Controversial Funerals and Concepts of the Person among the Paiwan, Taiwan

Tradition and Christianity: Controversial Funerals and Concepts of the Person among the Paiwan, Taiwan

Tan, Chang-Kwo


This essay focuses on the complex and dynamic relationships between tradition and Christianity among the Paiwan, an Austronesian-speaking people in Taiwan. The ethnography comes from a village in which adherents of traditional religion and Presbyterians live in the same area, but in separate quarters. They have close kinship and marriage ties; however, the funeral constitutes a major area of controversy and tension. I juxtapose their contrasting representations of death and funerals to highlight their differences and trace the source of their conflict to the ideological domain of concepts of the person. I argue that the differences in their funerals can be explained by the differences in the ways the two groups articulate two versions of the person. Adherents of traditional religion emphasize the holistic side of the person by calling the spirit of the deceased to reunite with the living, whereas Presbyterians stress the individualistic side of the person and maintain the boundary between the living and the dead. I propose that it is the co-existence of two distinct concepts of the person and the different ways of articulating those concepts which shape the dynamic but often contradictory relationship between tradition and Christianity.

The relationship between tradition and Christianity has been a prominent theme in anthropological writings on Oceania. Since the early 1980s, scholars have begun to address the trend of burgeoning interest in the politics of tradition among Pacific peoples. The use of kastom, ‘custom’, as political symbol in Melanesia was examined in a collection edited by Keesing and Tonkinson (1982), and the relationship between kastom and Christianity is a central issue in many studies (Tonkinson 1982:302). The relationship has been either perceived as fundamentally antithetical (Jolly 1982), or as in a state of satisfying symbiosis (Burt 1982). This issue was further explored in Jolly and Thomas (1992), and various notions of kastom were considered, some of which were excluded (Otto 1992), and others of which combined with Christianity (Thomas 1992) were considered. The relationship between local tradition and Christianity has also played an important role in shaping the face of local Christianity. In a volume dedicated to ethnographic appraisals of Christianity in the Pacific (Barker 1990a), Barker proposes that Pacific Christianity can be better understood in terms of popular religions, which consist of a combination of indigenous and Christian ideas and forms (Barker 1990b:10-15, see also Chowning 1990). At the level of organizational distinctions and politics, the divisions of Christian and traditional domains may seem rigid, yet people are often involved in both domains and can tolerate considerable ambiguity and inconsistency (Macintyre 1990).

In this paper, I address the issue of the complex and dynamic relationships between ‘tradition’ and Christianity among indigenous peoples in the Pacific. Using the term ‘tradition’, I refer to the indigenous cultural and religious practices that existed before evangelization and persist in contemporary life, as well as a construction of the past (or continuity with the past) deployed for political empowerment. By ‘Christianity’, I mean the Christian ideas and rites experienced and practiced by indigenous peoples, as well as the social and political organization recognized as ‘the church’. I am particularly interested in exploring the relationship between tradition and Christianity as it is mediated by controversial funeral practices and underlying concepts of the person among the Paiwan, an indigenous people of Taiwan.

Taiwan is better known as ‘Formosa’ in the international literature of history and anthropology. The speakers of Formosan languages,1 the aboriginal peoples of Taiwan, inhabit the northern border of the areas in which the Austronesian-speaking population is distributed (Figure 1). Their population is about 43,0000, 2% of the total population in Taiwan; these aborigines constitute the minority ethnic groups in the contemporary political structure of Taiwan. Historically, their ancestors have had contact with Christianity since the Dutch tried to colonize Formosa in the 17th century. Mass conversion to Christianity took place after the Japanese colonial government retreated from Taiwan at the end of World War II.2 The majority of the aboriginal population has converted to Christianity. This population includes both Protestants and Roman Catholics, and today most aboriginal Christians are members of the Presbyterian Church, the largest and most influential Protestant denomination in Taiwan (Huang 1996). In many villages, however, we can still see aborigines resisting the campaign of evangelization and abiding by their ancestral customs. Under the leadership of traditional authorities, the performance of traditional rituals has become a highly-charged context for constructing cultural identity. Recently, these efforts have been encouraged by the Taiwanese government, which has come to recognize Taiwan as a multicultural society and has begun to endorse the expression of cultural differences through preserving local traditions.

During my fieldwork among the eastern Paiwan between July 1997 and November 1998, I found that funeral rituals have become a central site for constructing Paiwan identity for local adherents of traditional religion. From holding a wake for the deceased to calling his/her spirit back through shamans’ power, funeral rituals unfold as a series of events in which all that is positive in being a Paiwan is exhibited and celebrated.3 On the other hand, for the regional church authority the funeral is a serious matter because it explicitly addresses the fundamental doctrine of the resurrection of Jesus Christ and the afterlife of believers. Though local customs should be respected and tolerated, paganism and superstitions are not allowed to be mixed into Christian services. Nevertheless, many elements in traditional funeral rituals can be identified as pagan and superstitious by the church authorities. Funerals have thus become the locus of tension and controversy between adherents of traditional religion and Christians. In the following, I will describe the ethnography of a village in which adherents of traditional religion and Christians live in the same area but in separate quarters, and whose relationships and interactions offer a valuable opportunity to observe the complex relationship between tradition and Christianity.


The Paiwan, speakers of the Austronesian language of Paiwan, are the third largest aboriginal group in Taiwan, comprised of approximately 68,000 people living in the rural and mountainous areas of the southern end of Taiwan (Figure 2). Their settlements are scattered along Mount Tjagaraus (or Kavulungari), the legendary homeland of origin myths and the oral histories of many Paiwan tribes. Using this mountain as a reference point, the Paiwan divide themselves into several subsets. The ancestors of the eastern Paiwan are supposed to have migrated from the settlements of the northern or central Paiwan no later than the 17th century, according to Japanese scholars (Utsurikawa et al. 1935). The Paiwan are surrounded by Han Chinese and other aboriginal groups including Rukai and Puyuma. Inter-ethnic marriages, barter and trade have been frequent since the Tsing Dynasty (Tsai 1998).

Paiwan social organization can be best described as a ‘house-based’ society which has been widely documented in Southeast Asia (Levi-Strauss 1983, 1987, Fox 1987, Errington 1987, Garsten and Hugh-Jones 1995). There is no kinship organization like clan or lineage, and the house (umaq), including the material building as well as the inhabitants living inside, constitutes the basic and enduring social unit (Chiang 1993, Tan 2001b). As a form of property, the house is inherited ideally by the firstborn (vusam) regardless of gender, while other junior siblings will be married out and establish their own houses. Siblings born in the same house share a strong sense of solidarity even after they have been married out. In a village, houses are mainly connected through the idiom of siblingship, which can extend to include the entire community (cf. Garsten 1997).

The Taiban village in which I conducted fieldwork is located in the hills, 200-300 meters above sea level, near the Southeast coastal region of Taiwan. The village site can be reached by concrete roads along the foothill from a nearby town. Like other aboriginal villages, the settlements and farmland of Taiban are enclosed by Reservation Land. The policy of Reservation Land, which originated in the Japanese colonial policy of aboriginal land, has been implemented by the Taiwanese government since 1948. Its purpose is to protect the livelihood of aborigines and to foster the control of aborigines, a major consequence of which has been to transform their pattern of livelihood from shifting cultivation to settled cultivation. Under current regulations, everyone with the ability to work, male or female, is entitled to have 0.6 acre of wet farmland, 1 acre of dry farmland and 1.5 acres of forestland. However, it is not easy for young married couples to acquire enough land to survive if they want to stay as farmers in the village. As a result, about 20% of the population are migrant workers, who find their livelihood in big cities as wage laborers (Tan 2001a).

This village is comprised of two major settlements: Tjauqau, which has 49 households and 249 residents, and Laliba, which has 116 households and 420 residents in 1998. Historically speaking, Tjauqau and Laliba have different origins and their own systems of leadership, and are dozens of miles away from each other. It is the resettlement policy of the Taiwanese government that forced them to migrate to their present adjacent locations to form a single ‘village’, the basic administrative unit in the state structure. They are still separated by a valley, which marks the boundary of their religious differences. The majority of Tjauqau residents are adherents of kakudan and palisi, which are ancestral ‘laws, customs’ and ‘taboos, rituals’ passed down from generation to generation. They hold the indigenous leadership of chiefs and shamans in high regard; they interpret the meanings of kakudan and direct the performance of palisi. In Laliba, by contrast, ‘chief’ is only a nominal title, and shamans have all either died or converted. Most residents have turned away from their traditional religion and have accepted Christianity. The Presbyterian church is the center of social life, and its pastor and elders are leaders of the community.

Tjauqau and Laliba residents also have developed wide kinship networks and marriage ties since they have inhabited the old settlements, and now they are connected by a concrete road which fosters contacts between the two groups. They often receive invitations and attend one another’s events, and the funeral is one of the rituals they feel obliged to attend. Migrant workers, in particular, see returning home to attend funerals as a serious obligation and a channel for maintaining ties with their kinsfolk. However, as a result of their different religious affiliations, they hold different opinions on the right way to conduct a funeral. Presbyterians consider funerals held by Tjauqau followers of traditional religion to be ‘idolatrous’ and show their attitude of defiance and disrespect during funeral processions. Conversely, during Presbyterian funeral services, Tjauqau residents make noise outside the mourning setting rather than joining the congregation. They perceive church practices and ideology as a threat which will destroy their traditions, and vehemently criticize the Presbyterians for ‘abandoning the ancestral way’. The funeral thus constitutes a symbolic field, which both expresses and mediates the antagonism between Tjauqau adherents of traditional religion and Laliba Presbyterians.

I propose that Tjauqau can serve as an example of the type of Paiwan society in which traditional religion remains active, and ‘tradition’ is an effective symbol in local projects of cultural and political empowerment, whereas Laliba is the type of society in which Christianity is dominant in most aspects of social and religious life. In this article, I try to elaborate on the differences in ways of dealing with death and conducting funerals between Tjauqau and Laliba and relate the explicit domain of the differences in their funerals to the implicit domain of the differences in their concepts of the person.


Before proceeding further, I shall outline the theoretical framework which enables us to connect ethnographic material regarding death and funerals with the analysis of concepts of the person. Hertz (1907/1960) points out a well-known ethnographic fact: the representation of death varies from culture to culture. In modern Western society, it is generally accepted that death occurs in one instant, whereas in other societies such as the Dayak of Borneo, death is represented as a lasting procedure, a transitional process involving a number of stages of transformation. Bloch’s article (1988), ‘Death and the Concept of Person’, begins by describing the contrast between the ‘punctual’ view of death and the ‘transitional’ or ‘processual’ view of death as noted by Hertz, and further argues that these contrasting views of death are fundamentally tied to different concepts of the person. In brief, the ‘punctual’ view of death assumes the person to be a ‘bounded individual’, while the processual view assumes the concept of ‘unbounded person’ (1988:15).

In Bloch’s terms, the concept of the bounded individual assumes that people are separate entities, unique bounded units, which form a real unity that transcends all other divisions. An individual is in-divisible; one is either an individual with all the parts joined, or one is divided, or in other words, dead. The boundary between life and death is very sharp, and there is no room for the existence of any in-between states. In contrast, for an unbounded person, the unique combination of the individual is not of primary importance, and the boundary between oneself and others is far from absolute. The constituent parts of the person can have the potential for a life of their own, because they are part of a greater cross-cutting whole, e.g. a patrilineal or matrilineal descent group. This cross-cutting whole has a life independent from the life of individuals, so the moment of death may mark the end of a person’s individuality, but this will not mean total annihilation. The part of a person that remains in the cross-cutting whole can survive the break up of the individual and continue to live (1988:16-7).

The most significant argument Bloch advances is that we cannot assume a great divide between a holistic society, in which the concept of the unbounded person is prominent, and the individualistic society, which often refers exclusively to modern western society. He argues that individualism exists in all cultures, and what differentiates the western system of thinking from other cultures is not the presence of individualism, but the idea that persons are nothing but individuals. In other cultures, the person has not only an individualistic side, but also a holistic side. The coexistence of two distinct concepts of the person in society can be phrased as ‘the duality of the person’. The duality of the person, as Bloch observes, has important implications for funerary practices. He proposes the hypothesis that in a society with a universalist world religion, e.g. Christianity or Islam, the world religion will primarily deal with the individualistic side; the holistic side, on the other hand, will be dealt with through other categories of religious practice. He argues that,

The world religions bury the individual and send him to God and out of the social world. At the same time this expulsion purifies that part of the person which continues on earth and which will be re-used and reincarnated in other members of the corporate group to which the dead belong. This ‘re-use’ will then naturally concern religious practices which contrast with those of the world religions. (Bloch 1988:20)

For societies without world religion, he argues that funerals also have a dual task: laying the individual to rest and causing the holistic element to survive on earth.

Basically, I assume that Bloch’s elucidation of the association of the representation of death and concepts of the person can provide insight into Paiwan ethnography. It must be stressed, however, that my understanding of the ‘individual’ and ‘person’, two key terms in the study of concept of the person, is somewhat different from Bloch’s definition of ‘bounded individual’ and ‘unbounded person’. I tend to adopt Fortes’s approach, which defines ‘person’ as the part of a human being created and defined by the society, and ‘individual’ as the part determined by biological laws and subject to human mortality (Fortes 1973:247-86).4 As La Fontaine points out, this usage has the advantage of maintaining the distinction between person and individual, which allows us to use the term of ‘individualism’ or ‘individualist’ to refer to the Western variety of the concept of the person (La Fontaine 1985:126).

Bloch’s usage, in contrast, characterizes the notion of ‘bounded individual’ as the Western folk view of people and society, in which people are discrete, separate, and bounded units, which form a real unity transcending all other divisions. People are merely linked by ‘social’ or ‘moral’ relationships, but these relationships are not as real as those which exist between the component parts making up the individual (Bloch 1988: 16). In my view, the notion of the ‘individual’ does not necessarily imply the notion of ‘bounded individual’. The former is supposed to be a more general notion which can be found in every society, while the latter is a specific view of what the boundaries are between one person and another, which is dominant in some societies but submerged in others. Likewise, the difference between ‘person’ and ‘unbound person’ lies in the latter assuming a specific view of the boundaries between persons. The concept of ‘unbounded person’ assumes that the boundary between one person and others is not fixed and definite, but flexible and negotiable. The ties which connect different persons are not merely ‘social’, but are also as ‘real’ and ‘material’ as those linking up the different parts of the individual (Bloch 1988:17). The implication is that persons are not only ‘dividual’ or ‘divisible’, but also ‘multiple’ and ‘partible’. This is a theme elaborated in Strathern’s well-known writing on Melanesian persons (Strathern 1988).

More recently, Carsten (1997) addresses the issue of the cultural construction of boundaries between persons by examining kinship and notions of the body and the person in Langkawi, Malay. She concludes that both notions of ‘multiple’ and ‘partible’ identity and more individualistic concepts of the person can be discerned in Langkawi. She advances an argument somewhat like Bloch’s: that ideas of ‘unbound person’ and ‘bounded individual’ will both be present in most societies. Thus, it is misleading to adopt the simple dichotomy between the individualistic society and the holistic society, between ‘Western’ and ‘nonWestern’ societies’ notions of personhood.5 Following Bloch and Carsten, I postulate that concepts of individuality and relatedness both exist in a society, and the key issue is to find the ways in which they are articulated and reconciled. With this orientation in mind, I turn to describe the funerals among Tjauqau adherents of traditional religion and Laliba Presbyterians respectively. My aim is to examine whether the practices observed here can support Bloch’s theses about the dual task of funerals. I will further analyze the ways in which the two versions of personhood are articulated in the funerals in these two types of Paiwan society, as well as the differences in their style of articulation.


Funerals among Adherents of Traditional Religion

In the traditional view, when a person’s breath stops, his soul departs, and the body starts to become as hard as wood; this is regarded as ‘death’ (matsai) (Hsu 1994:117).6 Viewing the deterioration of the body will bring bad luck to the living, so the corpse has to be buried as soon as possible. The soul, however, still exists and lingers in the vicinity of the living. It will not go immediately to the residence of ancestral spirits, but will stay close to the site of death to watch the reaction of the living. If the bereaved family devotes itself to meeting the soul’s needs, the soul will not disturb the living. After shamans hold a sacrifice on behalf of the bereaved family, the soul will be sent off on its journey, passing through the boundaries of several worlds of tsemas, which are a variety of supernatural or spiritual being. The soul of a person who has a ‘good death’ (to be described below) will return to Naqemati, the Creator God, who decides the destiny of an individual when he/she is alive, and also decides where his/her soul should rest. A minority will go to the upper world, which is the residence of ancestral deities. Most souls go to i makarizeng, which is the underworld where most ancestors rest. In i tjarhi teku, the lower world, nakuyakuya tsemas (bad spirits) and qaqetitan (ghosts) linger. These spirits cannot go to the world where ancestral spirits rest, but must perpetually wander in the wilderness, in an eternal state of hunger and thirst.

Funerary forms vary greatly according to the cause of death, and Tjauqau residents distinguish two major categories: ‘bad death’ and ‘good death’. The former includes death on the road (car or motorbike accidents), falling from a tree, drowning, being bitten by a poisonous snake, or being struck by rocks. The worst forms of ‘bad deaths’ include: suicide; the death of the victim of manslaughter or murder; the death of a murderer; death in pregnancy or childbirth, so called ‘death of sin’ (Tang 1973:21-28). People believe that those who have had a ‘bad death’ will inevitably become ghosts and malicious spirits. As a result, shamans often refuse the demand from the bereaved family to perform funeral sacrifices, fearing that misfortunes will fall upon themselves (Hsu 1994: 122-3). This circumstance sometimes leads the bereaved family to turn to other religious resources. The bereaved family in Tjauqau, for example, are likely to contact the pastors in Laliba if their shamans let them down.

Dying in the house by natural causes, in contrast, is normally regarded as a ‘good death’. The elderly who die in their own houses of old age or prolonged illness, in the company of children and grandchildren around the deathbed, is a typical example of a ‘good death’. Dying in the hospital is not conceived of as a ‘good death’, so patients who are about to die will be sent home to take their last breath. If the patients are too weak to breathe by themselves, or have already stopped breathing, a life support machine will be employed to sustain their breath until their bodies are at home. According to local belief, those persons who die ‘good deaths’, either in a natural way or outside human control, will become ancestors. In order to lead the soul of the deceased to rest in the land of the ancestors, a series of death rituals have to be carried out according to local customs. The following is a brief account of the succession of mortuary rituals to be conducted in the case of a ‘good death’.

When a dying person breathes his/her last breath, close relatives, normally affines, wash and put new clothes on the body. Today, a long refrigerator, provided by Han-Chinese undertakers, is widely used to store corpses and slow down the speed of deterioration. Once the corpse has been moved into a refrigerator, it is kept in the living room of the house of the bereaved, and concealed from mourners by a white linen curtain. The face of the dead can be seen from a small window of the refrigerator, but most of the time the window is covered by a towel, unless the mourners want to look at the deceased’s face. The corpse is surrounded by several elders from the natal houses of the dead and his/her spouse, and the natal houses of the deceased’s parents and his/her spouse’s parents. During this time, a temporary soul altar (palisian) is set up by the deceased’s firstborn. The altar, basically a table with offerings, is located in front of the refrigerator, but on the other side of the curtain, so that every mourner can see it directly from the front yard of the house. An enlarged, black-and-white photograph of the deceased is put in the center of that table, along with an incense burner in front of the photograph and two candles on either side. Offerings of food and drink, such as cooked rice, pork, fish, and two cups of millet wine are placed around the incense burner. The heir bears the duty of looking after the soul altar. S/he should replace the food with new food at meal times, present offerings, talk to the soul of the dead, and watch the candles and incense so that they do not extinguish. During the wake, mourners will come and approach the soul altar to make incense and wine offerings, then enter the house of the bereaved to see the deceased’s face and comfort the bereaved family. The length of the wake varies according to the social rank and social status of the dead. Wakes for aristocrats and chiefs are longer than those for commoners, and a prominent official or politician can have a wake lasting up to several weeks.

The corpse, however, has to be removed from the house of the bereaved and buried in the graveyard, and the soul has to be sent off to Naqemati no matter how long the wake lasts. The rite of seeing off (pusau) ritualizes the living’s gesture of bidding the last farewell to the dead. During this rite, everything is arranged according to a clear spatial order, and everyone behaves according to a strict code of conduct.7 The corpse is moved to a coffin, which in turn is carried to the front yard and placed between a formal soul altar decorated with layers of flowers, in front of which is a big offering table. The spouse of the deceased is seated beside the coffin, while children, children’s spouses and grandchildren kneel down around the coffin, with boys on the right hand side and girls on the left. The elders from the natal houses stay inside the house of the bereaved. The mourners sit on chairs lined up in the front yard facing the soul altar, where the photograph of the deceased stands in the center. They are organized into small groups to make offerings and perform the etiquette of kneeling down or bowing, depending on their relationships with the deceased. The entire session of offering-making begins with the immediate family of the deceased (ta tsekelan), which includes the deceased’s children, grandchildren, and their spouses. It is followed by kin who are classified by the idiom of siblingship.8 The official of the highest rank present launches the subsequent part of the offering-making by non-kin, known as ‘Public Offering’ in contrast with the former part, known as ‘Family Offering’. The highest official is followed by civil servants, colleagues, associates, and friends of the deceased from throughout his life.

After all acts of offering-making have been finished, the focus of attention shifts to the coffin and the corpse inside, and the theme of separation emerges rather bluntly. All participants are given one last chance to view the deceased’s face before the coffin is permanently sealed. The mourners line up to throw flowers on the coffin one by one, while the bereaved family bend their bodies around the coffin and wail over the corpse. Female relatives of the deceased display a particular reluctance to let the deceased vanish from their sight, and the male undertaker and his assistants have to fight with them to seal the coffin. The undertaker always wins the battle by force, when the women exhaustedly relinquish by fainting and other means. The coffin is then carried by male cousins of the dead to the graveyard on the outskirts of the village and laid down in the grave dug by villagers. The personal belongings of the deceased are dumped into the grave, while eulogy scrolls and flower baskets used in the funerals are burned at the graveside. The bereaved family makes the last incense offering, the mourners do the last bow, and then they leave the deceased completely alone. Shamans pick up some thatch, tie it, and put it on the ground of the graveyard’s exit. When crossing the tied thatch, everyone spits backwards several times to prevent the deceased’s soul from following the steps of the living to return home.

The long procedure of dealing with the dead does not end with the burial; rather, the important and difficult work of sending off the dead to the land of ancestors and causing the dead to return has just begun. The house of the bereaved has to prepare at least two pigs to be sacrificed, as well as money and gifts for the shamans, priests, and chiefs who carry out the sacrifices. The participants in these ritual events are normally confined to close relations, up to the kin category of the second cousin (marhe sika tjarlaita). Early the next morning of the burial, the first sacrifice (patjuaia) is held. The primary action in this rite is giving the deceased the property that belongs to him/her, which is enacted in a symbolic way. A shaman uses a ceremonial knife to cut a piece of new cloth into the size of a palm, and to scrape the surface of every object in the house of the bereaved, including the sofa, TV, telephone, the bedroom furniture and the cookery in the kitchen. The shamans assemble all these bits and pieces of possessions and wrap them in ritual leaves that are used especially for containing offerings to the dead. In the company of a priest carrying a long knife, the shaman walks through the forest and goes to the vicinity of the graveyard. She places this bundle of symbolic property at a specific ritual site, and prays to the deceased that s/he has been given everything and has a larger portion than the bereaved ones have, so s/he should be satisfied and want nothing more.

From this rite to the next, which is also the last rite, there is a month interval. During this period, the bereaved family are permitted to go to work or visit relatives, but usually they are absorbed in the mood of mourning, and the natal houses of the deceased or the deceased’s spouse will invite them, in a half-coercive manner, to visit their houses now and then. In the last rite, kirasudj, the theme of reintegration is prominent throughout. This rite begins with the slaughter of a pig, the most often used sacrificial animal. For followers of ancestral customs, the body of a pig can represent different kinds of relationships with spirits and kin, and at the same time bind all these relatedness in ‘one body’. Hierarchical order can also be objectified through the division and distribution of the sacrificial meat (see Tan 2001a:138-9). A special portion, the neck cut into a circle, is presented to the chiefly house as a tribute (vadis), while the tail, the right front leg and the diaphragm are presented to the shamans as payment.

The major task of shamans in this rite is to invite the ancestors of successive generations to come by presenting them with a feast of sacrificial meat and wine. Shamans perform a seance (malada) to summon the ancestors, whose names and numbers vary according to context. Normally, the list will include the founding ancestors of the village, the ancestors of the chiefly house with which the house of the dead associates, and the deceased grandparents and parents (marhe tja vuva, marhe tja alak) of the bereaved family. With the assistance of other shamans, one shaman enters into trance and is possessed by ancestral spirits who speak through her mouth.9 There is a pattern of interaction between the ancestors and the participants, their descendants. Ancestors will first ask for a share of the feast, and then make lengthy moral lectures. Their descendants will feed (pakan) them by making wine and meal offerings, then respond to the ancestors’ commands and make requests for protection and blessing. Through the mediation of the shamans, this becomes an occasion for renewing the reciprocal commitment between ancestors and their descendants (Fortes 1987:296-7). Moreover, it helps to negotiate the gap between the ancestors’ expectations and the descendants’ conduct, and offers a chance for subsequent reconciliation.

The last spirit to appear is that of the recently deceased, whose appearance constitutes the most dramatic moment of the entire seance. The shaman falls into total silence for a few seconds, then collapses slowly or suddenly. Upon seeing this image, the bereaved family members react with their full emotions. Some rush to support the falling body, some try to resuscitate the seemingly unconscious shaman/deceased, while some cannot do anything but cry uncontrollably. A short conversation is then held between the shaman/deceased and the bereaved family. The deceased murmurs that s/he has met his/her parents and grandparents and is going to live with them. S/he does not want to leave the living, but this is his destiny (sepi), the will of Naqemqti. The bereaved will reply that they miss him and often dream about him. They will take care of his unfinished business and look after his children, so he can leave in peace. For the bereaved family, to see and to talk to the deceased through the mediation of shamans is a great comfort and relief. Their mourning may be liberated, and they can drink and eat without grief from that point on. After the seance, the sacrificial meat is cooked, served, and an interval feast begins. Everyone shares food and drink with the bereaved family. Remains of the bones are collected for the rite of papu luqem (strengthening power) in order to strengthen the power, strength and mutual bonds of the bereaved family.

After all ritual leaves have been wiped away and the payment to the chief and shamans has been settled, the final communal feast in the courtyard of the bereaved’s house begins. In addition to kins and affines, almost all the villagers are invited to join in. The chief and shamans will be invited to make some speeches first. They summarize ancestors’ promises and instructions given in the seance, and insist that they are still useful in contemporary life. Overall, reconstitution of the traditional hierarchical order is a central concern throughout the funeral sacrifice. This order is objectified in the division and distribution of sacrificial meat, and in the order of appearance of the ancestors in the seance. The chief and shamans, key figures in the traditional sacrificial system, occupy center stage. Their authority is respected, and their superiority in the social hierarchy is recognized and reinforced. It is a ritual-political frame which reverses the earlier funeral rites, in which traditional authorities are marginalized by civil servants and officials. In the communal feast, the village head and civil servants will follow the lead of the chief and shamans and support their leadership in preserving and revitalizing palisi and kakudan, which are proclaimed to be ‘the true way of the Paiwan’. The behaviors of Christians are evoked as the common target of mockery, and Christianity is criticized as a morally inferior way of life. Christian funerals, in particular, are condemned for betraying their commitment to the ancestors and forgetting the origin of the Paiwan. This rhetoric is also politically charged, because Tjauqau and Laliba residents compete fiercely for the post of the village head in local elections.

Funerals among Presbyterians

Laliba Presbyterians believe there is only one God (Katsemasari), and to worship any spiritual beings other than God is tantamount to committing the sin of idolatry. The implication is that Presbyterians should not pray or speak to the deceased, or do anything which resembles worshipping the deceased. Their attitude is shaped to some degree by church policy on the national level. In terms of the church hierarchy, the Laliba Church is a constituent part of the Eastern Paiwan Presbytery, which in turn belongs to the General Assembly of the Presbyterian Church of Taiwan. The church authority recognizes that Taiwan is a multicultural society, and that every ethnic group has a unique understanding of death and funerals. Therefore, local churches are permitted to design their own funerals according to their cultural symbols, on the condition that they adhere to the basic doctrine. In any event, they should refrain from paganism and superstitions, which are still prevalent among Han Chinese and aboriginal groups.

It falls on the shoulders of local church leaders10 to put these principles into practice. First and foremost, they have to draw a line between acts of pagan superstition and acts of genuine faith. The major issues which constitute the central part of the doctrine concerning death are: what happens to the deceased when s/he dies? and what is the relationship between the living and the dead? These issues are also the most troubling to the bereaved family and ordinary believers. In response to their inquiries, the church leaders often quote the ‘Jesus’ parable of the rich man and Lazarus’.11 They explain that there is an unbridgeable gap between the world of the dead and the world of the living, so the dead cannot return to the world of the living and the living cannot interact with the dead. Even if the dead have some needs and make some requests, the living cannot satisfy them; conversely, the dead cannot respond to the requests from the living. Hence, there is no need for the living to worry that the dead will be angry at them and bring them misfortune. Through these idioms and metaphors, church leaders are able to concretely define the right attitude toward the dead. The living should remember the dead in their hearts, particularly their good deeds and virtues, which provide examples for their descendants. However, there is no point in speaking to them or offering them incense and food.

Presbyterians are not inclined to distinguish ‘good death’ from ‘bad death’, though they do feel uneasy about accidental and violent deaths. The forms of funeral services remain the same regardless of the cause of death. Most Presbyterians, either seriously ill or badly injured in accidents, are treated in the hospitals established by Christian organizations before they reach the end of their lives. To preserve the last breath of the dying to be taken at home is regarded as a custom rather than a superstition, so the dying are usually sent back from the hospital to die at home. The procedure of setting a white linen curtain, washing the body, changing the clothes and placing the corpse in the refrigerator is similar to the Tjauqau’s ritual, because all actions are directed by the same Han Chinese undertakers. A table is also set for the enlarged photo of the deceased, but the objects placed on the table are different: in the Presbyterian case, there are only two candles, a Bible and a hymn book, and no incense burner, food or wine offerings. The custom of holding the wake is preserved, but the influence of Christianity is obvious. The pastor and elders come to lead the prayer meeting each evening, the purpose of which is to ask for God’s mercy to grant the deceased a peaceful rest, as well as for the Lord’s love to comfort the bereaved family. A small amount of money offerings are collected to aid in the preparation of the funeral. The generous provision of alcohol by the bereaved family is characteristic of a wake in Tjauqau, but here drinking wine is discouraged, and its provision is limited. This reflects the ascetic policy of the Presbyterian Church in general. The duration of the wake varies according to the social and church status of the deceased. The wake of a pastor or an elder will be longer, but usually it lasts no more than one week, followed by the Farewell Service.

First, in the early morning, the Coffining Service is held in the house where the corpse has been laid out. A small group of the congregation continues to sing and pray, and the pastor concisely explains that there is no need to fear death because God has prepared a home in heaven better than the home on earth for the deceased. Separation from the dead is only temporary, because all who believe will meet in heaven sometime. After this speech, the children and relatives of the deceased move the corpse into the coffin, and the undertaker begin to redress the corpse and make up its face. The bereaved families put some clothes and personal belongings s/he treasured around his/her body. Presbyterians are careful to specify that those objects are not for the journey of the deceased’s soul to heaven or the other world, because there is no such journey. It is simply for the peace of mind of the bereaved family, so that they will not be too upset upon seeing these things.

The Farewell Service is normally held in the front yard of the house of the bereaved family.12 For Presbyterians, the Farewell Service is the most important rite in all death-related services. It is a solemn occasion for expressing ‘fear and adoration of the Lord of life’ and ‘respect for the deceased and the ethics of filial piety’. Through Christian worship and prayers, the sense of loss is expected to be transformed into a sense of hope, and the sense of fear into a sense of peace. The church leaders are especially careful to distinguish pagan practices from Christian behaviors on this occasion. What Tjauqau adherents of traditional religion do is often criticized as being pagan, so believers are warned not to follow those practices. The bereaved families are instructed not to kneel down, hold incense or worship the deceased; mourners are instructed not to bow or speak to the deceased. Any act of making offerings is forbidden, so there is no ‘Public Offering’, which is an indispensable part of Tjauqau funerals. There is indeed one moment in the service, nevertheless, when all participants stand up in front of the picture of the deceased. This is the moment of ‘Silence’, during which everybody stands up, remembering the deceased and praying for the bereaved family. This simple act transmits the core message of the Farewell Service, that is, to remember the deceased rather than worship him/her.

Local Presbyterians often confront the tragic death of their relatives and neighbors. In grief and bewilderment, they often wonder why God let those unfortunate things happen to them. During the Farewell Service, the pastor will deliver a sermon to construct a biblical view of how to understand death in general and how to grasp the meaning of the death event in particular. A sermon serves as a kind of medicine to heal their broken hearts and to restore their damaged belief. The pastor will remind them of the promise of God that whoever believe in Jesus shall not perish, but shall have eternal life. Christians believe that Jesus died on the cross, and was resurrected three days later, so Jesus has defeated the dominion of sin and death. The speaker then draws the attention of the congregation to the specific context of the death event. It is emphasized that we should not look at this death event from a worldly view point, and not follow the pagans in distinguishing ‘good death’ from ‘bad death’. As far as the deceased is concerned, death is indeed a peaceful rest, a relief from the worries and burdens of the world. Upon reflection, we will be amazed to find that either the timing of death or the means of death will reveal the good and perfect will of the God Father (Kama Tsemas). Then, the congregation is led to remember the good deeds of the deceased that exemplify Christian virtues. The speaker emphasizes that his/her example of good faith should be remembered and followed by his/her descendants.

Through a succession of hymns, prayers, and sermons, the service is conducted to construct a shared idea and feeling that no matter how the deceased died, s/he will rest peacefully in heaven, and they can bid farewell without regrets or worries. At the end of the Farewell Service, mourners line up to see the face of the deceased for the last time. Before sealing the coffin, one of the elders leads a prayer, while the bereaved family gathers around the coffin to say goodbye to their beloved with intense emotions. The church leaders, trying to calm the bereaved family, say a prayer for raising up the coffin and lead the procession to the graveyard. After the coffin has been placed in the grave and the congregation has regrouped, the Burial Service is held. The pastor explains that human beings are created by God from earth, so burial is a return to the beginning of creation to be with God, then he throws a handful of earth on the coffin, saying, ‘ashes to ashes, dust to dust’. The bereaved family are asked to do the same, which is interpreted by the pastor as a sign of approval for the burial to proceed as well as a sign of not including any superstition in the burial. The service is concluded by a preaching based on Jesus’ words: ‘I am the resurrection and the life’. It is proclaimed that for Christians, death is not the end of life, because when Jesus comes again, he will judge the world and raise the dead. After prayers, the congregation leave the graveyard silently. As soon as they return to the house of the bereaved, the church leaders lay their hands on the heads of all members of the bereaved family, praying for God to give them strength and courage to overcome this difficult time, and wishing them success and prosperity in the future. The form and the purpose of this prayer is in some ways similar to the rite of papu luqem practiced by shamans at the end of kirasudj. Presbyterians also hold a communal feast at the end of all death-related services, beginning with the wake and ending with the burial. However, the wake does not end immediately after the burial. The wake is held for three nights, and church members continue to visit and comfort the bereaved family.

As I have mentioned, the sacrificial rite of kirasudj is held one month after the burial by adherents of traditional religion. Somehow influenced by this custom, local Presbyterians had held a Memorial Service one month after the burial up until 1993, when a young pastor changed it to forty days after the burial. This is a further step in the Presbyterian Church’s efforts to reform death rituals. Through making a clear distinction between the Christian funeral service and the traditional funeral sacrifice, church leaders are able to discourage believers from old, pagan ideas and practices and to implement the new, Christian way of dealing with death. This change finds its justification in the Bible, wherein it is said that after Jesus has resurrected, he appeared to his apostles after a period of forty days, then ascended to heaven (Acts 1:1-11). Without elaborating too much on the theological significance of these verses, church leaders point out that the reason for choosing ‘forty days’ is biblical, and explain that the Memorial Service is not only to remember the recently deceased but also to remember Jesus. However, ‘forty days’ is sometimes considered by believers to be the time interval for the deceased to ascend to heaven, therefore an implicit concern in the Memorial Service is to confirm whether the deceased has come to rest in heaven.

The Memorial Service is defined by the church authority as an occasion for fulfilling Christian filial piety through the public act of remembering the deceased. It is stressed once again that the purpose is to worship God rather than to worship the deceased, so conducts of paganism such as contacting the spirit of the dead or summoning it to come back are not allowed. The descendants of the deceased are reunited for some activities in the name of remembering him/her. They will go to the graveyard and visit the newly completed grave bearing the inscription of the deceased’s name and his/her picture. They make offerings of flowers instead of making food and incense offerings, and they sing and pray under the guidance of church leaders. After returning home, they hold a service which is open to all church members. During the interval of hymns and prayers, the congregation share stories about the deceased, and sing his/her favorite songs. Usually, the bereaved family is moved, and they express their attachment to the deceased. Then, some relatives and other families who have dreamt about the deceased since s/he has died share their dreams. Church leaders or church members experienced in interpreting dreams try to explain these dreams according to biblical principles, while the audience tries to figure out what they reveal about the state of being of the deceased. For example, if someone dreamt the deceased wore white and walked on an ascending ladder, it is usually interpreted as the deceased having reached heaven. Dreams which indicate the deceased is in a ‘good’ state are spoken out more often, for the peace of mind of the bereaved family. The Memorial Service ends with a communal feast. Through sharing a kind of round millet dumpling with pork inside (ginavu), participants reconsolidate their kinship and social ties, and the bereaved are liberated from their mourning and reintegrated into the church family in a joyful atmosphere.


From the above, we can see that Tjauqau adherents of traditional religion maintain a dual view of death. In some ways, death is regarded as an instant event, occuring when the breath stops and the body becomes cold and hard. The consciousness that death occurs at a specific time is reinforced by modern medical institutions. As previously mentioned, it is common among Tjauqau residents to use a life support machine to maintain the breath of near-death patients in the hospital. The machine is not turned off until the dying person is carried back, placed in the house and surrounded by family members. Then, a nurse disconnects life support and announces the exact time of death. To place the body in a refrigerator after death has been announced confirms the body as a lifeless object, and indicates that the irreversible process of deterioration has begun. The boundary between life and death is clearly drawn. In other ways, death is regarded as a continuous process, and the boundary between life and death is blurred. The adherents of traditional religion feel that they have connections and associations with the dead, and these links and ties do not end suddenly, but instead gradually fade away. For the bereaved family, these links are so strong that they have no way to get rid of them totally except to transform them into a new form of relatedness. This is reflected in their efforts to set up a soul altar during the mourning period and to devote themselves to tending to the needs of the deceased’s soul. They talk to it and feed it, as though the deceased were capable of listening and eating. The soul altar is an objectification of the unseen soul, and the focal point of social relations of the deceased with the mourners. In front of the altar, the mourners make offerings, in order to recognize that their links with the deceased have not yet been broken.

Furthermore, in order to transform a dead person into an ancestor, a spiritual being, the body element, which decays, has to be disposed of, while the soul element, which can survive without the body, has to be preserved. The funeral thus unfolds as a dual task of separating the corpse from the community at an earlier stage, while nurturing the continuing relatedness with the soul and causing it to return to the house it belongs at a later stage. In the first task, the separation of the corpse from the community is also a process of individuation of the person. Though the corpse is physically inactive, it is still perceived as a part of the person upon which kinship ties are located. As a result, it is surrounded by the bereaved family, who show their grief and devotion toward it. The corpse is also the site of a complex of social links the deceased has constructed throughout his life, so the mourners pay their respect to it. All these ties and links, however, have to be cut off from the corpse, which can then be treated as a non-living thing and buried in the graveyard. What we have seen in the rite of seeing off is a procedure of individuating the deceased through drawing boundaries. The living organize themselves as bounded groups according to their relationship to the deceased. The deceased, however, is no longer a part of each bounded group, but separated from all of them by the coffin, and differentiated from them by acts of offering-making. The separation and individuation reaches its pinnacle when the corpse is buried in the graveyard and left alone. In this sense, the corpse objectifies the part of the deceased that is a ‘bounded individual’.

The soul, by contrast, does not decay like the corpse, but undergoes a transformation. For adherents of traditional religion, the soul has to be sent off to the land of the ancestors so it can become one of them, but these adherents also wish to maintain a relationship with the soul so that it can be summoned to return. The rite of giving pieces of property to the deceased after the burial is crucial to this purpose. The meaning of this rite has to be understood in a temporal dimension. On the one hand, this rite is a gesture of separation from the dead by dividing property in the present; on the other hand, this rite is also a gesture of expecting the dead to return. By giving possessions of the house symbolically, this rite not only defines an orientation toward the deceased’s soul on its journey to the land of ancestors, but also initiates an exchange relationship with the dead. He is expected to return with gifts and blessings to his descendants after he settles in the land of the ancestors.13

In brief, the belief and practice involved in transforming the dead into ancestors in Pawian traditional religion is to assert that death is not a total annihilation, but a transition to regeneration and reproduction. In the final sacrifice, the deceased’s soul has been encouraged to return, and the reciprocal commitment between the living and the dead has been confirmed. In this sense, the returning soul of the deceased, with a trace of ancestorhood, represents the part of the deceased that is an ‘unbounded person’. S/he continues to ‘live’ as an inalienable member of the house, and the descendents of the house will continue to worship him/her.

Laliba Presbyterians, in contrast, view death predominantly as a punctual event rather than a continuous process. At the moment of death, the soul of the deceased departs his body and can rest in heaven regardless of the ritual actions of the living. Death creates an unbridgeable gap between the living and the dead, and their communication becomes impossible and pointless. There is no journey for the soul of the dead, so the funeral is oriented toward the comfort of the living rather than the benefit of the dead. In terms of the dual task of funerals, a Presbyterian funeral focuses on burying the individual, represented by his body, and sending the dead out of the social world. The second part of the dual task, which would cause the holistic element of a person to survive on earth, has been virtually suppressed by the church authority as a result of their intolerance of traditional funeral sacrifice. The ceremonial investment of Tjauqau followers of traditional religion, from offering pig sacrifice to the shamans’ seance, is criticized as being not only a waste of resources but also ‘idolatrous’.

This does not mean that Presbyterians feel they have no relationship at all with the dead. They feel as much attachment to the deceased as Tjauqau adherents do, yet they are not encouraged to express this sense of attachment through material forms or public rituals. Instead, they maintain an intimate relationship with the deceased in the realm of inner emotions, memories and dreams. The Farewell Service, in particular, emphasizes remembering the dead as its major theme, and is conducive to constructing a collective version of memories about the deceased. Remembering the deceased, however, should not lead to worshipping the deceased as an ancestor, so kneeling down and bowing to a picture of the deceased are forbidden by the church authorities. This is to ensure that in the relationship between the living and the dead, the living are agents while the dead remain patients, the target of the living’s agency (Gell 1998:21-3). The dead are supposed to be the object of remembrance and forgetting; they cannot bring fortunes or misfortunes to the living, nor can they play a significant role in human affairs.

Nevertheless, the dead come to talk to the living in their dreams, and the living cannot decide whether they would like to have dreams or not, or when and how the dead will arise in their dreams. Sometimes, the bereaved family will desire to dream about their beloved deceased because they miss them very much, especially after the burial and the Memorial Service, but the living are often disappointed to find that the dead do not have words for them. This shows that local Presbyterians still have the profound need to cross the boundary between the living and the dead and to communicate with the deceased. The church authorities tend to regard dreaming about the deceased as biblically sanctioned, as long as believers do not explain their dreams in terms of ancestor worship. They even encourage the bereaved or other believers to share their dreams in the public gatherings of the Memorial Service and anniversary. On these occasions, though the deceased do not return in a physical and visible form, they do return and regenerate in people’s minds through words, memories and mental images.


I have tried to explore the dynamic relationship between traditional religion and Christianity by juxtaposing the ethnography of two types of Paiwan societies. 1 have tried to relate the ritual practices of funerals among Tjauqau adherents of traditional religion and Laliba Presbyterians to their concepts of the person. The former envisions a dual view of death, and the funeral is an important cultural institution that deals with the end of individuality, as well as the continuity of the social personhood of the deceased. The case of Tjauqau thus supports Bloch’s thesis quite well with respect to the dual task of funerals in a society without world religion. The latter, however, does not appear to support Bloch’s thesis with respect to a society with a world religion. I have shown that funerals in Laliba focus on burying the individualistic side of the person, and are not concerned with perpetuating the holistic side among social groups. Bloch is right in pointing out that world religions primarily deal with the individualistic side, and that the holistic side involves religious practices which contrast with those of world religions. Yet, in contemporary Laliba, no such religious practices can be found as a historical consequence of conversion to Christianity. Even though there are such religious practices available in the neighboring settlement Tjauqau, Presbyterians are not inclined to adopt them because of their hostile attitude toward traditional funeral sacrifice. This example suggests that in a society that practices a world religion, we cannot assume that the indigenous religion remains alive, active or co-exists with the world religion in a harmonious and cooperative way.

Does the example of Laliba also suggest that Presbyterians are individualistic and do not have the holistic concept of the person? I maintain this is not the case. Though Laliba Presbyterians recognize that each of them is an individual before God, in their social lives on earth, they think of themselves as part of their houses and members of the church family. Both the house and the church are regarded as the whole to which they belong. The following question remains: why, in Presbyterian funerals, is the holistic side of the person suppressed, whereas in traditional funeral sacrifices, the holistic side is stressed? I think we have to recognize that two concepts of the person are articulated in the two types of funerals in different ways. In Presbyterian funerals, the individualistic concept of the person is placed in a superior status and expressed explicitly in public by stressing the boundary between the living and the dead, conversely, the holistic concept blurs this boundary, and it is relegated to the relatively private realm of memories and dreams. In traditional funeral sacrifices, the individualistic concept of the person is almost denied by calling the spirit of the deceased to reunite with the living. The breaking of this boundary through the power of shamans is a public spectacle and is celebrated as a quintessential symbol of tradition.

The controversy over funerals, therefore, has to be understood in terms of the differences in their ways of articulating two concepts of the person. I suggest in both Laliba and Tjauqau communities, we can find that two versions of the person co-exist and are of the same importance in everyday life. It is political-religious organizations, namely the Presbyterian church and the traditional sacrificial system, that value one over the other and construct the hierarchical order of articulation. The Presbyterian church defends the fundamental belief in only one God and renounces ancestor worship. Their funeral services are designed to disengage with any ancestral spirit by maintaining the boundary between the living and the dead. The concept of ‘bounded individual’ is thus put forward, while the concept of ‘unbounded person’ is suppressed. The traditional sacrificial system, in contrast, views communication with ancestral spirits as the primary obligation. Thus, the concept of ‘unbounded person’ is played out conspicuously and vividly in funeral sacrifices.

Can this essay contribute to the study of the relationship between traditional religion and Christianity in the Pacific? Tonkinson points out that an important variable relating to Christianity and kastom is whether or not Christian and kastom groups co-exist in the same area, and that historical circumstances are important in influencing the tenor of relationships between the two kinds of groups (Tonkinson 1982:303). This essay has provided an example of a Christian group and a tradition-oriented group co-existing in the same area, keeping kinship and marriage ties, yet still in strong disagreement on the matter of funerals. For a more precise understanding, their relationship cannot be defined as simply harmonious or antithetical, but alternately harmonious and antithetical in different aspects and contexts. Funerals constitute a major field of controversy and antagonism in their social and religious lives. I have proposed some explanations from the perspectives of cultural politics in Taiwan, the competition between religious organizations, the contradiction in the basic doctrines of the two groups and most importantly, the differences in their articulation of concepts of the person. Perhaps further investigation of the history of evangelization and experience of modernity in Taiwan is needed for a fuller explanation.

In her study of the relationship between Christianity and a persistent cargo cult in Misiman island, a southern Massim society, Macintyre concludes that we have to examine and explain not only the persistence of particular conceptualizations and beliefs, but also their coexistence as dynamic, but often contradictory cultural forces (Macintyre 1990:100). Through the ethnography of funerals in Tjauqau and Laliba, I have demonstrated the coexistence of two concepts of the person, and I propose that the different ways of articulation of those concepts are cultural forces which shape the dynamic, but often contradictory relationship between tradition and Christianity.


Fieldwork in Paiwan villages between 1997-1998 was sponsored by the Taiwan Ministry of Education. The Institute of Ethnology, Academia Sinica, provided the institutional support and funding for my postdoctoral research in 2002. An earlier version of this paper has been presented at the ‘International Symposium on Austronesian Cultures: Issues Relating to Taiwan’, held by the Institute of Linguistics (Preparatory Office), Academic Sinica, from 8th to 12th, December, 2001. I would like to thank Heather Horst, Ying-Kuei Huang, Shelly Errington, James Fox and Janet Garsten for their comments and encouragement. I am also grateful for the valuable suggestions made by one referee and by the editor. Finally, I thank my wife, Shu-Yuan Yang, for her continuing support.


1. Formosan languages are of special interest to Austronesian historical linguists, and recently some of them have begun to regard Formosan as key to the reconstruction of Proto-Austronesian (Pawley and Ross 1993). Historically, there were twenty to thirty Formosan language (ethnic) groups living in Taiwan (Li 1997). With the increase of Han immigrants from southeast China from the 17th century, many aboriginal groups were assimilated into Han Chinese culture. Now, there remain about ten aboriginal groups with distinctive linguistic and cultural traits.

2. In the church history of Taiwan, the first Protestant missionary, sent by the Reformed Church of Holland, arrived in Taiwan in 1627. This was part of the colonial project of the Dutch East India Company. The Dutch, defeated by a Chinese general loyal to the Ming Dynasty, retreated from Taiwan in 1661, after which there was no further evangelization for the next two centuries. In 1865, the mission of the Presbyterian Church of England commenced evangelizing activities in southern Taiwan, while the Presbyterian Church of Canada started its work in northern Taiwan in 1872. The Presbyterian ministers began to show an interest in spreading the Gospel among aborigines in the 1910s, but the Japanese colonial government, in fear that the propagation of Christianity might incite aboriginal rebellion, tried to block the missionaries. Evangelizing activities continued on a small scale, despite severe persecution. Soon after the end of the World War II, Presbyterian missionaries worked intensively in aboriginal societies, and the growth of Christianity was so rapid that some missionaries called it a ‘Twentieth Century Miracle’ (Vicedom 1967, Huang 1996:425-6).

3. This can be compared with Sinclair’s study of contemporary Maori funeral rituals, known as tangi-hanga ‘mourning ceremonies’, which have become an important constituent of Maori identity (Sinclair 1990).

4. On this issue, Fortes is much influenced by Radcliffe-Brown, who has clearly distinguished the ‘individual’ and ‘person’ as the following:

‘Every human being living in society is two things: he is an individual and he is also a person. As an individual he is a biological organism…Human beings as individuals are objects of study for physiologists and psychologists. The human being as a person is a complex of social relationships…As a person the human being is the object of study for social anthropologists’. (Radcliffe-Brown 1940:193-4)

5. Obviously this is a critique of scholars who endorse this great divide, such as Dumont (1985). This critique is echoed among scholars of Indian as well as other South Asian societies. For example, McHugh (1988:77) argues that while relatedness is of central importance in South Asia, it does not preclude the concept of the individual.

6. Hsu’s work (1994) is of particular relevance to this paper. She has recorded funeral customs and has discussed how the concept of the person is embedded in death rituals based on her fieldwork in Kulalau, a village in Central Paiwan. This village is widely regarded as a representative of the type of Paiwan society which preserves traditional customs well. I sometimes use her work to illustrate the ideas and practices of adherents of traditional religion.

7. Notably, the spatial order and the code of conduct are not necessarily of indigenous origin, but the result of borrowing and appropriation of the Han-Chinese funeral customs. Thus, this part of the funeral is not ‘traditional’ in a strict sense; perhaps ‘syncretic’ would be a better term, because it is a mixture of Paiwan and Han-Chinese customs.

8. The first sibling group is marhe kaka, siblings who share the same parents with the deceased, followed by marhe sika tsekel, first cousins who share the same grandparents, marhe sika tjarlaita, second cousins who share the same great grandparents, marhe sika tjatjevel, third cousins who share the same great-great grandparents, and finally, marhe kasusu, all cousins who are beyond the above categories. The last category can almost include all villagers.

9. Paiwan shamans are exclusively female.

10. The local Presbyterian church is governed by its ‘small meeting group’ (‘shiao-hui’ in Mandarin), which is made up of elders and their appointed pastor. The elders, four or six in number, are elected by church members every three years. My use of the term ‘church leaders’ includes both the pastor and the elders.

11. In the Gospel according to Luke, there is a story of a rich man who dressed in purple and lived in luxury, and there was a beggar, Lazarus, covered with sores and longing to eat what fell from the rich man’s table. The time came when the beggar died and the angels carried him to Abraham’s side. The rich man also died, was buried, and was tormented in hell. He looked up and saw Abraham far away, with Lazarus by his side. So he called Abraham to send Lazarus to dip the tip of his finger in water and cool his tongue. But Abraham replied ‘…between us and you a great chasm has been fixed, so that those who want to go from here to you cannot, nor can anyone cross over from there to us’. Then, the rich man begged Abraham to send Lazarus to warn his five brothers so that they could avoid going to hell after they died. Abraham did not grant this wish and he replied, ‘They have Moses and the Prophets; let them listen to them.’ (Luke 16:19-29)

12. In the case of a pastor or an elder, the service may take place in the church to honor their contributions to church ministry.

13. This cultural idea is clearly reflected in the Maleveq, a ceremony held every five years. During this greatest Paiwan ceremony, successive generations of ancestors are all invited to return to visit their descendants, while descendants ask for good luck and prosperity from their ancestors through the activity of impaling rattan balls (see Chiang 1993, Hsu 1994, Hu 1999, Tan 2001a).


BARKER, J. (ed.). 1990a. Christianity in Oceania: Ethnographic Perspective. Lanham, MD: University Press of America.

1990b. Introduction: Ethnographic Perspective on Christianity in Oceania Societies. In J. Barker (ed.), Christianity in Oceania: Ethnographic Perspective, pp. 1-24. Lanham, MD: University Press of America.

BLOCH, M. 1988. Death and the Concept of Person. In S. Cederroth, C. Corlin, J. Lindstrom, (eds), On the Meaning of Death: Essays on mortuary Rituals and Eschatological Beliefs, pp. 11-29. Gothenburg: Acta Universitstis Upsaliensis.

BURT, B. 1982. Kastom, Christianity and the First Ancestor of the Kwara’ae of Malaita. Special Issue of Mankind 13(4):374-400.

CARSTEN, J. 1997. The Heat of Hearth: The Process of Kinship in a Malay Fishing Community. Oxford: Oxford University Press.

CARSTEN, J. and HUGH-JONES, S. (eds) 1995. About the House: Levi-Strauss and Beyond. Cambridge: Cambridge University Press.

CHIANG, BIEN 1993. House and Social Hierarchy of the Paiwan. Ph.D. diss., University of Pennsylvania.

CHOWNING, A. 1990. God and Ghosts in Kove. In J. Barker (ed.), Christianity in Oceania: Ethnographic Perspective, pp. 33-58. Lanham, MD: University Press of America.

DUMONT, L. 1985. A Modified View of Our Origins: the Christian Beginnings of Modern Individualism. In M. Cattithers, S. Collins & S. Lukes (eds), The Category of the Person: Anthropology, Philosophy, History, pp. 93-122. Cambridge: Cambridge University Press.

ERRINGTON, S. 1987. Incestuous Twins and the House Societies of Southeast Asia. Cultural Anthropology 2:403-44.

FORTES, M. 1973/1987. The Concept of the Person. In Religion, Morality and the Person: Essays on Tallensi Religion, pp. 247-286. Cambridge: Cambridge University Press.

1987. Endpiece: Sacrifice among Theologians and Anthropologists. In Religion, Morality and the Person: Essays on Tallensi Religion, pp. 287-301. Cambridge: Cambridge University Press.

FOX, J. 1987. The House as a Type of Social Organization on the Island of Roti. In C. Macdonald (ed.), De la Hutte au Palais, pp. 171-78. Paris: CNRS.

GELL, A. 1998. Art and Agency: An Anthropological Theory. Oxford: Oxford University Press.

HERTZ, R. 1907/1960. A Contribution to the Study of the Collective Representation of Death. In Death and the Right Hand. pp. 27-86. Illinos: The Free Press.

HSU, KONG-MING 1994. The Change of Mortuary Institutions in the Kulau Village of the Paiwan: a Discussion on the Concept of Person. In The Ritual and Culture of the Kulalau Village of Paiwan. pp. 109-168. Taipei: Dau Shiang Press (in Chinese).

HU TAI-LI 1999. The ‘Text’ and Interpretation of the Five-Year-Ceremony of the Kulalau Village, Paiwan. In C-K. Hsu, M-R. Lin (eds), The Development of Anthropology in Taiwan: Ethnographic Studies. pp. 183-222. Taipei: Institute of Ethnology, Academia Sinica (in Chinese).

HUANG, SHIUN-WEY 1996. The Politics of Conversion: The Case of an Aboriginal Formosan Village. Anthropos 91:425-439.

JOLLY, M. 1982. Birds and Banyans of South Pentecost Kastom in Anti-Colonial Struggle. Special Issue of Mankind 13(4):338-356.

JOLLY, M. and N. THOMAS. 1992. The Politics of Tradition in the Pacific. Special Issue of Oceania 62(4).

KEESING, R.M. and R. TONKINSON (eds). 1982 Reinventing Traditional Culture: the Politics of Kastom in Island Melanesia. Special Issue of Mankind 13(4).

LA FONTAINE, J.S. 1985. Person and Individual: Some Anthropological Reflections. In M. Cattithers, S. Collins & S. Lukes (eds), The Category of the Person: Anthropology, Philosophy, History. pp. 123-140. Cambridge: Cambridge University Press.

LEVI-STRAUSS, C. 1983. The Way of the Masks. London: Jonathan Cape. 1987. Anthropology and Myth. Oxford: Blackwell.

LI, PAUL JEN KUEI. 1997. The Ethnic Groups and Migration of Taiwan Austronesian Peoples. Taipei: Popular Culture. (in Chinese)

MACINTYRE M. 1990. Christianity, Cargo Cultism, and the Concept of the Spirit in Misiman Cosmology. In J. Barker (ed.), Christianity in Oceania: Ethnographic Perspective, pp. 81-100. Lanham, MD: University Press of America.

McHUGH, E.L. 1988. Concepts of the Person among the Gurungs of Nepal. American Ethnologist 15:75-86.

OTTO, T. 1992. The Ways of Kastom:Tradition as Category and Practice in a Manus Village. Special Issue of Oceania 62(4):264-283.

PAWLEY, A. and ROSS M. 1993. Austronesian Historical Linguistics and Culture History. Annual Review of Anthropology 22:425-59.

RADCLIFFE-BROWN, A.R. 1940. On Social Structure. In Structure and Function in Primitive Society. London: Cohen and West.

SINCLAIR, K. 1990. Tangi: Funerals Rituals and the Construction of Maori Identity. In J. Linnekin and L. Poyer (eds), Cultural Identity and Ethnicity in the Pacific, pp. 219-236.

STRATHERN, M. 1988. The Gender of the Gift: Problems with Women and Problems with Society in Melanesia. Berkeley: University of California Press.

TAN, CHANG-KWO 2001a. Mediated Devotion: Tradition and Christianity among the Paiwan of Taiwan. Ph.D. diss., University of London.

2001b. Building Conjugal Relations: the Devotion to Houses amongst the Paiwan of Taiwan. In D. Miller (ed.), Home Possessions: Material Culture behind Closed Doors, pp. 149-172. Oxford: Berg.

2002. Syncretic Objects: Material Culture of Syncretism among the Paiwan Catholics. In Journal of Material Culture 7(2):167-87.

TANG, MEI-CHUN 1973. A Structural Analysis of the Burial Custom and Funeral Rites of Lai-I, an Aboriginal Village in Taiwan. Bulletin of the Department of Archaeology and Anthropology. National Taiwan University. No. 33/34:9-34.

THOMAS, N. 1992. Contrasts Marriage and Identity in Western Fiji. Special Issue of Oceania 62(4):317-329.

TONKINSON, R. 1982. Kastom in Melanesia: Introduction. MANKIND 13(4):302-305.

TSAI, KUAN-HUI 1998. The Establishment of Tribal Societies and Ethnic Relations of the Paiwan Aborigines (1630-1894). MS thesis, National Taiwan Normal University (in Chinese).

UTSURIKAWA, MIYAMOTO, and MUAUCHI. 1935. Genealogical and Classificatory Studies on the Formosan Aborigines. 2 vols. Taipei: Taihoku Imperial University (in Japanese).

VICEDOM, GEORGE. 1967. Faith that Moves Mountains. Taiwan: The China Post.

Chang-Kwo Tan

Academia Sinica, Taiwan

Copyright University of Sydney, Oceania Publications Mar 2003

Provided by ProQuest Information and Learning Company. All rights Reserved