14. Worshipping Plural Ethnic Ancestors: Mechanisms of Change among the Sinicised Kvalan

Shimizu Jun

In this paper, I discuss the process of change among the Kvalan, one of the sinicised plains ethnic groups of Taiwan, and try to show that the particular mode of accepting different ethnic cultures can be traced to the character of the underlying ideological framework on which the aboriginal society concerned is based.
Cultural and ethnic change observed among Kvalan descendants who live in the village of Shinshe is a case in point. The data derive from intensive field research carried out from November 1984 to March 1986, and from supplementary research in March 1987.
The Kvalan are little known to anthropologists. To start with, this paper describes their historical situation and their traditional ideas of kinship and ritual. This is essential for a further understanding of the mechanisms of recent cultural and ethnic change.
The following discussion mainly focuses on the way in which the Kvalan practice ancestor cult rituals of different ethnic origins. The analysis makes it clear that their traditional framework is capable of absorbing different systems of ancestor worship. The discussion also elucidates the mechanisms of change in ethnic identity among the Kvalan. Kvalan of mixed descent affirm their ethnic identity through the practice of ancestor cults in accordance with their bilateral kinship ideology. As a result, they have come to hold coincident plural ethnic identities. This observation suggests that a people's traditional ideological framework may persist even though their ethnic identity has changed.

History of the Kvalan

Taiwan is currently populated by Han Chinese and a variety of Austronesian ethnic groups. The latter, amounting to 2% of the total population have been roughly categorised into two groups, those who populate the plains area and those living in the mountains or on an isolated island.[1]
The plains groups have been affected by Chinese culture since the beginning of Chinese settlement in the Ming and Ch'ing Dynasties. They were called shu-fan in Mandarin, i.e. "ripe", "cooked" or "civilised" aborigines in contrast to the sheng-fan, the "raw" aborigines. In Taiwanese (Minnan Chinese), they are known as Peipo-a, literally "plains peoples".[2]
The Kvalan are one of the plains groups who live dispersed along the central and southern coastal area of east Taiwan. They have a history of assimilation to Han Chinese, and to other ethnic groups. The relatively complicated situation of the contemporary Kvalan derives from their historical situation.
Recorded Kvalan history began when Europeans came to build military bases in Taiwan for their expanding Asian trade. Aboriginal ethnic groups of the northern coastal areas formerly had contacts with the Spanish, who had established a foothold in Keelon in the early 17th century, and then with the Dutch who pushed the Spanish out of this area to control Taiwan in the 1640s (Ino, 1973 [1904]: 55-56; Nakamura, 1938: 239).
The Dutch compiled a census of the natives in the plains area in order to administer them. They also collected taxes from them (Nakamura, 1936, 1937, 1938, 1951) According to their reports, in the 17th century, the Kvalan population was located in the Ilan Plains on the northeast coast, settling in about forty villages. Other ethnic groups defined as sub-groups of the Basai, one of the other plains aboriginal groups, also lived in some of the villages in this area (Mabuchi and Utsusikawa, 1974 [1939]: 475-476).
After the victory of Deng Cheng-kong over the Dutch in 1662, Chinese immigrants, mostly from Fujian Province, increased rapidly. These immigrants, however, colonised the vast Western Plains at first. They came into close contact with the natives of the Ilan Plain when they started to settle in this area almost two centuries later (Ino, 1973 [1904]: 167-168; 1973 [1911]: 172-174).
The first colonists who invaded the Ilan Plain resorted to arms to quell the resistance of the natives. After they subjugated the natives of the entire plains areas, an increasing number of immigrants brought the land under cultivation in the following years (Negishi, 1933).
The aborigines adopted the higher technology of intensive rice cultivation involving irrigation systems from the neighbouring Chinese farmers. Even before the Chinese settlement, they cultivated rice crops (Murakami, 1972: 282, 285-286), probably of dry-field rice in a relatively extensive way, so that it is not difficult to imagine that they accepted the higher agricultural technology of their own accord, which might have accelerated their sinicisation.
The colonists gradually forced the aboriginal people to dispose of their land by taking advantage of their credulity (Negishi, 1933). By the beginning of the 19th century when the Ch'ing Dynasty established dominion over this hinterland (Chen, 1957 [1852]), the aboriginal villages had gradually lost their original character, and had disappeared through the process of sinicisation and the emigration of the residents who sold or forfeited their paddy fields to the Chinese.
The flow of emigration of the aborigines had two different directions, toward the mountainous area, and toward the southern coastal area where some of the descendants of these emigrants still live and preserve their own language and culture (Negishi, 1933: 526). The latter built Karewan Village in the northern hinterland of Hua-lien Bay during the first half of 19th century (Ino, 1973 [1904]: 303-305). Some Turubiawan immigrants, a sub-group of the Basai, also lived in Karewan Village. These immigrants were a stabilising influence on the surrounding villages of the Ami in this area.
No sooner did Chinese colonists come into this area than their arrival led to disputes with the aborigines. The Karewan villagers formed an alliance with the neighbouring Ami and fought against the Chinese colonists. But at last the Ch'ing army defeated the aborigines in 1878 (Ino, 1973 [1904]: 618-619). After the rebellion, the Chinese established their political power in this area, while the ethnic identity of the Kvalan was weakened by this struggle. The Kvalan moved again to the south and dispersed among the Ami in the villages along the coast. The distribution of the Kvalan has not changed very much since then.
During the Japanese occupation and under the Kuo Min Tang (KMT) government after World War II, interethnic marriages have increased among the Kvalan. Cultural changes have also occurred through the influences of neighbouring ethnic groups, in addition to modernisation and Christianisation. The data from Shinshe, one of the coastal villages, indicate that the process of change consists not only of the simple process of sinicisation, but also has more complex features.

Location of the village

Shinshe Village is located in Fongping Xiang, Hoa-lien Prefecture. The village is situated on a narrow terrace between the steep slopes of the Coastal Mountains and the Pacific Ocean. Though there is only a small amount of flat land around the village, several small streams supply sufficient water to cultivate rice crops in the terraced paddy fields spreading out over the slopes of the foothills.
The villagers mainly cultivate rice crops. The yields of other crops are not sufficient to be sold outside the village. The small amount of land around the village has limited the expansion of agriculture. Recently, younger people have often gone out to work in the large cities to increase their earnings.
During the Japanese Occupation most of the villagers were also engaged in coastal fishing apart from agriculture. The coastguards currently keep watch on the seashore, so the villagers cannot use the bay as a port, and fishing as an economic activity is stagnant. Shinshe has maintained the largest Kvalan population since the dissolution of Karewan Village in 1878. Shinshe literally means new village or new community in Chinese characters. It is called Patrogan by the residents.[3]
The area where the Kvalan live together is in the center of the Shinshe administrative village, containing Patrogan and other small hamlets. In this paper I will use the name "Shinsha" (the Japanese pronounciation of "Shinshe") to indicate the traditional area where the Kvalan villagers are most concentrated within the current administrative village.
The whole population of the traditional Shinsha area is about 500, and among them the descendants of the Kvalan number about 250. At the time of my field research, only six people could be recognised as pure Kvalan, and the rest of the villagers are partly descended from other ethnic groups, such as the Turubiawan, the Tangabulan (the sinicised descendants of the Siraya, Makatao and Taivoan who moved up from southwestern Taiwan) or the Chinese. In this paper I call all of those who have at least one Kvalan ancestor "Kvalan" for the sake of convenience. The problem of their ethnic identity will be discussed later.
Apart from the Kvalan, the residents of Shinsha are Ami who have bought some rice fields here, or those who have married into the Kvalan households. There are also some Taiwanese (a sub-group of Chinese who speak the Minnan dialect) and Hakka (another sub-group of Chinese) who are married to aborigines. After World War II, some 20 Chinese mainlanders settled there. They are retired soldiers, and except for a few who are married to natives, most of them live alone.[4]
In Shinsha, the Han Chinese have not been a majority group either formerly or currently, so that the influence of direct interaction or social intercourse with Chinese on the Kvalan has not been intensive. As a result, they were in a position to preserve their traditional culture until today. Recently, however, school education and the mass media have accelerated the acceptance of Chinese influence, especially by the younger generation. The sinicisation of the Kvalan is advancing mainly due to the influences from the modernisation of Taiwanese society as a whole.
The relatively complex ethnic and historical situation are reflected in the Kvalan's use of languages. In Shinsha people use Kvalan, Mandarin, Taiwanese, Japanese and Ami on a daily basis.[5] According to age, sex, education and social experiences of each individual, together with the ethnic constitution of the members of the home family, the kind and number of spoken languages differ for each person. Villagers often use different languages to talk to different people.
Outside the village, people rarely have chances to speak Kvalan, because this is almost the only village where the people speak Kvalan daily. Shinsha is surrounded by Ami villages, and the Kvalan are a disappearing minority group in the coastal area. Even in Shinsha the younger generations cannot speak Kvalan very well so that their language will become extinct there in the future as it has in the Ilan Plains.

Kvalan family and kinship

Kvalan, like other plains groups, have been given Chinese surnames since the Ch'ing Dynasty. Their Chinese names and surnames were recorded with Chinese characters in the Japanese census, although in the cases of the Ami and other unsinicised peoples (Takasago-zoku in Japanese) the aboriginal name was recorded by the Japanese in katakana script.
Among the Chinese surnames of the Kvalan in Shinsha, Li, Lin, Zhu, and Pan are the most popular. Although they have accepted the Chinese naming system, succession to a surname is sometimes independent of either patrilineal afiliation or the residential pattern of the parents.
The Kvalan household family is called salppawan which literally means the people who live in a same household. The salppawan family is defined as a group of kinsmen who live in the same household, who cook and eat meals together, and co-operate in the daily work for the household economy. A salppawan typically contains members of two or three generations.
The residence of a married couple is decided in consideration of both families' circumstances. The percentages of virilocal and uxorilocal residence were almost the same during the Japanese Occupation. Young couples tend to leave their natal households to form a new salppawan within several years of marriage. As a limiting factor, one of the married couples remains in the original household to take care of the parents in their old age. Social relationships are traditionally based on an ego-centered and bilateral ideology. They call bilateral kinsmen knaswani which derives from a Kvalan word for sibling. Knaswani is a category of personal kindred, extending to second or third cousins. People tend to forget their more distant relatives. Knaswani functions as an exogamous category in Kvalan society.[6]
The traditional concept of incest prohibition was strict. Villagers try to avoid marriage with knaswani. This has produced increasing inter-ethnic marriage among the Kvalan, given the population decrease inside and outside the village. The ethnic identity of the Kvalan descendants in Shinsha has become complicated through the process of interethnic marriage, which is reflected in the practice of ancestor worship.
Another rapid change has occurred in marriage. The rate of virilocal marriage has increased from about 50% to 90% during the last half century. Changes in the form of marriage have caused other forms of culture change. As the result of virilocal marriage, the process of the formation of a new household has become similar to that of the Han Chinese. Few sons marry out, as most of the male descendants are going to inherit their parents' property eventually.
Despite this change, the ideological framework of the villagers is still based on the ego-centered and bilateral organisational principles, and the concept of patrilineality has not penetrated their way of thinking. The following analysis of ancestor worship will explain people's attitudes toward their traditional kinship ideology.

Religious beliefs and the Palilin ritual

Religious beliefs have greatly changed in recent years. Most of the Kvalan's religious activities have been on the decline, since many villagers have accepted Christianity and rejected the worship of their own deities.
Rapid Christianisation, as elsewhere in the world, was a result of systematic propagation. The villagers were attracted to Christianity because of the free distribution of clothes, powdered milk, flour and so on by the Christian Churches. In Shinsha there are a Catholic Church and a Presbyterian Church. Among the seventy households containing Kvalan descendants, thirty-eight households are Catholic, and ten are Presbyterians. Of course there are some exceptions. Sixteen households practice Chinese ancestor worship, and members of one household believe in Tenri-kyo (the Laws of Nature, one of the new Japanese religions).
Almost all the villagers, on the other hand, including Christians, keep on practicing Palilin, the traditional ancestor cult ritual on New Year's Eve. People believe that the spirits of the dead who live in another world, spend their daily lives just the same as living people. These spirits are believed to visit their descendants on New Year's Eve. For every living individual the Palilin ritual is a chance to maintain a link with deceased parents and relatives.
The performance of the Palilin ritual is as follows. People practice the ritual shortly before New Year's day on a day that is recognised as the Kvalan New Year's Eve. At midnight, household members gather in the kitchen to worship the ancestors' spirits.
A small square wooden board is put on the oven or the windowsill. Two wine cups are placed on the board. These are the plate and cups for the ancestors. Different kinds of rice wine and Chinese glutinous rice cakes are used as offerings. Glutinous rice balls seem to have been used in the old days.
Every family member, in turn, offers rice cakes and wine, calling the ancestors' names, and praying for good health and happiness in a low voice. They eat and drink together as they make the offerings to the ancestors. In this ritual no one person represents all the family members. Everyone has their own range of ancestors to celebrate for themselves in accordance with ego-centered and bilateral kinship principles. The range is different for each person, although it may overlap to some extent.
Ideally, every individual has an equal responsibility to worship all bilateral ancestors, but actually the names of the distant ancestors are allowed to be forgotten. Instead of remembering all the ancestors names, people usually ask one of ancestors' spirits, during the offering, to come together with all the other bilateral ancestors whose names they do not know. Logically, in this way, all ancestors are invited to the ritual without omission.
There is no doubt that a household or family functions as an important unit in the Palilin, but the ultimate unit of the ritual is each individual who has an obligation to make offerings to his or her own range of ancestors. The ideas of bilaterality and ego-centricity have penetrated the ritual, so that the idea of a cumulative genealogy is not part of the people's way of thinking. Lacking the principle of ancestor-focus or lineality, the Kvalan cannot organise an external ritual group beyond the household family.
Some irregularity results from interethnic marriage, which creates an imbalance in the range of ancestors celebrated in the ritual. Contrary to the bilateral rule, ancestors of different ethnic groups are not worshipped on the occasion of the Palilin. It is believed that each of the ancestors prefers to be celebrated according to the traditional rituals of their own ethnic group, so unless they are Kvalan they do not come to eat and drink Kvalan offerings.
The Kvalan of mixed descent practice different ethnic rituals besides those of the Kvalan. In this sense, the bilateral principle is still retained outside the Palilin ritual. Apart from the occasional curing rituals of the Ami in which ancestors are worshipped by offering food, the Chinese and Turubiawan ancestor cult rituals are practiced separately annually during a period of a few days before the lunar New Year.

The Palilin ritual of the Turubiawan

In the case of the Turubiawan Palilin, the cult group is formed differently. It contains plural household families bilaterally connected to the same Turubiawan ancestor. In most cases it contains old parents and independent siblings with each of their own household members. Independent siblings sometimes continue to form a cult group even after their parents have died.
The Turubiawan cult group has a strict boundary line between members and non-members. On the occasion of the ritual, non-members should neither enter the house, nor watch the ritual. They believe that the ancestors' spirits can be wounded by the power of an outsider's soul, which may bring unhappiness or misfortune to the descendants.
Turubiawan descendants celebrate their Palilin on the day after the celebration of the Kvalan Palilin. They kill a rooster and prepare for the ritual in the morning. Members gather in a kitchen or in the living room before noon, and make offerings of glutinous rice cake and rice balls, two kinds of rice wine and some internal organs of the rooster.
The ways of preparing offerings, the kinds of organs used as offerings and the manner in which foods are offered are slightly different in each cult group, especially the way of killing the rooster. There are three traditional ways in this village: hitting it on the head with a hammer, knocking its head on the doorway, or burning it alive.
Descendants of a Turubiawan family usually inherit their parents' ways of performing the ritual, and those who perform the ritual in the same way, form a single cult group even after each of them has started a new household group. Descendants who have married out, on the other hand, join a different cult group and follow other ways of celebrating the ancestors.
When the ritual begins everyone calls on their bilateral ancestors in a low voice, but there is no need to remember the name of each individual ancestor. Kinship terms are used instead of individual names. The position of ego is no more emphasised than in Kvalan ritual. The use of kinship terms leaves the bilateral relationship with a specific ancestor unspecified.
On the other hand, the names of some members of the group who come to celebrate the ancestors but are too young to join in the ritual are introduced to ancestors by one of the older members. In contrast to the Kvalan ritual, membership of the Turubiawan cult group is affirmed on the occasion of the ritual, in which the bilateral relationship between the ancestor and each person is emphasised through offerings. For the Kvalan, each individual, even a small child, is the most important ritual unit as I mentioned above.
In addition, there is one more clear difference between the Palilin rituals of the Turubiawan and Kvalan. The Kvalan share a less clear boundary with the cult group than the Turubiawan. The household group functions as a Kvalan ritual unit, but there is no strict rule to keep away outsiders. Bilateral kinsmen who are not members of the same household sometimes visit to offer food, even at midnight or the next morning.
I was not prevented from seeing the Kvalan ritual, but was prevented from doing so by several Turubiawan cult groups. At last two groups allowed me to join the ritual during the first year of the field research, and I had a chance to see the two different kinds of Turubiawan Palilin. The next year, however, I was completely excluded by all Turubiawan cult groups, because some of their Turubiawan kin had criticised these groups for breaking the customary rule. They warned their kinsmen that they would meet with misfortune in the coming year.
It seems that the Turubiawan Palilin ritual and their ethnic identity have been maintained by their strict rule of forming an exclusive cult group. It is also the result of the Kvalan way of thinking in allowing people to celebrate different ethnic ancestors in their own ways. In some cases, however, we can see people's different attitudes toward the succession to the two Palilin rituals, which may derive from the differences in the ritual form and concept of membership in these two ethnic rituals.
The Turubiawan Palilin is in most cases succeeded to by children who do not marry out. A Kvalan individual who married into a Turubiawan household joins in their ritual. In-marrying Kvalan also continue to practice the Kvalan Palilin by themselves at first, and are succeeded by their descendants of mixed descent. But in the case of Turubiawan descendants who marry into a Kvalan household, or household of another ethnic group, most of their descendants cease practicing the Turubiawan ritual before long.
Turubiawan spouses are allowed to practice their Palilin ritual by themselves when they marry into a Kvalan family. But for the Kvalan, the ritual unit is, after all, every individual, not a household group. Thus it is not easy for a Turubiawan in-marrying individual to organise a new cult group by calling on all of the Kvalan household members. One of the villagers told me that his in-marrying Turubiawan grandmother used to practice the ritual alone. The situation seems to be similar in the cases of inter-ethnic marriage with other ethnic groups.
According to my field data, twenty-five cases of practitioners of the Turubiawan Palilin are people of Turubiawan descent married to a spouse of a different ethnic group, but living in their own natal family, and four cases are people who married out into a household of another ethnic group. According to the Turubiawan traditional rule, in-marrying individuals join their spouse's ritual group and follow the way of the cult. The rule seems to still operate in the case of inter-ethnic marriage.
Villagers who worship both Kvalan and Turubiawan ancestors practice the Kvalan Palilin at first, and then practice the Turubiawan Palilin the next morning together with some other household groups. In addition, those who also have Chinese ancestors, or sinicised ancestors, hold a Chinese style ritual for gods and ancestors at the lunar New Years Eve, with their own household members.
Actually, many villagers continue practicing plural ethnic ancestor cult rituals. In this sense, the bilateral principle is still retained outside the Kvalan's Palilin ritual.
But it is also possible to abbreviate some of the rituals while complying with traditional principles. Practicing plural rituals is sometimes troublesome, and descendants are allowed to forget about distant ethnic ancestors. People believe that they can discharge their duties by offering food to the ancestors in order to satisfy their hunger in accordance with at least one of the forms of the different ancestral rituals.
They feel that it is unfilial to leave their dead kin hungry, and believe that hungry spirits make their descendants ill to indicate that they demand an offering. Good relations between descendants and ancestors are built up by constant offerings of food on ritual occasions. This is the basic idea of the traditional ancestor worship among the Kvalan. Other ethnic traditions, Chinese, Ami and Turubiawan, also share the concept of food offerings.
On the other hand, there is no occasion to offer food to a deceased person in the Christian tradition, so Christians are anxious about deceased relatives suffering from hunger. This is the reason why most of the Christianised Kvalan villagers still practice the Palilin ritual every year. This phenomenon indicates that the traditional idea of Kvalan ancestor worship is still retained in spite of the recent Christianisation.

Sinicised ancestor worship

In the following pages I will analyze the case of the sinicised Kvalan and show their peculiar inclination to accept Chinese influence, which is based on their ideological framework.
I will focus on the ways of accepting Chinese ancestor worship using tablets. Worshipping ancestral tablets is originally based on the rules of Chinese patrilineal kinship ideology. How Kvalan react to these rules is an interesting instance of the encounter between contrasting kinship ideologies.
In Chinese ancestor worship using tablets, patrilineal kinship ideology is presented through its form and order. If it is accepted by a society based on a different ideology or different value system, a different form and order may emerge in regard to worshipping tablets. By analyzing specific case, we can grasp the character of the society that defines the direction of change.
Further, I will compare my field data with the case of the Puyuma, one of the other aboriginal ethnic groups, which shows us another mechanism originating within a different ideological framework through which the same Chinese custom has been accepted in a different way.
After the systematic propagation of the Christian faith in the 1950s, many villagers stopped practicing Chinese religious rites. Among the seventy households containing Kvalan in Shinsha village, sixteen currently practice Chinese ancestor worship, and eight have ancestral tablets within the residential unit. Only three of them, however, keep plural tablets. Each of the other five households keeps only one tablet for an individual ancestor.
For ancestors without votive tablets, villagers sometimes substitute incense pots or incense ashes, but these are only provided for recent ancestors up to a few generations back. Among eight households without any tablets, two use an incense pot as a substitute for tablets, but six have no substitutes. As a result they prepare offerings outside the door and call the names of the ancestors to whom offerings are to be made on the occasion of the annual rituals. People of these latter households do not mind the lack of ancestral tablets, neither do they plan to install them.
Contrary to expectations, the use of tablets is uncommon among the Kvalan villagers despite two centuries of sinicisation. According to the villagers ancestral tablets have never been actively introduced into this village.
The only exception was observed at the end of the period of the Japanese occupation, when a religious assimilation policy was enforced by the colonial government and all the village households had to set up kami-dana, a Japanese domestic shrine, and Japanese-style ancestral tablets. Those who worshipped the tapestry of Chinese gods were forced to renounce them. Those who had set up tablets were allowed to worship them.
Influences from Japanese religion, however, left no traces in Shinsha. As soon as the war ended villagers, dumped kami-dana and tablets so that the custom of worshipping tablets did not take root among them, except for those who had maintained the Chinese custom of worshipping tablets before the colonial policy of assmilation.
Even today, the Chinese festival of Qingming is flourishing and all the villagers, including strict Protestants, come to visit the graves of the ancestors. However, the custom of setting up tablets for the dead, on the other hand, has never become popular among them.
In comparison with the Kvalan case, the sinicised Puyuma show another variant of practicing ancestor cults with tablets. According to Suenari's report, one fifth of the 150 households in a Puyuma village had ancestral tablets in 1968, but by 1983, an increasing number of villagers, including even those practicing Catholicism, had tablets. Suenari also suggests that more Puyuma villagers will set up tablets in the near future without changing their traditional concept of ancestor worship (Suenari, 1983:136, and this volume).
In order to understand the peculiar situation of the Kvalan, we should pay attention to the motivation by which some villagers are made to set up tablets for ancestors.
Among eight households with tablets, three worshipped ancestral tablets even before the Second World War, and others came to set them up after the war. The immediate motivation of worshipping tablets is in most cases interethnic marriage. The situation occurs as follows.
In-marrying spouses, Chinese, or other sinicised aborigines who practice Chinese annual cults are always allowed to worship their own ancestors in the Chinese way. Sometimes they bring the tablets of parents or siblings, and worship them on the occasion of the Chinese annual rites. Mixed-descent children and sometimes Kvalan spouses also worship these tablets. After the death of an in-marrying person or mixed-descent child, new tablets are made for the dead in accordance with their religious custom, and their descendants come to worship them.
According to the rules of the Chinese annual rites, on the other hand, people only set up a table with offerings outside the door and call the ancestors' names, if they do not have any ancestral tablets. Villagers seem to prefer a simple way like this to invite the ancestors' souls in. It is very similar to the traditional way of inviting in the ancestors' spirits in the Palilin ritual.
The traditional idea of calling back the ancestors' souls seems to prevent them from recognising the necessity of setting up a tablet for the dead. Villagers tend to set up tablets for the ancestors coming from other ethnic groups like Chinese or sinicised aborigines in order to respect their religious customs. On the other hand they are not so eager to do this for the sinicised Kvalan ancestors.
Do they, then, worship these ancestors on the occasion of the Palilin? In some cases, people worship the same ancestor both in Palilin and in Chinese annual rites. Or they may omit the Palilin ritual. In general, people take account of the fact of whether an ancestor was accustomed to the form of Palilin ritual or not.
Worship is, for the Kvalan, after all, the offering of food. If a person had always joined in the Palilin ritual and was accustomed to eat Kvalan food while alive, he or she will be invited for the Palilin after death. If not, the descendant will not call this deceased person's name in the ritual, because the descendants assume that the soul of the ancestor would not come to eat Kvalan offerings.

Worshipping ancestors with tablets

Among the cases of establishing tablets for ancestors, some examples of worship deviating from Chinese norms can be observed. The following analysis of these deviations clearly indicates the differences in the ideological frameworks of Kvalan and Chinese.

Case 1

Members of this household are mixed-blooded descendants of Kvalan, Turubiawan and Ami. They practice Chinese ancestor rituals in addition to Kvalan and Turubiawan Palilin.
In this case, an ancestor's tablet and an incense pot substituting for another ancestor, are set on a shelf in the house set up in the kitchen, which is located in the inner part of the house. In Chinese custom, a domestic altar should be set up in the domestic hall, the nearest place to the entrance, while the kitchen is often used to set up tablets of differently surnamed ancestors who do not have a claim to be worshipped in the domestic hall.
For Kvalan, and also for Turubiawan, however, the living room is used for the traditional public religious activities like the curing ritual (Shimizu, 1986, 1989) or funeral rite (Shimizu, 1986, 1990, 1992) to which many other villagers are invited as guests, while the kitchen is traditionally a sacred place to worship ancestors, usually alone with the salppawan members. It is not unnatural for them to set up a Chinese domestic altar and worship ancestors in the kitchen.

Case 2

In this household, people worship Kvalan and Hakka ancestors using tablets. There are two tablet boxes, that of the domestic ancestors and that of the ancestors with different surnames, in the domestic hall.
A tablet dedicated to a dead baby is also stored inside the box of domestic ancestors. There was no special motivation such as being cursed by the baby's ghost, but the tablet was still made for the baby on her death. According to the Chinese rule, a dead baby is not to be worshipped unless the ghost of the baby is recognised to torment descendants by causing illness or unhappiness in order to obtain worship (Wolf, 1978:147-148). Kvalan villagers do not distinguish between the dead based on differences of sex, age and surname. Members of this household do not feel any inconsistency in worshipping the dead baby who had once been a member of the family.
Further, on the same domestic altar, there is another box with a tablet devoted to a son who succeeded to the surname of his inmarrying father, rather than that of his mother's family, but who died at the age of nineteen. This relatively new tablet box is placed next to the box belonging to the domestic ancestors. The location is on the inferior side of the domestic ancestors' box. Though this is the suitable side on which to set it, there is no board separating these two boxes.
According to Chinese custom, tablets of differently surnamed ancestors should be placed in other rooms, the kitchen or storeroom. In Taiwan, however, people sometimes place those tablets in the ancestral hall. They set the tablets on a different shelf, or on the same altar with a board placed to divide them from the tablets of domestic ancestors. People place these tablets according to their relative distance from domestic ancestors (Ahern, 1973: 130; Wolf, 1978: 155).
In this case, however, parents of the dead youth state that they are not eager to place his tablet in another room, because he was their own son. Also they do not recognise the meaning of, or the necessity for, putting it on a different shelf in the domestic hall, or of separating tablet boxes by setting a board between them.
Though the parents accept Chinese religion, they are more concerned about traditional kinship relations than about Chinese rules when they worship. Their way of thinking is still based on their own kinship framework.

Case 3

In the Lin household, people worship eight ancestral tablets divided into two boxes, the largest number of tablets in this village. The two boxes are on the same shelf in the domestic hall. The larger box situated on the dominant side contains tablets of four ancestors. Two of them are those of the founding couple of this household, a Kvalan male surnamed Lin and his Turubiawan wife surnamed Pan, who moved into this village from the Ilan Plains during the Japanese occupation. They were the grandparents of the current household head Lin A-lu. Two other tablets are of an adopted son of the founding couple, who was surnamed after his foster mother Pan, and of his wife. Tablets of Lin and Pan ancestors are in the same box.
The latter couple had no child and adopted a daughter, who inherited all of her foster parents' property when they died. She married out and divided the ashes of the incense pot to set up new tablets in her husband's household. So it is not necessary for Lin household members to worship Pan ancestors. However, Lin A-Lu never considered to cease worshipping them, because these ancestors were his kin, his uncle and aunt-in-law.
The small box contains four tablets, dedicated to Lin A-lu's mother who was an adopted daughter of the former couple, and to her in-marrying husband and first born son. Another one is a tablet dedicated to her common-law husband who came to live with her soon after her husband's death. Though it is said that he was not the biological father of Lin A-lu who was born as an illegitimate child, he lived throughout his life in this household, as if he were the actual father of Lin A-lu. The reason why this Hakka male was never entered in the family register is that he neither had children nor property of his own.
Thus, in this box, different surnamed ancestral tablets are mixed together, Lin's ancestors, her husband (surnamed Li), her eldest son (Li) whose succession to a surname also ignores the Chinese rule, and her common-law husband (Chang). From the standpoint of Chinese patrilineal rule, it is strange that the tablet of a common-law husband is stored in the box for the domestic ancestors. Lin A-lu states that he does not want to separate his "fathers" tablet from other ancestors.
Again in this case, actual relationship with ego takes priority over patrilineal ideology, as is usual with the Kvalan, although Lin A-lu knows a lot about Chinese religion. Nevertheless, the way he divides ancestral tablets into two groups does not conform to the Chinese rule, but is similar to the way in which an extended salppawan family is divided into two new salppawan..
Lin A-lu's mother was an abandoned baby found in Ilan and presumed to be a Taiwanese, and the biological father of Lin A-lu was also said to be a Taiwanese who had once stayed in Shinsha. All the villagers including Lin A-lu himself believe him to be a Taiwanese. However Lin A-lus way of thinking is Kvalanised to some extent.
From the Chinese point of view, these examples I mentioned above seem to deviate somewhat from the norms based on patrilineal descent ideology together with the system of inheritance. However, the villagers recognise a different principle as underlying the practice of keeping tablets, so they do not think of these examples as a kind of deviation.
This version of ancestor worship with tablets is based on the local traditional principles of egocentricity and bilaterality. According to the traditional rule, a deceased person who once lived as a virtual member of ego's family is worth worshipping. According to the bilateral principle there is no reason to separate ancestors with different surnames from domestic ancestors.
Moreover, the inheritance of tablets does not necessarily follow the same inheritance rules as property. It is also influenced by the traditional way of thinking. In the traditional system, succession and inheritance do not entail the obligation of worshipping ancestors. Property is divided among all the daughters and sons except those who have married out. On the other hand, all the children have an equal obligation to worship their parents after death, regardless of whether they have inherited property.
In comparison with the Kvalan case, the sinicised Puyuma show another variant of practicing ancestor cults with tablets. According to Suenari's report, thirty-three of the 150 households in a Puyuma village had ancestral tablets in 1968, but by 1983, an increasing number of villagers, including even those practicing Catholicism, had tablets (1983: 136, and this volume). The situation is in contrast to the case of the Kvalan. However, it does not mean that the sinicisation of the Puyuma is complete. Suenari has indicated some of the peculiar features of Puyuma use of ancestral tablets. The lineage shown on the tablets is shallow, covering only a few generations of ancestors. If a domestic group separates, segmentation of the ritual group does not occur. In addition, the in-marrying members often return to their natal households to take part in tablet rituals there. There are no rules in regard to who is to be in charge of the tablets. Sometimes it is decided by the state of the household economy, by asking the ancestors through divination, or by the inheritance of property from the ancestor.
Suenari indicates that the membership of a ritual group worshipping tablets is decided in a way very similar to that which determines the membership of the traditional ritual group worshipping a small shrine (Suenari, 1983: 126). The content of the Puyuma worship of ancestral tablets still retains the traditional concept of descent, that is, the ambilineal principle of tracing ancestors (Mabuchi, 1976: 91-104; Suenari, 1970: 107).
Though the Puyuma lack the concept of patrilineality, the custom of using ancestral tablets has diffused among them. Thus I return to the question of why ancestral tablets have not diffused so widely among the Kvalan as among the Puyuma. The difference may arise from basic dissimilarities of concepts of ancestor worship in both societies.
It might be easy for the Puyuma to accept the idea of organising a ritual group centered on tablets of common ancestors within a framework similar to that used in forming a ritual group centered on a shrine. Traditionally the obligation of worshipping ancestors was to some extent related to the inheritance of property, even though the connection between the tablets and the individual was traced ambilineally (Suenari, 1970: 96, 107).
On the other hand, the Kvalan may have some difficulty in forming an exclusive ritual group centered on a common ancestor without an ancestor-based descent ideology. According to the organisation of their traditional cults, the domestic family is not defined as an exclusive cult group, but rather every individual is recognised as an independent ritual unit, and everyone has their own range of ancestors to be worshipped.
Furthermore, the obligation to record the names of ancestors on tablets is difficult to reconcile with the strongly individualistic ideology of traditional religion. Accordingly, everyone worships by calling out the names of the groups of ancestors that they trace bilaterally back from themselves. In fact, it is not necessary for them to remember the names of all their ancestors, as is often the case in bilateral societies.
The cult offers the most rational way to worship ancestors bilaterally, and as long as the Kvalan villagers, even those who have Chinese ancestors, retain this practice side by side with Chinese ancestor worship, the patrilineal rule of inheriting tablets will not penetrate into their way of thinking. The maintenance of the cult indicates that the bilateral and ego-centered ideology does not change.
As Suenari suggests, more Puyuma villagers will set up tablets in the near future without changing their traditional concept of ancestor worship (Suenari, 1983: 136) However, the Kvalan will not set up tablets for their ancestors in increasing numbers unless their traditional concept changes fundamentally, even if their sinicisation proceeds in other aspects of religion.
Acceptance and diffusion of the worship of ancestral tablets is one of the criteria of sinicisation. However sinicisation is not a simple phenomenon. It displays a diversity of form and character. As I have shown above, it seems that the forms sinicisation takes largely depends on the character of the traditional concepts of each individual ethnic group.

Changing ethnic identity and mixed descent

Through the analysis of the Palilin Ritual and of the sinicised ancestor cult, we can observe how the framework of the bilateral kinship ideology determines the way in which different ethnic rituals are accepted in parallel with each other. It also creates the current change in ethnic identity of Kvalan descendants. According to their way of thinking, ethnic identity results from their bilateral blood relations with ancestors. Ties of blood are the most important factor when they think of their ethnic identity. Such a principle, however, causes plurality in ethnic identity among those of mixed descent.
As I already mentioned, those who have ancestors of different ethnic groups often practice plural ancestor rituals. These descendants reconfirm their plural ethnic identities by practicing these different rituals. It is the proper way for Kvalan to worship bilateral ancestors. Everyone affirms each connection to the bilateral ancestors on each ritual occasion.
Those who have plural ethnic ancestors have plural ethnic identities. A person who has an Ami father and Kvalan mother is recognised as both Ami and Kvalan. Those who have Kvalan, Ami and Turubiawan ancestors have identities of Kvalan, Ami and Turubiawan, though they are culturally Kvalan. They refer to their different ethnic identities according to time and circumstances. However, people do not intentionally choose only one of the identities as long as they are conscious of their own kinship relationship to ancestors of different ethnic groups.
On the other hand, more correctly, a person of mixed descent is recognised as neither a pure Kvalan nor a pure Ami. "We are mixed," said villagers, explaining their ethnic situation. They describe the six pure-blooded descendants as "pure Kvalan". These pure Kvalan are advanced in age, and their spouses are not pure Kvalan, so that no one will remain pure-blooded in the future.
The Kvalan have chosen their way of change in accordance with the traditional ideological framework of their value system. They avoid marrying kin, which has led to increasing interethnic marriage. Mix-blooded descendants celebrate their ego-centered group of bilateral ancestors by practicing plural ethnic rituals. Offering and eating food together with ancestors is the basic activity of their ancestor worship. Even Christian villagers do not want to cease offering food to their ancestors.
Their idea is based on three important factors: egocentricity, bilaterality and reconfirming kinship relations by eating food with the ancestors. These factors have actually functioned to change their ethnic identity, although these characteristic features of their value system remain despite the influences of sinicisation, Christianisation and modernisation.
The reason why these factors persist is that these are also the most basic factors in constructing the kinship and family system of the Kvalan. Both maternal and paternal kinship relations are important for individuals in recognising their own category of knaswani, personal kindred, and also in reconfirming the link between themselves and the ancestors. Links made by eating together are significant for them. Salppawan family members eat meals cooked in the same oven. When members divide it into two salppawan but still have to remain in the same house, they build one more kitchen, or at least use different ovens in the same kitchen, and members of two salppawan eat meals in different rooms. The corner of the oven is also recognised as a sacred place for the salppawan family to practice ancestor worship, on the occasion of which people eat offerings together with their ancestors.
Family and kinship structure have not changed greatly among Shinsha villagers until today. It is largely because the status of in-marrying members is relatively weak in the Kvalan family that spouses from other ethnic groups could not become a strong influence in changing traditional family structure or the framework of the kinship system. Chinese in-marrying husbands often introduce ritual activities, but they are in most cases lacking strong kinship ties or economic backgrounds of their own. The influences of Chinese social values on kinship and family have not been strong enough to destroy the Kvalans ideological base.
These factors may change in the future as ethnic change is promoted, but they will not necessarily change side by side with a change of ethnic identity. These factors in the Kvalans traditional value system could remain as a peculiarity or tendency of the local culture even after the Kvalan villagers have changed into members of other ethnic groups.

The ethnic environment of Shinsha

In the following discussion I will discuss a problem of ethnic identity from the standpoint of the surrounding ethnic environment that may influence the process of change and maintenance of ethnic identity in Shinsha.
The Kvalan descendants in Shinsha are not ashamed of being Kvalan. This has influenced the direction of their ethnic change. It reflects the ethnic environment around this village, which has helped to keep the Kvalan identity along with some traditional rituals until today. Descendants of the plains ethnic groups would not, in general, wish to be identified as other than Han Chinese, because the aborigines have historically been despised by Han-Chinese, as Fana (Taiwanese), which means barbarian or aborigines in general.
Chinese people in Taiwan are generally unconscious of the difference between "raw" (i.e. non-sinicised) and "ripe" (i.e. sinicised) aborigines. Plains aborigines, on the other hand, have a strong consciousness of themselves as not being "raw" aborigines. Their way of thinking derives from the administrative division of the aborigines during the Japanese Occupation. Sinicised plains aborigines were recorded as Heiho-zoku (Plains peoples), or Juku-ban ("ripe" aborigines) in the Japanese census, and unsinicised aborigines were recorded as Takasago-zoku (Takasago peoples), or Sei-ban ("raw" aborigines). This originated from the administrative classifications during the Ch'ing period.[7]
In Shinsha, the word Fana mainly signifies the neighbouring Ami. A mixed-blooded Kvalan woman mentioned her Ami mother and said, "My mother was a Fana." However, she never thought her Kvalan father was a Fana. Kvalan have thought of themselves as having a degree of cultural superiority over the neighbouring Ami. However, the villagers know that the word is discriminatory and they try not to use it in face to face situations. Actually the Ami have been one of the most popular ethnic groups from which the Kvalan choose their spouses.
The Han Chinese are not a majority group in this area, so that Kvalan descendants do not feel the necessity of concealing their aboriginal origin. The reason why a small group of 250 descendants has been able to maintain their ethnicity until today is due to the lack of one-sided pressure from Chinese given the circumstances of multi-ethnic residence in this area. If Shinsha villagers had always been despised by a large number of Chinese neighbours, their ethnic identification would show a somewhat different form. They might have emphasised their Chinese origin despite the bilateral principle, and would long before have lost their identity as Kvalan.

Other conditions of change among the Kvalan villagers

I have emphasised the deep-rooted character of bilaterality and ego-centricity among the villagers as deciding factors of cultural and ethnic change. I should mention additional conditions that may have encouraged the survival of these factors in spite of the process of change.
Social pressures from village organisation might have been originally weak among the Kvalan as is often the case with bilateral societies. In addition, it seems that the flexible relationship with other villagers based on ego-centered social relations has been strengthened through the process of change, inversely proportional to the break-up of the solidarity of their village organisation. The village festival for celebrating a good harvest disappeared before World War II in Shinsha, although ego-centered religious activities like curing rituals, ancestor worship and funeral cult have remained. People tend to identify themselves more through their relationships to their own bilateral ancestors than through membership of the ethnic group or the village community.
It was difficult for them to maintain social solidarity within their historical and social environments. First, because of the variations in places of origin and period of immigration, people could not rebuild a strong village organisation in this place. Shinsha is a colonial village that has a history of over 100 years. The first founding ancestors of the village came from the colonial Karewan Village, and some of the ancestors migrated directly from various parts of Ilan. The immigration of Kvalan continued intermittently throughout the Japanese Occupation. As a result, the social unity of the villagers has been relatively weak.
Second, the multi-ethnic situation, increasingly noticable since the end of the war, has promoted the break-up of the ethnic unity of the village. Alongside this, Ami immigrants and members of aboriginal ethnic groups have married into Kvalan families, Chinese mainlanders have settled in this village. After retiring from military service in the village garrison, some of them bought plots of land and built small houses in order to spend the rest of their lives in this village. The number of mainlanders was limited, as was their influence on sinicisation. However, their political background had a strong influence on village policy for a long time, and in consequence, made for strained relations with the natives, the results of which make themselves felt in village life to this day.
The development of the money economy has also encouraged the freedom of the individual in decision-making. Villagers do not exclusively rely on agricultural work as before. Rather they depend on wage work outside the village to some extent. Salppawan members have become less subject to the authority of the household head who owns the land. Currently, those who earn money are able to speak their minds more. Even unmarried youths can decide themselves how to use their own wages without asking their parents. Villagers tend to pay each other wages to build a house or grave instead of depending on reciprocal labour exchange. People do not like to work for others without getting the immediate benefit. Reciprocity among the villagers is changing into a flexible relationship based on the money economy. Kvalan often say that they are individualistic, a tendency which seems to have been strengthened by the process of economic change. Social control by the traditional village organisation has become weak as individualism has grown strong.
The increase in virilocal residence also results from economic change. Male villagers do not have to marry out because they can earn money themselves, even if the parents do not have enough rice fields to keep their own son in their natal household after his marriage. In-marrying husbands felt much more ill at ease obeying their wife and parents-in-law than in-marrying wives because a husband's labour in the fields is more essential than a wife's work. The change of residential form is not simply the result of the penetration of the Chinese patrilineal system, but a movement originating from economic change.
These specific factors should not be neglected when we discuss the changing culture and identity of the Kvalan people in Shinsha. However, taking them into consideration, the deep-rooted influences of their traditional framework conceptual cannot be ignored in discussing social and cultural change among them.
Bilateral ideology has some flexibility. Bilateral societies often have similar characteristics to patrilineal societies such as an inclination toward virilocal marriage and male inheritance of property. If the people of a bilateral society, the Kvalan for example, adopt features of Chinese family and kinship organisation, it does not necessarily mean that their society will change into a completely patrilineal society.
Patrilineality is not easily accepted in Shinsha. People tend to understand the Chinese family system within the category of kinsmen centered on ego. The increase of virilocal residence, and inheritance of property by male descendants which derive from the change of residential form does not indicate acceptance of a patrilineal ideology. As far as I observed, villagers cannot understand the idea of ancestor-focused social relationships at all.
Kvalan society is a typical bilateral society with egocentricity and bilaterality, which does not organise exclusive groups. Judging from the Shinsha case, the influences of the bilateral principle may not easily disappear because they have actively functioned as a basis for accepting elements of change.

Acknowledgments

The research described in this article was supported by the Toyota Foundation. An earlier version of this paper was published as "Kanka no mekanizumu [The Mechanism of sinicisation]" in the Bulletin of the National Museum of Ethnology, Special Issue No.14, 1991. I have also dealt with this issue in my ethnography of the Kvalan, Kuvalan zoku - Kawariyuku Taiwan Heichi no Hitobito [The Kvalan - A changing people in the Taiwan Plains area], Tokyo: Akademia Shuppankai, 1992.

Notes

1. According to the ethnic and linguistic classification by Mabuchi Toichi and by Tsuchida Shigeru, Taiwan aborigines are divided into ninteen groups. (Mabuchi, 1974: 508, Tsuchida, 1977: 80)
2. The non-sinicised peoples include nine ethnic groups: Atayal, Saisiyat, Bunun, Tsou, Rukai, Paiwan, Puyuma, Ami and Yami. The sinicised peoples include ten groups: Ketangalan, Vasay (Basay, Basai), Kvalan (Kabalan or Kavalan), Taokas, Pazeh, Papora, Babuza, Hoanya, Thao and Siraya (Mabuchi, 1974: 508, Tsuchida, 1977: 80).
3. The village administratively contains over 200 households with a population of around 1,200. At the periphery of the administrative division there are two Ami villages and one small Taiwanese village. These are relatively new colonial villages.
4. They seldom learn to speak either Kvalan or Taiwanese. They only speak various kinds of Chinese dialects. Their communication and acquaintances with the villagers are limited, so that they have not taken an important part in promoting the sinicisation of the villagers.
5. Older people over 50 years of age speak Kvalan daily, along with Taiwanese and Japanese, but they are not fluent speakers of Mandarin. Younger people who graduated from school after 1945, speak standard Mandarin, but they are not a good speakers of Taiwanese, Japanese nor Kvalan. Though some villagers who are married to Ami usually become familiar with Ami, all of the in-marrying Ami have come to speak Kvalan more fluently than their spouses speak Ami.
6. The kin of a husband and wife merely overlap with one another, so that paternal and maternal kin for their children are clearly divided. This produces a result which contrasts with the case of the Iban in Borneo. In Iban society people prefer to marry a cousin. As the result of marriage with cousins, the category of bilateral kin overlaps to some extent in the generation of their children, which prevents the ego-centered network of the bilateral kin from dispersing as it expands (Freeman, 1960: 75-76). In Kvalan society, the network of kinsmen disperses because of the rule of exogamy as the generations pass. Their kinship relations do not create such an intensive network of connections as that of Iban society.
7. The Kvalan ritual for the dead is held after burial. They recall the soul from the other world to serve it food and drink, and they pass on the belongings of the deceased to be taken to the other world. This ritual is held exclusively for the soul of the deceased: the corpse is forgotten soon after burial.
8. After 1945, a new administrative division was made under the new government, which divided the former Takasago peoples into the two categories of Mountainous Compatriots (Shang-ti shan-pao) and Plains Compatriots (Ping-ti shang-pao). Chinese are registered as plains people (Ping-ti ren), and their former Japanese subdivisions of Hukken-zoku (Fujian people, which indicates Minnan Chinese), and Kanton-zoku (Guangdong people, which indicates Hakka) have disappeared in the new census. Sinicised plains aborigines were able to be registered either as Plains Compatriots or Plains People. Most of the Kvalan villagers register as Plains Compatriots in order to take advantage of financial aid or other special privileges from the government. There are only a few people who want to change their registration to Chinese. Under the circumstances in Shinsha, people have lived as Kvalan with no strong desire to become Chinese or to conceal their aboriginal origin.