12. The Ethnic Identity of the She and the Cultural Influence of the Hakka: A Study Based on a Survey of She Villages in Chaozhou, Guangdong.

Segawa Masahisa

This article attempts to analyse the ethnic relationship between the Han Chinese and the She in Chaozhou District of Guangdong where I carried out a short survey of four She villages in July, 1990. The She in the Chaozhou District are almost completely sinicised, and there is no marked difference in their daily way of life from that of the Han Chinese majority around them. But in spite of their cultural assimilation, an ethnic boundary between the She and the Han is clearly maintained. To understand the mechanism maintaining this ethnic boundary, we must take into consideration the culture of the Han Chinese living around them. Han Chinese culture in the southeastern part of China is by no means homogeneous, and therefore sinicisation itself is not a single process.

Contact between the She and the Hakka in historical perspective

According to the 1982 census, the She dwell in Guangdong, Fujian, Zhejiang, Jiangxi, and Anhui with a total population of about 368,000. (For its distribution see Map 1.) General theory holds that this minority group living in the mountainous areas of coastal China has been under the strong influence of the Han Chinese almost from the initial stages of their formation as an ethnic group, and the She are usually counted as one of the most "sinicised" people among the minorities in south China along with the Yao, the Pai, and the Pingbu group of Taiwanese minorities.
According to reports written by Chinese linguists, more than 99% of the total population of the She live in Fujian, Zhejiang, Jiangxi, and Anhui Provinces, as well as Chaozhou district of Guangdong. They speak dialects of Han Chinese, and studies indicate that most of them speak the Hakka dialect or something very close to it. Only 0.4 percent of the She living in Huiyang, Haifeng, Zengcheng, and Boluo County of Guangdong are reported to preserve their "own" language which is not a Han Chinese dialect, but belongs to Punu of the Miao-Yao branch (Shih, 1988: 21). Amongst the She, traditional male costumes seem to have been almost completely lost. In some parts of Fujian and Zhejiang, and traditional styles of costumes are worn only be females (Zhongguou Shaoshu Minzu Shehui Lishi Diaocha Ziliao Conkan Fujiansheng Bianjizu, 1986). As for production technology, it is said that they practiced slash-and-burn cultivation in the past, but in the survey to identify ethnic groups in the 1950s, the She in all regions had already abandoned this type of agriculture.
Ambiguities are found in their self-identification. For example, it was reported that some of them had lost their ethnic identity as She and did not know what group they belonged to. Even in those places where they kept some sense of ethnic identity, the name they call their own ethnic group varies a lot; "Hode" (Huiyang and Haifeng of Guangdong), "Yao" (Zengcheng and Boluo of Guangdong, Ningde of Fujian), "Miao" (Ningde, Luoyuan, and Xiapu of Fujian), "Shanke" (lit. "guest people of the mountains", in Luoyuan and Xiapu of Fujian, Yanshan, Guiqi, and Xinguo of Jiangxi), and "Qianke" (lit. "earlier guests", in Pingyang of Zhejiang) are some examples (Zhongguou Shaoshu Minzu Shehui Lishi Diaocha Ziliao Conkan Fujiansheng Bianjizsu, 1986: 3, 23, 51, 91, 118, 143, 179, 196, 241.) .
Of course, there are some cultural elements such as the Panhu myth that are claimed to be common to all She.[1] But it is doubtful whether the whole population scattered around these rather vast areas would have had any clear sense of identity as a single ethnic group before Liberation. Their ethnic identity today seems to have been established, or at least consolidated, only after the new policy for minorities began in the 1950s.
Although specialists of Chinese ethnohistory have advanced different views on the ethnic origin of the She, it is generally accepted that they were closely related to the Yao in their origin, and at least by the Song dynasty they had come under the strong influence of the Han Chinese both culturally and politically.[2] But "sinicisation" might not be a simple process, because in southeastern China the culture of the Han Chinese in itself is by no means uniform.
This part of China is characterised by the existence of many different subcultures within the Han Chinese population. The Hakka may be the most important among the various subgroups of the Han Chinese when we consider the cultural and social relations with the She. As noted above, most of the She population in Fujian are reported to speak a Hakka dialect today, and it may be natural that the She had had contacts with the Hakka more often than with other Han Chinese groups because the former were living mainly in the hilly areas of southeastern China.

But it is also true that distribution of the She today does not coincide with that of the Hakka. In Guangdong, those areas which have been reported to have She inhabitants such as Chaozhou and Haifeng are found just outside or on the fringe of major Hakka areas. On the other hand, Meixian where the Hakka are most concentrated is reported to have no She. In Fujian too, the distribution of the She is concentrated in the northeastern corner of the region where the northern Min dialect is dominant, but in the southwestern part of the region around Longyan and Changting, both major Hakka areas, only small numbers of She are found. In Zhejiang and Anhui, there are no Hakkas.
The close contact of the She people with the Hakka and their cultural assimilation to the latter must be traced back to the historical past. Migration routes of these two groups previously drawn by ethnologists and historians suggest that they had contact with each other in an area near the border of Guangdong, Fujian, and Jiangxi Provinces in the Song, Yuan, and Ming dynasties (see Maps 2 and 3).
This area seems to have special importance for the formation of the ethnic identity of the She people as well as for that of the Hakka. According to an oral tradition widely held among the She living in northeastern Fujian and Zhejiang, their ancestors came from Fenghuangshan Mountain in Guangdong which is located on the border of Meixian and Chaozhou. They claim that this place is a homeland for them, for Panhu, the mythical founding ancestor of their entire ethnic group, lived and died on this mountain.
For the Hakka too, this area on the border of the three region is also a homeland from which they migrated to various places in and outside China. Furthermore, in historical studies on the Hakka it is generally accepted that the major parts of current Hakka cultural tradition and self-identity were formed after they arrived at this area, even though their ancestors might have come down from the northern part of China as Luo Xianglin indicated (Figure 2). We may conclude that the encounter of the She and the Hakka in this area in the Song and later periods had a crucial effect on the ethnic identities of both groups.
But today's distribution of the She and the Hakka population indicates that after the She had been strongly influenced by Hakka culture in the both the region and period mentioned above, they moved outside the core area of Hakka culture. It is as if they could retain their own ethnic identity only by migrating to non-Hakka areas. We need to examine more closely the actual cases of sinicisation or "Hakkanisation" of the She to understand how they came to maintain their ethnic identity vis-a-vis the Han Chinese.

Sinicisation of the She in Chaozhou District, Guangdong

Here I shall report briefly on the present situation of the She in Chaozhou District of Guangdong on the basis of my field research conducted in four She villages in the summer of 1990. As noted above, the She in Chaozhou District have been reported to be more sinicised than the She in other places, even though Chaozhou is very close to Fenghuangshan, the supposed homeland of all the She. For this reason I thought it to be strategically important to choose this place for the observation of the assimilation of the She people to the Han Chinese culture.
Chaozhou is a old city located on the lower reaches of Han River and served as a political and economic centre for the whole area of eastern Guangdong until it was superseded by Swatow situated thirty kilometers to the south. Today, Chaozhou City as an administrative unit has a population numbering over 1,200,000, most of which are Han Chinese who speak the Chaozhou (or "Teochiu") dialect. Only six villages in the northern part of the Chaozhou city are inhabited by She. I surveyed four of these six villages: Shanli, Wanyao, Lijiangkeng, and Shiguping.
All four of these villages are located on steep hillsides, and slopes around the hamlets have been made into terraced fields for crops such as wet rice, tea bushes, and pineapple. Each of these villages is composed of a tightly nucleated settlement, the population size of which varies from 320 to 360. Wanyao and Lijiangkeng are single-surname villages named Lan and Lei respectively, and all of them are She. Shanli, on the other hand, has seventy-two households, of which sixty-four are surnamed Lei, while the remaining eight are Han Chinese surnamed Huang. In Shiguping, out of its 350 residents about 300 are She, all of which, except one household surnamed Lei, are surnamed Lan. The remaining fifty residents are Han Chinese surnamed Wu who came into the village after Liberation. Therefore, the two latter villages are multi-surname as well as multi-ethnic in their composition. Lan and Lei are two of the four surnames that according to She mythology are descended from Pahhu.
Except for Shiguping which specialises in the growing and processing of tea leaves, the livelihood of these villages depends on wet rice cultivation in terraces around the village. But recently pineapple has become increasingly important as a cash crop. Slash and burn cultivation has not been practiced within living memory.
The history of these villages is by no means clear, but fragmentary evidence taken from their oral traditions suggests that their ancestors settled in the present village sites relatively recently. For example, the people surnamed Lei in Shanli village claim that their ancestors first settled down in the village about 180 years ago. And villagers of Lijiangkeng remember that the village had been inhabited by Han Chinese surnamed Li before their ancestors came to live there.
The people surnamed Lei of Shanli, the people surnamed Lan of Wanyao, the people surnamed Lei of Lijiangkeng, and the people surnamed Lan of Shiguping all have their own ancestral halls. These halls are completely the same as those of Han Chinese villages nearby in regards to architectural style, ancestral tablets, and incense burners, as well as in their social functions. Each of these She villages also has a small shrine called Laoyeguan which is dedicated to such Chinese deities as Xietiandadi, Sanshanguowang, and Guandi, and shrines of this kind and style are very popular in Han Chinese villages in this area too.
She inhabitants of Shanli, Lijiangkeng, and Shiguping have ancestral paintings (zutu). The theme of these paintings is the Panhu myth which explains the origin of the four surnames of the She people, Ban, Lan, Lei, and Zhong. In the past, the paintings were kept by the zuzhang or the lineage heads, but today they are under the control of village representatives. These paintings are hung on the walls of the ancestral hall in each village for ancestor worship at the new year.
As for annual rites, most of them are the same as in Han Chinese villages; for example, Chinese New Year, Qingming, Zhongqiu, Dongzhi (winter solstice) and the minor rituals performed on every first and fifteenth day of the lunar calender. Almost the only ritual that seems to be peculiar to the She of this area is Zhaobing which was performed in the winter time before the Cultural Revolution. The timing of the performance varied from village to village; between Dongzhi and the Chinese New Year in every year (Shanli); between the autumn harvest and the Dongzhi every three years (Lijiankeng); on the last day of the twelfth lunar month of every year (Shiguping). In each village they needed to invite a religious specialist called fashi to perform it.[3]
At least from their appearance, these She villages have no special characteristics which distinguish them from Han Chinese living in the villages around them. Although these She villages are all located on hillsides surrounded by extensive terraced fields, such a landscape is by no means confined to She villages in this mountainous part of eastern Guangdong. Nearby villages, in which all the residents are identified as Han, are exactly the same as these She villages. Strangers who visit these villages without any preliminary information would never dream that they belong to a non-Han minority group.
On the other hand, as far as their own consciousness is concerned, the villagers seem to consciously identify themselves as She today. At least toward outsiders such as the government officials and ethnologists, they express their ethnic identity very positively. This attitude may result from the policy for minorities adopted by the Communist government. Today, She in these villages clearly have an advantage over their Han neighbours in regard to their access to electricity, piped water, transport infrastructure, and higher education. Furthermore, birth control regulations do not apply to them. These advantages have encouraged them to stress and positively identify themselves as She.
In the past, their own ethnic consciousness seem to have been very obscure. It was not until the 1950s that the term "She" came to be used as the name for their own ethnic group in this area. Before that, they did not have any particular term for addressing their own group, though it is said that a contemptuous name as Goutouzu (people with dogs' heads) was once used for them by their Han Chinese neighbours.
Just before the communist policy for minorities was implemented, the sinicisation of the She people in this area was very extensive. In addition to customs, their social institutions also seem to have been assimilated to the Han Chinese. For example, they had patrilineages whose collective power was visualised in the ancestral halls. Lineages were segmented into several branches called fang, and the lineage and their branches possessed some common properties. Their villages are of the single-lineage village type, which mostly consists of members of one powerful lineage, only occasionally with optional members with other surnames. Usually, village affairs were managed by the elders of the lineage. This type of village is typically found in the Han Chinese of southeastern China, and was adopted by Freedman as the basis of his model (Freedman, 1958; 1966).
Even such a ritual as Zhaobing may have been borrowed. According to the information which I got from the staff of the Guangdong Institute of for the Study of Chinese Nationalities, the ritual of Zhaobing has much in common with a Taoist ritual called Zuosheng which was performed periodically by Hakka villages in the Meixian district of Guangdong before Liberation.
How, then, do the She conceptualise the cultural differences between themselves and the Han Chinese? Asked about this point, She informants almost without exception said that there is essentially little difference between the Han and themselves in respect to the daily way of life. But at the same time, they pointed out a few elements which they regard as features peculiar to their ethnic group in comparison with the Han Chinese. The first one is their possession of zutu which explains the origin of their ethnic group from the mythical dog Panhu. The second one is a taboo against eating dog meat which is probably related to the Panhu myth. The third is a taboo against using treadle-driven mortars and pestles to pound the rice. The fourth is "their own language" which they claim to be different from dialects of Han Chinese.
Villagers still paying great respect to zutu and carefully preserve them. Of course, there may be a practical reason as well as a religious reason for this kind of respect. Apparently, when judging the validity of local claims to being She, the Chinese authorities and their collaborating ethnologists regarded the possession of zutu as the most reliable indicator.
Eating dog meat was regarded as a taboo in the past, and this taboo is still followed rather strictly at least among the elder generation, although many young people will usually partake of it when they are outside the village. On this point the She people of these villages seem to be in marked contrast to the Han Chinese people of southeastern China who often eat dog meat and the villagers themselves are well aware of this difference. They explain that they cannot eat dog meat because their ancestor Panhu was a dog. But it is reported that the She in some other places do not have such a taboo (Zhou, 1982: 104), so this taboo does not seem to have been shared by all She people living in different places.
The type of rice pounder is also regarded by the She as an important indicator of the difference between the Han and themselves. Han Chinese peasants in this area used to pound their rice with a treadle-driven mortar and pestle, but it was taboo in the She villages to use this tool; apparently they always pounded the rice in a mortar with a hand-driven pestle. Some explain that the She people hold rice in high respect and do not dare to tread on it. Others explain that it is because the shape of the pedal pounder resembles a dog.
The fourth item, "their own language" is the most complicated and the most important. Chinese ethnologists have reported that the She in Chaozhou District have already lost their original language, and today speak a Chinese dialect which is very close to the Chaozhou or the Hakka dialects. They communicate very well in the Chaozhou dialect with local Han Chinese people, but maintain that they speak "their own language" among themselves. Those few words which they gave me as samples of "their own language" very closely resemble the Hakka dialect.
Therefore we may conclude that the older descriptions in the reports mentioned above are essentially correct. But when asked about possible relations between "their own language" and the Hakka dialect, they strongly denied any connection between the two. From their own point of view, their language is peculiar to their own ethnic group, and it cannot possibly be similar to any Han Chinese dialect. They claimed that there were no Hakka Chinese in the vicinity of their villages and that there was no chance for contact with them. Because local Han Chinese speak the Chaozhou dialect and do not speak the Hakka dialect, they regard their language as an important item which demarcates the ethnic boundary between the She and the Han.

The ethnic boundary and cultural interaction between the She and the Hakka

As noted above, although there are many facts which indicate that Hakka culture exerted a strong influence on the She in the past, the present distributions of these two groups do not overlap. It seems as though they are trying to avoid each other. This may be explained partly by the driving force of Hakka migrants who expelled the She people from the places they previously inhabited in today's major Hakka areas such as Meixian and Zhangting, and drove them into the northeastern part of Fujian and further north. But this hypothesis is too simple when we consider the extent to which the She have been assimilated to Hakka culture into consideration. Isolation or escape from the Han Chinese could not have been the only factors that enabled them to survive as a discrete ethnic group.
The She people in Chaozhou, as noted above, speak a language which they believe to be "their own language" but which is actually very similar to the Hakka dialect, and Han Chinese people living around them are almost without exception speakers of the Chaozhou dialect. This case suggests that the maintenance mechanism of the She ethnic boundary might be dependent on the difference between the Hakka culture to which they had once been assimilated very closely and the other Han Chinese local cultures which are dominant in the areas where they live today. In other words, the cultural heterogeneity of the Han Chinese in the southeastern part of China seems to have played a vital role in the maintenance of ethnic boundaries between this highly sinicised minority people and the Han Chinese.
This hypothesis can be applied to the She in northern Fujian and southern Zhejiang too, whose language is also reported to be very similar to the Hakka dialect, and where the Han Chinese majority around them are speakers of the Min dialect or the Wu dialect. So we can suppose that the She people who moved out of the core areas of Hakka culture after being strongly influenced by it survived as a discrete ethnic unit owing to the difference between the version of Han Chinese culture which they had adopted and that of the local Han Chinese majority.
On the other hand, those She who continued to live within the core areas of Hakka culture, found it difficult to keep their ethnic boundary intact and were inevitably assimilated into Hakka culture. This hypothesis on the maintenance of the She/Han ethnic boundary seems to be useful at least in explaining why the distributions of the She and the Hakka do not overlap today.
Recently, in Huiyang, Meixian, Longyan and other districts, people who had previously been classified as Han Chinese applied to be registered as She. At present the authorities seem to be treating these claims very prudently, and only acknowledge the claim when it is validated by some objective facts such as the possession of zutu. This suggests that in the core areas of the Hakka culture, those people who had originally been She were absorbed into the Han Chinese majority more easily than other places. On the other hand, in non-Hakka Han Chinese areas the She people were in most cases already classified as a non-Han minority in the 1950s, even though their level of assimilation to the Hakka culture was pronounced.
Although it is not a well-defined concept even among specialists in ethnological or anthropological studies of China, the term hanhua or "sinicisation" is often used in describing an aspect of the cultural relations between the Han Chinese and minority peoples. This term usually refers to the process by which minorities accept or are forced to accept Han Chinese culture. At the same time, we usually presuppose that the "Han Chinese culture" in this context is a single body of cultural traditions characterised by Confucian morals, Taoist concepts, Chinese bureaucratic ethics, and so on. From the very general point of view, such a conceptualisation of Chinese culture may have some value. But when we look at the actual process of contact between the Han Chinese and the minorities, we find that the culture of the Han Chinese people is not necessarily homogeneous. Especially in south China, where a number of local versions of the "Han Chinese culture" exist, "sinicisation" is by no means a simple process.
On the other hand, we must also ask the reason why there exist so many versions of Han Chinese culture in south China in the first place. At this point we have to take into account the cultural influences exerted by the minority peoples on the Han Chinese.
In the ethnohistorical studies of the Hakka Chinese since Luo Xianglin, the social continuity and the cultural orthodoxy of the Hakka have been always emphasised. For followers of this school of thought, the Hakka people are the legitimate descendants of the ancient Chinese in the Yellow River Plain as a centre of the Chinese civilisation both in a cultural and a genealogical sense. But even Luo admitted that the encounter with the She people could possibly have had some impact on the formation of the cultural tradition of the Hakka. "It can be said that the peculiarities of the Hakka mainly derive from such factors as their historical migration, their own social tradition, and the natural environment of the land where they settled down, but the acceptance of the She's influence is also a factor that we can not regard as utterly unimportant, in just the same way as we cannot ignore the question of the Cantonese people in Guangdong and Guangxi receiving the influence of the Baiyue or Burman tribes" (Luo, 1933: 76).
Besides language, there are many cultural similarities between the Hakka and the She. Such traits as the active participation of women in manual labour and the tradition of singing folk songs at celebrations are common to both groups, and there is also a similarity in their burial practices. In addition, their kinship systems have much in common. She villages used to be composed of a single lineage, the headquarters of which was located in their ancestral hall. Although villages of the single-lineage type are by no means rare throughout the southeastern part of China, small scale single-lineage settlements with their own ancestral hall are especially common among the Hakka. As far as I have been able to determine in my own field observations in the New Territories of Hong Kong, modest size local lineages seem to be found more often among the Hakka than among the Cantonese, while extensive higher-order lineage organisations or villages without lineage organisations are frequently found among the latter (see Segawa, 1992: chap. 2, chap. 3).
Such distinctive development of lineage-based social systems among the Hakka was sometimes interpreted as resulting from their isolated and self-sufficient economic condition in mountainous areas in the past (see Nakagawa, 1980: 77). And as for the similarity found in the social organisations of the She and the Hakka, we usually suppose it to be the result of the She's assimilation to Hakka culture. We tend to regard it as evidence of She's profound sinicisation. But, at the same time, we need to consider the possibility of the influence of She social tradition on the Hakka.
Takemura Takuji has argued in his book that the hill tribes of east Asia such as the Yao and the She were generally characterised by the strength of their genealogical concept, and he surmised that they clung to this concept because it was the most effective method for self-identification and mutual communication given their highly mobile and dispersed way of life (Takemura, 1981: 109-111). So, the lineage-based social organisation of She village can not be seen as a mere copy of the Han Chinese pattern but rather it seems to be founded in their traditional lifestyle in the mountainous areas of south China.
Such possibilities of the She exerting a cultural influence on the Hakka seem to have been investigated only on a very limited scale. There seem to be two reasons for this. First, up to now ethnological studies in mainland China have usually been concerned almost exclusively with minorities and there the major focus has been to identify the cultural traits which are thought to be peculiar to each minority group. On the other hand, traditional customs of the Han Chinese have usually been regarded as outside the range of ethnological studies. Consequently, little attention has been paid to the cultural interactions between the minority groups and the Han Chinese.
The second reason is that the ethnohistorical study of the Hakka carried out by Luo Xianglin tended to estimate the possibility of She influence on the Hakka as minimal. Luo's study was motivated very strongly, at least in its initial stage, by the desire to disprove the prejudicial assumption that the Hakka were not Han Chinese. Of course, as noted earlier, Luo also admitted that Hakka might possibly have been influenced by the She people during their historical contact. But he presupposed that the only possible interaction between them was intermarriage between Hakka men and She women and that the cultural influence of the She on the Hakka, if any, would have been negligible, because the cultural superiority of the latter was quite self-evident (Luo, 1933: 74-75).
This pattern of inter-ethnic marriage between Han Chinese and the minorities has also been reported for other places (e.g. Hsieh, 1979: 42-43), and it might have been one of the more important channels through which the Han adopted elements of the minority cultures. Yet as for the historical process of interaction between the She and the Hakka, we do not have enough evidence to claim a widespread pattern of intermarriage.
For Luo whose historical study of the Hakka was mainly based on an analysis of their genealogies, this hypothetical model of interaction between the She and the Hakka was a very convenient one, because it enabled him to explain how She elements infiltrated into Hakka culture without undermining the glorious continuity of the patrilineal descent of the Hakka since ancient times. But the genealogical books are not mere records of objective facts in the past. They were usually compiled in accordance with the wish of family members to confirm their identity and pride. Therefore, even if the written genealogies are not fictitious, it is highly probable that they represent only certain aspects of past social conditions.
Of course we have to depend on some historical documents in order to examine the sinicisation process of the She in the past. But it seems that another effective way for this kind of study is to check the local variation of Han Chinese culture in detail, and then compare it with the distribution of the She people at present and in the past. Because our knowledge has been very limited in regard to such variation until recently, studies of the cultural interaction between the Han and the minorities have also remained at a general level. In practice, Han cultural traditions vary a lot from one place to another even within a dialect group such as the Hakka or the Cantonese. Today, Han Chinese communities in mainland China have become more open to field research. The accumulation of more precise data on the local traditions of the Han Chinese in the future, may enable us to completely revise our view of the ethnic boundary between the Han and the She.


1. According to this myth, their founding ancestor Panhu was a pet dog of the legendary emperor Gaoxin and the emperor gave his daughter to him after he helped him greatly in the subjugation of his enemies. From this marriage three sons and one daughter were born, who became the founding ancestors of the four surname groups of the She: Ban, Lan, Lei, and Zhong.
2. For example, in Outline of the History of the She People it is supposed that both the She and the Yao were descended from the Wulingman tribe of Hunan in Han and Jin period (Shezu Jianshi Bianxiezu, 1980: 20). On the other hand, Jiang regards the She as the descendants of ancient Baiyue tribes (Jiang, 1980) .
3. For more information on this ritual, see Shi (1989: 151-153).

Return to Contents page

Updated 4 June 2020