The Art of Neighboring beyond the Nation: Ethnic and Religious Pluralism in Southwest China

Abstract

Northwest Yunnan is nested in the border areas of Tibet, Myanmar, and Southwest China. The religiously and ethnically diverse region has astonishingly seen a lack of “conflict”, as is often assumed in regions of ethnic and religious differences. This paper argues that there is an organic form of pluralism through frequent inter-ethnic and inter-religious marriages, multi-lingual daily interactions, and strategic ethnicity registrations. Ethnic and religious boundaries are made permanently or temporarily permeable through the celebration of boundary-crossing rituals such as weddings and funerals and other shared experiences such as collective labor and migrant work. Despite an increasingly strong push to be integrated into the state power through various top-down developmental projects, minority peoples here still use kinship, collective rituals, and other shared experiences to foster group formation that is fluid, porous, and malleable, instilling empathy and obligation as the basis of this pluralistic borderland society. This organic form of pluralism presents an alternative to the nation as the standard modern form of community. This paper ultimately argues that this specific type of plurality requires us to think beyond the normative liberal notions of religious tolerance and diversity that are still promoted within the frame of the exclusivist nation-state.

1. Introduction

On Tuesday, 28 January 2013, two weeks before the Chinese New Year, the market day of a remote mountainous town Bingzhongluo in Southwest China, horrific news dimmed the holiday cheer: a villager was killed. Adju, a 17-year-old Protestant Lisu girl, just came to meet me at the market. She urged me to go back to her village: “My mother’s brother’s son has killed a man! All the relatives in my family have gone [to deal with it]. My grandmother is crying, and my parents are helping with the funeral”. Running back to her village, we heard the siren and saw a police car speeding away with Adju’s cousin in it. When we reached the dead man’s house, the funeral preparation had already begun. Four monks were chanting sutras in a backroom in which a small altar had been set up. A large crowd of people gathered around the central fireplace (kitchen/living room), including the nipa (Lisu word for healer and ritual specialist). A female neighbor was making xiala (a large pot of home-brew liquor cooked with diced chicken meat). Some villagers were sitting by the porch and the courtyard, drinking beer and liquor made from corn, waiting for their turns to do the chores. The Protestant villagers sat at the furthest end of the courtyard. They were offered soda since alcohol and tobacco are forbidden for the Protestants. Four men were busy making a coffin on the empty lot behind the house. One of them was the Protestant preacher. Three pigs were killed, and a few men were busy dissecting the pigs and cooking meat mixed with blood in a gigantic wok outside the house. The livers, which were considered the best part of the pig, were cooked separately for the monks. A few women were cooking a large quantity of rice in the neighbor’s yard. The food for the Protestants was cooked (and eaten) separately because of their dietary restrictions, which ban the consumption of blood and alcohol.

Two days earlier, Adju’s cousin Ade had come back down to the village for the Chinese New Year after weeks of tending cattle in the mountains. He invited his good friend Gida and another friend over to drink alcohol—a most commonplace social gathering in the village. They drank the whole evening until deep into the night. All of them got utterly drunk. Seized by excitement, Ade picked up a piece of firewood and waved it at Gida, hitting him right in the head. Gida fell immediately, bleeding profusely. Shocked at the sight, Ade suddenly sobered up and realized what he had done. He managed to bring Gida to the hospital, but Gida died very soon afterward. The next day, Ade called the police and turned himself in. To my surprise, there were no disputes or even quarrels around this event. Nobody interpreted it as a religious conflict between a Christian (Ade) and Buddhist (Gida) family or an ethnic conflict between a Lisu (Ade) and Tibetan/Nu (Gida). Nobody except the anthropologist paid attention to the ethnic or religious backgrounds of the two people involved. Most people seemed to have gotten used to this form of alcohol-induced violence due to the prevalence of alcohol use in this region (Lau 2023). People’s central concern, instead, was that each household in the village was supposed to send at least one person to help the family in need, regardless of one’s ethnic or religious backgrounds or whether there had been any unpaid debt or past grudges.

This essay asks the following question: how do people of different ethnic and religious affiliations co-exist in close interaction with one another? And what alternatives does this place present for the forms of governance offered by the modern nation-state? While most of the world’s conflicts today are attributed to such salient differences as ethnicity and religion (Tambiah 1989), Bingzhongluo seems an anomaly. With a total population of 6461,1 among whom there are over seven ethnic groups (Nu, Tibetan, Lisu, Dulong, Bai, Han, and Hui2) and adherents to four “types” of religions (Buddhism, Protestantism, Catholicism, and indigenous animism), why are there minimal conflicts in this region? Following Simmel, Coser (1956) argues that conflicts serve to solidify groups and maintain boundaries. This study examines the other side of the coin, which is how groups mediate boundaries, thus forestalling conflicts. In other words, this essay explores the mechanisms people employ to maintain peace and inter-group cooperation in such a pluralistic society, thus offering a possible solution to the violence and conflict caused by the rigidity and monolithic ways of modern state formation.

2. Pluralism and Boundaries in Zomia

“Ethnic pluralism” is sometimes seen as a product of postcolonial societies in need of solving the problems of integration and citizenship of different social groups that were formerly called “tribes” (Jenkins 2008, pp. 17–42). “Multiculturalism” (Buenker and Ratner 1992; McCarthy 2009; Winter 2011) was later coined (and criticized) by various states as a political institution to solve the problem of pluralism. These approaches highlight cultural differences among different ethnicities. My approach instead continues the Barthian emphasis on everyday transactions and negotiations of boundaries instead of the cultural content of ethnicity (Barth 1969).

Research on religious pluralism has experienced several waves. The first wave treats pluralism as a way toward the secularization and privatization of religion (Berger 1967; Bellah 1967). The second wave depicts a religious market in which greater competition and less state regulation lead to greater religious participation (Stark and Finke 2000). The third wave shifts the focus to the civic implications of religious pluralism, especially in relationship to democracy (Shweder et al. 2002; Banchoff 2007; Eck 2001). While Wuthnow (2004) famously distinguishes the descriptive term “diversity” and normative term “pluralism”, Beckford (2003, pp. 74–81) adds one more philosophical/ideational dimension to it, claiming that pluralism can be a value system as well as a phenomenon and a set of institutional arrangements (Beckford 2010, p. 218). Bedford ultimately points to the “treacherous” nature of the term religious pluralism (Beckford 2010, p. 217). Leading two national pluralism projects in the U.S. and Netherlands, respectively, Eck (2001) and Lars et al. (2012) differentiate diversity and pluralism, with the former being the descriptive fact and the latter emphasizing the relations, interactions, and engagements of diverse groups (emphasis mine). Almost all the above-mentioned research on religious and ethnic pluralism takes place in western modern nation-states, where the privatization of religion is closely linked to the making of modern national citizens. This article takes the diversity–pluralism differentiation by Wuthnow and Eck and Lars et al. as the starting point and uses ethnographic data of everyday transactions in the borderland villages in Southwest China to explore how the diverse groups co-exist peacefully and interact on a daily basis. While the modern nation-state treats diversity as “counting bodies” of each religious organization with clear boundaries, I examine the social mechanisms that enable ethnic and religious pluralism in these villages, that is, peaceful interactions among diverse ethnic and religious groups that transcend boundaries.

The Chinese state is no different from the modern Japanese state, as described by Yijiang Zhong in the same special issue, in that it instates clearly defined religious groups as constituents of the state. As the basic state policy, the Chinese government recognizes 56 official ethnic groups and five official religions—Daoism, Buddhism, Islam, Catholicism, and Christianity (Protestantism). Ethnicity and religion are two primary markers based on which the Chinese government makes policies toward minority areas such as Bingzhongluo. Most academic studies have been conducted along those ethnic and religious lines with focuses on one ethnic or one religious group. However, as Jean Michaud notes in his description of Massif, “In cultural terms, these highland societies are like a cultural mosaic with contrasting colors, rather than an integrated picture in harmonized shades” (Michaud 2010, p. 206). This cultural “mosaic” calls for more studies that examine the plurality of such regions, named “Zomia” by the geographer Willem van Schendel (2002) and later made popular by the political scientist James Scott (2009).

Scott famously attributes agency to the people in Zomia by offering an account in which they actively evade the state, a strategy that he calls “the art of not being governed”. But it is mistaken to envision that Zomia is devoid of state governance. Scott limits his study of Zomia to the beginning of the twentieth century and worries that the stronger states around it would threaten its autonomy. Many studies have followed this line and see the Chinese state as encroaching into such areas through communication technologies, infrastructure building, and education and economic development projects (see, for instance, Harwood 2014; Wu 2019; Rippa 2020). My study was conducted at the beginning of the 21st century, one century later, after the missionaries had combed through the mountains and an increasingly strong Chinese state governed at a much closer range; still, the mechanism of pluralism persists. As Michaud points out, “contemporary connections may just as well refashion Zomia once again, rather than efface it” (Michaud 2010, p. 214).

My findings show that in this part of Zomia, ethnicity is largely held abeyant in people’s daily interactions, but religious adherence, checked by kinship, is key to their self-organization. Following Seligman and Weller’s groundbreaking work Rethinking Pluralism (Seligman and Weller 2012), I propose that ethnic and religious boundaries are made porous through alternative notation systems, collective rituals, and shared experience. However, the picture is not always rosy. Nation-states in general tend to construct new or solidify old boundaries. The media and academia, on the other hand, may also reinforce the state-imposed categories by becoming cultural activists or celebrating the “cultural revivals” (McCarthy 2009) staged by the governments, often for economic reasons. If Zomia is “a society that maintained itself by the active exchange of goods, people, and ideas without a unified territory” (Scott 2009, p. 48), I argue that the state-directed projects, though powerful, are part of such “goods, people and ideas”. They are authoritative and compelling but not conclusive or exhaustive. Through what I call the “mercurial” group formation that is fluid and dynamic, the Zomia society is capable of absorbing and transforming even the most powerful external forces.

Let me begin by introducing the ethnic and religious diversity in Bingzhongluo and then discuss how the boundaries between those seemingly separate entities are soft, porous, and sometimes irrelevant to the villagers. I conclude by highlighting the centrality of rituals and shared experiences for transcending boundaries and the possibility of an empathy-oriented way of dealing with differences and argue that such boundary-crossing and boundary-blurring mechanisms are the basis of an alternative organizing principle to the modern nation-state.

3. Ethnic and Religious Diversity in Bingzhongluo

The township of Bingzhongluo administratively belongs to Gongshan County of Nujiang Lisu Autonomous Prefecture in Yunnan Province. It consists of 32 multi-ethnic and multi-religious villages bordering Tibet, Myanmar, and India. The angry River Nu (Nujiang in Chinese or the Salween in Southeast Asia) roars through the heart of Bingzhongluo. Surrounded by two snow mountains, Gaoli Gongshan and Biluo Snow Mountain, Bingzhongluo is known for its inaccessibility, utter poverty, and ethnic tourism (Harwood 2014). Based on the research conducted over a span of five years (mainly between 2008 and 2013 with shorter follow-up visits), this article uses ethnographic data mostly from Bingzhongluo, supplemented with material from nearby villages in the Nujiang Prefecture.3

According to the 2011 census, 3275 people registered as Nu ethnicity in Bingzhongluo,4 making it the largest ethnic category in this township. The name of Bingzhongluo is Tibetan, meaning “a Tibetan village”. Its earlier Nu name was Changpu Tong, which means a village full of Changpu, or Calamus (sweet flag). Although the “original” inhabitants of Bingzhongluo are unknown, the Nu people claimed to be the oldest residents in comparison to other groups. Gongshan County is the only Dulong and Nu Autonomous County in China. Altogether, there are less than 30,000 Nu in China, making it one of the smallest ethnic groups. An equal number of Nu live within the Myanmar border. Around half the Nu people in Bingzhongluo are Buddhists. The others are Catholics who were converted by the Tibetan Catholics.

Bingzhongluo was under the rule of the Tibetan Tusi for several centuries.5 Historically, flows of Tibetans settled in the lower banks of the Nu River and intermarried with the local Nu or Dulong women. The 540 Tibetans living in Bingzhongluo today (2011 census) roughly fall into two groups. One group was composed of faithful families who followed the Buddhist monks to settle here in the 18th century to serve the monastery. The descendants are still devout Buddhists today and share close kinship ties. The other group came from Dimaluo, a Catholic community south of Bingzhongluo. They gradually settled in Bingzhongluo in recent decades in pursuit of land and business opportunities. Some of them are migrants due to the construction of a hydroelectric dam in their villages.

Most of the Lisu migrated to Bingzhongluo from Weixi County in Diqing Prefecture on the other side of the Biluo Snow Mountain in the late Qing and early Republican periods (the turn of the twentieth century) due to warfare. Some of them migrated here during the Land Reform years of the 1950s from Puladi, a steep town downstream, in pursuit of flat lands. By 2011, there were 2015 Lisu living in Bingzhongluo.6 Although the Lisu have lived here for less than a century, they are the second-largest ethnic group. The Lisu also has the highest percentage of Protestants. One local pastor estimates that around 30–40% of Lisu here are Protestants, though still much lower than the estimated 70–80% in the Lisu-majority villages further south.

The 413 (2011 census) Dulong people living in Bingzhongluo came from Dulong Jiang, which is a township in Gongshan County. It is named after the Dulong River, which is the origin of the Irrawaddy in Myanmar. Even though Dulong Township is blocked from the outside world for half of the year by heavy snow, there are many interactions and intermarriages of the Dulong with other ethnic groups in the rest of Gongshan and even upstream of the Tibetan villages and downstream of the Lisu villages. Some of the Dulong have converted to Protestantism. According to the Lisu, the banks of the Dulong River are dotted with nameless tombs that belong to the numerous Lisu traders who entered Dulong Jiang to spread the gospel.

Besides the four major groups, Nu, Tibetan, Lisu, and Dulong, the small number of Han (152) and Bai (11) are mostly businesspeople working in the market town, running restaurants, hostels, supermarkets, vegetable stands, and other shops. A few Han men have married local women in recent years and built houses in the villages. But they are segregated from the rest of the villagers. Only two Hui households moved to Bingzhongluo in the past year. Both operate restaurants in the market town. Pigs are very important to the livelihood of villagers, so the Muslim Hui rarely enter villages.7 Like the Han and Bai, therefore, the Hui do not participate in the village life. This article thus focuses on the four dominant groups, Nu, Lisu, Tibetan, and Dulong.

Religion is regarded as a very important component in the ethnic identification project of the Chinese government since it correlates with ethnicity. But, as I show below, this is not the case here. Three major institutionalized religions are found in Bingzhongluo: Tibetan Buddhism, Catholicism, and Protestantism.8 Even though the latter two are Christian, both the government and local people treat them as separate religions. Therefore, I follow their usage in this essay. There are a few healers/shamans who are called nipa by the Lisu and Namusa by the Nu, but since most of them are no longer active, their influence has diminished a great deal.

Tibetan monks came to missionize Nujiang in the 18th century. It is said that a very famous living Buddha from Dege, Xikang District (the present-day Sichuan Province), named Du Gongjian, came to spread Tibetan Buddhism, but he failed among the Lisu-majority regions of Nu River. So, he came upstream to Gongshan and established a temple in 1766 in an empty lot of flat land, surrounded by what he called 10 sacred peaks. Soon devout Tibetan families settled around the temple.9 He named the place Lama Zhong, which means the village of monks. As oral history has it, a very powerful living Buddha, who played with water in his hands as if it had shapes, won the faith of the local Nu people. By the end of 2011, the official record of Bingzhongluo reported 2600 Buddhists. The liturgical language of the Buddhist temple is the Deqin Tibetan dialect, which is different from the locally spoken Tibetan. The living Buddha and the Sutra teacher came from Deqin County of the Diqing Tibetan Prefecture. The rest are all from villages in or around Bingzhongluo. Among the 42 registered monks of the temple today, around one-third are Tibetan by official registration. The rest are registered as Nu. To recite the sutras, however, one must learn spoken Tibetan from Deqin and written Tibetan. To be ordained, one must study in the monastery in Deqin for some years and become well versed in scripture and the Deqin Tibetan dialect. Each year, some Buddhist families in Bingzhongluo, usually of Nu and Tibetan origins, send their children (both girls and boys) to the temple to study written Tibetan from the Sutra teacher.

The Catholic missionaries from the M.E.P. (Les Missions étrangères de Paris) and the Order of the Grand St. Bernard of Switzerland came to evangelize Tibet beginning in 1846 (Lim 2013, p. 112) and established the Tibetan diocese in 1854 by Charles Alexis Renou from M.E.P.. The French missionary Jean Charles Fage came to Bingzhongluo in 1854 and started teaching Catholicism to the Nu and Tibetan slaves of the rich Tibetans (Gros 2011) and converted them. Fage’s Chinese last name is Xiao, and still today, there is a famous clan among the Catholic Tibetans with the last name of Xiao. A typical feature of Catholics in China (Madsen 1998; Wu 2001) is that it often transmits within the lineages. Besides the Xiao clan, the other two Catholic clans are the Hu (tiger) and Xiong (bear) clans. Many Catholics have intertwined kinship relations despite their ethnic backgrounds. But, the liturgical language of the Catholic Church is Tibetan, supplemented by mandarin Chinese. Even though the Bible is translated into Tibetan, Chinese Bibles are used because most Catholics do not know written Tibetan, which is a skill reserved for the Buddhists. By the end of 2011, Bingzhongluo reported 889 Catholics, mostly of Tibetan or Nu or mixed ethnicity. There are only a few Dulong or Lisu Catholics, often by marriage.

The Britain-based China Inland Mission10 and U.S.-based Assembly of God11 first converted the Lisu to Protestantism in the early 1900s. J. Russell Morse, originally affiliated with the United Christian Missionary Society (UCMS) and later with Churches of Christ (Foster et al. 2004, p. 36), was the leading figure in spreading the mission among the Lisu after he failed to evangelize among the Tibetans. The Morse family later became the most well-known missionary among the Lisu of Southwest China and Northwest Thailand. The China Inland Missionary James O. Fraser created the Lisu orthography in collaboration with Karen assistants who spoke Lisu.12 The Fraser Bible is the most used Bible among the Lisu throughout the world. The liturgical language of Protestant churches is Lisu. Even the Nu and Tibetan Protestants deliver sermons and sing hymns in Lisu when they attend the services. By 2011, the official record showed 806 Protestants in Bingzhongluo. The actual Protestant body is much bigger if unbaptized believers and children are included.13 However, as I shall show later, official membership means little in this context.

The indigenous religious practices were almost gone when this study commenced in 2008. I could find only one such specialist in Bingzhongluo. A descendant of the early Tibetan Buddhist families closely connected with the monastery, Grandpa Tah used to be a monk but started seeing spirits and acquiring healing powers in his late teens. He became a healer soon after the temple was forced to shut down and the monks were expelled during the 1960s. Now he also goes to the temple during festivals and often works side by side with the monks at funerals (as shown in the example at the beginning of this article). The relationship between indigenous religions and the organized religions is beyond the scope of this article, but Buddhism is quite tolerant of indigenous practices. After state oppression of all religions between the 1950s and 1970s in China, the institutionalized religions have been revitalized since the early 1990s. Partly due to the state’s negative attitudes toward “superstitions” and partly due to the organizing capacities of institutionalized religions, practitioners of indigenous religions have diminished significantly today. For the purpose of this article, I focus mainly on the organized religions.

This picture of religious diversity is of recent formation. The migration and conversion history is only slightly over a hundred years. During these years, conflicts do exist, and personal grudges are commonplace. Nevertheless, they are rarely framed in ethnic or religious terms. Historical textbooks describe this region as conflict ridden. The Lisu rebellions against the Qing government were frequently narrated in official accounts of how the Lisu were fighting against feudal oppression before the Communist Revolution.14 The Tibetans and Tusi were often referred to as the evildoers. They charged heavy taxes and kidnapped slaves and women from the Nu and Dulong villagers. In the Chinese official accounts, the famous tattoo faced Dulong women (wenmian nv) are said to have tattooed their faces in order to escape being captured by the Tibetan Tusi. Interestingly, alternative explanations abound. Some villagers hold the view that the tattoo-faced women used to be the nobility in the Dulong society: “Those who have tattooed faces got to keep slaves and they didn’t have to work in the fields”. While the official account emphasizes inter-ethnic and class struggle, the local view emphasizes intra-group differences.

Regarding inter-religious relations, Baihanluo Jiao’an15, the famous battle between the Catholic Church and the Tibetan Buddhist temple in 1905, never ceases to capture the fascination of academics and officials alike. The church established by the M.E.P. missionary Annet Genestier in Baihanluo village was burned down by the monks and other adherents of the Tibetan Buddhist temple. Genestier, whose grave still stands in the churchyard of today’s Bingzhongluo, appealed to the Qing government and successfully obtained a large sum of compensation from the temple. The Catholics regard this as their final triumph, but the Tibetan Buddhists regard their action as heroic behavior against the imperialists. Some monks still tell this story today with nostalgia of their glorious past. Gros (2011), however, reveals that the Catholic missionaries were acting like merchants. They were initially well received by the Buddhists, and it was their interception of the original rental and loan systems between the tenants and debtors (the temple and Tibetan, Naxi, and Han official/merchants) that caused these conflicts. The fact that later scholars and government officials tend to see such events as religious conflicts reflects more about their own ideological tendencies than about the event itself.

Therefore, despite the fact that these historical accounts depict a much more turbulent picture than this paper presents, it does not contradict my thesis. First, most of the conflicts, upon closer examination, did not result from ethnic or religious differences but from the power relations between the officials and the local people or clashes of economic interests. Second, these events demonstrate that violent conflicts are possible if certain boundaries are consolidated to form interest groups. Thirdly, what this article describes constitutes only a slice of history, and that slice is unstable since it springs out of complex local interactions, migration history, and external forces (such as missionaries and state policies toward different groups) and, therefore, is subject to constant changes. Echoing Leach’s classic study (Leach 1954), this study shows not a state of equilibrium but rather a process that might recur and reshuffle constantly. Instead of working with the image of a mosaic or melting pot (Bibby 1990; Wong 2008), or a “bricolage” (Hervieu-Léger 2005) or “patchwork quilt” (Wuthnow 1998) model (Banchoff 2007, p. 6), I prefer a mercurial model—one that is decentralized, collaborative, and capable of branching and merging. While the melting pot implicates assimilation and being absorbed into something larger and more powerful, the mercurial model emphasizes decentralized and collaborative efforts. While the bricolage emphasizes individual choice and freedom, the mercurial model emphasizes group formation that is fluid and porous. The “patchwork quilt” metaphor works well in immigrant societies like the U.S., where the boundaries among different groups are more fixed, but the mercurial model I use here allows room for branching and merging, which will be further elaborated later in the text. It is sufficient to say here that when inspected at any historical moment, certain groups appear to be entities, but they are always entity-forming/transforming. The significance of highlighting such temporal systems lies in how it may serve as a window through which we learn how to live with differences and how to maintain peace when much of the world constantly falls prey to ethnic and religious conflicts.

4. Two Intertwined Notation Systems: Religion and Ethnicity

Ethnicity and religion have been two notation systems that government, media, and academia choose to understand the world today. As Seligman and Weller (2012) have argued, notation is both necessary and dangerous, because it helps us understand the world, but it also creates boundaries. Despite the effort of earlier scholars who espouse a fluid nature of ethnic boundaries (Barth 1969, 1984), ethnic and religious boundaries throughout the world are increasingly perceived as “brick” walls instead of “cell” walls (Seligman et al. 2008, pp. 93–97) through pure notation systems. In China, the state-initiated Ethnic Classification Project and state-sanctioned five official religions (Buddhism, Daoism, Islam, Catholicism, and Protestantism) further solidify the boundaries among ethnic and religious groups.

The Ethnic Classification Project (minzu shibie) in the 1950s, soon after the founding of the People’s Republic of China, was very well studied (see Tam and Wu 1988; Dikotter 1992; Harrell 1995, 2001; Mackerras 1995; Tapp 2002; Yang 2009; Mullaney 2011). However, this was not the first attempt to notate peoples of China. As early as the 3rd to 2nd century BCE, the Han Dynasty court historian Sima Qian already recorded the names of barbarian peoples living in the present-day Yunnan and Guizhou provinces (Yang 2009, p. 10). Every dynasty or government uses notation as a form of bio-politics. The Ethnic Classification Project by the PRC government, which recognizes a finite 56 ethnic categories out of over 400 proposed names, was the first large-scale, nation-wide project that invoked modern “scientific” and ethnological methods involving large numbers of scholars and intellectuals in the country. Although this was also the first time each of the recognized group was given equal status and representation in the People’s Congress, Nicholas Tapp calls this process an “internal ‘self- Orientalizing’ mission designed to homogenize and reify internal cultural differences” (Tapp 2002, p. 65). The system is justified by “science”, which makes all the “cultural walls” appear so objective and realistic that some of the created groups internalize the identities they are given (Kaup 2000). Of the 56 ethnic categories, 25 peoples were recognized among over 200 group names proposed in Yunnan Province alone (Mullaney 2011).

Once created, those categories become mental tools for policy makers, local government officials, and cultural activists/revivalists alike. Most scholars of Chinese minorities feel that the 56 ethic categories do not do justice to the reality, but by advocating more groups, they merely install more walls using the same measures by the state, such as language, customs, and shared history. So far, the ethnic groups in China remain at 56, and that is the picture with which most Chinese citizens learn to imagine China. Every citizen in Bingzhongluo has an ID card that prints his or her ethnicity. They are aware of the ethnic categories, but, as I show in the following section, they have viewed them as nominal and performative categories.

In the PRC, the Religious Affairs Bureau and the Ethnic Affairs Bureau have long been one administrative unit. While ethnic groups enjoy certain rights, especially representation in the local or central government, there are no specific rights reserved for religious groups. In Bingzhongluo, however, the three institutionalized religions as ways of organizing daily lives have been largely recognized by villagers. The religious life in the villages is structured around various religious holidays. The Buddhist holidays are four times a year, all according to the Chinese lunar calendar: the Chinese New Year, the Lantern Festival (15 January), the Fair Lady Festival (15 March), and the Mid-Autumn Festival (15 August). Those occasions normally involve large crowds gathering in the temple compound, getting blessings from the living Buddha, and watching the monks perform Mahakala dances and recite sutras. The Protestants worship five times a week: Friday and Saturday evenings and three times on Sundays. They celebrate Easter, Thanksgiving (any Sunday after the autumn harvest), and Christmas (3 days). The Catholics gather once a week on Sundays and have big celebrations on Easter and Christmas Day. Therefore, the public celebrations of each religious group mark the calendars of the villages. Moreover, all the life rituals (birth, wedding, and funeral) are celebrated with fellow villagers and relatives. These ceremonies carve out overlapping public spaces within and across the villages.

Since around 70% of villagers subscribe to one religious group or another, their public life beyond the household is often structured around religious events. The state selects ethnicity as a marker and bases policies such as education and job opportunities on this notation system. Villagers, however, highlight religion as their own notation. The rest of the paper describes how shared experience and collective rituals de-notate ethnic categories and transcend religious boundaries without sacrificing commitment.

5. Porous Ethnic Boundaries and Shared Experience

During the Ethnic Classification Project, the Chinese intellectuals and government took a very “primordial” view of ethnicity: things such as blood, language, customs (clothes, songs, and dances), and religions/rituals were considered intrinsic to group cohesiveness. In the 1990s, with the promotion of ethnic tourism, each group was granted standardized costumes, songs, dances, and even languages and holidays.16 Bingzhongluoers, especially local government officials and school children, are required to learn “The Eight Standard Ethnic Dances”,17 taught by the government-trained dancers. While tourists and researchers criticize the official standardizations as ridiculous, they, too, search for the “true” ethnic blood, “real” ethnic clothing, “original” songs and dances, and “authentic” ethnic cuisine and art, resulting in rigid boundaries and essentialized identities.

Villagers of Bingzhongluo are experts in the shared usage of cultural objects (Seligman and Weller 2012, p. 161). Almost all people in Bingzhongluo share similar housing and clothing styles, as well as the dietary habits of the Nu people, despite their proclaimed ethnicities. If one tries to find the textbook-version “Lisu style house” or the “Nu-style house”, one will fail miserably. There are only two kinds of houses: the “traditional” wooden houses covered with slates and newly built brick houses covered with asbestine roofs. The so-called “traditional Nu house” has very narrow windows and extremely low gates since “the Nu believes that the evil spirits might enter the house if the openings are wide”.18 In one village, the lowest gate is found in a Protestant household. The husband is half Dulong and half Lisu, and the wife is Lisu. If we dwell on meaning, then we are likely to reach the absurd conclusion that the Protestant Lisu/Dulong household embraces the “primitive religious belief” of the Nu people. The villagers know that what matters only lies in how long ago the houses were built. The houses built over 10 or 20 years ago may have narrow openings. The new ones abandon such practices since those are considered old fashioned and inconvenient.

All Bingzhongluoers drink tea in the morning, regardless of ethnicity, prepared the same way as the Tibetan yak butter tea Yak butter, however, is expensive and hard to obtain since only the pastoral Tibetans know how to make it. So, the villagers replace it with pressed oil from the seeds of lacquer trees. The Nu, Dulong, and Lisu, including the local Tibetans, all consume lacquer oil tea.19 Since it is considered especially nutritious, all village women must eat chicken soup cooked in this oil for a month after giving birth.20 The famous “pipa pork”, which is considered a typical Nu food in the official account of the Nu “customs”, is again consumed by all families. To prepare for it, the pig’s intestines and fat are taken out, then the pig is stuffed with herbs inside the stomach, mouth, and nostrils, stitched up, and then dried flat above the fireplace for weeks to months before being cooked and consumed. These dietary habits show that the villagers share the same “usage” of foodstuff, regardless of their ethnic categorizations.

Clothing is no different. Most villagers wear things that are very similar to the Han people nowadays. When introduced to one of the oldest living Tibetan women from the original families who came with the Tibetan Buddhist missionaries, I was dismayed to find that she was wearing Nu-style clothing. A Nu girl once went shopping for her holiday wear and came home with Tibetan-style clothes. Responding to my look of puzzlement, she said, “It’s pretty!” The Lisu clothing, on the other hand, has become the church wear for many people, regardless of their ethnicity. The same applies to festival dances that are organized by the government since everybody in the village is multi-lingual not only in terms of the languages they speak but also in terms of the songs and dances they perform. Whenever the government officials or the tourists require an “ethnic” style of reception, any villager can put on any ethnic clothing and perform any designated ethnicity. Whereas the tourists feel upset about the lost authenticity, the villagers feel very proud of their versatility. Again, the ethnic symbols, though distinguishable, are open to usage by any member of the society.

As Oakes (1998) and Notar (2006) note, it seems that only the tourists are distressed about authenticity. It does not bother the villagers, not just because they think of it as a source of income, but because authenticity is not a concern for people who have shared usages of the “cultural stuff”. When a tourist complains of the loss of the authentic, they are searching for a kind of “romanticized or demonized” (Lutz and Collins 1993) image, museumified in the “traditional” primordial forms. But, for the villagers, it is essential to keep the boundary open and porous. During the Chinese New Year celebration of 2013, for instance, the village children performed the Korean Pop dance “Oppa Gangnam Style” next to the grandmothers’ ethnic dance remix. No one feels uneasy or in-authentic.

Besides lifestyle, the villages are also flexible in choosing marriage partners. The majority of the marriages are mixed. One Lisu man, questioned about his wife’s ethnicity, replied, “She was Akha but now she is Lisu”. Unsatisfied, I probed further. He answered, “Since she is living with me now in all Lisu ways, so she is Lisu”. When it comes to the children’s marriage partners, villagers often say they want people who are “hard-working” or “respectful to the elderly”. The younger generation often says “love” or “getting along well”. Nobody mentions ethnicity. One six-year-old boy of Han and Lisu mix felt amused by all my questioning of people’s ethnicities on one occasion and said, “I belong to the Bingzhongluo ethnicity”. How to choose the ethnic registration for the children of such mixed marriages is a different story.

Let me introduce the story of one “typical” Bingzhongluo family’s choices. Teacher Yang is of Bai origin from a village downstream in the Nujiang Prefecture. Upon graduation from the Teacher’s Secondary School in the 1980s, he changed his ethnic registration to Han, which was considered an upgrade at that time. He was assigned to work in Bingzhongluo, where he married a Lisu woman. When their sons were born in the early 1990s, he registered them as Han. However, when their sons were going to college, he regretted his decision since ethnic minority students get extra points in the entrance examination. Furthermore, Nu and Dulong, being the smallest ethnic groups, have the most benefits, such as priority in civil servant recruitment. Therefore, Teacher Yang went to the security bureau to change their sons’ ethnic registration. The local government rejected their request, to his greatest dismay. “Why cannot they be Nu/Dulong?” Teacher Yang was rather indignant, “They were born in the Nu and Dulong Autonomous region! They think our son cannot speak Nu? It was because they were in school too much. If they stay in the village for a month, he will be fluent as the others”. The government officials are equally upset, “These minorities! They are so clueless, but they are equally cunning! They see being a Nu/Dulong has benefits and they all want to be, but at least prove that they have somebody within three generations who are of Dulong or Nu origin! They cannot! And they want to change their ethnic registration. If everybody is like that, it will be chaos!” Where the government officials and villagers differ is how they perceive ethnicity. For the government, it is blood and the intrinsic characteristics related to blood. But for the villagers, ethnicity is not an essentialized category but a performative one—if one can perform Nu, why cannot he be Nu? Therefore, the villagers are not dishonest. They think they can outsmart the system since it is based on something they can perform.

Therefore, ethnic conflicts are rare since ethnic boundaries are porous. Ethnicity rarely interferes with daily transactions since nobody has a monopoly over the usage of shared cultural symbols. However, it does not mean they are not recognized. Ethnic categories are evoked when any interactions with the state take place, for instance, going to school, getting a civil servant quota, getting subsidies, etc. Moreover, stereotypes do exist. The Han and Tibetans are considered arrogant and mercantile since both groups are resourceful and economically better-to-do. One wealthy villager who is also politically well connected is referred to by others as Han or Tibetan randomly. He himself even changes the narrative whenever he sees fit. The Lisu would say the Nu are slow; the Nu would say that the Lisu sell their daughters due to their demand of high bride price; the Dulong are said to be short and pretty. But, as I have demonstrated, people intermarry so much that when they stereotype one group, they are also laughing at themselves. Moreover, these stereotypes do not guide people’s actions. One Lisu woman who says the Nu is slow treats her Nu daughter-in-law and her family very well. She is considering having another Nu daughter-in-law. Teacher Yang’s son tells me that the Dulong have no plans and just go with the flow. One of his best friends is Dulong, and sometimes he is envious of his friend since the Dulong family does not put so much pressure on the future of the son, unlike his Bai father, who acts like a calculating Han person. Therefore, ethnic categories for the villagers in Bingzhongluo are not essentialized categories, inseparable from the person, but performative categories that are easily transcended.

6. Kinship Overriding Religious Boundaries

Unlike ethnic categories that are imposed on the people, religious adherence is a self-selected way people organize themselves in Bingzhongluo. Following Richard Madsen’s account of Chinese Catholics, Francis Lim uses the term “religio-ethnicity” to refer to religious communities that act as ethnic groups. Using Protestants as an example, Lim identifies five phases for the formation of the religio-ethnicity, in which language, place of residence, and kinship all play important roles (Lim 2013, p. 118). This way of seeing religious communities, instead of the state-designated ethnic categories, as the organizing principle for public life is useful since religious grouping seems more stubborn and enthusiastically embraced by the villagers.

Different from ethnicity, religion is a concern when it comes to choosing marriage partners: “It is better for one household to be of one religion, because it is more convenient”. That is probably because religions often prescribe dietary habits and lifestyles. Protestantism in this region forbids drinking, smoking, or eating blood or the meat of dead animals. The Buddhists, while having no dietary restrictions, make offerings to spirits before they eat or drink, which is considered non-acceptable by the Catholics, who equally observe no dietary restrictions. Such dietary differences cause inconveniences for a married couple if they subscribe to different religions. However, marriages between two people of different religious backgrounds do happen frequently—one party simply converts to the religion of the other. The rules are also flexible. One Nu Protestant man married into a Buddhist family in Bingzhongluo. Soon after the marriage, he converted to Buddhism. He recalled: “I used to be quite religious. During my army years, I would still say my prayers at night. But ever since I married my wife, I had to change because her entire family is Buddhist. They make offerings to deities during holidays and go to the temples for blessings. It is not convenient [for me to stick to my Protestant practices] anymore. Plus, we plan to open a restaurant/hostel for tourists. They often expect us to drink with them. I cannot make my wife entertain them alone. So, I had to drink. When you start drinking, you feel ashamed to attend the church”. The man does not exhibit guilt in converting to Buddhism, but he still thinks Protestantism is a good religion: “You can get eternal life and not drinking is better for one’s health”. Since there is no discrimination attached to people who drop out of certain religions, nobody ostracizes him from the community. Instead, people show respect for his sense of responsibility toward his family.

Another case also involves a man who converted from Buddhism to Catholicism by marrying into (shangmen21) his wife’s Catholic family. He says,

Because I married into my wife’s family, they requested that I convert to their religion. So, I did. I had to learn how to sing the hymns and make sense of what they say in the church on Sundays, but it was no big deal. I think Catholicism is good—the funerals are much less expensive [than the Buddhist ones] and they go to the tombs to remember the ancestors every year. In Buddhism, we must do three years of rituals after the funeral, killing pigs and inviting the monks, that sort of things. It is expensive but after three years, we stop going to the graves. But Buddhism is also a good religion. My parents are still Buddhists. Sometimes I still go home when there is a celebration. The atmosphere feels good, although I don’t worship idols any longer.

Since he is well educated, the man is an important person in the Catholic community, entrusted with the duty of doing the accounting for the church. This case shows that conversion does not incur hostility, and it does not downplay one’s sincerity or commitment to religion either.

In both cases, the men converted to the wives’ religions, but there are also cases when the wife converts to the husband’s religion. After her husband died of heavy drinking, the Lisu woman Simei married a Protestant man and joined the church. She says, “It’s good to be a Protestant. Drinking is very bad for you. My ex-husband drank so much that he threw up blood before he died. Now all my children are Protestants”. Sometimes, one chooses not to participate in anything. Ahni became Protestant in her teens through her brother, who converted and was elected to be the village Bible teacher. After 10 years of a failed marriage to a Han man in Northern China, Ahni came back to the village and began dating a Nu man, the son of an elder monk who was forced to marry during the Cultural Revolution. The son later also became a monk.22 Since her marriage to the monk husband, Ahni was too ashamed of the smell of alcohol on her body smell that she stopped going to the church. As Lu (2011) notes in his study of Lisu Protestants in Fugong County of Nujiang prefecture, the Protestants can easily distinguish their own kind by the lack of alcoholic smell that is often found on non-Protestants, who are usually heavy drinkers. Since Ahni lives with a family of heavily drinking Buddhists, she starts to feel that her body odor now is unfit for the church. Her brother comments on this: “What can I do? Even if I don’t like her to marry again, but she wants to! I can do nothing but let her be. It’s her life”. He continued, “Of course, it is inconvenient now for her to come to the church. She has her family to take care of. Everybody understands that”. So, his emphasis is not on how she may have sinned by leaving the church but on how staying in the church causes discord in her family. Since she also feels very troublesome to be a Buddhist due to the frequent and expensive rituals, Ahni shows a disinterest in anything religious now. And it is perfectly fine for people who choose not to belong to any religious community, even when their family members are devout.

The cases of people changing from one faith to another can go on. The flexibility demonstrates that even religious boundaries can be crossed with little animosity attached. The only exception is probably Catholicism. I personally do not know of any cases of Catholics leaving the church, although people tell me it also happens occasionally. The reason is that Catholicism is so intertwined with large lineages and kinship networks. By leaving the church, one is almost leaving behind all his family ties. What happens more often is whoever, man or woman, marries a Catholic, converts to Catholicism. I know of two Catholic Tibetan sisters. Each married a Lisu and a Bai man and both husbands converted to Catholicism before the wedding.

The above cases show that religion, being a prevalent marker of one’s group adherence in contrast to ethnicity, is also flexible without tarnishing the sincerity of the individuals. In the first case, the Lisu men who convert to Buddhism from Protestantism feel not only inconvenient but also embarrassed to go back to the church after participating in the family sacrificial rituals and drinking. In Ahni’s case, she could choose to remain Protestant, but both her preacher brother and she think that it is not convenient to join the church after the marriage to a drinking monk. For her brother, sticking to the family religion is important. For her, she feels “ashamed” of attending the church after having the wrong body odor. They are both actually “sincere” actors in the sense that they drop out of a religion when they cannot adhere to it on a practical level. This formation of religious boundaries is better illustrated with the mercurial model, which allows flexible conversion rules and portrays the fluidity in religious adherence. The next section analyzes how religious rituals act as mechanisms of transcending the geographic boundaries of villages and how secular collective rituals act as mechanisms of transcending religious boundaries.

7. Collective Rituals: Transcending Religious Boundaries

Rituals delineate boundaries and establish interpenetrations in which “otherness is possible” (Seligman and Weller 2012, p. 98). As I have shown, each religion has its calendar of celebrations around the year. One feature of those celebrations is that most of them go beyond village levels. Almost every village has Catholics and Buddhists, but there are only four Catholic churches and one Buddhist temple in a total of 32 villages. That means all those celebrative occasions involve large groups of people traveling between villages. Due to the bad road conditions and difficult accessibility, sometimes the journey must start a few days before the celebration starts. For Protestants, each church takes turns hosting Christmas for the entire township. Every year, around five days before Christmas, one can see villagers traveling in small groups, carrying food, clothing, and sometimes bedding on their backs. Those are worshippers from surrounding villages traveling to the same village church to celebrate Christmas. The celebration lasts for three days. Once they arrive, they sleep in their relatives’ houses or other fellow sisters’ and brothers’ houses. Often, their relatives are not Protestants. Thus, religious celebrations not only allow villagers to cut across territorial boundaries of village borders but also provide opportunities for inter-ethnic and inter-religious exchanges. Viewed on a macro level, periodic religious celebrations throughout the year, moving people from one village to another, create another system of interactions, with different centers, in addition to the marketing system as described so well by Skinner (1964). Parallel to Skinner’s hierarchical marketing system, trans-village religious interactions create a mercurial image of convergences and regroupings.

Other collective rituals, such as weddings and funerals, are also important components of village life. Such occasions further create bridges that cross over religious boundaries. The funeral introduced at the beginning of this article is a suitable example. Though the monks are invited to host the main service, the Protestants are fully integrated into the entire funeral process. They can participate in the making, carrying, and burying of the coffin. Though they are offered non-alcoholic drinks and their food is cooked separately, their labor is equally rewarded and welcomed. Preparing two kinds of beverages and foods has become the feature of most collective village rituals in the villages: Protestants and non-Protestants, respectively. There is no taboo against the participants of any religious affiliation. If the host family is Protestant, drinks and foods are prepared their way. And the rest of the villagers are happy to celebrate on their terms.

In a Protestant wedding I attended, half of the guests were relatives and fellow villagers who were non-Protestant. The other half were people of the same church group. The food in the banquet was prepared in the Protestant way (no alcohol, blood, or dead animal meats). The guests happily celebrated with fruit drinks and soda. During the ceremony, everybody was invited into the church. The guests did not feel the pressure to convert, but they were interested in what the church looked like, and some of them listened to the preaching with curiosity and fascination.

Rituals like this create a safety zone in which differences are temporarily lifted and boundaries (if any) made permeable. Buddhists can go into the church without feeling embarrassed. The Protestants can peek into the Buddhist sutra chanting without feeling any breaching of the rules. It enables an open border that temporarily allows the flowing in and out of individuals without causing discord. Indeed, there are a few cases of conversion resulting from such ritual attendance. Adhi was one of them: “I went to a relative’s wedding in a church thirty years ago. She had converted [to Protestantism]. I still remember seeing a pastor, nicely dressed, preaching to us about eternal life. I liked what he said. After coming back from the wedding, I started going to church myself”. Adhi later converted his wife to Protestantism and raised three children in a Protestant family. This case shows that such collective rituals carve out “cell walls” on existing religious boundaries. Each ritual occasion creates an alternative layer of contact and convergence. Each assemblage of people is constantly in flux, forming a dynamic mercurial fashion. Multiple cross-cutting ritual occurrences propel junctures of mergers and divergences.

Sometimes, even the state is inserted into such collective rituals, and villagers accommodate that element just the way they handle other differences. One such example was the Catholic funeral of a schoolteacher. Teacher Zhang served the village primary school all his life. When he passed away, his family planned a Catholic funeral. The Education Bureau of the county government wanted to honor his life-long service by giving him an official memorial ceremony. When I arrived on his funeral day, his coffin, open casket, was lying on the floor of the family room. His sons, wife, and sister were sitting around the coffin. Each guest was greeted at the door, offered alcohol (or tea to the Protestant guests), and invited to sit with the family. A monetary gift was given to the family. At around 11am, the officials arrived, bringing paper flower bouquets that were typical of Han Chinese funerals. The family then closed the coffin and moved it to the courtyard. The chief of the Education Bureau moderated the standard official funeral by reading out Teacher Zhang’s life history in the government record. Then, people were instructed to lower their heads and remain silent to the standardized official funeral music for three minutes. The service lasted for about 20 minutes before the officials quickly left. Then, a Catholic priest took over and covered the coffin with a black cloth and a piece of wooden cross. Praying the rosary, Catholics carried the coffin into the church, laying it in front of the altar. This was followed by an hour-and-a-half Mass. Afterward, just like in any other village funeral, friends and relatives carried the coffin to the pre-selected graveyard on a hilltop and buried it in the ground, using concrete to seal the grave and erect a tombstone. Then people drank together (and the Protestant relatives were offered sprite, instead of alcohol) and a big banquet was served.

In a sense, the state is one of the many “differences” the ritual overcomes. Not only is the boundary between the living and dead crossed and the boundary between Catholic and non-Catholic bridged, but also the gap between the state and people are temporarily mediated. The state is no longer an aloof or dominating actor. By joining certain ritual occasions, the state is taken as one trajectory of “otherness” that is successfully incorporated into the village life worlds.

Besides collective rituals that blur the boundaries among seemingly segregated religious communities, other ritualized village obligations serve as additional means through which villagers on different sides of the religious compounds are brought together. This is most evident in the organization of economic activities. Bingzhongluo is among the poorest parts of China. A local government official estimated that the annual personal income was around RMB 3000–4000 (roughly USD 420–570 at today’s rate) in 2014. Most villagers grow their livestock, grains, and vegetables. The source of cash income used to be purely ad hoc: matsutake mushrooms and herbs they collected from the mountains. In the past decade, however, more cash crops have been introduced through government-initiated agricultural projects. In 2019, some villages reported an average annual income of RMB 8305 (USD 1130).23 This is beyond the poverty line of RMB 7000 set by the government. Despite the introduction of the cash economy, many traditional forms of shared economic activities persist. If a new house is being built in the village, each family is obligated to supply a capable laborer to help—no payment required. The host family only supplies food, usually meat and alcohol (tea for Protestants). If one family withdraws from participating, others can cease their obligation toward this family when it is in need. Planting and harvesting are labor-intensive endeavors. The entire village works in the collective fields of the village, one patch after another, regardless of who owns which patch. Wild mushroom or herb collection is the major source of cash income. To do this, one stays in the damp and high-altitude mountain forest for at least two continuous weeks. It is essential to group up to fend off possible dangers and to share the workload. The collectors usually form groups of five or six relatives or neighbors.24 In the past, the tea-horse trade routes were full of teams of small groups of villagers who collaborated in such expeditions. Pursuing migrant work in the coastal cities or Tibet is becoming increasingly popular today. Contrary to one study that found that the Yi brokers in the Pearl River Delta use ethnicity as an instrument to take advantage of their co-ethnic migrant workers (Ma and Su 2022), young people in Bingzhongluo prefer traveling together in small friendship groups, sometimes called “sworn brotherhood” or “sworn sisterhood”, regardless of one’s religious or ethnic backgrounds. These small voluntary networks of people support one another financially, physically, and psychologically during the migrant work periods. Such shared experiences (Seligman and Weller 2012) of collective labor and group expeditions, like the collective rituals, foster another layer of the mercurial collage.

8. Conclusions

This essay explores mechanisms that enable ethnic and religious pluralism—“a systematic set of rules governing inter-ethnic social encounters” (Barth 1969, p. 16) and ways of engaging with differences without sacrificing “deep religious commitments or secular commitments” (Eck 2001, p. 71). Following Seligman and Weller (2012) and using ethnographic data of everyday transactions at the China–Myanmar–Tibet borderlands, I discuss how alternating-notation, shared experiences and collective rituals render ethnic and religious boundaries soft, malleable, and porous. I propose a mercurial model of understanding social organization in this society. This decentralized model, capable of branching and convergences, demonstrates multiple layers of associations that are made possible through different ritual and social contexts.

Different from multiculturalism as a regulatory state policy that is adopted by many modern nation-states, the pluralistic social formation described here is a result of autonomy and local governance. Multiculturalism evokes mosaic or patchwork metaphors that emphasize the co-existence of entities with clear boundaries. This becomes the dominant model of religious and ethnic policies around the world because it creates ethnic and/or religious enclaves, each prioritizing their vertical ties with the state. However, this model also creates ethnic and religious conflicts that have plagued many nations. Pluralism, on the other hand, underscores exchange, reciprocity, conjoining, and ultimately blurring boundaries and creating ambiguities.

Such boundary-crossing reminds us of Victor Turner’s theory of liminality and communitas. In this account, the liminal stage is “necessarily ambiguous” (Turner 1969, p. 359), and such ambiguities allow the temporary lifting of structural constraints that are imposed on people (such as hierarchies and identities) to create a sense of belonging to a common humanity. “Communitas breaks in through the interstices of structure, in liminality; at the edges of structure, in marginality; and from beneath structure, in inferiority” (Turner 1969, p. 372). Thus, Turner calls it anti-structure. Though Turner points out that communitas is a “subjunctive mood” (Turner 1969, p. 372), its psychological power on the individual is strong and lasting. Equally powerful, the mercurial model of group formation that I propose here, however, differs from communitas. Turner’s communitas assumes that there is one structure and one anti-structure. In Bingzhongluo, there seem to be multiple structures that create multiple positionalities: one may be a poor farmer, but he is also an eloquent pastor in the Protestant church, as well as the brother-in-law of a Buddhist monk. One may be a retired government official but also the accountant for the Catholic church. One may be a village cadre but also the husband of a wife who has severe illnesses and sponsors a healing ritual that is cohosted by the spirit medium and the Buddhist monks. Such overlapping identities do not just lift the boundaries within a time frame of liminality and seal it again. Instead, group formation itself is contextual and dynamic, which I call mercurial. Whatever commonalities the context calls for, a new group is formed, such as a ritual community for a funeral or a sworn sisterhood that allows mutual aid in their expedition to seek migrant work. Instead of communitas that may cause violent eruptions, such fluidity is what sustains the pluralistic society.

With such fluidity, however, I am not arguing that ethnic or religious identities are irrelevant. On the contrary, the villagers are keenly aware of the two important notation systems and use them fluently to their advantage. I do not mean a kind of selfish utilitarian rationalism but a Barthian understanding of rationalism that helps the actor maximize his or her interest in specific moments of social transactions (and this interest may even be ultraistic). What this paper argues against is an essentialized view of ethnic or religious identities that are closed and act as puppet strings that determine people’s behavior. If modern nation-states make up their ethnic and religious policies under such an apparatus, it is bound to cause conflicts that in turn further encapsulate people in their ethnic and religious cocoons. How people in Bingzhongluo maneuver such categories with both respect and fluidity may provide a breakthrough to this padlocked way of thinking.

Echoing the other essays in this special issue, I argue that the harder the boundaries the nation-states reinforce by various ethnic and religious policies, the more ethnic and religious conflicts there are. Contrary to the intuition that strong local autonomy and local governance are going to weaken the state, I argue that it may better facilitate state governance. Indeed, the Chinese government is most “successful” in governing Yunnan, in comparison to other ethnic regions in China, not because the most hardline policies are adopted here but because this region has a long history of local governance, one that gives the area most autonomy and their own obligation-based public life that cuts across ethnic and religious barriers. In other parts of the world, where the hardest boundaries are enforced (such as today’s Gaza), there are the most conflicts. As Turner suggests, communitas is often followed by more rigid structural control (Turner 1969, p. 373). What is needed is not a movement that calls for unifying humanity but rather a Zomia mindset that allows group formation to be mercurial. Zomia, as we have seen, is not devoid of the state; instead, it has learned how to live with the state and make the state one other such group formation indicator that is ready to be replaced by others when the circumstances arise.