Mythogeographies of anthropological knowledge: writing over the lines and footsteps of history in Southwest China
Abstract
enIn this article, I delve into the field diary of Ma Changshou – a major Chinese ethnohistorian and social anthropologist active between the 1930s and 1960s – to show how his journeys through Liangshan, a mountainous land in Southwest China inhabited by the Nuosu-Yi, led to a new kind of anthropological knowledge. Ma worked under the theoretical, methodological, and ideological assumptions of his era, where he strove to assemble his findings on Nuosu-Yi kinship relations and textual sources into a linear historical narrative, even though the Liangshan of his day was not fully controlled by a centralized state bureaucracy. Retracing Ma's journey in mythogeographical fashion, I propose that the travel-fieldwork trajectories of China's Republican-era social scientists and their interpretation of data obtained from native chieftains, bimo ritualists, and other members of prominent Nuosu-Yi clans have shaped today's knowledge of Nuosu-Yi history, society, and culture. Notably, Ma's mythogeography fleshed out two different essentialized but intertwined ways of seeing the past: that of state-bureaucratic societies and that of genealogy-based societies like the Nuosu-Yi. Mythogeographies like this can give rise to more than new ways of performing genealogies; they can throw light on the anthropo-history of China and the world at large.
Résumé
frMythogéographie des connaissances anthropologiques : écrire par-dessus les lignes et empreintes de pas de l'histoire dans le Sud-ouest de la Chine
L'auteur se plonge ici dans le journal de terrain de Ma Changshou, grand ethnohistorien et anthropologue social actif entre les années 1930 et les années 1960, pour montrer comment les voyages de celui-ci à travers le Liangshan, une région montagneuse du Sud-ouest de la Chine peuplée par les Nuosu-Yi, a donné naissance à une nouvelle forme de connaissance anthropologique. Le travail de Ma était sous-tendu par les suppositions théoriques, méthodologiques et idéologiques de son temps et il s'est donc efforcé d'assembler ses connaissances sur les relations de parenté des Nuosu-Yi et les sources textuelles en un récit historique linéaire, alors même que le Liangshan n’était pas complètement contrôlé par une bureaucratie d’État centralisée à l’époque. Retraçant le voyage de Ma à la façon d'une mythogéographie, l'auteur suggère que les trajectoires de voyage et de travail de terrain des chercheurs en sciences sociales dans la Chine républicaine et l'interprétation faite par celui-ci des données obtenues des chefs locaux, des praticiens des rites bimo et autres membres des clans éminents des Nuosu-Yi ont donné forme à nos connaissances actuelles de l'histoire, de la société et de la culture de ce peuple. Il est intéressant de noter que la mythogéographie de Ma a donné naissance à deux manières différentes, essentialisées mais entremêlées, d'envisager le passé : celle des sociétés d’État bureaucratiques et celle des sociétés généalogiques telles que les Nuosu-Yi. Les mythogéographies ne font pas que susciter de nouvelles manières d'examiner les généalogies : elles peuvent aussi faire la lumière sur l'anthropo-histoire de la Chine et du monde dans son ensemble.
‘Over here’, I am told by a Nuosu-Yi friend, who points to a cluster of tall old trees offering much-needed shade in the hot summer afternoon. Surrounded by chirruping birds, buzzing insects, and lush vegetation, he tells me about the Lili chieftain ancestors’ tombs. We are standing where the fertile Jianchang Valley1 reaches its widest point, with mountain ridges towering above on both sides, just north of Xichang,2 the administrative centre of the Liangshan Yi Autonomous Prefecture3 in the southwest of Sichuan Province, People's Republic of China (PRC). On this stretch of land that Chinese historians call Anning chang,4 and Nuosu-Yi know as Lurjjo gatuo,5 rises Jiaoding Mountain6 with a settlement at its foot that until recently was called ‘Niubashi’.7 My companion continues his explanation. ‘The last of the chieftains who inherited the local offices (Ch. yamen 衙門) came here every year to clean these tombs and offer sacrifices. Their gravestones remained in pristine condition until the seventies’, he adds, lighting a cigarette. Pointing eastwards to the river, he says, ‘but then the government built the weir with canal systems and embedded the shattered gravestones and their foundations into its embankment’. Decades earlier, a text had been carved into three of these seven tombstones to reveal parts of the Nuosu-Yi Lili clan genealogy, whose members are thought to have ruled Liangshan for hundreds of years through the imperial and republican period.
The Yi nationality (Ch. Yizu 彝族) is an ethnopolitical category created during the campaign for the recognition of nationalities in the early 1950s by the ruling Communist Party of China (CPC). Once referred to by the old pejorative exonyms of Yi (夷) and Lolo (猓玀), which have been replaced by the allegedly neutral Yi (彜) character, the Yi nationality encompasses dozens of linguistically and culturally related ethnic groups spread across Southwest China, and especially Sichuan, Yunnan, and Guizhou provinces. Anthropologist Lin Yaohua, who built on previous linguistic analyses during the recognition campaign, dubbed the Liangshan Yi – who self-identify as Nuosu (Nuo. ꆈꌠ) – the ‘archetypal Yi’ (Mullaney 2011: 112). However, his decision was undoubtedly influenced by the fact that the Nuosu are the most populous Yi and by the state-stipulated Marxist-Leninist ideology and its Marxist historical materialism, which positioned the Nuosu-Yi as a backward ‘slave society’ in social evolutionary terms. Paradoxically, the ‘slave society’ label ascribed to the Nuosu-Yi, whose practice of keeping slaves and captives was dismantled during the Democratic Reforms of the late 1950s, endowed them in the 1980s with the discursive power of being the most ancient and authentic Yi people.
At the heart of Nuosu-Yi society is their genealogical social order (Nuo. cyvi ꋊꃤ) that divides them up according to their essentialistic theory of bone colour and hardness. This essentialism enabled a minority of black and hard-bone endogamous aristocrats (Nuo. nuoho ꆈꉼ; Ch. Heiyi 黑彜) with a hereditary superior status to dominate the white and soft-bone non-aristocrats (Nuo. quho ꐎꉺ; Ch. Baiyi 白彜). Later scholarship on the Nuosu-Yi further hierarchized the two groups in line with the ‘slave society’ label. Beneath the nuoho were the quho, who were divided into the ranks of free commoners (Nuo. qunuo ꐍꆈ) and two strata of enslaved bondservants who performed light (Nuo. ggajie ꈝꏦ) and hard (Nuo. gaxy ꇤꑭ) labour. The aristocratic nuoho are often conflated in national as well as international scholarly accounts with the supreme sovereign rulers (Nuo. nzymo ꌅꃀ), who are still commonly thought to have the hardest bones (see Lu 2001: 69; Hein & Zhao 2016: 274, 287). Many of those referred to as nzymo by the Nuosu-Yi were, however, appointed by the Chinese imperial court as native chieftains (Ch. tusi 土司) – a situation that has thrown doubt on the universal hard-or-soft-bone distinction. The court incorporated select members of both nuoho and quho strata into its native chieftain regime (Ch. tusi zhidu 土司制度) to act as intermediaries between the local population and the state. It was these nzymo who seeded the court's language, rituals, and ethics among ethnic people in the empire's borderlands.
While the collective cultural identity of the Yi nationality started to solidify in the 1980s under the unforeseen leadership of prominent Yi scholars, the roots of their discipline of Yi studies (Ch. Yixue 彝學) lie in the Republic of China (ROC; 1911-49) and are intrinsically connected to the research practices of Han-Chinese scholars (Kraef 2014) intent on indigenizing Western theories and methods of social science in China. One of the most influential of these Han scholars was Ma Changshou (馬長壽,1907-71), a sociologist domestically trained in the early 1930s at the National Central University in Nanjing, who reconstructed the genealogy of the historically most influential native chieftain's office in Liangshan, Lili nzymo, from the depths of history. Like the characterization of the nzymo, this history became common knowledge in China (e.g. Jiang 2013: 577-655) and international anthropology (e.g. Harrell 2001: 85-6). Ma was thought to be more of an ethnohistorian until thirty-five years after his passing, when his published fieldnotes (Ma, Li & Zhou 2006) covering two research journeys to Liangshan (in 1937 and again between 1939 and 1940) revealed him to be an avid ethnographer. The distinctive Nuosu-Yi script8 appears on numerous pages of Ma's diary, having travelled there from the scroll-books (Nuo. teyy ꄯꒉ) of Nuosu-Yi ritualists (Nuo. bimo ꀘꂾ) who often served in the nzymo offices as educators and clerks (Wen 2020) but kept their writings secret from outsiders and laypeople. Ma was among the few outsiders who managed to penetrate deep into the feared area of Liangshan – at that time filled with opium growing and anarchic gun violence (see Lawson 2017) – and persuade some bimo to share their anxiously guarded knowledge. He was forced, like others who attempted visiting Liangshan at the time, to stay perpetually on the move and perform his fieldwork in a ‘culture as travel’ manner (Pan 2002). The westward trajectory of his unidirectional, one-month journey through the heart of Liangshan – walking and sedan-chair-riding through the Shuonuo plateau9 – became a transformative personal experience and a founding pillar of anthropological knowledge connected to contemporary Liangshan.
Much has been said about the critique of China's social engineering projects and the related practices surrounding the construction of its multi-ethnic nation, but considerably less attention has been paid to the scholarly influenced formation of the most basic Nuosu-Yi social rankings. The same people who stood behind the making of China's minority nationalities, including Ma and his research team, set out to derive, synthesize, and essentialize the Nuosu-Yi social rankings from their internally diverse and fluid cosmology. Later, these social rankings were internalized rather unreflexively by the emerging Nuosu-Yi academic and political elite, who, having assumed the total authority of the scientific practice behind them, popularized this vision of themselves through their everyday conversations with Nuosu-Yi laypeople. As a result, the Nuosu-Yi ethnohistories and, as I am about to show, anthropo-histories were derived from their cosmology just as their ethnopolitical identity was derived from it during the recognition campaign. It might seem that this process became a fait accompli in the manner of the mythologies and legends surrounding modern European nations, but the mythogeographical travels behind the Nuosu-Yi stories remain surprisingly fresh after almost a century. Revisiting these anthropo-histories evokes their mythogeographies in potent ways that make them ‘participate, anew, in the everyday’ (Desjarlais 1992: 215). Crucially, the precise directions that scholars take (see Wang 2017) during their mythogeographical journeys shape the epistemologies behind their knowledge production and the uptake of that knowledge by others. Mythogeographical wanderings through space intimately connect the practices of anthropologists and their subjects – including shamans and other ritualists such as the Nuosu-Yi bimo – who travel through physical and metaphorical landscapes during rituals. Often when anthropologists and ritualists travel ‘from river to forest, they sing of their journeys’ (Desjarlais 1992: 210) in the texts they perform orally or write down. Approaching the story of Ma and his research partners through my own journeying, I propose in this article that multiple layers of fieldwork, writing, writing over, and knowledge-making lay the foundations to diverse mythogeographies of anthropology, even as they throw light on the construction of anthropo-history at large.
The linear trajectory of Ma Changshou converges, diverges, and intersects with those of others across time, space, ideologies, and texts. Ingold points out the ubiquity and mundanity of lines as being on par with ‘the use of voice, hands and feet’ (2007: 1) because human beings generate lines wherever they go. Ma's line drawn by his sedan chair and feet reflects the way he chose to siphon the heterogeneous fragments of mostly Nuosu-Yi genealogical data into his own unilinear and unidirectional writing, which suited the common approach to historical anthropology within the ROC. In what follows, I argue that Ma's remarkable endeavour – for which he utilized his ‘body as a reader uses his eyes’ (Mueggler 2005: 729-30) as he moved from Leibo10 to the Jianchang Valley (Fig. 1) – created a new epistemological line of ‘seeing’ that informed his perspective on the genealogy of the Lili nzymo and a large portion of history connected to Liangshan. This history has continued to take shape over the decades following Ma's initial journey, when he covered the backbone of his research with the flesh of his data with the view to synthesizing it, often together with interlocutors and informants encountered along the way, both during and after his second and third (1957) research trips to Liangshan. Informed most notably by Lewis Henry Morgan's then-popular Ancient society (Ma et al. 2006: 289-91), Ma managed to assemble the cosmological vision of the pre-ROC dynastic imperial bureaucracy, the anthropo-historical practices prevalent during the ROC, and the PRC-era Morgan-inspired ideology of Marxist social evolutionism.
Ma's assemblage may contain a lot of ‘in between the lines’ data discoverable only after long-term immersion in the geographical, historical, and political landscape of Liangshan, but it speaks to anthropology as a whole. Being a historian-turned-ethnographer myself, I animated the fragmentary data of Ma's travels during my mythogeographical journeying by foot or public transport between 2009 and 2019 – and later by jeep in 2023. Following in Ma's footsteps imparted me with something of ‘a paranoid, exploratory, detective-like approach to space and place’, in which I prioritized ‘anomalies and “in-betweenness”’ and worked through these ‘voids’ (Smith 2011: 268) to assemble the ‘apparently missing layers of contestation between the “geographical environment”, “emotions”, and “behaviours”’ (Smith 2015: 167). Imitating Ma's footsteps while exploring new paths carried me through Liangshan's landscape as I walked my mind (Ingold 2010) through pages of historical and anthropological texts. My sensory experience with the surroundings actualized my lines of thought and shifted the relations between these interconnected realms. Like the materiality of the landscape, the sheets of paper on which I jotted down lines became a terrain in which utterances and their written forms ‘intersect[ed] and neutralize[ed] one another’ (Kristeva 1980: 36). Maps of intertextual relations, when layered one over another by anthropologists and their subjects, may become the ‘myths’ that facilitate the productive tensions between diverse paradigms of knowledge production (Navaro-Yashin 2009: 7). Reading between the lines of Ma's fieldwork diary and the written or oral accounts of the Nuosu-Yi reveals that the selection and arrangement of ‘facts’ about the past have shaped both the anthropo-histories of the Nuosu-Yi – which reflect the indigenization of Western sciences in China – and the indigenous Nuosu-Yi genealogies themselves. This reading illuminates two contested ways of seeing the past: the historical-cum-anthropological view of state-bureaucratic societies and the genealogical view of societies like the Nuosu-Yi. The ruins of nzymo offices, their immediate surroundings, and especially the (hi)stories attached to them, form important intersections between these contestations.
Between the lines of Ma's writing lurk the stories of ethnic communities that oscillate between dwelling within reach of a bureaucratic state, hiding away from it, or ignoring it wilfully. As a ‘captive guest’ (Swancutt 2012) of the Nuosu-Yi, Ma became one of many pawns in the competitions for power between various headmen and their script-possessing bimo clerks, who stuck to their offices during the last gasp of the native chieftain regime, the eventual collapse of which made many of them roaming ritualists (Wen 2020: 281-9). So, while Ma creatively curated his sources to fit the ROC's and later the PRC's anthropo-historical enterprise that constructed subjects for ideological and political purposes, his Liangshan Nuosu-Yi history is ultimately the history of his interlocutors who overwrote the pre-ROC imperial bureaucracy. Ma achieved this feat in concert with his Nuosu-Yi research partners by repeatedly inquiring into, and later reinventing, certain details about their common practice of comparing genealogies when encountering a Nuosu-Yi stranger (Harrell 2001: 91). Genealogies are recited orally and form the basis of the Nuosu-Yi face-to-face community (Bamo 2004: 152), but they are also recorded in writing.
Like the Nasu-Yi from Guizhou, Nuosu-Yi have line-like (Ch. zhixianshi 直綫式) genealogies that record their family lineages (Ch. jiapu 家譜) but are open-ended, cyclical (thus repetitive), multidirectional and time-independent. They contrast with the tree-like (Ch. shuzhishi 樹枝式) structures typical of the linear imperial bureaucratic records that document the whole clan (Ch. zupu 族譜) (Wen 2020: 370, 378).11 Both the Nuosu line-like and open-ended ways of seeing are constructed essentialistically, but by no means inhabit separate realms, as evidenced by the four-volume genealogical compilation written by Nuosu-Yi scholars and edited by Qomo Chyho (2007) with local government support. Between the 1980s and 2007, these scholars ran broad and repetitive field surveys, obtaining written and oral genealogical accounts that they put together during numerous symposiums. In the preface (Qomo 2007: 5-13), they pledged their allegiance to Marxist historical materialism and acknowledged the genealogy of Confucius as a source of inspiration. This compendium linearized the heterogeneous and fluid Nuosu-Yi genealogies and may well have constrained their performative strength. Ironically, these scholars wanted to perform their exceptionality – with some even wondering whether their ancestral clans were somehow related to the clans of ‘Evans-Pritchard's Nuer’ – but in the end their work became similar to the state-bureaucratic genealogies. Multiple clans have since published their own versions of genealogies (e.g. Ddisse & Ddisse 2016), not only as a matter of prestige (Swancutt 2016: 141), but presumably also to counter this newly emerging genealogical grand narrative and to reignite the performance of their genealogies in this altered setting. Scholars like Ma Changshou have, then, unleashed essentialist constructions in their ethnographies about ‘the other’ that parallel the ‘indigenous essentialism’ (Scott 2016) that ethnic populations such as the Nuosu-Yi have mobilized to construct images of themselves (see Tapp 2002).
In what follows, I narrate a similar story of contested genealogical writing, which, however, points at the making of anthropology. I follow Ma on his ethnographic journey through the depths of Liangshan's forests, plains, and steep mountains to show how his vision of Liangshan was formed before taking us to the offices of several native chieftains who curated Ma's vision of the Nuosu-Yi social order. Next, I show how Ma assembled his mythogeography of the Nuosu-Yi on a further fieldwork trip, which I juxtapose with my own travels through and reading of the Nuosu-Yi lineages and anthropo-history of Liangshan. Finally, I conclude with reflections on the importance of mythogeographies for the anthropological enterprise.
Ma's trajectory: from Leibo to the Jianchang Valley
Scaling Shuonuo
Ma opens his diary (Ma et al. 2006: 17-62) with a record of his travels from Chengdu, through Pingshan,12 to Leibo, a garrison with a mixed Han-Chinese and Nuosu-Yi population. With the help of local authorities, he recruited an interpreter and an ROC-allied Nuosu-Yi nuoho guide named Niqo Bitu.13 Both were opium addicts. In Ma's view, entering the Shuonuo plateau was like travelling to a foreign country. He had to exchange his ROC banknotes for silver ingots and stock supplies, employ twenty carriers and seven sedan porters, and enjoin several Han-Chinese merchants to form his entourage of around forty people, who were then good to go.
They left Leibo on 7 March 1937. Later, they reached the Wujiao14 settlement, which was the last frontier of state-controlled land. Beyond this point lay an illegible landscape dotted with Nuosu-Yi personal names and toponyms that Ma transcribed into Chinese characters and placed into the streams of sentences filling his notebooks. Advancing through settlements often inhabited by Bitu's relatives and smaller garrisons, they reached the territory of the feared unruly Ggenra15 clan. At the entrance to a village where one of Bitu's nephews lived, the group ran into an armed and drunk Ggenra bimo, who claimed to have killed a native chieftain as well as a ‘foreigner’, the French missionary Baptisin Biron,16 who was beaten to death two years earlier by a hundred-person-strong Nuosu-Yi mob in neighbouring Mabian17 (IRFA 2021). Suddenly dropping down on one knee, the bimo pointed a gun at Ma's face, ready to shoot, but Bitu was able to disarm the bimo and point his own gun against the ritualist's forehead. Ma was worried the situation would spiral out of control, but after the bimo sobered up he came to the home of Bitu's sister, where the team was preparing to spend the night, and apologized with a live chicken brought as a token of peace.
Over the following week, Ma's team made their way through curvy steep paths, primordial forest, and snowdrifts to Shuonuo, a no-man's-land. Descending on its northwestern fringe, they arrived in Njiyi Texy,18 a place full of watchtowers and unexpectedly affluent households. They entered Limu Moggu19 (Fig. 2) – ‘Limu’ being an abbreviation of ‘Lili muddi’,20 the land of Lili nzymo – and reached Bitu's natal village of Vatuo.21 Here, the Han-Chinese merchants decided to leave Ma's team and return to Leibo. Ma also encountered problems with his sedan chair on the steep path and had to walk when one of the porters injured his leg. To make matters worse, at Vatuo, Ma's opium-addled guide needed to invite a bimo to conduct a purification ceremony for having brought a Han-Chinese deep into the mountains, which, it was feared, might cause great misfortune. After the ceremony, they continued to Lietuo22 on the opposite slope, where other problems ensued. The sight of Ma's body being carried on the shoulders of others resembled the moment in many Nuosu-Yi funerary rites when the corpse is carried to the cremation pyre. This was not an auspicious sight and Ma had to get down from his sedan chair again.
Nevertheless, three days later, Ma made Lietuo the base for his first long-term observation and, as he did elsewhere, set about mapping kinship relations: who belonged to the nuoho and quho strata and how the various local clans, among others the Shama23 and Hma,24 got along. Lacking knowledge of the Nuosu-Yi language or script, Ma needed the assistance of the bimo to interpret their texts. His guides introduced him to Ozzu,25 who had previously worked with Biron. In Ma's observation, Ozzu was shy but proud – maybe even arrogant – yet ‘not a person with nothing in his head’. Ma presented Ozzu with cloth and alcohol as payment for rewriting a scroll-book containing origin narratives (Nuo. bbopa ꁧꀿ). Ozzu always wrote down a given segment on a clean scroll, explained its content to Ma, and then chanted it.
Later, another bimo named Qubi Yiery,26 the son of a locally known ritualist, appeared at the door. Ma asked him to provide a genealogy of the Gguho27 and Qonie28 clans, which were allegedly the first to enter Liangshan from what is now northeast Yunnan. When Ozzu learned that Yiery was working for Ma, he got furious, claimed that Yiery lacked knowledge, and left abruptly. ‘They do not understand my mission, this is all research material’, Ma noted in his diary (Ma et al. 2006: 45), without grasping the inter-clan rivalries behind this show of anger. He added that although talented, Yiery's behaviour was strange. The delegation learned that Yiery had stolen their metal pot and several other items while repeatedly demanding more payment for his services. Over the course of six days, he provided a copy of one scroll-book to Ma along with his insight into the Nuosu-Yi language and marriage customs, but in the end he left angrily as well.
During his eight-day stay, Ma attended two funerals but soon found himself lamenting ‘how great it would be to stay six months or one year to immerse oneself in the local life’ (Ma et al. 2006: 46) as a war between two local clans forced his team to leave. They went to Chezzy ladda Valley,29 where the British adventurer John Weston Brooke was killed in a scuffle on Christmas Eve in 1908 after gravely disregarding the local customs (see Fergusson, Brooke & Meares 1911). Following the stream, they entered the northern fringe of Limu Juojjo30 plain but had to traverse the territory of the Hma clan, who intermarried with the Asho,31 to reach the walled garrison of Zhaojue.32 Ma managed to negotiate their toll down from three silver ingots to one and a half, sealing the deal with an ROC flag as a present. Passing through Limu Zhuhxi,33 the team was overjoyed that they had survived the feared Shuonuo.
Reaching Zhaojue
‘According to the Yi custom, tumu is one class higher than the nuoho’, Ma noted as they marched further towards Zhaojue (Ma et al. 2006: 55). In his later analysis of the Nuosu-Yi hierarchy, he explained that the tumu (Ch. 土目) are appointed relatives of nzymo (Ma et al. 2006: 354). Crossing the Bbaqi34 clan-dominated plain, they encountered heavily fortified clusters of houses that echoed the design of the garrison. On 2 April, the delegation reached a dilapidated Zhaojue, and Ma recounted its history across several pages, though his account of the Nuosu-Yi submitting to the imperial court in the mid-nineteenth century (Ma et al. 2006: 56-7) was probably an official version of what locals understood quite differently. Roughly three decades before Ma, the French military official-adventurer Vicomte Henri d'Ollone (1910: 76-7) had arrived in this area from Xichang and was astonished there were no ‘Chinese mandarins’ in the city. D'Ollone collected vernacular accounts about the substantial Han-Chinese presence in Zhaojue during the mid-Ming Dynasty before continuing eastwards through the more navigable territory of Shama nzymo, which was separated from Shuonuo by a deep gorge. Settlers had allegedly petitioned the imperial bureaucracy to grant the place city status and built a wall around it, a symbol of the empire's presence. But in 1868, after the Nuosu-Yi had become more formidable, the imperial forces suffered a defeat at Chezzy ladda and withdrew from the region (d'Ollone 1910: 101-9). Echoes of this came to light nearly a century later, in the 1950s, when victorious CPC officials sorted out the local archives and found that 95 per cent of the material had been lost to the region's perpetual warfare (SSZXBW 1999: 544). Both a patchy historical note in a gazetteer from the early twentieth century (Xu 1996 [1920]: 4.47-8) and signs of (re)construction connected to local temples and shrines (Xu 1996 [1920]: 2.24) – yet another symbol of the empire's presence – suggest that losses to archives occurred repeatedly. D'Ollone, Ma, and others possibly had access to various sets of narratives that often conflicted with each other, but it was the lack of any central rule that led to the absence of an authoritative account.
During Ma's eleven-day visit, the garrison was run jointly by the appointed magistrate and the tumu of the Hma and Bbaqi nuoho clans. His initial venture into nearby settlements for research turned to glorious failure. When he wanted to measure the inhabitants’ physical features in line with the then-current anthropological practice, his subjects became terrified of his implements and ran away in horror. His next research attempt was successful as he managed to recruit a local bimo who was labelled ‘trustworthy’ due to his venerable age of 60. This bimo provided Ma with copies of scroll-books presenting his clan's bbopa of the universe and genealogies of all the important nzymo in the region. Through these scroll-books, Ma gathered a particular vision of regional local knowledge, which decades later entered the scholarly and standardized Nuosu-Yi Book of origins (Nuo. Hnewo teyy ꅺꊈꄯꒉ, see Bender, Aku & Jjivot 2019).
Ma's team then journeyed to Jianchang Valley. Having left the gates of Zhaojue, they soon found themselves in the territory of the Asho nzymo, who intermarried with the Bbaqi. However, when they arrived at the River of Three Bends,35 called Hxuoggur ladda36 in Nuosu-Yi, Ma found not only that the Asho and Bbaqi were enemies on the opposite bank of the river just as d'Ollone (1910: 73) had noted, but also that different family lineages within the Asho clan held grudges against each other. Ma's Bbaqi escort accompanied the team only as far as the riverbank before leaving them to be received by his foes. On 16 April, after forty-one days on the road, Ma recorded that the Daxing chang37 that they had finally reached was to the Jianchang Valley what Wujiao is for Leibo: the last frontier of the state's control.
Chieftains along the Jianchang trading route and connections to Zhaojue
Between Tianba and Jydagu
Ma met Ling Guangdian38 (locally pronounced as ‘Leng’), the nzymo from Tianba,39 in a temple when visiting the sacred Daoist Lu Mountain40 that towers above Qiong Lake near Xichang. Leng was a famous intellectual, modernizer, and educator knowledgeable in both Nuosu-Yi and Han-Chinese ways whose yamen stood in the northern stretch of an old trading route that connects Sichuan to Yunnan. He had studied at Nanjing's prestigious Whampoa military college and tirelessly petitioned the ROC government for political recognition of the southwestern ethnic population (Ma et al. 2006: 62-70; Wen 2018). This encounter proved crucial for Ma, who, after returning to Chengdu, evaluated his materials. ‘The bimo books contain mostly blessings and curses, but between their lines dwells important historical material’, he noted. So, Ma met with Leng in Chengdu again, who helped him to tell the difference between the Nuosu-Yi local language variants. Ma also recruited fearless Nuosu-Yi students to measure their physical features. In 1938, still unable to make sense of the Nuosu-Yi script, he employed a bimo from Vytuo lurkur41 – a place between Xichang and Tianba – to assist him with the interpretation of scroll-books. However, Ma considered this bimo’s knowledge shallow since he could not persuasively explain all the scroll-book content (Ma et al. 2006: 71-2). This led him to plan a second trip to Liangshan.
From 6 December 1939 to 7 April 1940, Ma aimed to ‘live like the locals’ and practise longitudinal participant observation (Ma et al. 2006: 73-96). He made his main fieldsite the Tianba, which lay at the heart of the Jjieggur Galo42 area and was ruled by both Leng Guangdian and Ling Bangzheng,43 whose offices in Sheni44 and Jydagu45 were just a couple of kilometres apart. Leng became Ma's other important interlocutor after Bitu. Upon arriving in Tianba, Ma visited Ling Bangzheng, a direct descendant of the Synzy clan46 who had formerly administered a stretch of Vytuo lurkur (Fig. 3). Ling provided Ma with a genealogy of his lineage and Ma visited the tomb of his most famous ancestor, Ling Cheng'en,47 who had gained immortality in imperial history by eliminating the youngest of three leaders of the Taiping rebellion (1850-64).
In contrast, Leng Guangdian could not recount his genealogy as it was lost during the turmoil of the early 1920s along with his insignias, which, however, he did replace after returning from Whampoa in the late 1930s. Since Leng's close relatives could only provide genealogies going back four or five generations, Ma did not feel it was worth recording them. Leng claimed in his memoirs that his Syrpy clan came from the Wumeng mountain ridge48 (Ling 1988: 1) stretching from Guizhou to Northern Yunnan, which is colloquially referred to as muvu (Nuo. ꃅꃴ, ‘sky’) because its relatively high altitude makes it seemingly touch the heavens. Multiple Nuosu-Yi scholars who follow the semantic-centric method of Chinese historiography, which attributes common origins to common phonology (Harrell 1995), told me that Wumeng's toponym derives from the historically dominant clan of Uomur.49 Many nzymo see their origins in the genealogies of the mythical Yi rulers (see Wu 2001) of Wumeng, which are equated with Zzyzzypuvu,50 a mythical place of Nuosu-Yi origin to which the souls of the dead return during post-mortuary rites. Yet while the nzymo clans of Leng Guangdian and Ling Bangzheng were once related by marriage, they stopped intermarrying when a nzymo family from a neighbouring valley that could not prove its bone hardness was permitted under the patronage of Ling Bangzheng's ancestor Ling Hanping51 to seal a marriage with the Syrpy. Generations later, this decision was still questioned by the descendants of Synzy, who were suspicious that this marriage had softened the Syrpy clan's bones (Ma et al. 2006: 90). In turn, the Syrpy claimed that Ling Hanping made no mistake when permitting the marriage and that his consent was proof of their bone hardness. Clans originating from Wumeng were, then, careful when forming bonds and each proclaimed their own legitimizing narrative. The story of their origin in Wumeng endowed them with power, prestige, and authority to the extent that the Synzy even proclaimed their Wumeng ‘heavenliness’ through an inscription on the arch over the entrance to their yamen (Fig. 4). Their status heavily influenced Ma's writings.
Leng Guangdian provided Ma with bimo clerks who helped him interpret the material in his possession. Ma then oscillated in the ensuing weeks and months between the households of Leng and Ling, collecting material, often by chance. When attending a post-mortuary rite, he ran into a knowledgeable Jjike52 bimo conducting the event. On another occasion, he met a local winner of kenre (Nuo. ꈍꎞ), the competitive disputations that showcase a person's knowledge of bbopa and genealogies, to whom he attributed a deep knowledge of the primordial Gguho and Qonie clan genealogies. Ma felt that he was uncovering the underlying structure of Yi history with each encounter that brought him new material or led him to (re)interpret what he already possessed.
Walking through the official bloodline-based kinship
With new information in hand, Ma drew partial conclusions about Yi society and history. The backbone of his analysis was his division of the clans into those already classed as nuoho, quho, nzymo, and bimo (Ma et al. 2006: 218). He observed that only the clans of rulers and ritualists were mentioned in scroll-books. The rest – and especially those of the quho, who Ma felt often emulated the nuoho practice of preserving genealogies – were passed on orally (Ma et al. 2006: 203-4). Ma knew that Ling Bangzheng's lineage, the mightiest nzymo family of Liangshan for many decades, ran the yamen in Jydagu and inherited the offices of Lili in Xichang and Anning chang. Notably, Ma was also familiar with the gravestones in Niubashi (Ma et al. 2006: 347-50) but took the information they contained for granted. One of these gravestones named An Pupu,53 the first Pacification Commissioner of Luoluosi,54 who received the surname ‘An’ (‘Tranquil’) when appointed by the court in the late thirteenth-century Yuan Dynasty, as the earliest ancestor of the Lili. Speculatively, the Nuosu-Yi name of Lili nzymo and the ‘Luoluosi’ toponym could have been derived from the territory that Chinese sources reported as ruled by the Lianglin55 (Hu 1981: 78) and/or Luolan56 (Jiang 2013: 578-9) clans.
Considering the material in Ma's hands, the story spanning from An Pupu to Ling Bangzheng seemed coherent. When the Ming Dynasty overthrew the Yuan, Liangshan was a place of contention between the former and emerging elites. An Pei,57 another ancestor mentioned on the Niubashi gravestones, managed to defeat a Mongol general, Ürüg Temür, in competition for the title of native commander of Jianchang58 garrison (Taizu, Yao & Shen 1991 [1586]: 3.19-20). His office probably stood close to the garrison, as suggested by the location of its ruins in today's Hedong59 (Fig. 5), across a seasonally active river outside the walled old town of Xichang. After An Pei, records concerning the An tusi (aka Lili nzymo) became extremely scarce between the fourteenth and eighteenth centuries as the court repeatedly failed to secure the Jianchang Valley under direct imperial rule due to local resistance. In one rare instance, the thin records reveal an inter-clan struggle between the An/Lili and other local clans possibly connected to the Wumeng area, which in 1609 sought to usurp their position in Jianchang (Gu 1982 [1630]: 458.3).
After the Qing Dynasty overthrew the Ming, more records started to appear as the empire sought to integrate native domains by force into its ‘interior’ (Ch. neidi 内地). Again, in Liangshan, the process stalled, and the court had no option other than to appoint a new, smaller, and more disparate group of native chieftains, one of which was based in Hedong.60 The Lili and Synzy clans with tusi-nzymo offices along the Jianchang Valley's trading routes had a long history of virilocal and uxorilocal intermarriages, and also married with different clans such as the Shama (Zheng & Yang 1996 [1942]: 8.18-19). Between the eighteenth and early nineteenth centuries, relatively small stretches of the flat land surrounding the Jianchang trade route were incorporated into the interior (Jiang 2013: 611; ZDYLDG 2000: 825-66), with the empire supporting the construction of a new office in Anning chang that lay closer to a volatile unruly area. The prominence of the Synzy clan rose through the deeds of Ling Cheng'en, but the Lili clan was in sharp decline as its last legitimate male heir, An Shaohui,61 died young in 1862. Unable to pacify the region, the widow of Shaohui married Ling Hanping (Ma et al. 2006: 83), weakening the Lili clan and their offices, which were absorbed by the powerful Synzy clan. Ling Bangzheng was its last appointed tusi-nzymo. Eventually, a second yamen was built in Anning chang to house the expanding family (Fig. 6). Nothing was left of the Lili but the gravestones at Niubashi.
Ideologized histories, genealogies, and their nzymo
Never entirely retreating from field to office, Ma Changshou viewed fieldwork as an opportunity to correct or replenish his knowledge. He sought to work on the history of the ethnic groups he had surveyed throughout his career, and especially the Nuosu-Yi, after leaving Liangshan in 1940. Once the ruling CPC had implemented the Democratic Reforms in 1957, which toppled the power of the nzymo and nuoho clans, Ma got the chance to return to Liangshan and oversee a survey team that headed to Limu Moggu and the site introduced to him by Bitu around today's Jiukou Township.62 Here, Ma's team collected oral accounts, the most widespread of which described local clans toiling for the mighty Lili nzymo. Part of this account included the claim that the original office of the Lili was located in Jiukou until the Hma and Alu63 clans chased away a nzymo called Lili Savie64 (SSBZ 2009: 134–5). Invested in the ideological vision of a linear history shaped by peasant uprisings against landlord oppressors – which included the diffusionist view that the tusi-nzymo had been pushed away from Limu Moggu – Ma filled the more than 400-year gap in the patchy Liangshan historical records with oral accounts from the Hma and Alu clans. He located An Pupu and his yamen in Jiukou and, making his best guess, assigned them historically to the early years of the Yuan Dynasty before suggesting that the Lili clan was chased to Hxuoggur ladda in Zhaojue over the next century, and then again chased to Hedong outside of Jianchang (Ma & Li 1987: 106-12, 122-5). With this move, Ma merged the mythogeography of his own research trajectory with the genealogy marked on the gravestones of Niubashi and gave birth to his still authoritative history in which the Lili dominated Liangshan.
Ma's account assumed physical form through two steles erected in the early 2000s in Hxuoggur ladda, which claim that the four sandstone pillars behind them are remnants of the office of the Lili nzymo before his clan retreated to the Jianchang Valley (Fig. 7). The dating and ethnicity of the actors in the alleged nzymo retreat were questioned numerous times in the early 1980s. The ethnologist Li Shaoming pointed out that before reaching Zhaojue, d'Ollone managed to obtain a rubbing of a now-lost stele with alternative records calling An Pupu the early Ming Dynasty chieftain (Li 1980: 50-1, 55). Discrepancies are also found in the contemporary gazetteer, which dates the nzymo retreat not to the fourteenth century, as Ma did, but to the mid-sixteenth century (SSZXBW 1999: 5). An Pupu's name in fact appears only in sources of the late Ming and Qing Dynasties, or in Republican gazetteers, hundreds of years after his death. As in Ma's case, this information was likely copied from the gravestone inscriptions of the tusi-nzymo genealogy in Niubashi. While it is possible that the clan abandoned cremation for standardized burial in a coffin with a tombstone, under the pressure of the empire's etiquette and rituals (see Mueggler 2014), An Pupu and the other Lili nzymo clan members mentioned in the Chinese sources are altogether absent from the Nuosu-Yi genealogies (see Qomo 2007: 4696-710).
None of the bureaucrats, surveyors, explorers, or anthropologists (Ma included) who visited Hxuoggur ladda noticed the alleged local Lili's yamen. A British missionary doctor residing in the Zhaojue area in the 1940s pointed only to a nearby walled hamlet which he called ‘Small Zhaojue’ or ‘Old Fortress’ (Broomhall 1953: 63, 86). Similarly, Zeng Zhaolun (2012 [1947]: 183-5), a Han-Chinese surveyor passing through Zhaojue at the time, reported that the Asho clan were the local rulers and resided in a house with a watchtower inside a walled village. Fast-forward to the 1980s and a member of the Bbaqi clan recounted the story of a Syge65 tusi who was driven out of his office in Hxuoggur ladda and replaced by the Asho clan (Baqie & Baqie 1995: 36-9). The surname Lili is also missing in the original account about Syge and was perhaps attributed posthumously, if somewhat paradoxically, by Li Shaoming, who edited Ma's text more than a decade after Ma's passing (see Ma & Li 1987). An ethnographic report collected in July 2013 that builds on Ma's work additionally claims, through the narration of an Asho nzymo clan member, that the whole clan arrived in the empty Hxuoggur ladda from Wumeng in the early eighteenth century (Hein & Zhao 2016: 279). Another angle to this story that I managed to collect narrates the history of the Ali66 clan nzymo, who was pushed from Hxuoggur ladda eastwards to today's Nanwa Township,67 where their yamen reportedly survived until the mid-twentieth century (Yang 1992: 134-5).
The legends surrounding Lili nzymo that Ma collected in Jiukou are likely drawn from the collective memory of the locals whose ancestors periodically chased away the imperial court as it tried to penetrate Liangshan and establish a shortcut between the trade routes (Wen 2020: 126-31) connecting Jianchang with resource-rich Leibo (Fang 1945: 14). Evidence for this may be found in the empire's expansion into Hxuoggur ladda, which, unlike Ma's westward trajectory, was likely east-bound and originated from the Jianchang Valley. Both Jianchang garrison and the yamen of Lili nzymo that stood outside its gates served as models for the newly built Zhaojue garrison. Alternatively, local legends may have equated the empire-supported Lili nzymo clan with the Ali, and later the Asho, clans, who intermarried with Jianchang's Lili (Baqie & Baqie 1995: 57-61) and became the nzymo of the yamen for Zhaojue garrison. Another possibility is that this building was claimed by a different clan when the Lili abandoned its yamen following the destruction of the Zhaojue garrison. Whatever the case may be, the central yamen of the Lili clan probably never left the Jianchang Valley and the story of its origin in remote Jiukou is even less likely.
Ma's neglect of the complex story about how the Lili came to Zhaojue – and his choice to weave only the Lili's retreat from Zhaojue to Xichang into the vernacular account of the retreat from Jiukou – shaped the received Nuosu-Yi history in a powerful way. An ideologically invested Ma appears to have downplayed the repeated destruction of Zhaojue's archive, the lack of any authoritative narrative, and the tendency for Nuosu-Yi oral narratives to hang on the threads of genealogies that are envisioned performatively, cyclically, and multidirectionally rather than in a linear, historical manner.
Already in the first half of the nineteenth century, Liu Wenwei,68 a career bureaucrat in Leibo, remarked that the Yi tusi intermarried only with other tusi (1968 [1831]: 14). Liu's finding suggests that the nzymo stratum formed gradually atop the aristocratic nuoho, even as the Nuosu-Yi cosmology absorbed this element of the empire's native chieftain regime. The possibility of nzymo marrying somebody with softer bones is still seen as unlikely in places that are ethnically relatively homogeneous, such as highland Hxuoggur ladda. But in the lowland Jianchang Valley, the bone hardness of local nzymo clans remains shrouded in suspicion (see Li 1980: 52-4; Schoenhals 2003: 26–7; She 1980: 45). The genealogy of the nuoho Bbaqi (Baqie & Baqie 1995: 57-61) includes clans that came under suspicion of having Han-Chinese origins due to Ma's 1957 survey (SSMSLD 1963: 49). A similarly complex situation arose at Tianba, where Ma noted that one stele in Nuosu-Yi script proclaimed the court-appointed tusi to be hierarchically above the nuoho, while another stele inscribed in Chinese refrained from mentioning this hierarchy and pointed only to the modus operandi of the chieftains (Ma et al. 2006: 350-3). This juxtaposition of texts, where one states something different to the other yet both act as translations for each other, is often considered a sign of the incommensurability between the Han and Yi cultures (Mueggler 2021; Wen 2020: 183-90). But this is not necessarily true in Liangshan, where multidirectional narratives may give rise to opportunities for contesting the bone hardness of anyone, including the mythological ruling lineages that arrived in Liangshan from Guizhou and Wumeng (see Bender et al. 2019: 76-85). Put differently, just as ‘the genealogies of winners tend to survive (and to be improved), those of losers tend to vanish (or to be recast)’ (Ong 2002 [1982]: 65). Or, as the popular Chinese saying goes, ‘the winner is king and the loser is foe’ (Ch. shengzhe wei wang baizhe wei kou 勝者為王敗者為寇).
Conclusion
As with any mythogeographical knowledge, the anthropo-history of the Nuosu-Yi remains open-ended even though their genealogies have been written over many times. The epistemology of Ma Changshou is similarly open-ended, despite the many lines and footsteps that he traced across the ruins of three different political regimes: imperial, republican, and the period of high socialism in Communist China. Following the trajectory of his own body through space, Ma crafted a historical, linear, and unidirectional brand of anthropological knowledge. Had he journeyed in the opposite direction during his initial fieldwork or travelled in further directions during the decades he surveyed Liangshan, the history of the Lili native chieftain may have turned out differently. His admirable affinity for fieldwork contributed not only to the construction of a pan-Yi and Nuosu-Yi identity but also, perhaps unwittingly, to uncovering how the Nuosu-Yi breathe life into their multidirectional genealogies by performing them orally. Ma's career reveals two intertwining ways of seeing the past. Whereas he followed the blueprint of a bureaucratic state that sought to be anchored in historical time, the Nuosu-Yi – who often oscillated between state control and anarchy – take a genealogical and multilinear approach that, being cyclical, is not actually dependent on time. Retracing Ma's footsteps through the nzymo offices where these lines intersect shows the ‘subjectivities and residual affects that linger, like a hangover, in the aftermath of war or violence’ (Navaro-Yashin 2009: 5).
Crucially, the mythogeographical approach of walking or travelling through physical landscapes and texts, tracing and retracing routes, discovering dead ends, following desired paths, and talking with people to reconfigure the detritus left by anthropological predecessors (Smith 2015: 167) is one of the only means we have for illuminating our shared anthropo-history that is both written and written over. The intensity, repetitiveness, and multidirectionality of this wandering is formative to anthropological research. Ma's endeavour to capture the historical soul of Liangshan by bringing it into dialogue with his own mythogeographical journeying has shaped more than the Chinese ethnology, ethnohistory, and anthropology of the wider Yi nationality; it has encouraged the absorption of Western concepts by scholars in China and by the Nuosu-Yi. Once Ma and his colleagues had laid the foundations for the anthropo-history of the Nuosu-Yi, it was only a matter of time before anthropologists, Yi scholars, and the Nuosu-Yi themselves would harness his work and send it travelling through unreflexive channels and the ‘hyper-reflexive feedback loops’ (Swancutt & Mazard 2016) that enliven international anthropology. Seen in this light, Ma has become a major anthropological figure who holds his own next to his contemporaries, including Fei Xiaotong and Bronisław Malinowski (see Harrell 2015).
Mythogeographies of anthropology stretch far beyond the previously unimagined routes and multiple meanings that emanate from places designated as heritage sites or forgotten for other reasons (Smith 2011: 266). What makes these mythogeographies powerful is their penchant for connecting, intersecting, and leapfrogging across anthropo-historical time and imagination. Like the waterway-reinforcing detritus of the Lili clan's inscribed tombstones, these mythogeographies have found their way – through Ma Changshou and his team – into our shared pool of anthropological knowledge.
Acknowledgements
Parts of this article were presented at the 10th International Convention of Asia Scholars in Chiang Mai, Thailand (2017), and the 6th Conference of the Asian Borderlands Research Network – Borderland Spaces: Ruins, Revival(s) and Resources in Bishkek, Kyrgyzstan (2018), with support from the Hong Kong Ph.D. Fellowship Scheme (no. UGC/GEN/456/08, UGC/GEN/456/5/09, PF15-16536). I am indebted to Molo Yygumo ꃀꇈꒉꇴꂾ (Ch. Mao Xiaoying 毛曉英) and Jjivo Yyzu ꐚꃮꒉꊥ (Ch. Jiwu Yizuo 吉伍依作) for their help with reconstructing the Nuosu-Yi personal and place names hidden behind the Chinese characters; Liu Xinyi (劉昕怡) and Jakub Hrubý for the consultation of imperial and republican Chinese language sources; Liu Xinyi and Ivana Frolíková for visual material; and David Kurt Herold, Katherine Swancutt, and three anonymous reviewers for their excellent suggestions and guidance. Research for this article was supported in part by the Lumina Quaeruntur Fellowship for Prospective Researchers of the Czech Academy of Sciences (LQ300211901). This article is an outcome of the COSMOVIS project, which has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (Grant Agreement No. 856543). Ethical approval for the research has been granted by the Research Committee of the Hong Kong Polytechnic University, the Czech Academy of Sciences, the College Research Ethics Committee at King's College London, and the ERC.
NOTES
Biography
Jan Karlach is a research associate of the ERC synergy grant ‘Cosmological Visionaries’, which focuses on animism and climate change. Together with the Nuosu-Yi of Southwest China, he researches their cosmology, the construction of their historical as well as their contemporary identity, and their negotiations of Chinese state-driven modernity.
