Who Are the Hakka?

As the following chapters demonstrate, "Hakka" defies any attempt to arrive at a single, simple definition. To some, including many in Shung Him Tong, the name "Hakka" evokes pride and patriotism and connotes Chinese origins of high status dating back to when Henan province was the "cradle of Chinese civilization" in the fourth century A.D. To others the term can evoke shame and embarrassment. Spoken by a non-Hakka, it can suggest poor, uneducated country bumpkins—connotations similar to those of "Oakie" and "hillbilly" in the United States.

Today, the name "Hakka" is commonly used in Guangdong, Hong Kong, Taiwan, Southeast Asia, North America, and other regions throughout the world to refer to the approximately seventy-five million members of this Chinese ethnic or subethnic group. "Hakka" is the Cantonese (or Yue) pronunciation of a term that translates literally as "guests" or "stranger families," or less literally as "settlers," "sojourners," "immigrants," or "newcomers." Scholars speculate that the name originated in the descriptive term used before the seventeenth century in Chinese population registers to distinguish recent immigrants from earlier Yue inhabitants (Leong 1980). The label "Hakka" indi-

cated the relatively recent arrival of a group of people in Guangdong as compared with the longer tenure of the Cantonese-speaking inhabitants. The word "Hakka" also connotes a transitory or temporary social status; indeed, many people, including members of some of the oldest lineages in the New Territories, are thought to have once been Hakka who crossed the "ethnic boundary" and assimilated into the larger, dominant Cantonese-speaking group for a variety of political and economic reasons (Baker 1966; Faure 1986; J. Watson 1983).

When the term became fixed as a group label is a debated issue (M. Cohen 1968; T. Hsieh 1929; Leong 1985; see chap. 2), but by the nineteenth century it clearly distinguished a group of people who were not Punti, that is, not "indigenous" or "native inhabitants" of Guangdong province in southeastern China. By the early 1920s, due in large part to the establishment of the Tsung Tsin (Chongzheng) Association in Hong Kong, and the United Hakka Association (Kexi Datonghui) in Shanghai and Canton, the name became more widely accepted as an ethnic label. These two associations, founded by Hakka elites and intellectuals, were highly successful in their goals to unify Hakka organizations worldwide and to promote Hakka solidarity, but perhaps somewhat less so in their attempt to foster a public understanding of Hakka culture and identity (T. Hsieh 1929; Leong 1985). In 1921, in response to a Shanghai Commercial Press publication of The Geography of the World that erroneously described the Hakka as non-Chinese, the United Hakka Association held a conference in Canton attended by over a thousand angry delegates representing Hakka organizations worldwide (ibid.). The result was a forced retraction of the offending phrases. Since then Hakka associations continue to regularly publish and assert their views regarding Hakka identity, which are discussed in the following chapter.

In the People's Republic of China today, the Hakka are officially recognized as members of the Han Chinese minzu or nationality, but in nineteenth-and early twentieth-century China their Han status was debated and not widely acknowledged. At that time violence often broke out between Hakka and non-Hakka (cf. M. Cohen 1968; Leong 1985; W. Lo 1965), and Hakka, with their different language, clothing, and cultural practices, were often accused of being non-Chinese barbarians of lowly tribal origin. Although this accusation is now uncommon, and linguists and Hakka historians have demonstrated that it has little basis in fact, it still underlies the gravest insults that are directed toward the Hakka in Hong Kong today.

Anthropologists have used many different approaches to analyze or attempt to "define" ethnic groups and their identities. One older approach that has been aptly criticized tries to demarcate an ethnic group on the basis of a distinct culture or a list of distinguishing cultural traits or markers. Language, place

of origin, clothing, food, religion, and numerous other cultural criteria have all been used to define ethnic groups. The case of the Hakka demonstrates the problem with such an essentialist approach.

No single cultural trait or group of traits can determine who is Hakka and who is not. Some Hakka may suggest that all Hakka speak the language, come from Meixian, or cook stuffed tofu. Non-Hakka might suggest that all Hakka are poor and darker skinned and wear circular black-brimmed hats. However, many of those who are considered Hakka, or who consider themselves Hakka, do not fit these criteria. In Hong Kong today, "Hakka" refers sometimes to people who speak or whose ancestors spoke Hakka language, yet not all of these would identify themselves or be identified by others as Hakka. Conversely, many who no longer speak Hakka still consider themselves to be Hakka; the young people of Shung Him Tong are a case in point.

Place of origin presents a problem in defining the Hakka because although some proud Hakka say their people originated in north central China as early as the fourth century A.D. , this is where all Chinese claim their origin. As Skinner (1977) and Leong (1980) have pointed out, the Hakka are the only Chinese "ethnic group" not to be named after a single place of origin or native place they can call their own. The Cantonese come from the area around Canton; the Chaozhou from the synonymous region, and likewise for the Shanghainese.

Although many Hakka identify Meixian or the wider region of Meizhou as the Hakka "heartland" or the core of Hakka culture, the areas that Hakka identify as their native places are as widely dispersed as is the Hakka population. Hakka are most densely concentrated in northeastern Guangdong, east of the North River, in the mountainous, less fertile region of Meizhou prefecture that includes seven predominantly Hakka counties surrounding Meixian. But sizable Hakka populations are also located in southwestern Fujian, southern Jiangxi, and eastern Guangxi, on Hainan Island, Hong Kong, and Taiwan, and in lesser numbers in Sichuan and Hunan. Overseas, Hakka can be found on virtually every continent, from South and Southeast Asia and the South Pacific to Europe, North and South America, Africa, and the Caribbean. This wide migration—discussed in the following chapter—makes it all the more impossible to identify any one set of cultural traits that can be labeled as Hakka.

Certain settlement patterns have been identified as typically Hakka, but these demand special qualification and contextualization. As relatively late arrivals in many of the regions of China where they settled, Hakka were generally forced into the more hilly, less productive, and less desirable lands. Such was the case in Guangdong, Guangxi, and the New Territories of Hong Kong, where the Punti or Cantonese had already settled the more fertile river valleys, and also in Taiwan where the Hokkien (Min speakers) owned the better land (cf. Lamley 1981; Pasternak 1983, 1972).

During the eighteenth and nineteenth centuries, Hakka settlements in some regions of Guangdong could be distinguished from those of the Punti. There the Hakka often lived in small numbers, sparsely dispersed throughout the hills on land that they often rented from Punti landlords. In contrast, Punti were more likely to live in densely populated towns or in large, single-surname villages surrounded by their fields. In other regions Hakka and Punti occupied separate villages in the same areas (M. Cohen 1968).

There are also certain architectural styles that appear to be uniquely Hakka, but again these point to strictly regional variations. In southwestern Fujian and northern Guangdong, Hakka built "roundhouses," circular, multistoried, fortresslike dwellings designed for defensive purposes with walls of adobe or tamped earth nearly a meter thick (Knapp 1986:45–49). At one time, in certain locations, Hakka communities might have been identified by their architectural styles (see Naquin and Rawski 1987:180), or by the fact that their villages were more dispersed than those of the Punti (M. Cohen 1968; Eitel 1867), but today in Hong Kong these are no longer valid distinctions and a Hakka house or village can generally not be distinguished from a Cantonese one. Hakka now reside in villages that were once Punti, people who were once Hakka are now identified as Punti, and members of both groups often live side by side in both urban and rural areas.

In many ways Hakka and non-Hakka Chinese in Hong Kong today are virtually indistinguishable. Most dress the same way, speak Cantonese, and eat virtually the same foods. Although some people say that the Hakka are shorter and darker skinned and have larger feet, it is impossible to distinguish them from other Chinese on this basis. In terms of clothing, a few older women still wear a black cloth draped over their heads (similar to a nun's wimple) and tied over the top and behind the ear with an embroidered band (Blake 1981, 105–10, 150–51; E. Johnson 1976a; see plate 3). Others wear what is called the "Hakka hat," a flat circular hat with a black cloth "curtain" around the brim; but as often as not these women either are not Hakka or do not identify themselves as such.

The "Hakka hat," the most public symbol associated with the Hakka, is worn by many Hakka and non-Hakka women who work outdoors. It serves, like the stereotypes of large feet, muscularity, and dark skin, as a class marker: those who do hard labor are naturally more tanned and more muscular and do not wear dainty shoes. Tourist brochures and postcards reinforce the stereotype of Hakka by labeling women farmers and construction workers wearing the "Hakka hat" as Hakka. In Shung Him Tong village, women are rarely seen wearing such a hat, although several older women wear the black head cloth with embroidered band that they say is the only "real" remaining clothing marker of Hakka identity.

In Hong Kong, there are few indicators of whether a person is Hakka, other than hearing him or her speak the Hakka language, which is increasingly rare since many Hakka speak Cantonese in public. When they are away from their village, the people of Shung Him Tong are no exception. The majority of their interactions with people outside of the community—with the exception of their association with members of other Hakka churches and the Hong Kong Hakka Association—are without reference to their Hakka or Christian identities. Only occasionally, I was told by several people under forty, does one discover that a coworker or classmate is also Hakka. For the teachers, politicians, and businesspeople I spoke to, the majority of their interactions outside of the village are with non-Hakka non-Christians, and they do not make it a point to let others know that they are Hakka. Most of the time they "pass" as Cantonese or Hong Kong Chinese. Christian identity may be more apparent because of such outward markers as jewelry and desktop decorations, and because it is a more common topic of conversation. As will become more apparent in the course of this book, the church is the main context in which Hakka identity is important for the people of Shung Him Tong.

Although there are few visible markers of ethnic identity in Hong Kong, in some specific locations or contexts ethnic identity may be presumed. Some villages, such as Shung Him Tong, are known to be Hakka by those who live in the vicinity. Certain prominent individuals are also known to be Hakka. The winners of the "Miss Hong Kong" competitions of 1986 and 1987, for example, were both reported in the local newspapers as being Hakka. This knowledge is often only noteworthy to other Hakka. Certain restaurants may be billed as Hakka restaurants—though these are not as popular or as conspicuous as the large, elaborate, popular Chaozhou restaurants that were established in the 1980s. As some Shung Him Tong villagers explained, Hakka restaurants are not tourist attractions, but Hakka people know where to find them. At Luen Wo and Cheung Wah, the two markets closest to Shung Him Tong, it does not appear that most people conduct business along ethnic lines (see also Blake 1975, 1981; cf. E. Anderson 1968). Most shoppers would rather look for the best bargain or the freshest products, or return to the same vendor whom they trust, regardless of ethnic identity.

Although some anthropologists, Hakka historians, and nineteenth-century missionaries have presented evidence of distinct Hakka religious practices, styles of clothing, foodways, and architecture, these are regionally or generationally specific and cannot distinguish Hakka from others except in a particular historical and cultural context. One of the objectives of this book in considering the many beliefs regarding Hakka cultural differences is to avoid an essentialist, reductionistic portrayal of the Hakka as much as possible. The case of the Hakka supports the view that "ethnic groups are categories of ascription

and identification by the actors themselves" that cannot be reduced to a static list of traits (Barth 1969:10). Furthermore, ethnicity is situational, structural, and interactional. Dual Hakka and Christian identity takes on great relevance to residents while they are in Shung Him Tong, but with their classmates, coworkers, and colleagues outside of the village this identity is far less an issue.

Another general approach to the study of ethnicity has been labeled the "circumstantialist" or "instrumentalist" approach, in which ethnicity is viewed as primarily a manipulative or political strategy.^[3] Typical of this genre is Abner Cohen's (1969) work on Hausa immigrants in Yoruba towns in Nigeria, in which he demonstrates how the Hausa use religion to strengthen the ethnic boundaries between themselves and the Yoruba in order to protect the political and economic interests of the Hausa community. Similarly, Barth has used an ecological analogy to suggest that different groups maintain structural boundaries in order to maintain their monopolies of particular occupational niches (1969).

Leong (1980, 1985) has illustrated how an instrumental approach is applicable to the Hakka, particularly during the nineteenth and early twentieth centuries when Hakka could perceive themselves as sharing an economic niche in competition with the Punti. Blake has also convincingly demonstrated the way in which, during the 1970s, political and economic distinctions in the Hong Kong market town of Sai Kung lined up along ethnic/linguistic lines that were "negotiable" and situational (1975, 1981). Today there is no question but that Hakka associations, including the Hakka church, continue to exist and thrive in vastly different settings, serving a variety of the political, economic, and other needs of their members. As I relate in Chapter 2, Hakka identity was the basis of a number of political and economic groupings during the nineteenth century. In Chapter 3 I show how ethnic and religious solidarity served Hakka interests in the establishment and foundation of Shung Him Tong.

However, today in Shung Him Tong, Hakka identity no longer serves the instrumental economic or political interests that it did in earlier decades. That is not to say that members of the community do not share certain political interests, but in Shung Him Tong, as in many parts of Hong Kong today, ethnicity has decreased in political relevance. Most occupations and economic interests no longer break down along ethnic lines to the extent that they once did, and there is little competition with other ethnic groups for economic resources. During the past twenty or thirty years, with the decline of ethnic tensions and of Hakka/Punti economic competition, one should expect, according to Barth, "either no interaction [between ethnic groups], or interaction without reference to ethnic identity" (1969:18). Neither has been the case for the Hakka of Shung Him Tong.

Regardless of how useful an instrumental approach may appear, it too often overlooks the less material, more subtle features of ethnic identity and ethnic groupings that help explain the recent "reemergence" of ethnic affiliations in places such as eastern Europe and the former Soviet Union. An instrumental focus does not help us understand periods during which ethnicity is not blatantly political. As I describe in Chapters 4 and 5, economic and political ethnic interests in Shung Him Tong wax and wane, and today they are not as clear, uniform, or crucial a concern as they were during the early decades of this century. But ethnic identity in Shung Him Tong persists—although not in a static and timeless way—because it is linked to the unique history of the community and the particular pairing of Hakka identity with Christianity.

Abner Cohen's (1969, 1974) and Barth's (1969) insights into how ethnic boundaries and interests are created and maintained through cultural symbols are nonetheless very pertinent to the case of Hakka Christians. If the Hakka of Shung Him Tong have a "political strategy," it is to maintain power and control over the meaning of their own identity. At present, that in itself is the goal, rather than the means to another more explicitly political or economic end. The "power" of Hakka Christians in Shung Him Tong is not merely over resources or politics in the narrow sense but over the symbols that define them as a group.

The church is what provides Hakka Christians with a context in which to construct Hakka history and, more importantly, the authority by which they can promulgate and take control of their own identity. The "resources" that they attempt to control are not primarily economic but cultural and religious—if saved souls can be counted as such. As discussed in Chapter 4, Hakka identity plays a central role in attracting converts to the congregation.

Another anthropological approach to ethnicity has been labeled the "primordialist" or "sentimentalist" approach (see Bentley 1981, 1983). The emphasis is not on ethnicity as a strategy but on the basis of the shared identity—on the idea of shared history or ancestry, whether real or imagined. The idea that Geertz called "primordial attachments" (1973a) is useful in an analysis of Hakka identity because it points to history, process, and cultural symbols. In contrast to the situation during the nineteenth and early twentieth centuries, when the instrumental aspect of Hakka ethnicity was most evident, the contemporary situation calls attention to the cultural and symbolic aspects of ethnicity (see chap. 6). Shared history and a common "primordial past" are the central themes of Hakka identity in Shung Him Tong today.

Keyes is one of several scholars who have taken the position that ethnicity should not be seen as either primordial or instrumental but entails both facets (Keyes 1981; see also Bentley 1981, 1983, 1987; Harrell 1990; Nagata 1981).

Focusing on the primordial aspect of ethnicity, Keyes argues that ethnicity "derives from a cultural interpretation of descent" (1981:5, 1976). While he views ethnicity as "a form of kinship reckoning," he stresses that "it is one in which connections with forbears or with those with whom one believes one shares descent are not traced along precisely genealogical lines" (1981:6). In the case of Hakka Christians this is an important point. Hakka pay particular attention to Hakka history—which serves as their "genealogy"—in order to reinforce their claims to Chinese identity.