In a large, almost empty conference room inside the rundown Liuku Hotel, Lañi, an elderly Nusu man, slowly climbs steps up to the theatrical stage at the front of the room.1 He is wearing a handmade woven jacket with narrow blue-and-white stripes, a self-consciously ‘ethnic’ garment (Ch. minzu fuzhuang). A machete hangs at his waist. Lañi seizes the machete, still in its holster, and strikes it nine times on the stage floor. He is miming the way a Nusu shaman would open the ground of the burial plot to open the path for the deceased into the ‘shadow realm’, or mhade.2
In the ritual unfolding of a Nusu funeral, the yãn-hla (soul or doppelgänger) emerges from latency into full personhood, supplanting the corporeal existence of the deceased. Through transformations enacted in death and fire, the deceased and his or her belongings become ontologically other. This process of metamorphosis entails a geographical movement as well, as they set off for the land of the dead (mhade). Lañi, onstage, is demonstrating the beginning of this journey for a group of researchers who traveled from Kunming, the capital city of Yunnan in Southwest China, to the province’s remote northwestern frontier to study endangered traditions of ethnic minority music and dance. In his home village of Khrada, however, Lañi ceased his ritual activities decades ago when he converted to Christianity.
The cleavage between the living and the dead, mapped out in separate geographies and enacted in the ritualized transformation of the person, reflects the poly-ontological character of Nusu animism, which I explore in this article. For Lañi and other Nusu participants in the research encounter, it is echoed in a further ontological divide, between themselves and the researchers documenting their cultural artifacts. More than a matter of linguistic or social differences, for Nusu, different ethnicities, like different states of being, are essentially and irreducibly different, but interactions can occur across ontological divides. The key to this is metamorphosis.
Metamorphosis and Animism
This article takes metamorphosis as a starting point to explore poly-ontological animism among the Nusu in Southwest China. The transformation of persons and ideas enables productive exchanges across boundaries in a fundamentally plural socio-cosmic order. In looking at animism, my focus is on how Nusu understand the invisible dimensions of personhood. Dualistic paradigms—viewing the body as ‘clothing’ for the soul, for instance (Viveiros de Castro 1998: 471)—are inadequate to describe Nusu ways of thinking about the self, whose ‘soul attributes’ possess corporeal as well as spiritual qualities.
The anthropology of animism has long enjoyed a mutual rapport with Euro-American philosophy. Exchanges between these disciplines have infused new energy and ideas into theoretical models on both sides. Consider Deleuze and Guattari’s (1980) exploration of non-dualism, for instance, or Viveiros de Castro’s (1998) description of Amazonian perspectivism. Yet in the heady meeting of anthropological and philosophical minds, there is a risk of making fleshand-soul animists into pure thought exercises (Starn 2011: 193). Michael Scott argues that our theoretical models require careful calibration as they “are not context-free universal decoders” (see Venkatesan et al. 2013: 306). By situating Nusu animism in the ethnographic encounter, I hope to depict the subtlety of their soul theories, showing that animists engage in philosophical explorations that reflect back on anthropological beliefs.
The ethnonym ‘Nusu’ refers to roughly 8,000 people who speak a TibetoBurman language and live primarily in Nujiang, a steep mountainous prefecture bordering Myanmar in the west and the Tibetan Autonomous Region in the north. Due to its geographical proximity to these two politically sensitive regions, Nujiang was militarized and subject to intense ideological scrutiny from the Chinese Communist Party during the Mao era. I conducted 12 months of fieldwork in Nujiang in 2006 and 2008 and additional fieldwork among Nusu living in Kunming during the same period. I spent most of my time in the villages of Khrada and Uvri, which cling to the sides of the Nu mountain range. This location is regarded as a remote and exotic frontier, attracting visitors in search of untouched natural wonderland, but in fact it saw radical transformations during and after the Mao era. The tree line has shifted, new roads crisscross the river gorge, and a massive, controversial dam project is underway (Litzinger 2007). The greatest shift is perhaps in the area’s ideological points of reference.
Historically, Nusu ritual life centered on shamanic practitioners called yüigu or yüigusu, meaning ‘a person who makes sacrifices to the yüi’, the miscellaneous term for spectral beings such as spirits of the forest and human ghosts (see He Shutao 2000: 845–847). In their rituals, shamans divined the spectral causes of illnesses and offered animals as sacrifices across ontological boundaries. From the 1940s to the mid-1950s, Nusu came into contact with evangelists from the neighboring Lisu ethnic group and began converting to Christianity in large numbers (Mazard 2014). This coincided with the beginnings of Chinese ethnographic research in Yunnan, which defined animism in terms of yuanshi zongjiao (primitive religion) (Yunnansheng Bianji Weiyuanhu 1981: 113). People in Southwest China came to understand their ritual practices as ‘backward’ (luohou) and insufficient (see Swancutt, this issue). From 1958 onward, episodes of political turmoil interfered with shamanic and Christian practices, as well as with ethnological projects. The intense politicization of discourse during the Mao era—and the violence surrounding it—made Nusu people intensely aware of the ideological implications of discourse and ritual life (Mazard 2011).
This brings us to the concept of hyper-reflexivity developed in this issue, which is essential to understanding the transfers of people, objects, and ideas between ‘natives’ and ‘anthropologists’, to employ a well-worn dualism. Reflexivity has infiltrated animism and anthropology alike as Nusu thinkers engage with outsiders whom they consider ontologically other. The encounter between animism and two powerful ideologies in Southwest China, that is, socialism and Christianity, has opened up unstable philosophical ground for Nusu people, who are now questioning and redefining their beliefs. Metamorphosis—in this case, the metamorphosis of animism itself—allows exchanges to occur across ontological boundaries, and anthropology is implicated in the transformation.
The ‘Algebra of Souls’
Nusu assume that their social interactions take place in a context of plurality, in which ordinary people experience only one facet of a poly-ontological social order. I draw the idea of poly-ontology from Michael Scott’s (2007) work on the Arosi of the Solomon Islands. Scott deploys this concept to resolve the contradictions between Marilyn Strathern’s (1988) relational, anti-essentialist model of Melanesian sociality and the forms of essentialism that Scott (2007) discerns among the Arosi. Arosi differences are grounded in a “cosmos in which the parts precede the whole” (ibid.: 10), and their cosmogenesis stories point to an “original plurality” (ibid.). Moving on from the anecdote above about a staged ritual in the Liuku Hotel, I argue that Lañi and other Nusu participants view their transactions with the researchers through the lens of poly-ontology, as participants in two separate realms that can communicate only through a carefully managed transformation of words and ideas.
For Nusu, multiple forms of being comprise the inhabitants of multiple, communicating ontological realms. These include multiple ‘types’ that a mono-ontological outsider, such as a non-Nusu anthropologist, might identify as ethnicities, souls, soul attributes, ‘gods’ or ‘ghosts’, each associated with its own geographies and capacities. This conceptual framework has implications for our theory of personhood. Rather than being a monadic individual, or even a relational dividual (Strathern 1988), a person can, under certain conditions, evince different aspects of the self, with multiple ontological identities. Seen from another perspective, ontologically different persons can converge into one identity.
I employ the term ‘algebra of souls’ to describe the Nusu understanding of personhood because it suggests the complexity and transformability of this poly-ontological order and of other forms of animism as well. In mathematics, algebra offers a language for resolving the co-existence of known and unknown elements. An algebraic equation, static on the page, illustrates the movement and transformation of terms. Nusu personhood is reckoned complexly, as in algebra. Some of its attributes remain unknown or possibly in flux, while ontological shifts may bring unresolved elements to the fore. Nusu think of soul attributes in terms of fractions (fragmentary selves), doubles, and latent possibilities. Unforeseen events, shocks such as death or major illness, and vivid emotional states can unbalance the equation of self and shift the number and state of one’s soul attributes. In algebra, equivalent values may take different forms on two sides of an equation, and something similar is at work in Nusu ways of understanding persons, whose multiple attributes, some of them spectral, may co-exist on two sides of an ontological divide. To illustrate this, let us first examine the work of Nusu funerals.
Crossing Ontological Divides
Nusu mortuary rituals involve the management of people and objects transitioning from the realm of the living into the afterlife. For non-Christians, this afterlife is mhade, the shadow realm, where the dead go about their business much as they did when alive, pursuing the same occupations and daily chores. In mhade, as in the realm of the living, people cultivate crops to feed themselves. The dead, like the living, can be wealthy or poor. Items belonging to the dead are sent to mhade to ensure their comfort and to prevent the yãn-hla (soul or doppelgänger) from being drawn back to the family home to use favored items, unable to see the difference between the living world and the shadow realm, unaware that he or she is dead.
During the days following a burial, the deceased still return to their former homes to partake in meals with the living in spectral form. Their presence is often sensed in disembodied perceptions, such as through the sound of chop-sticks scraping against a bowl. Funerals end with the yãn-hla lõng, the ‘soul-awaiting day’, when the deceased’s presence should be felt for the last time. This occurs on the seventh day for women and on the ninth day for men, counting either from death or burial, depending on the family and village. Mueggler (2014a: 200) has also documented the custom of holding seven- and nine-day vigils elsewhere in northwest Yunnan. Wellens (2010: 251n14) notes that “[t]he symbolic association of the number seven with female and nine with male is widespread” among speakers of Tibeto-Burman languages. In Nusu communities, one explanation for the seven- and nine-day vigils is that women and men possess so many souls, a point I return to below.
Managing the belongings of the deceased is a way to maintain souls in their proper place after death. They are detached from the household and their former social relations, as these items form a link that may bring them into dangerous proximity with the living. There are several ways to achieve the transition from the realm of the living into the shadow realm. Items can be buried directly with the deceased or placed on or adjacent to the tomb. It is common to see porcelain bowls and liquor bottles holding the remnants of rice and a strong grain alcohol (Ch. baijiu) placed on the lip of a tomb. Some families also place food next to the deceased during the wake, to sustain them in their transition to the afterlife, as I saw done for a young Christian woman from Uvri village who died at the age of 27 after a sudden illness.3 In addition, fire acts as a means of transport between this world and mhade. Mourners may burn common-use items that are too bulky to fit inside the tomb, such as bedclothes and mattresses. A former shaman also told me that the practice of tomb burial replaced the earlier custom of cremation, which has historically been practiced among ethnic minority groups in Yunnan (Mueggler 2014b: 20–22). Perhaps the deceased themselves once reached the afterlife through means of fire.
Fire can transform objects into a form that the dead can receive. Charcoal or charred wood is money in the shadow realm, and it may be placed in the tomb with food and other important items. When I first asked my Nusu friend Ayima about the significance of charcoal, I employed the term ‘symbolize’ (Ch. xiangzheng). “Does charcoal symbolize money?” I asked. Ayima corrected me: “Charcoal does not symbolize money. It is money for the deceased.” Some families prefer to buy Chinese paper money at the lowland market and burn it after burial. In each case, fire acts as a transformative element. More broadly, in Nusu social life, fire is an agent of destruction and transformation. Yet it also enables social relations of a certain kind. In swidden agriculture, fire destroys vegetation in order to fertilize crops. It enacts creative destruction, initiating cycles of productivity and reproduction: fields lie fallow when their fertility decreases before new vegetation is burned off to make them fertile again. In the household, the fire is at the center of comings and goings in the social space surrounding the hearth. Visiting friends and relatives are told to mi hla, draw close to the fire, where people share gossip and tall tales, drink liquor and eat amödjioguei (maize porridge), pop corn and roast tidbits of meat in the burning embers.
Fire anchors the gra, the cooking tripod and embodied presence of founding ancestors within the house, the link between the living and the dead. The three legs of the gra represent generations of ancestors: aya, abaw, and api (grandmother, grandfather, and great-grandparents). The gra is fed bits of food—lumps of maize and rice that are placed at the top of each of its three legs. Nusu do not practice ‘ancestor worship’ in the sense that anthropologists have employed the term among Han Chinese. Yet ancestors are, in a sense, present in the gra that encompasses the fire. Among the Premi, who also speak a Tibeto-Burman language, Wellens (2010: 120) finds that “the hearth and fireplace with its iron tripod … make up the locus of worshipping the ‘ancestors’ (bap’u) and divine beings of the mountains, water, wind, heaven, and earth.” For the Nusu, the ancestors persist through the hearth, not as known and named persons, but as generic and unnamed elders in a de-individuating remembrance that doubles as an act of forgetting (Carsten 1995; Vitebsky 2008: 245). Elsewhere in northwest Yunnan, effigies take the place of the dead and are gradually moved away from the living, eventually to be discarded (see Mueggler 2001: 71–72).
Acts of giving to the dead, which maintain their relations with the living, also simultaneously contribute to dissolving their social relations. The deceased belong to their own ontological plane, whose geography communicates with that of the living, but whose essence remains fundamentally incompatible—the living cannot live off scraps or swap charcoal money for goods at the market. The yãn-hla is continuous in life and in death, yet living and dead persons are essentially different categories of being, not just different in outer appearance, in spite of shared attributes of personhood. This essential difference is echoed in the ontological divisions between human and non-human persons, the living and the shadow realm, Nusu and non-Nusu.
Nusu refer to souls of the living and the dead indiscriminately as yãn-hla. However, the yãn-hla may become separated from the living self even in life, and in death it entirely supplants it. The yãn-hla is only one of several alternative, fragmentary selves that partake in the identity of their owner, while sometimes exhibiting their own agency. I suggest that we think of Nusu yãn-hla and other kinds of doppelgängers (Hultkrantz 1953; Willerslev 2007) and fragmentary selves as soul attributes, latent and malleable components of the person that, in becoming manifest, may acquire their own agency, acting without their owners’ knowledge. An important corollary to this is that a Nusu person may be perceived as a different ontological type, endowed with different capacities, when certain of her or his soul attributes are present or absent.
One good example of how a Nusu person may become a different ontological type can be found in the difference between women and men, who possess different numbers of souls. Women, who have only seven, can never become yüigu (shamans), although they can acquire the skill to perform certain rituals. The two additional souls possessed by men endow them with the full potential to become yüigu. Meanwhile, men and women endowed with second sight, that is, seers or mia-vr-su (people who see through), possess a predatory soul attribute, the kösu, which can attack and consume human victims. Fear, envy, bereavement, and other affective crisis states can furthermore trigger the inception or appearance of an otherwise latent soul attribute. If one should suffer an emotional disturbance, such as the death of a loved one, the yãn-hla can separate from the body-self and wander away: its owner overcome with grief, the yãn-hla seeks out the deceased in the shadow realm. The owner of the soul remains in possession of his or her wits and personality, but will gradually weaken and die if not reunited with it, similar to soul loss in Chinese folk belief (Harrell 1979: 524–527). Nor can a yãn-hla survive for long in the realm of the living without its corporeal owner, which disappears from the living world within a few years of its owner’s death. Non-Christian families sometimes leave food and alcohol at their relatives’ tombs for some time following the death, typically three years, after which it is not necessary. Even very unhappy or confused ghosts leave the realm of the living after a few years, at most five to seven, I was told. This illustrates the ontological rift between the living and the dead, the impossibility of co-existence due to an essential multiplicity (Scott 2007). By contrast, among other native societies in Southwest China, such as the Nuosu (Swancutt 2012: 67–69) and the Lòlop’ò (Mueggler 2001: 250–284), ghosts can interact with the living for many years after death, whether as benevolent ancestors or malevolent predators.
For the Nusu, human and non-human persons can be multiplied or split into ‘algebraic’ soul fractions in soul loss and in other affective states. The yisu is a soul attribute that emerges from envy. It appears when a person intensely desires an item that belongs to someone else. The yisu attacks the target of envy, sickening his body-self and consuming his yãn-hla. To prevent this, a person will sometimes offer up a belonging as a gift—whether a bracelet, jacket, or penknife—as soon as someone else expresses admiration for it, however innocuously. However, the emergence and attack of an yisu cannot always be avoided, as it can be triggered simply by another’s youth and beauty. Because he or she has no knowledge of its actions, the person who has given rise to the yisu is not considered morally responsible for it in the same way that one is morally responsible for a deliberate curse, which would have a similar effect on the victim. This is something like the view of witchcraft in Evans-Pritchard’s (1937) classic study of the Azande.
The difficulty of translating concepts like yãn-hla and yisu, Pedersen and Willerslev (2012) argue, lies in their ‘fuzziness’. We could gloss them as ‘corporeal souls’, but this is a hybrid term for what are discrete and coherent entities. Pedersen and Willerslev (ibid.: 465) maintain that “in the ‘animist’ concept of ‘soul’, we find the same fuzzy boundaries between self and other, human and inhuman, inner and outer, that are said to characterize postmodernity in the advanced and sophisticated, globalizing West.” Their observation reflects a recurrent theme in the recent anthropology of animism: its capacity to confound dualistic concepts such as the notion of an exterior body and an interior soul (see also Vilaça 2005). I would go further, arguing that the notion of ‘fuzziness’ does not denote any lack of lucidity in animist thought. Rather, it points to the unsatisfactoriness of ‘non-fuzzy’ concepts such as ‘body’ and ‘soul’. Nusu have no difficulty understanding the human-inhuman nature of personhood, or the spiritual-material qualities of their soul attributes. There is nothing inherently ‘fuzzy’ to them in the idea that a spectral being consumes a person’s flesh when it attacks his or her soul, or that one self can emerge from another self that contains the possibility of yet other latent selves.
Soul attributes such as the yisu possess varying degrees of agency and personhood. They sometimes act as true doppelgängers, separating from the consciousness and moral core of their originators, even attacking their loved ones. The terms yisu and kösu build on the linguistic root su, meaning ‘person’ or ‘people’, underlining the potential to achieve personhood. Other kinds of su—for example, what an anthropologist might refer to as other ‘ethnic groups’—are endowed with different ontological attributes. The Bai (Miwa, in Nusu) are the originators of a predatory emanation that attacks Nusu children. Spectral beings, such as the mikhru who inhabit mountains and forests, possess their own predatory, metamorphic qualities. The essential difference of non-human personhood is situated in its corporeal-spiritual attributes, such as the ability of mikhru and certain other non-humans to transform or disguise their appearance at will, or to imbue foods from the forest with their displaced agency. Thus, the distinction between different types of beings lies not so much in their interiority as it does in their latency, in the kinds of soul attributes they possess. The presence or latent possibility of their existence marks both intra- and inter-species differences in personhood (cf. Swancutt 2007: 242–243).
Since the reflexive turn in late-twentieth-century anthropology, early theories of animism have been critiqued for their social evolutionism and the Judeo-Christian assumptions they brought to the study of soul beliefs (Harvey 2006: 3–12; Willerslev 2011). Nevertheless, early anthropological studies sometimes touched on the non-dualistic, complex nature of animistic personhood—what I call its ‘algebraic’ qualities. In Primitive Culture, Tylor ( 1920) endows souls with material as well as spiritual qualities and with the potential to exhibit independent agency while acting as the double of their owner (see Pedersen and Willerslev 2012: 466). Durkheim ( 2001: 49) draws attention to these qualities in The Elementary Forms of Religious Life. In The Uncanny, Freud ( 2003: 142–143) describes spectral doppelgängers as a “defence against annihilation” that nevertheless become “object[s] of terror.” He sees them as liminal beings between life and death, as Richardson notes in her ethnography of technological animism (this issue). However, the theoretical potential of these descriptions remained latent until the recent revival of anthropological interest in animism. New ethnographies have challenged us to explore animistic person-hood in its full complexity.
In the Nusu socio-cosmic order, a shaman (yüigu) develops the ability to negotiate ontological boundaries and the ‘algebraic’ qualities of animistic personhood: human/non-human, visible/invisible, living/dead, and, by extension, Nusu/ non-Nusu. Ordinary people cannot address spectral beings like the mikhru. Shamans possess the linguistic skills to do so, typically claiming the ability to speak numerous languages and to communicate with spectral beings in their language of origin. Shamans can ritually retrieve lost souls and identify the beings responsible for soul attacks. They address sacrifices to spectral beings, who consume invisible aspects of the animal offering and leave the flesh for the humans to eat. Shamans also know how to open paths into the afterlife, and some can transform into snakes. They manage the transformations that occur at ontological boundaries, which enable communication between different ontological categories. Perhaps for all of these reasons, very few Nusu men who perform rituals today call themselves ‘shamans’. People speak of shamanism mostly in the past.
In Nusu funerals, the cremation and entombment of items belonging to the deceased is important because of the malleability of human souls and their movement and transformation between this realm and the shadow realm. Christians, however, learn new doctrines concerning personhood and the afterlife. In this view, the soul is monadic and acquires a transcendence that is supposed to protect it from the predations of spectral beings on earth and ensure its swift transport to the afterlife. Nusu Christians speak of the ‘eternal soul’ as a single entity that does not leave the body until death. This ‘eternal soul’ is the locus of Christian disciplinary practices (e.g., the avoidance of alcohol and cigarettes) that are intended to guarantee the practitioners’ proper standing in relation to God and, hence, their ability to reach Heaven after death. Conversion to Christianity thus involves “the cultivation of a conscious, reflexive self … that must serve as its own moral regulator,” as Chua (2012: 517) notes for the Bidayuh of Malaysian Borneo. Yet church doctrine often proves difficult to reconcile with what Nusu Christians already know about souls, creating dissonance between what they say they do and what they feel ‘should’ be done to ensure the proper transformation of persons and objects into the shape they must take in the afterlife. Layou’s funeral in late 2007 displayed some of these compromises with religious orthodoxy.
From Multiplicity to Dualism
Layou had been sick for some time when he died in 2007, and his funeral was a lively affair. In his youth, he had been a successful yüigu, but he converted to Protestantism around the time of the Communist takeover and had already been a Christian for many years by the time of his death. When the Communists initiated campaigns against religion in 1958 (Mazard 2011), the family was one of the few in Khrada to hold secret worship meetings in their homes. Layou went on to become a popular and respected preacher in his old age, trading his former skill in ritual for success of another kind. Preachers, like yüigu, are expected to ‘speak well’, that is, to possess the gift of charismatic speech.
I went to Layou’s funeral with his grandson Youlin, who was my host in Nujiang. A significant number of the inhabitants of Khrada and many of Layou’s family members who had traveled from other towns and villages were also in attendance. After the wake, a procession transported Layou’s body to the gravesite some distance from his home, in the semi-forest at the northwestern edge of the village. We saw a group of young men putting the final touches to the raised concrete tomb, freshly built. The men pulled the coffin up into the vault while, in light rain, the church-goers in the audience opened their hymnals and began to sing. Layou’s kin then placed his favorite belongings in the tomb. Youlin proferred his grandfather’s chipped enamel tea mug and a few other forgotten items that he had rushed back to the house to fetch. A couple of mourners started a fire to burn the items too bulky to place in the tomb—Layou’s mattress and some of his bedding and clothing. The damp weather delayed the fire at first. As mourners filed back downhill to the funeral feast, a few people remained behind to finish sealing the tomb and ensure that the reluctant fire fully consumed the belongings of the deceased.
The careful way that Layou’s relatives dealt with his personal belongings reflects the untidy reality of a Christian funeral set in the context of animist ontological assumptions. By the time Layou died, I had been living in Khrada for several months and had participated in a number of funerals in varying shades of Christianity. Subscribing to the anthropological teaching that my task as an ethnographer was to discern consistent ritual patterns across varied and often messy real-life ritual events, I had endeavored to develop an idea of the ‘normal’ structure of a Nusu funeral. I expected to find a clear division between Christian and non-Christian funerals, reflecting the differences that Nusu Christians strive to assert in their everyday lives. But while the religious identity of the deceased was never ambiguous, I found that I could not conclusively sort funerals into Christian and non-Christian types. Christian doctrine certainly opposes the burial or burning of personal items belonging to the deceased, as it suggests that the mourners do not trust in the transcendence of the soul. Burning Layou’s items was an ambiguous act. Perhaps it was intended to send the items to mhade, the afterlife, with different connotations for Christians and non-Christians. Or perhaps it was only to ensure that the belongings did not remain where the soul of the deceased might return to fetch them. In any case, it was certainly not an orthodox Christian act.
I returned to Layou’s house nine days after his burial for the yãn-hla lõng (soul-awaiting day). Most of the out-of-towners had already left, but many of the original mourners from Khrada and some other relatives and friends of the deceased filled the house during the day and night of the ritual. There was more drinking, pork, and card games, as there had been during the wake. Later I mentioned the event to my friend Ayima, and she flatly contradicted me: “That’s impossible! Christian funerals can only last three days. That’s the rule.” Christians are not supposed to await the soul’s return, since they should believe that departed souls cannot return to the realm of the living after death. There is, in other words, a transition from ‘spectral attributes’ (with their latent multiplicity) to a unique, transcendent, morally constituted soul. This lays the blueprint for a clear body-soul dualism: the body is vulnerable to disease and death, while the soul is not.
In practice, however, Nusu Christians act as if souls and selves remain multiple, ‘unstable’ (Vilaça 2005), subject to fragmentation and metamorphosis. Yazhu, my first host mother, is a prominent member of the Khrada church and is married to a well-known local preacher and evangelist. Like many other Christians in Khrada, Yazhu describes events such as soul loss and attacks by spectral beings (yüi) as matters of simple fact. She often warned me against walking the mountain roads at night because of the presence of ‘bad people’. Eventually, when I failed to take the hint, she clarified that she meant yüi, which she has encountered in various spots around Khrada. She refused to let her daughter take her grandson, then a year old, to the forest higher up on the mountain, a popular spot for young people’s hunting and leisure expeditions, because of the known presence of mikhru, to which infants are particularly vulnerable—in spite of the fact that Yazhu proclaims her belief in prayer as an infallible protection against yüi. Yazhu related vividly the encounters she has had with disembodied souls, such as her mother-in-law’s yãn-hla, which visited her shortly before her mother-in-law’s death. As is often the case, this ‘soul encounter’ (Chua 2011) manifested itself as fragmentary sensations: a sound as if her mother-in-law were knocking on the door and a sense of her invisible presence.
Such somatic experiences of yüi confront Nusu Christians with the question of how to reconcile their empirical reality with contradictory religious doctrine, a question that leads to an inherently reflexive approach to both the new and the old. As Webb Keane (1997: 677) observes: “[M]issionary activity and religious conversion have been a constant inducement to self-revelation and reflection. Missionization demands an enormous amount of talk, as preachers, converts, and the unconverted are compelled to explain themselves to others, to explain others to themselves, and even to explain themselves to themselves.”
Consider the dialogue below, which follows the death of an Uvri man who had been attacked and slowly driven insane by a mikhru over a period of several years. The mikhru eventually killed him in the forest and left behind his partially consumed corpse. As is typical in such cases, the people who told me about the event made no distinction between the mikhru consuming the man’s body and consuming his soul, since they are part of the same process. However, for two Christians, Ayima and her cousin Ahuo, this raised an ontological problem, since the soul is supposed to transcend material existence. Employing Chinese terms for soul attributes (see Harrell 1979), Ahuo concedes that the hun, which could be translated into English as the ‘living soul’ (or ‘vital principle’), might be consumed by a mikhru. But, he argues, this would not affect the fate of one’s ling, one’s eternal soul.
What the superstitious say is that they [the mikhru] seize our soul … they attack our soul! That way, an illness appears in our physical body … That’s what the superstitious say … But what we Biblicals say is, we humans can’t live anymore after our soul is gone from inside our physical body. It will die. Because, when [a mikhru] has gone away with our soul, that’s how it is— [interrupts him in disagreement] But that’s how it is! In Chinese, ‘departed soul’ [ling] and ‘living soul’ [hun] are two different things, right? Yes. That’s right … Maybe what [mikhru] take is our ‘living soul’. A ‘departed soul’ [means] a person can die, that’s what it says in the Bible. Whatever the Bible says is the truth.
What the superstitious say is that they [the mikhru] seize our soul … they attack our soul! That way, an illness appears in our physical body … That’s what the superstitious say … But what we Biblicals say is, we humans can’t live anymore after our soul is gone from inside our physical body. It will die. Because, when [a mikhru] has gone away with our soul, that’s how it is—
[interrupts him in disagreement]
But that’s how it is! In Chinese, ‘departed soul’ [ling] and ‘living soul’ [hun] are two different things, right?
That’s right … Maybe what [mikhru] take is our ‘living soul’. A ‘departed soul’ [means] a person can die, that’s what it says in the Bible. Whatever the Bible says is the truth.
Faced with doubt, Ahuo cuts short his exposition on the soul and falls back on a statement about the Bible’s infallibility that Ayima cannot challenge. Ahuo explicitly locates the soul inside the body and refers to non-Christians as ‘superstitious’. The nature of the soul, its identity (single or plural), and its relationship to the body remain in question. This ontological uncertainty underlines the failure of Christian doctrine to account for what Nusu already understand about the soul. This discussion, like Yazhu’s accounts of her ‘soul encounters’, directly touches on the question of soul ontology, indicating the incomplete success of missionary attempts to establish body-soul dualism through ritual and doctrinal measures, and the resulting reflexivity and ‘post-modernist’ multiplication (Gerholm 1988) in Nusu beliefs.
The soul-awaiting ritual, burning of personal items, and other practices that I witnessed during and after the funeral of a prominent preacher challenge the church’s doctrine that the soul, after death, immediately and irreversibly departs from the body to take its place in one of two afterlives (heaven or hell), both understood as morally transcendent domains. Nusu Christians are called upon to forswear practices like the yãn-hla lõng that refer to a divisible personhood, an overlap between the body and its soul attributes, and a fluid interconnection between the living and the shadow realm. But then they must grapple with body-soul dualism, which, to them, has become a ‘fuzzy’ framework of interpretation.
“We Don’t Call It Primitive Religion Anymore”
Perhaps it is because Nusu view the world as poly-ontological that they are predisposed to reformulate their socio-cosmic assumptions when they interact with those whom they see as ontological outsiders, including representatives of the Chinese state. Ethnographies of ethnicity and religion in China have brought to light the complexity of the relationship between the Chinese body politic and nationalities such as the Yao (Litzinger 2000), Dai (Davis 2005), Tibetans (Makley 2007), and Premi (Wellens 2010). Following in the vein of post-colonial theory, this research has shown that power is best understood not as an externally applied force, in the sense of ‘oppression’, but rather as a deeply embedded aspect of social relations. Ethnic minorities have been “active agents in the reimagining of the post-Mao Chinese nation” (Litzinger 1998: 241) and in their own reform-era cultural reinvention (ibid.; see also Gros 2012; Schein 2000), including the reinvention of religious life (Litzinger 1998).
Over the past sixty-odd years, there have been several shifts in Chinese state policies toward ethnic minority (shaoshu minzu) religions in Southwest China (see Swancutt, this issue).4 In the mid- to late twentieth century, Nusu rituals were officially classed as ‘primitive religion’ (Ch. yuanshi zongjiao) and then attacked as ‘feudal superstition’ (Ch. fengjian mixin) before being re-evaluated in the 1980s onward as ‘nature religion’ (Ch. ziran zongjiao) and ‘animism’ (Ch. wanwu you ling), or the teaching that “everything has a soul” (Swancutt, this issue). Through these recent discursive shifts, Chinese ethnologists have moved away from the social evolutionism of Marxist-Leninist anthropology to seek ways of recognizing and even respecting indigenous religion. The political sensitivity that continues to surround manifestations of indigenous culture within the People’s Republic of China has fostered the careful cultivation of state discourse among both anthropologists (Mullaney 2004) and cultural representatives of ethnic minorities, such as tour guides, folk performers, and native intellectuals.
As members of the Nu nationality, one of China’s smallest officially recognized ethnic groups or xiaoshao (small minorities), Nusu lack the clout of larger groups who have mustered their political representation and intellectual elites to create official spaces for their reinvented religious identities in the post-Mao reform era. Litzinger (1998), for instance, documents the revitalization of the deity Pan Wang as a Yao cultural icon, while Swancutt (this issue) examines the role of Nuosu ritualists and intellectuals in state-sponsored research on Nuosu religion. Yet Nusu are also active participants in the reinvention of animism, albeit on a smaller scale.
In 2006 to 2008, the researchers whom I met and collaborated with in Yunnan were intensely aware of the politically ‘sensitive’ (Ch. mingan) nature of projects concerning minority ritual and religion. Fieldwork in Nujiang was already considered mingan because of the controversy surrounding a series of planned hydroelectric dams all along the Nujiang river from the Tibetan border to the fringes of Kachin State (Litzinger 2007). The term ‘sensitive’ metonymically suggests topics that must be approached with caution, often indirectly. In spite of its sensitivity, however, Nusu animism has been reinvented—in the field by its practitioners, as well as in ethnographies and video recordings. Practitioners and researchers alike must navigate difficult political waters with a tricksterish skill, transforming the meanings of terms to disguise sensitive ideas (Mullaney 2004), redefining identities, and carving out cosmological spaces at the margins of a hegemonic apparatus of state. A shamanic ritual might be enacted and recorded as a ‘folk dance’, for instance, as was the case with the video recording session at the Liuku Hotel.
Let us then return to the event described in the opening paragraphs of this article—the ritual performed onstage in the Liuku Hotel. The recording session was organized by a team of Chinese researchers who traveled from Yunnan’s capital city to the province’s remote northwest in late 2007, ostensibly to record the ‘folk songs’ and ‘folk dances’ of the ethnic minorities of Nujiang Prefecture. They devoted several days to the Nusu participants, who, like all the performers, were asked to dress in ‘traditional’ outfits to represent their nationality. At one of the recording sessions, Lañi and another elderly Nusu man demonstrated a soul-calling chant. Lañi later acted out the mortuary ritual where he struck the machete on the ground to open the path for the yãn-hla into the afterlife. Onstage, he was pretending to be a shaman. He was also pretending not to be a shaman. In front of the stage, brandishing a video camera, microphones, and a boom stand, the researchers pretended that this was not a ritual—and that it was not not a ritual. Both parties were navigating ambiguous boundaries of being and becoming, action and mimicry.
Lañi was in his mid-eighties when I interviewed him in December 2007, a month after the recording session. When he was a young man, his first wife was sick for three years. His father, who was a shaman, conducted sacrifices of pigs, chickens, and two heads of cattle, but she did not recover. Afterward, Lañi thought that perhaps the sacrifice had not been carried out well, or perhaps they were too poor to do enough. He became a shaman sometime later, and then, at a point that remained unclear in our interview, he converted to Christianity. His second wife was a Christian as well. In the 1980s, she became a follower of a millenarian religious leader in Khrada and remained so until her death in 2008. Millenarians differed from other Christians in Khrada, eschewing the use of medicine in favor of prayer. Their religious differences were a cause of unhappiness for both spouses, leading to domestic strife. Lañi’s wife would argue with him, pressuring him to convert. By 2007, then, when Lañi acted the part of the shaman for the recording session, he had not been a shaman for quite some time, and Christian religious debates loomed large in his personal life. Nevertheless, people in Khrada knew that Lañi had passed on some of his knowledge and skill to one of his daughters and that she was active as a ritualist, perhaps putting into question his identity as a non-shaman. The recording session placed him in the equally tenuous role of a politically neutral shamanic folk dancer.
After Lañi’s recording session, I asked Zhou, the Chinese team’s leader, about the place of animism in his research on the Nusu. I employed the term ‘primitive religion’ (Ch. yuanshi zongjiao), which is commonly used among anthropologists and other visitors when discussing indigenous cosmologies in Yunnan. Zhou quickly corrected me: “We don’t call it primitive religion anymore. Now we use the term ‘nature religion’ [Ch. ziran zongjiao].” Skirting the sensitive topic altogether, Zhou explained that he wanted the recordings to become a permanent document of Nujiang’s folk music and dance. He also planned to introduce cultural elements into the university admissions procedure for ethnic minority candidates. His idea was to mandate that all such candidates must recite at least one of the ‘folk songs’ belonging to their nationality in order to pass the admission test. The songs would be checked against a list drawn from research projects like theirs. He complained that the Nusu performers were almost all old men and women, and he expressed concern that these traditions might disappear in Nujiang’s current phase of development. Indirectly, the team leader was drawing a connection between the preservation of animism and the strengthening of minority identity.
The researchers could not address the cultural life of shamanic rituals or their animistic connotations. They staged Lañi’s rituals in the guise of a performance, devoid of ritual intentions or effects, like the denatured rituals that occur in the Beijing Ethnic Culture Park as demonstrations of ethnic minority traditions (Makley 2010). Yet this was a way for them to mark out a space for ritual and to preserve it as an aspect of minority identity. Approaching animism from the perspective of ‘folk dance’ allowed the researchers to create possibilities for its study and documentation in a political context where it remained officially banned.
Amid these ambiguities, one of the Nusu participants was tasked with navigating between the shamanic and ethnological realms. Youlin, whose grandfather Layou was buried a few weeks earlier, was invited to organize the performers and act as interpreter for them and the researchers. Youlin supports his family through a variety of sources: he farms, operates one of Khrada’s hostels, acts as a tourist guide, performs as a singer and dancer in county events, and holds office as a village councilor for the administrative village of Laomudeng. In most of his roles, Youlin acts as a gatekeeper between outside visitors and the village (or image) of Khrada and its Nusu inhabitants, adeptly molding himself to visitors’ expectations. He honed his intercultural skills during his first job at a cultural park similar to the one described by Makley (2010) in her analysis of post-Mao ‘national exhibitionism’. At a time when Southwest China is banking on its cultural capital for economic development, Youlin displays “a trickster-like ability to negotiate between the new market system and the old parallel culture of socialism” (Pedersen and Højer 2008: 82).
During the 2007 research project in Liuku, Youlin acted as a kind of ‘cultural shaman’, mediating between Nusu and non-Nusu. Most importantly, he translated the text and content of Lañi’s shamanic ‘dances’ into Chinese-language description—into a cultural form that he knew would be both acceptable and comprehensible to the researchers. This meant that the researchers were never confronted with the differences between the socio-cosmic assumptions underlying the ritual performances and the understanding of animism that they brought with them into the field. Like the yüigu who speak to spectral beings, or the kin who bury coal as money for the deceased, Youlin deployed transformative techniques to make interaction possible across ontological divides. Metamorphosis—whether of forms or ideas—is the fundamental precondition of exchange in any poly-ontological world. As Scott (2007: 18) puts it: “[A]ctors engaged with poly-ontological assumptions must create unifying relations among multiple pre-existing categories of being. In so doing … they must also find ways to preserve their distinctive identities without rupturing the ties they have formed and reverting to primordial disjunction.” Whereas Nusu shamans enact transformative exchanges with spectral beings, Youlin and other masterful cultural interpreters perform such exchanges with ethnologists, whom the Nusu situate in another ontological domain.
The ‘algebra of souls’ describes a socio-cosmic order in which personhood is subject to plurality and fragmentation. Unknown aspects continuously emerge from and propel the person throughout her or his lifetime, engaging visible and invisible aspects of self. Nusu persons are “irreducibly plural and capable of transformation,” as Carlos Fausto (2011: 581; my translation) says of the powerful beings depicted in the ritual masks of the Upper Xingu. Among the Nusu, this applies to ordinary men and women as well as to shamans, seers, and spectral entities. Human and non-human persons possess a “non-dual plurality” (ibid.), containing unknown fractions and doubles, multiplying and shifting in a non-dualistic matrix.
To this dynamic uncertainty is added the complexity of other perspectives on the animistic self, resulting from Nusu participation in Christian and socialist representations of their reality. Animistic ideas are reformulated in and through Christianity, as well as in political and anthropological encounters. When Lañi and others provide researchers with cultural prestations, they are acutely aware that these have ideological implications (Mazard 2011). Ideas move hyper-reflexively across ontological boundaries. I view Nusu rituals, whether staged or ‘real’, as ontological statements formulated in this sensitive context.
As Nusu animism has become hyper-reflexive, a new ontological uncertainty has emerged, beyond the uncertainty of soul algebra. Christian doctrine asserts the unity and transcendence of the soul, which Nusu understand as partly corporeal and dangerously divisible. Socialist modernity competes with sacrifice and prayer, and ethnology reinvents animism through its theoretical trajectory. Nusu people who participate in these competing representations in various ways therefore understand their practices and beliefs from points of view that are already reflexive, implicitly responding to the cosmological and ideological implications of ritual discourse. Their rituals are organized around multiple negotiations and transformations between domains: the realm of the living and the shadow realm, shamanic practices and the values of Christianity, political history and anthropological classifications. Nusu situate these multiple realms of meaning in a poly-ontological order. The key to their co-existence lies in metamorphosis, a necessary condition of exchange across ontological divides.
The ethnography presented in this article was made possible by grants from the Firebird Foundation and the University of Cambridge. Many thanks are due to Ayima, Youlin, Lañi, and other Nusu hosts and friends who showed me the meaning of patience and generosity. Material that eventually made its way into this article was first presented in a talk entitled “Ghosts of Ritual” at the Magic Circle seminar, held at the Scott Polar Research Institute, University of Cambridge. I wish to thank Piers Vitebsky, the organizer of the seminar, for his valuable feedback and the anonymous reviewers for their comments on the article. In addition, Katherine Swancutt, co-editor of this special issue, suggested new lines of thought and gave encouragement throughout the writing process. Erik Mueggler’s work in northwest Yunnan provided inspiration, particularly with regard to the importance of ghosts.
I employ pseudonyms for Nusu names throughout the text.
Unless otherwise indicated, all non-English terms are in Nusu.
The bowl of rice and meager serving of meat placed by her head when she was lain out for the wake would have attracted disapproval among the Christians of Khrada, who are perhaps more sensitive to church prohibitions on such practices.
CarstenJanet. 1995. “The Politics of Forgetting: Migration, Kinship and Memory on the Periphery of the Southeast Asian State.” Journal of the Royal Anthropological Institute 1 no. 2: 317–335.
ChuaLiana. 2011. “Soul Encounters: Emotions, Corporeality, and the Matter of Belief in a Bornean Village.” Social Analysis 55 no. 3: 1–17.
ChuaLiana. 2012. “Conversion, Continuity, and Moral Dilemmas among Christian Bidayuhs in Malaysian Borneo.” American Ethnologist 39 no. 3: 511–526.
FaustoCarlos. 2011. “Le masque de l’animiste: Chimères et poupées russes en Amérique indigène.” Trans. Emmanuel de Vienne. Gradhiva 13: 48–67.
He Shutao. 2000. “Nuzujuan.” [Nu Nationality Chapter] Pp. 826–938 in Zhongguo Ge Minzu Yuanshi Zongjiao Ziliao Jicheng: Naxizu Qiangzu Dulongzu Juan Lisuzu Nuzujuan [Collected Materials on Primitive Religions of All Chinese Nationalities: Naxi, Qiang, Dulong, Lisu, and Nu Nationality Chapters] ed. He ZhiwuQian Anduan and Cai Jiaqi. Beijing: Zhongguo shehui kexue chubanshe.
- Search Google Scholar
- Export Citation
)| false . He Shutao 2000. “ Nuzujuan.” [Nu Nationality Chapter]Pp. 826– 938in [Collected Materials on Primitive Religions of All Chinese Nationalities: Naxi, Qiang, Dulong, Lisu, and Nu Nationality Chapters], ed. . , He Zhiwu , and Qian Anduan Cai Jiaqi Beijing: Zhongguo shehui kexue chubanshe.
KeaneWebb. 1997. “From Fetishism to Sincerity: On Agency, the Speaking Subject, and Their Historicity in the Context of Religious Conversion.” Comparative Studies in Society and History 39 no. 4: 674–693.
LitzingerRalph A. 2000. Other Chinas: The Yao and the Politics of National Belonging. Durham, NC: Duke University Press.
- Export Citation
)| false Litzinger, Ralph A. 2000. . Durham, NC: Duke University Press.
LitzingerRalph A. 2007. “In Search of the Grassroots: Hydroelectric Politics in Northwest Yunnan.” Pp. 282–299 in Grassroots Political Reform in Contemporary China ed. Elizabeth J. Perry and Merle Goldman. Cambridge, MA: Harvard University Press.
MakleyCharlene. 2010. “Minzu, Market and the Mandala: National Exhibitionism and Tibetan Buddhist Revival in Post-Mao China.” Pp. 127–156 in Faiths on Display: Religion Tourism and the Chinese State ed. Tim Oakes and Donald S. Sutton. Plymouth: Rowman & Littlefield.
MazardMireille. 2014. “The Art of (Not) Looking Back: Reconsidering Lisu Migrations and ‘Zomia.’” Pp. 215–246 in Globalising Migration History: The Eurasian Experience (16th–21st centuries) ed. Jan Lucassen and Leo Lucassen. Leiden: Brill.
MuegglerErik. 2014a. “‘Cats Give Funerals to Rats’: Making the Dead Modern with Lament.” Journal of the Royal Anthropological Institute 20 no. 2: 197–217.
MullaneyThomas. 2004. “Ethnic Classification Writ Large: The 1954 Yunnan Province Ethnic Classification Project and Its Foundations in Republican-Era Taxonomic Thought.” China Information 18 no. 2: 207–241.
PedersenMorten Axel and Lars Højer. 2008. “Lost in Transition: Fuzzy Property and Leaky Selves in Ulaanbaatar.” Ethnos 73 no. 1: 73–96.
PedersenMorten Axel and Rane Willerslev. 2012. “‘The Soul of the Soul Is the Body’: Rethinking the Concept of Soul through North Asian Ethnography.” Common Knowledge 18 no. 3: 464–486.
StarnOrin. 2011. “Here Come the Anthros (Again): The Strange Marriage of Anthropology and Native America.” Cultural Anthropology 26 no. 2: 179–204.
SwancuttKatherine. 2012. “Fame, Fate-Fortune, and Tokens of Value among the Nuosu of Southwest China.” Social Analysis 56 no. 2: 56–72.
VenkatesanSoumhyaKeir MartinMichael W. ScottChristopher PinneyNikolai Ssorin-ChaikovJoanna Cook and Marilyn Strathern. 2013. “The Group for Debates in Anthropological Theory (GDAT), the University of Manchester: The 2011 Annual Debate—Non-dualism is Philosophy Not Ethnography.” Critique of Anthropology 33 no. 3: 300–360.
- Search Google Scholar
- Export Citation
)| false . , Venkatesan, Soumhya , Keir Martin , Michael W. Scott , Christopher Pinney , Nikolai Ssorin-Chaikov , and Joanna Cook Marilyn Strathern 2013. “ The Group for Debates in Anthropological Theory (GDAT), the University of Manchester: The 2011 Annual Debate—Non-dualism is Philosophy Not Ethnography.” 33, no. 3: 300– 360.
VilaçaAparecida. 2005. “Chronically Unstable Bodies: Reflections on Amazonian Corporalities.” Journal of the Royal Anthropological Institute 11 no. 3: 445–464.
VitebskyPiers. 2008. “Loving and Forgetting: Moments of Inarticulacy in Tribal India.” Journal of the Royal Anthropological Institute 14 no. 2: 243–261.
WillerslevRane. 2011. “Frazer Strikes Back from the Armchair: A New Search for the Animist Soul.” Journal of the Royal Anthropological Institute (n.s.) 17 no. 3: 504–526.
Yunnansheng Bianji Weiyuanhu ed. 1981. Nuzu shehui lishi diaocha [Investigations on the Social History of the Nu]. Kunming: Yunnan minzu chubanshe.