hxala Shyge himself has written in an appendix to his unpublished document that the genealogical records of Hxala were lost three times. They were allegedly lost first in 1645, when the anti-Ming/anti-Qing forces of Zhang Xianzhong 張憲忠 pillaged Sichuan and burned the temples in Mahu, including the ancestral temple of the Hxala on the Jinsha’s southeastern riverbank. Their second loss is thought to have occurred in 1730, when the office of the Hxala clan was burned down by the rebelling Yi mob. Their genealogy then finally disappeared during the Cultural Revolution’s campaign to “Destroy the Four Olds” 破四舊 in the 1960s.Footnote ²² However, an online website featuring an article written by two alleged descendants of the Hxala nzymo from Yunnan Province mentions a quasi-historical document called The Record of Merits and Deeds of Mahu Prefecture’s Hereditary Officials of the An Clan (世襲馬湖府職官安氏宗籍功德實錄記)Footnote ²³ written by Hxala clan member An Zhou 安洲 in 1868, which neither Hxala Shyge nor I have been able to get hold of. The historical credibility of both the website and this document remains unclear, not least because when the document was written in the nineteenth century, it was narrating events that allegedly happened two millennia earlier without providing any further sources for the author’s claims. Similarly, a record from the Ming Dynasty’s Jiajing-era gazetteer for Mahu Prefecture, written in 1555, claims that the Hxala had already resided in the region since the Han Dynasty for fifty-eight generations.Footnote ²⁴ This assertion also does not seem to be based on any plausible historical records. Both accounts probably can be attributed to the snowballing of reproduced assertions from individuals of different time periods who wanted to impart their own clan with historical authority. There are far more authoritative records about the Hxala clan residing in the Mahu area since the Tang and Song dynasties, which, however, do not mention their Han-Dynasty origin.Footnote ²⁵ Nevertheless, An Zhou’s document, which Hxala Shyge never saw and quoted only from a second-hand account similar to that appearing online, sides with the gazetteer and claims that the history of Hxala begins in 135 BCE when the Han Dynasty established the local ancient trading route called Bodao 僰道. An Zhou’s document also mentions Hxala clan members, who in the years 620 and 757, respectively, received the court’s insignia confirming their authority over their lands and the local population.Footnote ²⁶ The first date and name that I was able to confirm in the dynastic chronicle of the Song Dynasty comes from 974, when Dong Chongxi 董舂惜 (Nuo. Hxala Yydi ꉐꇁꒉꄂ), a “prince” (Ch. wangzi 王子) of thirty-seven tribes of the Mahu bu, brought horses as tribute to the imperial court.Footnote ²⁷ More detailed records started to appear only in the early Hongwu era of the Ming Dynasty. Everything began with the ceremonial journey made by the Hxala officials of Mahu to the seat of the imperial court, where they presented tribute and received, in line with Confucian etiquette, valuable objects in the form of caps, belts, and other garments placed on top of their most important gift––their official insignia presented on documents and as seals. This was supposed to prompt them to continue learning court rituals and spreading them among their subjects.

Hxala Shyge’s narrative coalesces with various official sourcesFootnote ²⁸ claiming that, in 1371, the acting circuit native administrator An Ji 安濟 (Nuo. Hxala Boshy ꉐꇁꀨꏂ) sent his son An Ren 安仁 (Nuo. Hxala Bodi ꉐꇁꀨꄂ)Footnote ²⁹ with a larger delegation of his subordinates––including, for example, Wang Qi 王麒 of Nixi, the chieftaincy closest to the Mahu administrative center—to officially submit to the court. As a result, the circuit was transformed into a native prefecture (Ch. tufu 土府) and the headman of An Ren’s lineage was confirmed as both the tuguan bureaucratic native official and the holder of its native prefect title (Ch. tuzhi 土知, military rank 4a).Footnote ³⁰ Officials from Nixi and five other sub-regions also gained the title of chiefs (Ch. zhangguansi 長官司, 6a). Two years later, An Ji informed the emperor that he was in poor health and requested his replacement by An Ren, which the court approved. In 1374, however, An Demin 安德珉, a nephew of An Ji from the same patriline, sent his younger brother to the court, possibly in hopes of switching the court’s alliance to his intra-clan lineage. The imperial officer told him that his uncle had already sent his son three years earlier, who had been confirmed as successor. His actions were deemed inappropriate and the horses he had brought as tribute were refused. An Demin did not give up. The competition for the native prefect post continued for more than a decade. In 1379, he brought the precious timber of the local xiang nanmu tree 香柟木 (Lat. phoebe zhennan), and in 1383 he brought eighteen horses. For both, he received a reward of reciprocal value in clothing, money, and rations (rice). In 1403, however, An Ren re-emerged as the sole holder of the native prefect of Mahu, possibly coming out as victorious from this intra-clan struggle for power.

Horses and the rare wood of phoebe zhennan were the two most precious commodities of Mahu. The local timber was used to craft wooden pillars in the Forbidden City built between 1406 and 1420. Under An Ren, the Yidu Mountain (夷都山, today’s Wuzhi Mountain 五指山) became a center of timber extraction. The Yongle Emperor issued an edict ordering it to be renamed as Spirit Wood Mountain (神木山) and to have the Spirit Wood Mountain Shrine (神木山祠) built on it for yearly magnanimous offerings,Footnote ³¹ which according to locals in contemporary Zhongdu Township (中都鎮), continued along with the timber extraction and trade until the Cultural Revolution in the twentieth century. At the very beginning of this centuries-long practice, the Hxala got very close to the court. They also extended this proximity to their subordinate chiefs through marriage. An Jun 安濬, who succeeded An Ren in 1407, together with Wang Feng 王鳳 of Nixi, built a Buddhist Temple of Heavenly Tranquility (天寗寺, also known as the Cooling Temple 清凉寺) on the northwestern bank of the Jinsha River. Other accomplishments are attributed to An Hong 安洪 who inherited the office from An Hao 安灝 in 1452 and significantly assisted the provincial governor of Sichuan in taming the borderlands, for which he received an official cap and a belt of the Grand Master of Palace Accord (中順大夫, 4a). Together with Nixi chieftain Wang Mingde 王明德, An Hong subsequently cast a bronze bell (see Figure 4) which was later placed in the Temple of Eternal Longevity (萬壽寺) located near the Nixi chieftaincy as another Buddhist place of worship. Yet, like his father, he died prematurely, and in 1470 the office ended up in the hands of his oldest son, An Ao 安鰲 (Nuo. Hxala Mynge ꉐꇁꃋꉬ).

Figure 4. One of the bronze bells of the Mahu area, now stored in Lengyan Temple (楞嚴寺, also called Leiyin Temple 雷音寺) in Zhongdu (Courtesy of the author, 2023).

This development might have been an ideal example of the transformation process through which the Hxala and other native officials were ideologically incorporated into the Chinese realm by way of Confucian rituals that interconnected with the religion(s) that its rulers promoted. However, An Ao was described in the official sources as a cruel leader who inflicted great suffering on his people. He allegedly overtaxed them, behaved obscenely towards the local women, and even sought to use a hundred-legged nightmare charm summoned by a “bad monk” (Ch. yaoseng 妖僧) to kill people. This anecdote is almost akin to some of the strongest cursing rituals conducted by the Nuosu-Yi bimo ritualists, who were notoriously present at the office of the nzymo and were responsible for rituals, ceremonies, and record-keeping. Whether some of them appropriated the dark sides of Buddhism and/or Daoism through the network of temples and integrated them into their own rites remains a practicably unanswerable question, due to lack of evidence, but one that cannot be ruled out. Harmonious relationships between An Ao and his subordinate chiefs also deteriorated as he sought to send an assassin to murder Wang Daqing 王大慶 of Pingyi. After Wang escaped, the assassin killed his younger brother. As a result, the vice-governor of Sichuan Province ordered An Ao to be executed by a thousand cuts (Ch. lingchi 凌遲).

A short report from a certain Liu Zhong 刘忠, preserved in the Ming-Dynasty’s Jiajing-era gazetteer for Mahu Prefecture, portrays An Ao in a much more favorable light. According to the author, An Ao built houses, granaries, temples, and libraries, and again provided the “spirit wood” to rebuild Nanjing’s Ming Xiaoling 明孝陵 Mausoleum of the Hongwu Emperor after it burned down.Footnote ³² One cave near Nixi hosted a statue of An Ao on a horse from 1488.Footnote ³³ This could suggest that An Ao was respected by his own people and that the imperial court, which was the ultimate winner in this situation, had used him as an excuse to frame the Hxala and take their land. In 1495, the Mahu Native Prefecture passed through the gailiu process into the hands of the empire’s career bureaucrats, erasing the “native” word from the name of the administrative unit. Thereafter, unlike those of their subordinate chieftaincies, the Hxala clan members were stripped of the tuguan position and banished from the Mahu prefectural seat to the region’s western fringe, closer to the untamed core of Liangshan.Footnote ³⁴

The different details and circumstances of An Ao’s overthrow in many official records point to the utmost importance of this event. On Ming-Dynasty maps, the lake in the peripheral region to which the Hxala were banished––the biggest body of water in the area––is depicted as equal in size to those from different parts of the empire that are, in reality, much larger. According to Tan Ying,Footnote ³⁵ the size of the lake on the map shows the importance of Mahu Native Prefecture to the imperial court as it lay in the vicinity of both the north–south overland and west–east river trading routes. Before Zhuge Liang’s campaign, the lake was called “Submerging Pool” (Linchi Qian 临池潜) because its waters are carried by a rapid underground stream to the Jinsha River. Later, it carried the name of “Heavenly Lake” (Tianchi 天池).Footnote ³⁶ The stories about the mythological creature with a dragon’s head and a horse’s body (Ch. longma 龍馬) living in its waters were possibly invented later by Han settlers coming from the inner fringes of the Chinese empire, who took the phonetic transcription of Mahu (“Horse Lake”) administrative unit literally and attributed it to the lake, gradually forgetting the connection of the whole region to the Moho clan. Following An Ao’s execution, the emergence of the Hxala as the dominant power in this peripheral region had, from the Nuosu-Yi point of view, overwritten the story of the Moho clan with that of the Hxala. This was, among other things, indicated by the fact that the lake is still called “Hxala Lake” (ꉐꇁꎿꀸ) in Nuosu-Yi while many other local topographical features carry the Hxala clan’s name. The imperial court played a significant role in this by granting the Hxala this territory.

Authority: Abandoning officialdom, embracing nativeness

Since the sixteenth century, Huanglang 黃螂, a settlement on the northern fringe of Hxala Lake, has appeared as a toponym in the records together with the Ningrong Police Office (寧戎司巡檢) title, which the court established there.Footnote ³⁷ Huanglang is often considered to be a phonetic transcription of HxalaFootnote ³⁸ and thus to mark the location to which the clan was banished, but many sources suggest that during the Han Dynasty, the broader Leibo area belonged to the somewhat phonetically similar Tanglang County (螳螂縣),Footnote ³⁹ and that the two toponyms are related. However, the actual position of Tanglang was on the southeastern riverbank near the contemporary seat of Qiaojia County (巧家縣), almost two hundred kilometers southwest of Huanglang as the crow flies.Footnote ⁴⁰ In fact, Tanglang County most probably derived its toponym from Tanglang Mountain (堂狼山). Its ridges are perceived by the Liangshan Nuosu-Yi as Zzyzzypuvu ꋪꋪꁍꃴ, the legendary place of origin from which the six sons (Ch. liuzu 六祖; Nuo. Jjumu sse fu ꐧꃅꌺꃘ) of the primordial ancestor Apu Jjumu ꀉꁌꐧꃅ––representing the six different historical branches (or original clans) of the Yi––all dispersed in different directions before establishing their domains in Yunnan, Guizhou, and Sichuan.Footnote ⁴¹ Equating Tanglang with Huanglang is thus problematic, and probably a result of a snowballing mistake made by authors in later centuries.

In 1523, the court bestowed the lowest military title of Native Police Officer (Ch. tuxunjian 土廵檢, 9b) on An Ao’s son, An Yu 安宇. For many years, he served in the Sichuan army as a tubing native soldier, as a reward for his achievements during the pacification of the borderlands. However, such an appointment in Huanglang was not enough for An Yu, who sought to reclaim the lands and position that once belonged to his clan. The authorities discovered that he had even met with the native prefect Cen Meng 岑猛 in a far-off area of today’s Guangxi, where An Yu once served, presumably to gather inspiration on how to reclaim the land under the rule of the Hxala clan. Three years after An Yu’s appointment, Cen Meng led an unsuccessful rebellion against the empire, and the imperial officers kept an eye on An Yu to prevent him from doing something similar.Footnote ⁴²

One of An Yu’s descendants, An Xing 安興 (Nuo. Hxala Shypo ꉐꇁꏂꁈ) was much more successful. Between 1551 and 1566, Pushur Yobo ꁊꎿꑿꀨ (Ch. Pushu Yuebo 普書約伯) of the Hnielie clan (ꅮꆽ; Ch. Ninai 膩乃), a subordinate of the Qiongbu chieftaincy (邛部長官司) in Liangshan’s northwestern Yuexi 越嶲 area who traced the ancestry of his people to the primordial Moho chieftain,Footnote ⁴³ rebelled against the Ming Dynasty. He was followed by his son Yobo Sanjie ꑿꀨꌐꐷ (Ch. Sajia 撒假), who in 1587 promulgated himself as King of the Western Lands and Peaceful Heavens (西國平天王). An Xing, who inherited the Native Police Office in Huanglang, along with the chief of the Azho clan, Yang Jiuzha 楊九乍 (Nuo. Azho Jjizha ꀋꍐꐛꍆ) of Leibo, supported the self-proclaimed local ruler. Their rebellion became known as the “Upheaval of Three Heroes” (三雄之亂).

All three families of the main perpetrators were related through intermarriage. “Mahu is my old home, I must retrieve it,” An Xing allegedly stated when receiving the insignia, thus continuing the office and following the wishes of An Yu. The “heroic” deeds of the trio included killing almost three thousand local imperial military personnel, kidnapping men and women, and pillaging settlements. The description of An Xing in the official sources oscillates in style between his depiction as a crooked villain similar to An Ao, and his portrayal as tall, dashing, and courageous. Despite his being an enemy of the empire, the local gazetteer acknowledged his local heroism. Through Sichuan Province and with the help of Wumeng’s native troops, the imperial court put together an army consisting of forty thousand soldiers. The army recruited people who could interpret between the Chinese and Yi languages and waged psychological warfare to quickly obtain the heads of the insurgents. However, the native officials and their native troops from the area often could not be reliably summoned by imperial command. For example, Wen Anmin 文安民 of the Manyi chieftaincy, who was formerly subordinate to the Hxala clan, was supposed to mediate between the imperial army and An Xing; instead, he revealed the army’s plot to An Xing, enabling him to escape.

The terrain around Huanglang was notoriously harsh and remains so today. The rugged karst hills full of caves, which the Yi knew thoroughly well and in which they often found refuge, forced the imperial army to use scaling ladders. The darkness of the caves, their eerie silence, and the cold wind emanating from their seemingly bottomless depths terrified those unaccustomed to them. The unyielding fog further hindered their visibility and contributed to the disorientation of the imperial troops in the terrain. The battles were extremely bloody, with the constant moisture preventing wounds from healing, which caused rapid infections and inflammations. Nonetheless, the imperial army captured Sajia and executed him by mincing (Ch. hai 醢). Meanwhile, An Xing and his younger brother retreated to the Qylemu area ꐓꇌꃅ (Ch. Qulimi 祛裡密), a sub-region of Tieggurmu ꄟꈯꃅ (Ch. Tiangumi 天姑密) in Leibo. While on the run, he tossed silver ingots on the road to slow down his pursuers. The army, however, captured his mother, Lady Yang (楊氏) of the Azho clan, and also his wife, Lady Sha (沙氏) of the Shama clan.

One morning, a few days later, several soldiers reported a collective dream in which the spirit of the Eastern Han general Guan Yu 關羽 advised them how to capture An Xing alive. They succeeded the same day when An Xing became stuck while clearing his way through a thick bamboo forest and, surrounded by imperial soldiers, surrendered. The biggest headache for the army was Yang Jiuzha. They sought to trap him through his captured wife, Lady Ma (馬氏). If Jiuzha would surrender, they promised, the rest of his family would be spared––a typical strategy that had also been applied to the Hxala after the episode with An Ao, so as not to alienate the rest of the local population. Jiuzha’s and Lady Ma’s young daughter, for whom the wife of the villain on the run had a special affection, was initially held hostage in Mahu’s prefectural seat. The prefect returned her to Lady Ma. Allegedly touched, she secretly consulted an elderly bimo, who imitated the sounds of a black bear, used his fortune-telling books, and carved a ritual tablet which he splattered with (probably chicken) blood to track Jiuzha down. When on the verge of being captured, Jiuzha fiercely resisted. In the process, he got shot in his left rib. His injury was serious, and he succumbed to it on the way to the military camp.

Three heroes hence perished one after another. Hundreds of their followers were captured alive, and over one thousand of their accomplices were beheaded. Then in 1589, the Huanglang township was turned into a garrison that was reinforced by a new wall. A shrine for Guan Yu was erected, with sacrifices to celebrate the assistance of his spirit while chasing after An Xing. The Yidu chieftaincy that was closest to the principal extraction site of the “spirit wood” was renamed Zhongdu. To accommodate the souls of the fallen imperial soldiers, a shrine was also built in Manyi chieftaincy together with a high protective wall. The Hxala, along with their formerly subordinate chieftains of the Azho clan from Leibo, once again lost their insignia in their respective domains and their official positions as native headmen recognized by the empire. They retreated westwards, deeper into the mountains of untamed Liangshan, and out of reach of the imperial bureaucracy.Footnote ⁴⁴

Land: Back to officialdom, two Jinsha riverbanks and the Mitie massacre

As a result of their falling out with the empire, the genealogies of both the Hxala and Azho nzymo disappeared from the official records for more than a hundred years, as they embraced a native way of life––albeit one influenced by their collective memory of officialdom. Their retreat caused a significant gap in understanding the history of the Mahu area. After the fall of the Ming Dynasty (1644), the new Qing regime eventually tightened the rules significantly for the native officials’ succession and education. Those native officials who did not observe them were threatened with an immediate gailiu process. Already during the Ming, the Hongwu Emperor wanted the succession to be patrilineal, so that only sons through the male line were eligible to hold the post of native chieftain. Furthermore, he heavily sponsored the buildup of Confucian charitable schools to spread the court rites to the borderlands, using a tactic similar to that unleashed in the Mahu area. However, Hongwu’s orders were executed rather leniently. During the Qing, the Shunzhi (1643–1661) and Kangxi (1661–1722) emperors were notably more stringent in the implementation of succession rules for native officials, whose domains they wanted to integrate more quickly into the empire’s inner borders.Footnote ⁴⁵ The situation of the Yi, who practiced different customs of succession and alliance forming, became more precarious. Besides the fact that women could potentially be appointed as clan leaders or native chieftains, they often practiced both uxorilocal (Ch. ruzhui 入贅) and levirate (Ch. di wang xiong ji 弟亡兄繼) marriages that tied them to their own succession processes, which conflicted with the rules stipulated by the imperial court. In 1664, for example, the Yi native official An Kun 安坤,Footnote ⁴⁶ referred to in the fieldwork notes of state surveyors from the late 1950s as An Gang 安崗––a possible phonetic transcription of his Yi name into Chinese characters––rose up from the influential Anre clan (ꀋꎞ) in northwestern Guizhou Province’s Shuixi 水西 (located westwards from Wusa Native Prefecture (烏撒土府)) to initiate a large-scale rebellion. An Kun’s rebellion was later subdued by the Qing general Wu Sangui 吳三桂, who went on to occupy the broader region during his own anti-Qing revolt. Allegedly, after An Kun’s defeat, death, and the gailiu of his centuries-old native chieftaincy, one of his offspring named Anre Ayy ꀋꎞꀉꒉ (Ch. An Heji 安合吉) became heir to a native chieftaincy of the Shama nzymo clan residing in the depths of Liangshan. The transfer happened through uxorilocal marriage, which solved the Shama clan’s problem of the absence of a male heir, as Anre Ayy continued the Shama nzymo chieftaincy.Footnote ⁴⁷ Compared to northwestern Guizhou, and even central Mahu during the Ming where the Yi communities had long imitated imperial bureaucratic structures, Liangshan represented a leaderless environment without the firm grasp of imperial power. Here, the Yi nzymo often clashed with free aristocratic nuoho and other regional quho strata, and succession patterns were less influenced by imperial rules and rites.

Huang explains that the possible reason behind the Qing’s mounting presence in its southwest borderland was the court’s increasing dependence on copper imported from Japan for domestic technological and military production.Footnote ⁴⁸ When the Tokugawa regime began to put restrictions on copper exports in the early eighteenth century, the court was pushed to extract precious metal deposits in Southwest China. Parallel to this process was the large-scale dismantling of the large native domains containing these copper deposits, which enabled the court to access them directly and at speed. After dismantling the larger domains, the tusi––or military native officials––were re-installed in those native administrative units where the populations still resisted this development. However, these units were significantly smaller and more scattered than those of their predecessors. Through this gradual process of absorbing the native domains into the empire, their resistance was supposed to have been blunted and eventually brought to an end.

The copper mining areas were located in Dongchuan 东川, Wumeng, and Zhenxiong 镇雄 (formerly Mangbu 芒部) prefectures, all on the Jinsha’s southeastern riverbank and all under the rule of native officials. The reappointed chieftains had to make another journey to the capital to submit to the new emperor, which gave the Hxala and Azho clans the chance to reclaim some of the insignia formerly held by their ancestors. While some of the lower chieftains of Mahu went to the court during Shunzhi’s rule, the descendant of An Xing, An Yongchang 安永長, had to prove his loyalty by “civilizing” (Ch. huayu 化諭) local Miao natives living around his domain. For this merit, in 1695, he obtained the Qing-ubiquitous military title of Native Battalion Commander (Ch. tuqianhu 土千戶, 5a) and became a tusi, or military native official. Remarkably, the land where he operated, and over which he gained official permission to rule, lay in Huixi 檜溪 (Nuo. Agga lietuo ꀋꈜꆻꄧ) on the southeastern riverbank, just opposite Huanglang. This was because after An Xing’s capture, the surviving members of the principal ruling Hxala clan lineage fled there. In official records, the territorial administrative unit in Huixi was even named after An Xing. According to Hxala Shyge, who drew on his clan’s collective memory of their return to nativeness, his ancestors spent the century as Nuosu-Yi chiefs dwelling in the Liangshan highlands on the southeastern riverbank without the court’s insignia but ruling over their population and developing their agriculture and pastoralism.Footnote ⁴⁹ Eight years after An Yongchang, the Azho clan through Jiuzha’s descendant Yang Lagui 楊喇哇 (Nuo. Azho Lagi ꀋꍐꆿꇜ), also submitted to the court. He was appointed as the newly established Leibo Qianwanguan Principal Chief (雷波千萬貫正長官司, 6a) in the southern part of the region.Footnote ⁵⁰

But the dismantling of native domains accelerated again with the ascension to the throne of the Manchu Yongzheng Emperor (1722–1735), whose loyal Manchu prince Ortai 鄂爾泰 turned his focus in 1727 towards the Wumeng and Zhenxiong native prefectures. However, Ortai underestimated the complex Yi blood relations that often stretched over numerous large Yi domains, as was the case for northwest Guizhou’s Anre Ayy, when he later became nzymo of the Shama clan. The spiral of horrific violence into which most of the Liangshan Yi clans were drawn, including the Hxala and the Azho, began with a seemingly mundane event preceding Ortai’s arrival. Lu Yongzhong 祿永忠, a minor native official of the Leibo chieftaincy on the northwestern riverbank,Footnote ⁵¹ which was probably recreated after the ousting of the Azho clan following Yang Jiuzha’s capture, passed away. Following the levirate custom, his widow, Lady Lu 祿氏, was married off to his older brother Lu Yongxiao 祿永孝, who was at the time a native supervisor (Ch. tumu 土目) in Mitie 米貼, a region subordinate to Wumeng on the southeastern riverbank. Lu Yongxiao was involved then in an inter-clan feud with the lineage of Lu Dingkun 祿鼎坤, paternal uncle of the successor to the position of native prefect in Wumeng, and its current chief, Lu Wanzhong 祿萬鈡. However, Lu Wanzhong was still an infant and ineligible to inherit the office. Sensing an opportunity, Ortai and his subordinates investigated the poorly maintained genealogical documents of the native officials in the prefectural archives and found many gaps and breaches in their rules of maintenance and in the succession process. This gave Ortai a legal pretext for the deposition of the whole Lu native chieftain clan. Lu Dingkun also seized the moment to usurp power for himself and get rid of Lu Yongxiao by manipulating the documents for Lu Wanzhong’s succession. He inscribed Lu Yongxiao into them as a successor, facilitating his arrest and transfer along with hundreds of other native officials to prisons in the provincial capitals of Sichuan and Yunnan for the alleged fraud of manipulating the succession rules. Further investigation revealed that Lu Yongxiao had nothing to do with Lu Wanzhong’s case. The officials, however, refused to release him. Ni Tui 倪蛻, an aide to officials in Yunnan who left a firsthand account of the events through his personal writings, claims that Ortai’s administration––in line with their objective––were able to conveniently charge Lu Yongxiao with illegal conduct (Ch. bufa 不法) for the possession of two chieftaincies and the abduction of his younger brother’s wife. Lu was then executed by hanging.

These events naturally resonated in Liangshan. More so when Lu Dingkun escaped to the highlands surrounding Mianshan 冕山, relatively near to the Jianchang 建昌Footnote ⁵² Garrison, which secured another major trade route on the western fringe of the Yi-dominated Liangshan. On the cusp of spring 1728, Ortai’s subordinates were sent to investigate Lu Yongxiao’s people in Mitie. Lady Lu was meant to face a similar fate to her husband. After general Guo Shouyu 郭壽域 reached Mitie with five hundred of his men, he ordered the establishment of a provisional military camp near some uncannily quiet and empty settlements. Only at the door of Lu Yongxiao’s house did they see a woman shouting: “Release my man, so he can return home!” Identifying her as Lady Lu, Guo’s men went after her, but she managed to slip away. Later that night, around the third watch (between eleven o’clock at night and one o’clock in the morning), Guo’s men were obliterated in an ambush set by the local Yi. Guo himself was stoned to death. Only one person managed to escape and provide testimony to his superiors. Lady Lu had forded the Jinsha to its northwestern bank and journeyed further into the depths of Liangshan. She sought help within the extensive network of her relatives, mostly lower-ranking official supervisors and members of the tusi genealogies (Ch. tumu and tushe 土舍) of the court-appointed native officials. These included not only people from Huanglang and Leibo––and especially the Qianwanguan Azho clan’s chief Yang Mingyi 楊明義––but also those residing on the land governed by the Shama and the Dechang near Jianchang garrison. On the southeastern riverbank, natives from Tundou 吞都, close to An Yongchang’s domain, were especially fierce when clashing with the imperial army. Unlike his ancestor An Xing, in this conflict, An Yongchang of the Hxala clan stood firmly on the side of the empire and assisted its armies, thus going against Yang Mingyi of the Azho.

Views on the following events differ. Ortai and his administration saw the native officials in a similar light to the official depiction of An Ao, as evil barbaric villains who through blood alliances conspired against the emperor, hid traitors, and blocked rations for the imperial soldiers. Their subordinates, Ortai maintained, were stupid, and were capable only of following their leaders. Infuriated by the events surrounding Guo’s failed campaign, Ortai ordered the generals Zhang Yaozu 張耀祖 and Ha Yuansheng 哈元生 to drown the locals in their own blood. Ni Tui’s unofficial personal account lays out unbelievable cruelties and sides with those affected. Soldiers allegedly cut off the limbs and cut open the stomachs of all those who inhabited the area, regardless of whether they were Yi natives, Han settlers, or people of other ethnicities that were coined much later. Their campaign left over thirty thousand dead, with others surviving only by crossing the river in places stretching dozens of kilometers and vanishing in the landscape of Liangshan. In May of 1728, Lady Lu was captured and executed. Ortai maintained that the locals got what they deserved. He blamed the Tundou natives, who allegedly provided the main support to Lady Lu, without which she would never have dared to hurt imperial soldiers. The end of the principal phase of the uprising was followed by years of aftershocks.

An Yongchang showed great skill in subduing the Tundou. Symbolic respect for his deeds from the imperial court was, however, outweighed by the harsh reality that followed two years later. During another upheaval, An Yongchang was murdered together with most of his nuclear family by a member of genealogy of the local Yi chief from Tundou, Muggur Syge ꃅꈯꌦꇰ (Ch. Mugu Sige 木谷四哥). During this obvious act of revenge, his office was also burned to the ground, leaving only one heir alive. Other people who had previously fled to Liangshan and had been involved in this act were massacred by the imperial army, which arrived at the scene right at the moment when its enemies attempted to wade the river back onto the Liangshan side. Countless dead bodies floated on the river’s surface, turning its brown waters red.Footnote ⁵³

Some years after the Mitie massacre, all lands of the former native prefectures on the southeastern riverbank were transferred from Sichuan’s administration to that of Yunnan Province. Wumeng was turned into Zhaotong Prefecture (昭通府), Mitie was renamed Yongshan 永善 (“Eternal kindness”), and Mahu Prefecture was made a part of Xuzhou Prefecture (徐州府) with its center in Yibin. Nominally under the governorship of career bureaucrats since 1495, the western fringe of what had formerly been Mahu remained volatile for centuries. In the relatively isolated garrisons of Huanglang and Leibo, the provincial government established the Anfu 安阜營 and Pu’an 普安營 permanent military campsFootnote ⁵⁴ that kept attempting to keep the region in check. Mimicking the acts of his predecessors, after the “Upheaval of Three Heroes” in Huanglang, in 1731 general Huang Tinggui 黃廷桂 erected the Guan Di Shrine (關帝廟) in Leibo.Footnote ⁵⁵ The stepmother of Yang Mingyi, Lady Yang Sha (楊沙氏), took over the position of Qianwanguan chief. Later, she even obtained the title of Native Supreme Battalion Commander (Ch. tuqianzong 土千總).Footnote ⁵⁶ A woman holding the title of chieftain represented a clear breach of the Qing succession rules. However, due to the instability of the region, the Qing court had to, again, become flexible when some of its rules proved incompatible with the social order of the Yi. Following the Mitie massacre, the Hxala clan members of Tieggurmu on the northwestern (Liangshan) side of the Jinsha River reappeared on the stage through the submission of An Bao 安保. Initially, his submission did not result in the issuance of insignias. In Huanglang, the small Ggaha ꈜꉱ clan related to the Hxala with whom it shared a common ancestor, obtained similar recognition through the submission of Guo Bao 國保.Footnote ⁵⁷ Yet, the Qing court made relatively small concessions that broke the authority of the native chieftains over their vast lands and shrank their domains where necessary. All the court did was make exemptions to some of its succession rules and reappoint some chieftains in the borderlands of their formerly much vaster administrative units. Nevertheless, the non-rebellious members of the Hxala, Azho, Ggaha, and some other clans managed to reappropriate their tusi status and authority––especially over their own people—on account of their long-term loyalty to the imperial court and their historical influence in the area. Following this, and throughout the second half of the Qing rule, the official records offer very little detail or even mention of the Mahu native officials.

Material culture: Tombs facing the sky

After the Mitie massacre, the tombstones for the native chieftains of Mahu became nearly the only source of information about them. Nonetheless, the design and layout of the chieftains’ tombs reveal their appropriations of both officialdom and nativeness––and they do this even more eloquently than the inscriptions on the tombstones themselves. To make this clear, I devote several paragraphs here to the broader context of the burial rites, as seen through the eyes of the imperial court and local Yi communities across Southwest China.

The official imperial rites throughout the Ming and Qing stipulated that the dead body of an official should be buried in a coffin placed inside of a proper tomb adorned with a tombstone. However, many of the Yi communities practiced cremation and did not erect tombstones. The precise origins of cremation among the Yi, who were not the only ones in the area to practice it, as Buddhists also cremate, is unknown. Cremation has remained essential up to the present, especially for the Nuosu-Yi, because without it the souls of the dead cannot returnFootnote ⁵⁸ to their place of origin, called Zzyzzypuvu, through a complex post-mortuary ceremony (Nuo. nimu cobi ꆀꃅꊾꀘ), and from there ascend to the ancestral heavens of Ngemu Puggu ꉬꃅꁋꈬ (also called Shymu Ngehxa ꏃꃅꉬꉎ). Burial without cremation was habitually reserved for those sick with leprosy,Footnote ⁵⁹ who could not reach the heavens and whose souls would turn into various malicious ghosts.

However, the Yelang Kingdom (夜郎; in sources also known as Zangke 牂柯), which was the earliest locally recorded greater polity, might have imitated some of the rites of the neighboring Chinese empire at various stages of its own formation and expansion. Active from the third to the first century BCE, and gradually expanding across the Yunnan-Guizhou plateau, the Yelang Kingdom is often considered a proto-state polity of at least the semi-sedentary ancestors of the Yi, who seemed not to have exclusively cremated the bodies of deceased elites but buried them in the ground.Footnote ⁶⁰ After the conquest of the Han Dynasty, the Yelang declined and many of its former inhabitants possibly took refuge in the vast Wumeng mountain range, which spans from northwest Guizhou to northern Yunnan and runs along the Jinsha River that borders Liangshan. There, they may have started to practice cremation to leave minimal traces of themselves, as they were practically on the run from the empire. In his Book of Southern Barbarians (蠻書), the Tang-Dynasty author Fan Chuo 樊綽 talks about the cremation practice associated with the “Black Barbarians” (烏蠻) who had spread across the highlands of what is today western Guizhou and Liangshan to resist the state control. The more empire-oriented burial customs were associated instead with the “White Barbarians” (白蠻) who lived at least partly within the reach of the court’s power.Footnote ⁶¹ Perhaps attracted by the numerous advantages of interacting with the empire, the so-called White Barbarians either stayed in the lowlands after the fall of the Yelang or were lured back there from the highlands, to which they might also have escaped. The Ming, the Qing, and even the Republic of China repeatedly sought to ban cremation among the Yi.Footnote ⁶² While the pressure to abandon cremation eventually pushed numerous Yi communities in Southwest China to adopt a coffin-and-tombstone burialFootnote ⁶³ at some point or another, some sought ways of accommodating both the ritual requirements of the imperial court and their own local cosmological visions of the world and afterworld. The highland Liangshan Nuosu-Yi chose to continue practicing their migratory lifestyle and cremation. But many of the Yi staying or returning to the lowlands neither returned to the Yelang’s or the empire’s soil-burial practice, nor did they exclusively continue cremating without a tomb.

The appropriation of two distinct burial rites that adhere to both officialdom and nativeness produced a hybrid approach to sending off the dead, whose remains were deposited in a specific round-shaped tomb called xiangtianfen 向天墳 in Chinese, which literally means “a tomb facing the sky.” This name was derived from the theory that in certain parts of Southwest China, the tombs were built with the Yi lunisolar calendar in mind, as a way of connecting the land to celestial beings.Footnote ⁶⁴ Whether adorned with a gravestone or not, xiangtianfen tombs usually did not contain a coffin with a body, but rather the cremains (Ch. guhui 骨灰) of one or often multiple bodies. In some instances, the cremains contained whole bones or their fragments, which did not completely burn to ashes during the cremation process. Interestingly, the Nuosu-Yi language does not have a direct translation for xiangtianfen and therefore refers to them as coqyrho ꊿꐗꉻ, “a cremation site,” or by the more specific and poetic expression “nzymo qyrfur” ꌅꃀꐗꃝ, which captures both the sense that these round tombs were reserved for the nzymo elite and the notion that their ashes were “married” (Nuo. fur ꃝ) both to the land they ruled and to the Ngemu Puggu/Shymu Ngehxa afterworld.Footnote ⁶⁵

The absence of an exact Nuosu-Yi equivalent term for xiangtianfen suggests that they were extremely rare in the core of Liangshan, with most of them being found in the transitory region of Leibo. Cremation in Liangshan is still undertaken today on pyramid-like structures made of wood, which somewhat resemble the tumulus-shaped xiangtianfen, but leave no traces behind since the cremains are often placed, left, scattered, or even interred by a tree in the forest close to the dwelling of the deceased or in a bamboo grove behind the house. Perhaps, though, a remarkable exception emerged from the aforementioned case of the uxorilocal marriage of Anre Ayy, originally from the Shuixi of Guizhou, into the Shama nzymo clan that resided in the core area of untamed Liangshan. In today’s Animi Village (阿尼米村; Nuo. Anyimu ꀋꑍꃅ) of Guli Township (古里鎮; Nuo. Ggunyi ladda ꈬꑌꇁꄿ), close to one of the former offices of Shama nzymo, are two clusters of considerably old round xiangtianfen (see Figure 5). Despite lacking a tombstone to verify the origin of the cremains contained within their inner rings, which are made of stone blocks, the existence and position of these xiangtianfen speak volumes. Being the burial sites of important people, perhaps even the nzymo, they appear to confirm the orally transmitted genealogy of the Shama clan that mentions Anre Ayy’s arrival as well as the origin of many of Liangshan’s inhabitants further to the east, which was the source of the major waves and flows of migrations that filled the area’s ridges and ravines. Going by this orally transmitted genealogy, it appears that these migrants brought their hybrid burial rite with them. This long link to the east testifies to the expansion of local Yi polities, which had Liangshan as their westernmost limit. With the weakening power of the native chieftains during the Ming and Qing, Liangshan turned from a buffer zone into an ideal hideout.

Figure 5. A tomb in one of the two tomb clusters known as the Guoaonaiju Stone Block Tombs (果奧乃居石板墓) and the Longgou Stone Coffin Tomb Cluster (龍溝石棺墓群), located near the former office of Shama nzymo, which I associate with the story of Anre Ayy (Courtesy of the author, 2023).

Besides xiangtianfen, the hybrid and “in-between” nature of their associated burial practices was recorded in a story about the troublesome and long-term burial of Yang Shijin (楊石金; Nuo. Azho Mabo ꀋꍐꂸꀨ), the Azho clan’s most officially celebrated hero who accomplished great deeds for the empire. Already adorned with a fifth-grade military rank, in 1867, Shijin perished in Huanglang while assisting the imperial army’s punitive expedition against the notorious Ggenra ꈦꎖ clan, which fiercely resisted the oversight of other clans and the Qing court alike.Footnote ⁶⁶ Shijin’s body was carried away from the battlefield in a litter, later put into a coffin, and remained on display on (or in) a local mountain (or cliff). According to the Yi custom, until revenge was taken, the burial could not be finalized. The son of the deceased, Yang Deshou 楊德壽 (Nuo. Azho Hoho ꀋꍐꉺꉻ), was an infant when his father died. He was not only too weak to take revenge on the Ggenra clan himself but, allegedly, also did not have enough power to keep his slaves, who absconded in great numbers. Two years later, a delegation of provincial officials arrived to resolve the matter. Different clans and their various lineages submitted to them, in an attempt to gain the recognition of the court, appropriate some of its power, and project it upon their local subjects. However, the Ggenra clan chose to follow the different Yi custom of seeking mediation with the officials, and they offered one thousand silver ingots as well as numerous cows and sheep to the Azho clan as a settlement. For Deshou, this was not enough. He wanted five hundred slaves and more ingots. The matter dragged on. The officials lost their patience and left, obviously lacking sufficient leverage to resolve the problem to their satisfaction. After Deshou’s premature death, his son Yang Zhongting 楊忠廷 (Nuo. Azho Sizu ꀋꍐꌋꊥ) was, again, an infant, and hence too weak to settle the matter. Deshou’s older brother, who had been delegated to settle this matter, failed as well. All these years, Shijin’s body still lay in the coffin, which made the clansmen very uneasy. In 1877, after a decade, his body was finally cremated as per the Yi burial custom.Footnote ⁶⁷ The location of Shijin’s xiangtianfen remains unknown. According to Yi sociologist Liu Yaohan, a xiangtianfen standing south of today’s Leibo county-town does belong to the Azho clan, but not to the branch that acted as native officials, because its design does not feature a slot for the insertion of a tombstone.Footnote ⁶⁸ The presence or absence of a tombstone slot, then, also says something about the position of the deceased within the local hierarchy, where the deceased with a tombstone leaned more towards officialdom, and those without it or without a grave leaned altogether to nativeness.

In Leibo’s Tieggurmu on the northwestern riverbank, where some of the Hxala clan members resided after the “Upheaval of Three Heroes” in 1589, many xiangtianfen and their tombstones have succumbed to the passing of time. Along one of the roads from Huanglang to Leibo, which passes through Tieggurmu, the landscape opens onto a handful of vistas to the opposite, southeastern riverbank. Sheltered by a grove close to the area of the present day Yatang Village (堰塘村), several such tombs are falling into oblivion with empty tombstone slots and pillaged inner rings. Through brush, ink, and woodblock prints, however, some of the information and epitaphs on the tombstones made it into the pages of local gazetteers, where they are preserved. The scarce records suggest that members of the Hxala clan occupying lands on both riverbanks were in close touch, as An Zhenji 安貞吉 (whose Nuosu-Yi name is unknown, suggesting his distancing from nativeness) attended a military school in Yongshan. His successor An Guoxi 安國璽 even obtained the court-issued insignia of the tuqianzong military title for pacifying the borderland.Footnote ⁶⁹ In nearby Qylemu, two well-preserved xiangtianfen rest in the field (see Figure 6). Although not containing any of the names of the aforementioned native chieftains, their tombstones display Nuosu-Yi as well as Chinese characters for both surnames under which their inhabitants were known––Hxala and An––which points to their membership in the local nzymo genealogy and their position in between officialdom and nativeness.Footnote ⁷⁰

Figure 6. Hxala tomb in Qylemu (left) with a bi-lingual tombstone (right) (Courtesy of the author, 2014).

In Huixi on the southeastern riverbank, where the main nzymo lineage of the Hxala clan migrated after 1589, their presence is considerably more palpable. The graveyard is hidden together with the native chieftain’s office in the old city amid the newly built blocks of flats and a playground. It is designed in the sumptuous style typical of imperial officials, with monolingual, Chinese-language tombstones showcased at the center and framed by ornate archways (see Figure 7). The only sign of nativeness in this design, typical of officialdom, is a massive round xiangtianfen that stands behind each tombstone. Local historians, again, managed to copy some of the information from the sandstone tombstones before the characters faded away. The records reveal that after An Yongchang’s death, his son An Tianzhu 安天柱 inherited the office, which he resettled from the mountain slope to the lowland near the Jinsha River. Hxala Shyge’s material offers a supplement to these terse records, with the claim that in 1738, the inheritor of his office, An Qingchao 安慶朝, built and ran a Confucian charitable school to educate the local population. Later, in 1802, An Qing 安清, whose tomb dates to 1823, inherited the local office. Like his relatives in Tieggurmu, he eventually gained a hereditary military rank. In 1811, he and several other prominent Hxala clan members erected stone pillars (one of which still stands) to commemorate the deeds of his ancestors. Hxala Shyge further claims that since 1822, starting with the rule of An Xiangheng 安像恆 and later An Ruitu 安瑞圖, Huixi became a market town, where the incoming Han population was welcomed to settle because the local Yi were keen on trading with them. The construction of a caravan station along with a city wall and numerous shrines and temples marked these developments. In 1870, An Shaoyang 安少陽 was the last one to officially inherit the office. Towards the end of the Qing-Dynasty’s rule, most of Mahu’s native chieftaincies started to slowly slip out of existence, officially due to the absence of suitable male heirs. The Qianwanguan’s chieftaincy, for example, produced a locally famous female chieftain Yang Daidi 楊代蒂 (Nuo. Azho Dadi ꀋꍐꄊꄂ), who in later life saw herself as a symbolic inheritor of the chieftaincy. The Hxala clan did not have problems with their heirs. However, the swansong of the office came with the decline of the Qing Dynasty and the rise of an entirely new, Eurocentric republican political system. The diminishing influence of the office under the new regime symbolically passed to An Shizhen 安世珍, then An Jiazhen 安家珍, and, finally, to Hxala Shyge’s father, An Dazhang 安大璋, who was known locally as an elementary school principal. He died in 1941 when Hxala Shyge was only three years old.Footnote ⁷¹ Paradoxically, however, when putting aside the waning political influence of the Hxala clan and focusing on its culture, it appears to have come full circle. Under the republican system, the Hxala clan returned to the state it found itself in at the beginning of Ming Dynasty, when it was acculturating itself to the Confucian, Buddhist, and Daoist rites that were culturally so close to officialdom that they may have significantly suppressed many elements of Hxala nativeness. Yet as evidenced by the Hxala clan history, when necessary––and especially in times of defense––they could reconstruct and reignite their nativeness, provided there was space for it to take root. With the exertion of a little will and creativity, and acts of appropriating the court’s material culture, a space could always be produced. This remained true of the cultural-historical heritage of the native chieftains even after the collapse of the empire.

Figure 7. Hxala tomb in Huixi (Courtesy of the author, 2016).

Genealogies: Entangling the official and the native

Under the Republic of China, and over the course of its first three decades, all offices of the former Mahu region gradually lost influence in terms of governance over the land or population. However, their long-term ties to the borderland areas made them interesting for the kind of anthropology conducted under the Kuomintang-ruled Chinese state, which inherited the colonial-explorative, theoretical, and methodological modus operandi of its Euro-American contemporaries. The Republic of China needed to tap into its resource-rich borderlands, and Southwest China was no exception. Two of the many scientists and explorers that visited Mahu were Ma Changshou, a sociologist by training, whose objective was to survey the customs and habits of the Nuosu-Yi in the perilous and virtually impenetrable Liangshan, and Fang Zhuangyou, an ethnohistorian in the field of borderland studies, who was interested in the genealogies of the ruling families.

On his way to Leibo, Ma Changshou traveled in the company of a teenage Yang Daidi and her family members, the only remaining direct offspring of the Azho nzymo clan. Ma managed to obtain the Azho genealogy in Yang Shijin’s now non-existent ancestral temple, which had been built inside the walls of the Leibo garrison, and phonetically re-write it from Nuosu-Yi into Chinese.Footnote ⁷² More or less by coincidence, during his stay in Leibo, Ma met the opium-sodden Hxala Tiehxa ꉐꇁꄟꉐ (Ch. Hala Tiha 哈拉提哈), who presented him with an oral performance of his lineage’s genealogy that he knew by heart. Hxala Tiehxa mentioned to Ma that the earliest history of his clan was connected to a clash with the imperial army that took place in caves. Ma thought he was referring to the era of Zhuge Liang’s southern expedition and concluded that, given the timespan covered by the genealogy and the paucity of names it contained, the records must contain vast gaps.Footnote ⁷³ But Hxala Tiehxa’s genealogy, which Ma recorded, began with An Xing. So when Hxala Tiehxa referred to the clash with imperial armies, he meant the events surrounding the “Upheaval of Three Heroes,” which became the earliest collective memory of the part of the Hxala clan that had remained in or returned to Tieggurmu after 1589––that is, those clan members who were dwelling between officialdom and nativeness. Ma’s record, which to my knowledge is the only recorded orally performed genealogy of the Hxala clan of the northwestern riverbank, was thus mistakenly––perhaps due to Ma’s lack of access to the Guangxu-era Leibo gazetteer––represented by the sociologist as incomplete.

Fang Zhuangyou’s work, which was carried out a couple of years after Ma’s, somewhat emulated the compiling practices of the Chinese historians and bureaucrats of the past. Like his predecessors, Fang used the available genealogies of the native chieftains to uncover some of their interpretations of local events. However, the major difference was that Fang strove to work with the official records as well as the written and oral histories of the Nuosu-Yi, which brought local perceptions into play. Some former chieftain families, such as those of Manyi and Zhongdu, were willing to share their genealogical records. But Fang started to encounter problems at Nixi, where he got into a heated verbal exchange with a family member who returned home in the evening just as Fang was rewriting the genealogy of his ancestors.Footnote ⁷⁴ Not long after, he arrived at the Azho ancestral temple in Leibo at a time when everybody was busy with some local festivities, and was ignored by Yang Daidi’s closest family members who claimed that they had no written genealogy to share. Nonetheless, over the following days and with the help of the local council secretary, Fang gathered four people (including one bimo) in a local teahouse facing the council building, where they made a collective effort to put together the Azho genealogy, albeit in the absence of any Azho clan member.Footnote ⁷⁵ Ignoring the Hxala clan members from Tieggurmu on the northwestern riverbank, Fang chose to visit some of the living descendants of the Hxala clan from Huixi on the southeastern side at their homes. Yet they also told him that the genealogy was not in their possession. Hxala Shyge’s claim that the genealogy was lost during the Cultural Revolution, which happened roughly two decades after Fang’s visit, suggests that the Hxala clan did in fact have the genealogy at this time, but that like the Azho clan, they were unwilling to share it. Because Fang did not find it in the prefectural archive either, he had to cobble it together like his imperial predecessors with the help of temple merit steles (Ch. gongdebei 功德碑) and tombstones.Footnote ⁷⁶ After finishing his work, Fang compared his results with those of some of his predecessors and lamented. All the recorded genealogies––not only the official bureaucratic Chinese and Nuosu-Yi versions but also the individual accounts from each of these two sides––significantly differed from each other.Footnote ⁷⁷

Since genealogies were necessary to verify the ruling family’s legitimacy, and their existence was a precondition for succession, whenever they were missing it was up to clerks, historians, or the Nuosu-Yi bimo record keepers to reconstruct, if not reinvent, themFootnote ⁷⁸. These reconstructed genealogies were often based on the creative reinterpretation of collective memories. Both Ma and Fang were nevertheless fortunate, as some of their predecessors had failed to obtain any information, as evidenced by the personal writings of Assistant Prefect Liu Wenwei (劉文尉) from 1832, who was stationed in Leibo in the first half of the eighteenth century and had investigated the situation and records of the Azho clan. “When Yang Yinglong (楊應瀧) was inheriting the office, the order was given to investigate his genealogy. His family, however, already had disappeared from the official records. Somebody from their clan [was therefore asked to] write it down [from memory],” Liu begins his remarks. “The Yi do not study Chinese characters. I also heard that in the past, the Yi characters appeared in a column on the right and resembled the foot[prints] of a crow,” the author adds, indicating that some official records might have been bilingual. “Nowadays, the characters are carved into a wooden tablet,” he continues, indirectly pointing to the script’s shifts between ritual and bureaucratic usage when seen in the context of the recorded wooden tablets used by Yang Jiuzha’s wife, Lady Ma, during the “Upheaval of Three Heroes.” Liu disappointedly notes that “All Yi whom I know and asked told me they do not have [a genealogy],” before concluding that: “Those who submitted to the imperial court’s governance did not learn their own script or Chinese. Therefore, they cannot recognize a single character.”Footnote ⁷⁹ Liu’s account brought to light another paradox of being caught between officialdom and nativeness––forgetting or abandoning the native Nuosu-Yi form of writing (or going without it if there were no knowledgeable bimo at hand) while also failing to absorb the Chinese official one. His account also points to the possibility that the Nuosu-Yi script might have fallen out of its original official use (presumably in some of the former Yi kingdoms) only to have been appropriated later and exclusively for native rituals. Most importantly, though, his account shows just how flexible the (re)inventing and passing on of genealogies was in Leibo during Ortai’s investigation in Wumeng and Zhenxiong native prefectures. This flexibility in fact contributed to the partial illegibility of native genealogies to the officials of the bureaucratic Chinese state.

The cases of Fang Zhuangyou and Liu Wenwei fleshed out the difficulties faced by many when cobbling together historical records. Liu Wenwei’s record testifies to the fact that the Azho native chieftains either did not want to share their genealogy with Liu or the clerks in the archives or that they had lost their genealogy and perhaps even the ability to record it at one point. This opened up a space for the Azho clan to manipulate the succession process. Unlike Ma Changshou and Ma Xueliang 馬學良, who both managed to obtain quite a lot of texts during their research, most scholars and bureaucrats who went to Liangshan a century earlier, such as Liu Wenwei, had to make do without the knowledge that genealogies imparted. Many Nuosu-Yi clans remained relatively secretive about their genealogies and texts, and especially those kept by the bimo, until the 1950s. This changed after the 1980s in particular, when the branch of Yi ethnohistory (“Yiology,” Yixue 彝學) run by scholars that included the emerging Yi academic elites, turned the sharing of the texts from a matter of intra-clan transmission to one of public dissemination through vernacular and scholarly academic writing. Within the last few decades especially, the practice of publishing clan genealogies in both Chinese and the Nuosu-Yi language became a means of boosting the prestige of a clan or its lineage.Footnote ⁸⁰

Yet, the individual and regional versions of these genealogies have long differed from each other within the Nuosu-Yi language accounts. The discrepancies between them become more pronounced when looking at the overlaps between the Nuosu-Yi and Chinese-language materials. Take, for example, the Nuosu-Yi language records of the Hxala nzymo’s clan. In the local government-sponsored volume of the Sichuan Nuosu-Yi clans and lineages edited by Qomo Chyho, for which a large team of Nuosu-Yi fieldworkers have conducted an in-person survey that dates back to the 1980s and has run across multiple years in various parts of Liangshan, the Hxala clan is represented exclusively by the nzymo lineage of Huixi. It begins with a handful of ancestors carrying Nuosu-Yi names. However, from the early Ming Dynasty onwards, the names in Qomo Chyho’s edited volume match the Chinese names quoted in numerous official historical sources. The monikers for the native prefect An Ji and all those who come after him are, then, only phonetically transcribed from Chinese into the Nuosu-Yi script.Footnote ⁸¹ Another source makes use of a reconstructed Hxala genealogy in which certain names are supplemented with their Nuosu-Yi equivalents.Footnote ⁸² As Hxala Shyge admitted, his Nuosu-Yi language skills are only rudimentary because he grew up mostly with the Chinese language. So when the people conducting Qomo Chyho’s survey came, he simply recited the Chinese names of his ancestors found in the official records that he put together when working as a researcher in Liangshan’s Museum of the Slave Society of the Yi Nationality (Ch. Liangshan Yizu nuli shehui bowuguan 涼山彝族奴隸社會博物館) located in Xichang.Footnote ⁸³ The Nuosu-Yi language transcription done by Qomo Chyho’s team and later reproduced by others is another example of the kind of mutual appropriation that may lead––in a snowballing manner––to the creation of hybrid forms of officialdom and nativeness.

Histories and identities: The localized state and the Nuosu-Yi over the longue durée

Official and native genealogies are the principal lines along which the history and even the hybrid form of the still-inherited chieftain’s identity in Mahu unfolds. It is therefore not surprising that the (re)construction, (re)presentation, and (re)positioning of official and native genealogies shapes the ways in which local histories are being written and told. One of these histories is the official narrative connecting the Mahu area to Zhuge Liang’s southern expedition. While the historical sources dedicated only one short sentence to the clash of the famous Shu-Han general and the local chief Moho,Footnote ⁸⁴ the Ming-Dynasty novel Romance of the Three Kingdoms (Ch. San guo yanyi 三國演繹), which came out one millennium after this historical event, stretched this brief mention into a relatively long, eventful chapter.Footnote ⁸⁵ Echoing this story, the Leibo gazetteer and the gazetteer of Sichuan Province, both of which date to the Qing Dynasty, mentioned a place called Wumasi 五馬寺 that lies in the mountains above Hxala Lake, where in 1732 a local stonemason discovered four bronze drums hidden in a karst cavern. The drums are thought to originate from Zhuge Liang, who allegedly gave them to Moho and his people as an insignia permitting them to keep ruling their lands after he and Moho allegedly befriended each other. The discovery happened immediately after the Mitie massacre, and the drums were stored in the newly erected local Guan Di Shrine in Leibo.Footnote ⁸⁶ This official story in the Mahu area clearly combines narratives from official historical sources and popular literature. However, the snowballing of stories happened several times over the longue durée and is still ongoing.

Traces of these events are found today just in front of Dahai Village (大海村), where the local Ggaha nzymo had his office, and where a few dozen meters into the spectral mist, calm, and light blue waters of the lake, there is a small islet colloquially known as the “sea hump” (Ch. haibao 海包). Yuan-Dynasty official sources claim this islet was a stronghold of the “barbarian king” (Ch. manwang 蠻王).Footnote ⁸⁷ Nowadays, the islet hosts a Buddhist Hailong Temple (海龍寺), which features statues of Moho and his people in the main hall, while a statue of Guanyin dwells in the innermost hall. The pillars of the main hall are adorned with the Nuosu-Yi script (see Figure 8). A recently erected stele made of concrete claims the temple was built in 1589, immediately after the “Upheaval of Three Heroes” was subdued, and that it was the Hxala clan that hybridized their cultural practices with that of the empire and built this as the sole Yi temple in all of China.Footnote ⁸⁸ However, there is no record in the official sources for the year of the Hailong Temple’s alleged establishment.Footnote ⁸⁹ Only a bronze bell similar to those cast by Mahu chieftains in other temples confirms that the place was “rebuilt” in the late eighteenth century during the rule of the Qianlong Emperor––although the temple was absent from all official and non-official records at that time.Footnote ⁹⁰ Notably, the Guangxu-era Leibo gazetteer lists it among its “ten views,” maps, and leisure historical sites (Ch. guji 古跡), but does not present it as an official state-run temple (see Figure 9).Footnote ⁹¹ A brief mention in a writing from the Republic of China also mentions a “villa” (Ch. bieshu 別墅) on the islet, but not a temple.Footnote ⁹² It seems, then, that a much bigger player entered the stage who wanted to make use of Moho’s story centuries after Pushur Yobo of the Hnielie clan claimed ancestry to him during the “Upheaval of Three Heroes.”

Figure 8. Hailong Temple with statues of Moho and his people (Courtesy of the author, 2011).

Figure 9. An illustration of Ma Lake with its “sea hump” and the temple from the Guangxu era (Qin and Wan, Leibo ting zhi, “Tu kao” 20a–21b).

Examining some of the recent writing in local periodicals, it becomes evident that both the connection between Moho and the alleged kingship of manwang and the later design of the temple are recent embellishments to the narrative on the part of writers and policymakersFootnote ⁹³ with no grounding in historical evidence.Footnote ⁹⁴ This story, which purports a historical reconciliation and harmony between an important Chinese historical figure and the local ethnic population, was most probably written to promote the PRC’s ethnic policies concerning the “unity of nationalities” (Ch. minzu tuanjie 民族团结). The story has also contributed to the development of the local tourist industry. Some of the local sticky stories even reached the opposite riverbank, where they have clung to an extraterrestrial object. During my visit to the now insignificant and small village of Mitie, the older villagers claimed that a meteorite found on the premises of their settlement was in fact a cannonball that landed there during Zhuge Liang’s campaign in Mahu. Here, the Moho (hi)story––which had been previously overwritten by Hxala-derived toponyms—managed with state support to both resurrect and sideline the remarkable events surrounding the local Hxala, Azho, and other clans. The reconciliation story of Moho and Zhuge Liang serves the desired goals of the PRC state better than the convoluted history of how the Hxala and Azho clans appropriated elements of officialdom and nativeness.

Despite being somewhat bitter that the Hxala clan’s glorious past has not attracted more fame in today’s China, and that its deeds are falling into oblivion even in the region the Hxala formerly governed,Footnote ⁹⁵ Hxala Shyge sees his family and himself as having been good servants of the Chinese state throughout the longue durée. He firmly rejects that his history is the product of Chinese imperial colonialism. In so doing, he takes a similar stance to many Yi intellectuals, who tend to position themselves as an integral (and even co-founding) element of broader Chinese civilization,Footnote ⁹⁶ albeit for reasons that go with (rather than against) the recent political and ideological attunement of the PRC.Footnote ⁹⁷ Hxala Shyge’s own life journey is similarly complicated. As a teenager, he participated in the Democratic Reforms in Liangshan that freed all the people formerly enslaved by the powerful Nuosu-Yi. His Hxala nzymo clan was among those who owned many slaves. Later, and much like its imperial predecessor, the Communist Party of China appointed some of the nzymo who pledged their loyalty to the new regime to local government positions. Hxala Shyge therefore worked in government offices, schools and research institutions. In his writing, he tends to downplay the parts of his clan’s history in which his ancestors were seen as villains rather than heroes and to emphasize the parts where they practiced orderly officialdom. Interestingly, he entirely omits the Hxala clan branch from Leibo’s Tieggurmu since they were much closer to Liangshan’s anarchic nativeness and may even have sided with the rebels during the Mitie massacre. Despite being fluent only in the Sichuanese Chinese language variant, Hxala Shyge proudly observes most of the Nuosu-Yi customs, including the art of reinterpreting his genealogy. He has appropriated elements from the Chinese state’s officialdom and his own Nuosu-Yi nativeness into the story of his clan and lineage.

Numerous local (hi)stories underpin the ideological positions of the PRC over the longue durée, while others problematize it. Some aspects of the Hxala clan history, including the fact that they inhabited both banks of the Jinsha River where they acquired their official insignia and clung to officialdom and nativeness, go against the grain of a core concept in the current PRC ideology––Marxist historical materialism and its evolutionary model of social development. Evoking the inner and outer borderlands of the empire, which became especially visible after the Mitie Massacre, the Yunnan riverbank was called “inside of the river” (Ch. jiangnei 江内), while the riverbank in Sichuan’s Liangshan was referred to as “outside of the river” (Ch. jiangwai 江外).Footnote ⁹⁸ Everywhere inside of the river was considered closer to officialdom, order, and authority––the purported evidence being the monolingual xiangtianfen tombstones in Huixi––which the PRC portrays as having been part of the “feudal stage” of society. By contrast, everywhere outside of the river was considered closer to nativeness, as shown by the bilingual xiangtianfen tombstones in Tieggurmu and Qylemu. Beyond this point, and further to the west, lay an illegible and untamable realm of chaos that allegedly dwelt one step below feudalism at the stage of a slave society. The Jinsha River in this vision serves as a time-space divide between feudalism in much of Yunnan and a slave society in Sichuanese Liangshan.

However, this notion of (historical) time is problematic. Although the people on both riverbanks inhabited different, even if often entangled, worlds, as shown by their burial practices and genealogical recordkeeping, they were contemporaries, and they sometimes came from the same clan. People on the two riverbanks often interacted with each other and traveled back and forth across the Jinsha, especially in the nineteenth century. It would even have been possible to have come eye to eye with each other across the Jinsha gorge. The Yi certainly enslaved people on both sides of the river, as the gazetteers have vividly portrayed on numerous occasions. Yet deterministic assumptions about their so-called phase of social development deny their coeval status.Footnote ⁹⁹ This perception of the Yi continues to cause a whole range of social problems today, including their stigmatization and discrimination as people who are allegedly backward.

Remarkably, the banks of the Jinsha River and the collective memory of catastrophic violence associated with them probably contributed to some of the most distinctive features of Nuosu-Yi culture, society, religion, and everyday life in Liangshan. As SwancuttFootnote ¹⁰⁰ points out, the lives of Nuosu-Yi are filled with thresholds that need to be crossed, often with the assistance of bimo ritual specialists. A case in point is the popular yyrku ꒌꈎ ritual, in which the bimo calls back the soul of a person that became detached from the body of its carrier when attacked by ghosts. Many of these ghosts include the souls of the deceased imperial soldiers and other non-Yi who died violently in Liangshan and did not receive proper death rites. Soul loss tends to cause a plethora of problems related to health and other matters for the Nuosu-Yi, which I often saw bimo address by weaving a ryji ꏜꏣ rope made of fresh grass and placing it over the household threshold of the person hosting the yyrku ritual. These bimo pointed out that many kinds of grasses they use during their rituals grow only on riverbanks and that in ancient times these ropes were used to wade the Jinsha River, including in moments when people had to flee from their rivals and/or larger conflicts. The blueprint for the abstract Nuosu-Yi “threshold of the cosmos,” as Swancutt calls it, which resonates with other thresholds in the landscape––from the household thresholds crossed by lost souls in the yyrku ritual to the Jinsha River crossed by countless Nuosu-Yi over time––could well be found in the accumulation of past historical events. That said, without the role of the empire in the Mahu borderlands and elsewhere in Liangshan, the Nuosu-Yi of today wouldn’t possess the same, in their words, “excellent culture” (Ch. youxiu wenhua 優秀文化).