Chapter in Imaging China: Regional Division and National Unity

Ed. by Shu-min Huang & Cheng-kuang Hsu.

Taipei: Institute of Ethnology. P. 43-80.

From the Qiang Barbarians to the Qiang Nationality:

The Making of a New Chinese Boundary

Wang Ming-ke

This chapter has two purposes. First, we study how the mountain people living along the upper Min River Valley in northwestern Sichuan have gradually gained their Qiang 羌 ethnic identity in the last half century as the result of selecting and remodeling their own history, folklore, legends, and historical as well as ethnographical information provided by the Han Chinese and foreign missionaries. Second, we examine how the geographical and ethnic concepts of the Qiang have been re-delineated and the pertinent histories have been rewritten by the Han Chinese to define and redefine their ethnic boundaries for more than 3,000 years, especially in the recent 100 years in which traditional China came to be a nation-state and the Han Chinese the majority population in a nation ¾ Zhonghua minzu 中華民族 ¾ in a modern sense.

In the 1930s, Maurice Halbwachs suggested that memory is a collective social activity, and that all social groups and institutions have corresponding social memories. Social organization provides a persistent structure to which our memory must conform. To reinforce the sense of oneness, members of a social group preserve their collective memories in, or renew their collective memories through, mnemonic media such as artifacts, rituals and writings (Halbwachs 1992: 38, 119; Coser 1992: 22). Since the 1980s, questions such as how societies constantly imagine or invent their own pasts to enhance group feeling among their members (Hobsbawm 1983; Anderson 1991), and how societies have reconstructed or reinvented their pasts to meet social changes (Schwartz 1990; Fentress and Wickham 1992) have become major concerns among scholars of various backgrounds.

The narrative aspect of the past is also noted by students of ethnicity and nationalism. One of the conclusions that primordialists and instrumentalists in the anthropological studies of ethnicity of the 1970s reached was that common origins are important in solidifying ethnic bonds. Furthermore, these so-called common origins need not be rooted in a factual past but can be a subjectively reconstructed past, corresponding to what Clifford Geertz termed “assumed blood ties” (1973: 261-62) and to Charles Keyes’s notion of “cultural interpretation of descent” (1981: 5). Over the years, more scholars have turned their attention to the study of how “history” is imagined and selected by various ethnic/social groups to solidify group attachments (Tonkin, McDonald and Chapman 1989; Smith 1987: 191; Roosens 1989; Hobsbawm and Ranger 1983; Anderson 1981). There is also a similar tendency among many anthropologists “to do ethnography of the historical imagination” (Comaroff and Comaroff 1992). In this respect, the case of the Qiang with its history of historical imagination through 3,000 years in Chinese documents provides an excellent opportunity for examining how 'history' has been rewritten and reinterpreted by the Han to redefine their ethnic boundaries, and how events and legends of the past have been selectively learned and reinterpreted by the natives to create their Qiang identity.

Today the Qiang is one of the 55 officially recognized ethnic minorities in China, with a population of approximately 200,000 living mainly in Wenchuan 汶川, Lixian 理縣, Maoxian 茂縣, Songpan 松潘 in Aba Zangzu Qiangzu Autonomous Prefecture 阿壩藏族羌族自治州, and Beichuan 北川 in the Mianyang 綿陽 region, all in northwestern Sichuan province. Although now a minor segment of the population, the Qiang is commonly believed to be a very old people whose history can be traced to the Shang dynasty (Gu 1980; Ren, Li, & Zhou 1984). It is also believed to be a once strong and populous group whose offspring can now be found among, besides the Qiang in northwestern Sichuan, the Han, the Tibetans, and many minorities in southwestern China (Ran, Li, and Zhou 1984).

From another angle, however, the Qiang nationality is a new ethnicity. Many scholars argue that ethnic self-identification and memory of common origin among natives, and ethnic self-awareness, are the necessary hallmarks of the making of an ethnic group. In this case, the people we call Qiang acquired and accepted the ethnic label Qiang zu 羌族 (Qiang nationality) only in the last half-century. Further, the memories of their past ¾ the history of the Qiang ¾ have been learned, selected, interpreted, and distributed by Qiang literati who became a core group in native society no earlier than the 1960s. Nevertheless, the Qiang today are definitely an ethnic group with a common self-appellation (Qiang zu or erma 爾瑪 [1]), historical memories (even some natives are arguing over the details of these memories and still others are encouraged to learn them) and ethnic self-awareness.

This chapter examines three historical processes. The first is a long historical process from the Later Han dynasty (A.D. 25-220) to the Qing dynasty (1644-1910). Within this period, the concept of the Qiang ¾ meaning ‘those people in the west who are not one of us’ ¾ among ancient Chinese first served as a western ethnic boundary of the Huaxia (the ancient Han Chinese) referring to a huge range of people living mainly along the eastern edges of the Tibetan Plateau. But through interacting with the ancient Chinese and Tibetans, more and more western “barbarians” were classified by the Chinese into new ethnic categories and therefore were pulled out from the general and blurry category of the Qiang. Finally, during the Ming (1368-1644) and Qing (1644-1911) periods the term Qiang denoted only the non-Han people living in the upper Min River Valley and Beichuan area.

The second process is that, during the late Qing period (approximately 1840-1911), under the expansion of political and economic imperialism in China and her peripheral areas, the Han revolutionists and reformists adopted the concepts of nation and nation-state to build their new country, Zhonghua minguo 中華民國 (the Republic of China), and the nation, Zhonghua minzu 中華民族 (the Chinese nation). In this process of nation-building, the traditional peripheral non-Han aborigines, including the Qiang, and the people of subordinate polities under the Mandarin regime became ethnic minorities within the Chinese nation; the nation and nation-state can be seen as a new scheme within which the old image of central kingdom (Zhongguo 中國) surrounded by subordinate people and states can be maintained.

Third, having been identified and categorized as the Qiang nationality, natives began to learn more about themselves. The Qiang literati who have been trained in Chinese language and culture now functioned as the receivers, interpreters, and propagandists of native history and cultures; and then finally the Qiang became an ethnic group in a modern sense.

The Qiang 羌 in China’s Historical Memory

The Chinese have recorded and preserved memories about the Qiang for thousands of years. At about 1300 B.C., people of the ethnic label Qiang , a pictograph composed of sheep and man, had already been recorded on the oracle-bone inscriptions of the Shang dynasty. In the eyes of the Shang people, according to these inscriptions, they were a hostile population living in the west. During the Later Han dynasty (25-220), people with the ethnic label Qiang were widely distributed along Han China's western frontiers. Among them, the Qiang living in the upper Yellow River Valley in eastern Qinghai province are well known for their battles with China at that time (de Crespigny 1984).

The Formation of the Qiang Belt

Historians have pieced together fragments of historical data to reconstruct a whole, continuous history of the Qiang from the Shang to the Han dynasty, and to the present. The presumption underlying these studies is clear: both the Qiang of the Shang dynasty and those of the Han period were the same “people.” In my doctoral dissertation, I posed an alternative explanation: Since the word Qiang was an ethnic label by which the Chinese referred to some non-Chinese in the west, it might be describing a sense of otherness for the Chinese. As the ethnic entity of the Chinese was forming, Qiang came to denote a shifting ethnic boundary; as western populations assimilated into the Chinese, this boundary shifted westward. Finally, during the Later Han period, it reached the eastern edge of the Tibetan Plateau. This was, and still is in most cases, also an ecological frontier for the Han Chinese; the Chinese could not manage the lands and peoples beyond this frontier in the Chinese fashion, nor could the natives of these areas adopt the Chinese way of food production and social organization. The ethnic boundary has remained fixed ever since. Thus, during the period from the Shang to the Han dynasty, the Qiang was not a “people” that had continuity in time and space. Instead, it was an ethnic boundary, a concept among the Chinese. When we trace the history of the Qiang, we are tracing not the native history of a non-Chinese people but a portion of the history of the Chinese (Wang 1992: 98-140).

During the Later Han and Wei-Jin periods (25-220, 221-419), Qiang or Qiang ren (the Qiang people) were widely distributed along the mountainous fringes of the northern and eastern Tibetan Plateau, from the Kunlun Mountains 崑崙山 in Xinjiang province (East Turkestan), and eastern Qinghai area, to southern Gansu, western Sichuan, and northern Yunnan. A geographical or tribal label was added to indicate a specific Qiang people; thus, in this “Qiang belt,” there were, from north to south, Canlang Qiang 參狼羌, Baima Qiang 白馬羌, Baigou Qiang 白狗羌, Bailang Qiang 白狼羌, Qingyi Qiang 青衣羌, Maoniu Qiang 旄牛羌, among many others (Figure 3.1). In this period, the ancient Chinese tried to distinguish some ethnic or racial stock from the general concept of the Qiang. Therefore, there were the Hu 胡 in the southern end of East Turkestan, the Di 氐 in southern Gansu and northern Sichuan, and the Yi 夷 in southern Sichuan and northern Yunnan. However, the distinction between the Qiang and the Hu, the Di, and the Yi is not always clear in Chinese documents.

The Dissociation of the Qiang Belt

During the period from the Southern and Northern dynasties (420-589), to the Sui (589-619) and the Tang (620-907) dynasties, some relatively large political entities emerged in western China, such as the tribal alliances of Dangchang 宕昌, Dengzhi 鄧至, Dangxiang 黨項 and Tuyuhun 吐谷渾, and the Tufan 吐蕃 (the Tibetan Kingdom, also pronounces as Tubo). In Chinese records, the majority of the people under these regimes were the Qiang, Qiang ren 羌人 or zhu Qiang 諸羌 (various Qiang tribes) (Figure 3.2).

The rise of the Tibetan Kingdom in the 7th century is especially noteworthy. The rapid eastward expansion of the Kingdom caused a series of confrontations with Tang China along the eastern fringe of the Tibetan Plateau; this was a belt zone, extending from southern Kansu southward to northern Yunnan, whose inhabitants had been identified by the ancient Chinese as the Qiang. From the mid-7th century to the 8th century, the Tibetan Kingdom conquered almost all of this belt and many of the tribal confederacies of the Qiang, such as Dangxiang, Yangtong 羊同, Dangchang and Dengzhi fell as her victims (Figure 3.3).

After the collapse of the Tibetan Kingdom in the 9th century, and even after the rise and fall of the Western Xia Kingdom (1038-1227) whose inhabitants were mainly Dangxiang Qiang, the influence of Tibetan culture persisted in this peripheral zone. This is because most of the areas in this zone were still controlled and ruled by local chieftains who were (or ascribed to be) descendants of the nobility or generals of the previous Tibetan Kingdom. And more important, Lama Buddhism, in which Tibetan culture was rooted, spread throughout this peripheral zone during the Yuan (1280-1368) and Ming (1368-1644) dynasties. As a result, from the Song dynasty (960-1279) on, especially during the Ming and Qing periods, the most common ethnic label Chinese gave to the natives in this zone became fan 番 (Figure 3.4). This ethnic label referred to the people of Tufan, and to barbarians.

Thus, we may say that the rise and eastward expansion of the Tibetan political and cultural power in the west, and the concomitant confrontation with China along the Qiang belt, altered the Chinese concept toward non-Han people in this belt. From this time on, even its geographic position did not change much, this western belt zone as an ethnic boundary in the Chinese mind ceased to be maintained by the sense of otherness toward the Qiang but rather by the sense of otherness toward the Fan (Tibetans). This change was also illustrated by the name of a new frontier town, Linfan 臨番 (literally adjoining Fan) of the Tang dynasty, which was located near the abandoned city of Linqiang 臨羌 (adjoining Qiang) of the Later Han dynasty (Figures 3.2 and 3.3). The senses of otherness maintained by the concept of Qiang and by that of Fan in the Chinese mind were similar but different; since the historical memories behind these two ethnic labels are different. For the Han Chinese the Qiang represented a blurry ethnic boundary and the Fan a more clearly defined one. As more and more Tibetanized people [2] in the west were classified by Chinese as the Fan, a sharpened ethnic boundary delineating the concept of the Fan was gradually substituted for the blurry one in most places along the old Qiang belt.

There were still two other situations that caused people in the Qiang belt of the Later Han to gradually drop from this ethnic category. The first is the Sinicization of the land and the people. In the northern part of this belt, this situation occurred in the areas occupied by the Dangchang Qiang and later, the Western Xia Kingdom in Gansu (Ran, Li, and Zhou 1984: 179-90). In the central part of this belt, the same process occurred in Tianquan 天全 and Yajiang 雅江 areas in western Sichuan, an area whose inhabitants had been known as the “blue-clothes Qiang.” From the Tang dynasty on, these areas became the lands of imperial China and its inhabitants became registered civilians (Torrance 1920: 5-12; Ran, Li, and Zhou 1984: 100-101). Second, when the Chinese had better knowledge about the people, they distinguished the Yi 夷 in southern Sichuan and Northern Yunnan, who had earlier been identified as Maoniu Qiang, from the category of the Qiang. From the Tang dynasty on, as the result of the rise of the Nanzhao 南詔 (738-902) and the Dali 大理 kingdoms (937-1252 ) in the south, the Chinese had better motives and chances to acquire knowledge concerning people within and surrounding these kingdoms. It is in this political context that there came to be a certain number of official documents, personal travel notes to describe the non-Chinese living in southern Sichuan and northern Yunnan; this new ethnographic knowledge led the Chinese to distinguish various branches of the Yi in this area from the category of the Qiang (Figure 3.4).

From the 14th to the 19th century, even in some rare cases the terms Qiang rong 羌戎 (Qiang barbarians), zhu Qiang 諸羌 (many Qiang tribes), and more frequently Fan and Yi, were still applied to a wide range of western non-Han people in a generic sense, but only the Fan living in the upper Min River Valley and the nearby Beichuan region were recorded as Qiang, Qiang ren or Qiang min 羌民 (Qiang civilians). Of particular note is that during this period the Han always applied the term Qiang min to the more sinicized natives from Maozhou 茂州 to Guanxian 灌縣 who were registered civilians of local Chinese governments, while they used the Fan, sheng Fan 生番 and Fan Qiang 番羌 to refer to the less sinicized people in nearby Songpan, Diexi 疊溪 and Heishui 黑水 (MSL 1382.6, 1389.3, 1433.10, 1436.6, 1439.7). Nevertheless, in the historical documents of the Ming and Qing dynasties, Fan was doubtlessly the most common identification for the people living in the upper Min River Valley and Beichuan area.

From the 16th century on, a large number of Han immigrants or refugees fled into the upper Min River Valley and Beichuan areas. They were storekeepers, merchants, artisans, and civil or military officers living in towns, or farmers marrying native women and staying in villages. By the latter part of the 19th century, the Han-style Buddhist and Daoist religious practices and the Confucian ideology of ancestral worship became an inseparable part of the natives’ lives. It is at least from this time that many of the native leaders and leading families assumed Han origins while other assumed Tibetan origins (MZZ, 3; LFTZ, 4).

The Rma of the Upper Min River Valley

Why have the natives of the upper Min River Valley and Beichuan areas (Figure 3.5), and these people only, become in the eyes of the Chinese the Qiang nationality? The non-Han people of the Qiang belt have been gradually removed from this category as they were Sinicized, Tibetanized, or classified by the Han into the category of the Yi. Looking at “the past” (historical memories) both in writing and in oral tradition, and from ethnographic data, we know that both the Sinicizing and Tibetanizing processes have occurred in the upper Min River Valley and Beichuan areas. Nevertheless, as the following paragraphs show, a sort of local identity―rma identity―has distinguished each regional human group from their more Sinicized or Tibetanized neighbors.

The Qiang today are mountain dwellers. A fortress village, zhai 寨, composed of 30 to 100 households in general is the basic social unit beyond the household. An average of two to five fortress villages make up a village cluster (cun 村). The inhabitants of fortress village or village cluster have close contact in social life. Again, several fortress villages or village clusters in a small valley, known in local Chinese as gou 溝, along a mountain stream, comprise a wider society. In these small valleys, people cultivate narrow fluvial plains along creeks or mountain terraces, hunt animals or collect mushrooms and herbs (for food or medicine) in the neighboring woods, and herd yaks and horses on the mountain-top pastures.

The houses in the fortress village are built of stone and wood standing two or three stories high. All the houses are built together for the purpose of defense. In some village all houses face inwards, while the back walls, which are connected, form the outer rampart. The symbol and most outstanding feature of a fortress village are its defense towers. These towers were built of stone of about 15 to 20 feet across at the base, tapering to about half of this size at the top. Many exceed 100 feet. According to the natives, all these defense towers were built at least 80 years ago. The memories of the endless blood feuds between fortress villages, village clusters and gou valleys, and of the wars against Chinese troops and bandits are still fresh for the older generations. The geographical isolation, the relatively autonomous economy and the fear of outsiders, all make the fortress village a confined, solitary social unit. This isolation also resulted in the tremendous variation in the local dialects.

From the linguistic point of view, all modern Qiang people speak the Qiang language, which is a member of the Tibeto-Burman language family. However, for the people themselves, the local dialects are extremely varied; as suggested by the local proverb: “our language cannot go far.” Thus, the language of rGyalrong Tibetans was once the lingua franca of the leading class in villages ruled by headmen of Tibetan connections or origins. Nevertheless, now the “Han language” (Sichuan dialect) is the most common tool of communication among Qiang people from different places. In some villages near main highway along the Min River Valley, the “Han language” is the only language used.

The Han and native languages, along with the social memories each contained, provide two different social contexts within which the people built their identities. In the context of their own language, the concept of ma, rma, or its sound-alikes is the most common source of identity. In the context of the Han language, they identify themselves as Qiang zu (Qiang nationality). Now they equate the rma (or its sound-alikes) with the Qiang zu, but most of them also admit that they had never heard of the ethnic label Qiang zu, or even the Qiang, before 1950. Elders who live in mountain villages along the Heisui River can still remember that, the rma (or its sound-alikes) represented only the people of their own valley, by which the tshep (literally the barbarians upstream)[3] and the erh (the crafty Chinese downstream) were excluded. Thus the dwellers of a small valley who called themselves rma, were also identified by the people upstream as erh and by the people downstream as tshep. Meanwhile, it seems that not all dwellers in a fortress village, a village cluster or a gou valley hold the same opinion about who should be included or excluded from the category of rma; it was a matter of experiences interacting with outsiders and knowledge about the outside world. Therefore, the concept of rma changed from generation to generation.

In the Beichuan area, before the 1950s some families in mountain villages might have been aware of their non-Han origins since they had always been called by the people downstream as manzi 蠻子; that is, barbarians. However, again, most of them had never heard of the Qiang. Before the 1970s it was difficult to distinguish the manzi and the Han in some areas. From the lower Baicao River or the lower Qingpian River upstream, villagers always declared that their village or family was of Han origin, and all villages upstream were of the manzi. In fact, all these villages were practically identical in ways of subsistence, daily life, religion, and language (Chinese of Sichuan dialect). And most families kept a genealogical book that demonstrates their Han origins by tracing their motherland to Huguang 湖廣, or more specifically, Xiaogan 孝感 county in Macheng 麻城, Hubei province ¾ a factual or imagined homeland commonly claimed by the Han Chinese in Sichuan.

Thus, before the 1950s and before the concept of nation or nationality (minzu) entered these areas, not only did the identity of the Qiang zu not exist, but the distinction between the Han and non-Han was also obscure. In the upper Min River Valley, even the concept of “us” represented by the term rma (or its sound-alikes) was also a relative one; there were no geographically or culturally defined boundaries between the rma and tshep, and the erh. Thus, the concept of rma illustrates a relatively isolated but flexible self-identity of the local people.

Now it seems clear why the natives of the upper Min River Valley and Beichuan areas were the last Qiang people in the mind of the Chinese. Obviously, from the Tang dynasty on, these areas have been sandwiched between Tibet (Tubo) and China politically as well as culturally. The confined natural environment and the self-sustaining valley economy discouraged cross-valley unification in any political or cultural form; these people could hardly reach a common identity and political unity to fight against the Tibetan or Chinese invaders. Consequently, during the late Imperial and early Republic period, the cultural traits and political tendencies of the valley-dwellers in the area manifested themselves in a picture of continuous variation from the more Tibetan-like communities in the west down to the more Han-like communities in the east, thus becoming the last blurry boundary between the Tibetan and the Han Chinese.

The History and Ethnography of the Qiang Nationality from a Chinese Viewpoint

In the eyes of the Han Chinese, the non-Han living along the western edges were “others.” They were, and in a sense still are, a kind of ethnic boundary of the Chinese. For thousands of years, the Chinese have maintained and strengthened these ethnic boundaries by describing the alienness and barbarism of the non-Han in ethnographic writings, and recording the cost of endless raids launched by these barbarians throughout history. As I have argued elsewhere (1997a: 316-18), throughout the process of ethnic boundary-forming, a great degree of cultural variation existed among the geographically as well as socially distinguished population who identified themselves as the Zhongguo ren 中國人, literally the people of the Central Kingdom; for this reason, a series of alienating boundaries were always needed to strengthen the unity of all the people within these boundaries. In this boundary-keeping tradition, there was no need for the Chinese literati to write a history with a common belief of origins for all the “people of the Central Kingdom.” History, and the common belief of origins, was written only to reinforce the coherence of the upper class or leading families.

Nationalism and Modern Ethnic Minorities

Scholars specializing in the study of nationalism and ethnicity have pointed out the crucial role and narrative characteristics of history in constructing a modern nation (Smith 1986; Tonkin, McDonald and Chapman 1989; Silverman and Gulliver 1992; Duara 1995). During the late imperial period, Han Chinese literati were startled by the political as well as economic threats from western powers of commercial imperialism on the one hand, and stimulated by the concept of nation and the underlying principles of social Darwinism on the other. Thus, a very important work in their save-the-country-project was building a Chinese nation by reconstructing history and rediscovering the ethnography of the people. In an important work, Prasenjit Duara (1995) has shown how the Chinese literati adopted Enlightenment History in the process of creating a Chinese nation. Duara suggested two very important purposes of lineal history: to glorify the ancient character of the nation, and to emphasize the unprecedented nature, the modernity, of the nation-state (1995: 27-30).

In all “history” aimed at cohering people, “origins” always are decisive. The common belief of origins, no matter whether it involves a creative ancestor, a legendary homeland, or an important event, creates attachments among a people. This phenomenon is found among various cultural backgrounds, and is well known to students of ethnicity. In China, “to find a common ancestor” has been an important discourse in the process of nation-building during the late imperial era. In a 1998 essay, Shen Sung-chiao pointed out that the collective memory of “the descendants of Huang Di 黃帝 (Emperor Huang or Yellow Emperor)” now widely held by Chinese all over the world was created and modified through a series of arguments proposed by revolutionists and reformists in late imperial China. As he suggests, the legendary Huang Di, who had been a symbol of royal or dynastic tradition, was chosen by the Han literati as ancestor of all members of the newly created nation ¾ Zhonghua minzu. This discourse included the themes: (1) whether or not the northern-frontier people, such as Manchurians and Mongolians, were the offspring of Huang Di; (2) whether the nation should be built on the commemoration of Confucius (cultural nation), instead of that of Huang Di (racial nation); (3) if Huang Di’s offspring included some non-Han people, whether the Emperor’s direct heir, the Han Chinese, were superior to the non-Han descendants of the Emperor. All these discussions reflect necessary steps in the process of nation-building: strengthening primordial ties among the people by commemorating a common ancestor, modifying the boundary, and stratifying the inner groups.

After the establishment of the Republic, supposedly a nation-state, in 1911, the project of creating and solidifying the Chinese nation went on. The nation, Zhonghua minzu, now included the Han and all the people who had been treated by the Han as peripheral barbarians. For the Han Chinese, the traditional position of peripheral people as ethnic boundaries defining “the people of Central Kingdom” was still the same. Nevertheless, they now became part of the nation, and the “Central Kingdom” and its “boundaries” could be put together. This work has been conducted mainly by rewriting history and ethnography of the people concerned. It is in this context that much of the history and ethnography of the Qiang have been produced or reconstructed in the past 50 years.

The Quest for the Qiang nationality in the Former Half of the 20^th Century

In the period from 1911 to the mid-1940s, the leaders of the Republic confined their attention to four minority nationalities: the Mongols, Manchus, Tibetans, and the Hui; the numerous minorities of Sichuan, Yunnan, and Guizhou were basically ignored. These southwestern people were becoming assimilated into the Han, and were encouraged to do so (Mackerras 1994: 59-60). The Qiang min or Qiang ren in northwestern Sichuan, who had already been Sinicized to a certain extent before that time, was one such case. As a result, the Han literati made little effort to study the cultural uniqueness of the people. This explains why the earliest ethnographic field research concerning the Qiang was all conducted by western scholars and missionaries.

During the second decade of the 20th century, Rev. Thomas Torrance began his work among the population in the upper Min River Valley. He identified these people as Qiang and reconstructed their history from the Dayu 大禹 (Great Yu) of the Xia dynasty (presumably 23rd-18th century B.C.) down to the Qing dynasty (Torrance 1920: 4-14). However, believing that the Qiang were monotheists, he further suggested that the Qiang were descendants of the ancient Israelites (1932). This conclusion was obviously driven by evolutionist and diffusionist ideologies prevailing among ethnologists of his era. From 1925 to 1948, another western scholar, David Crockett Graham, made a number of field trips to the upper Min River Valley region. As his predecessor, he identified the people as Qiang and constructed a lineal history of the people by combining all the historical data quoted from Chinese historians’ work. But he also indicated that the Qiang were not monotheists; and he observed that, accepting the teachings of Torrance, some of the Qiang people had come to believe that they were the descendants of the ancient Israelites (Graham 1958: 98-101).

Two points here are noteworthy. First, both the work of Torrance and Graham were among the people living in Maozhou 茂州, Weizhou 威州, and Lifan 理番. We know that almost all the people in these regions had become Qiang min (registered Qiang civilians) during the Ming and the Qing dynasties; this explains why they were recognized by both scholars as the Qiang. Second, for the customs and religion of the Qiang, Torrance’s description stressed its uniqueness, as he believed the people were a “unique race” (1937: 36). However, according to Graham’s description these Qiang cultures were a mixture of the Chinese culture and their own, as he had noticed that the Qiang were being gradually absorbed by the Chinese (1958: 102). Nevertheless, Graham also noticed that there was great variation in language and customs among the Qiang of different localities (1958: 101).

Starting in the 1930s, several Chinese scholars had proceeded ethnographic as well as linguistic fieldwork in the upper Min River Valley. Among them, Hu Jianmin was probably the earliest. By describing the religions, customs, and economic life of the people, he also admitted that it would be difficult to distinguish which were the native traditions and which were those being influenced by the Chinese and Tibetan cultures (Hu 1944: 39). The vague boundary of cultural representation, especially in native dress, between the Qiang and the Han, and the rGyalrong Tibetans at this time was also illustrated by the report of the Frontier Service Team of Chinese Colleges. As the report stated: “The Qiang’s dress, whether that of men or women, has no special form. Therefore it is impossible to identify the Qiang by their dress” (Jiaoyubu Meng Zang Jiaoyu Si 1943: 23). The report further suggested: “since all cultural phenomenon are identical with those of (the Han of) the inner land and the differences are trivial, there is no way to identify the Qiang except by their language” (1943: 25).

Meanwhile, the Qiang in Hu Jianmin’s study was not merely a small group of contemporary people living in northwestern Sichuan; He connected them historically with all the Di-Qiang minzu 氐羌民族 ¾ a huge ethnic group, which as he believed, originated from the Qiang, Di or Di-Qiang in ancient times [4] and whose branches then become many non-Han peoples widely spread in western or southwestern China (Hu 1944). He further connected the Di-Qiang minzu with the Tibetan; the relationship between the two people, he believed, was physically but not linguistically proven (Hu 1944: 43).

Whether scholars were impressed by the uniqueness of Qiang culture or confused by its hybrid nature, all academic quests for the Qiang at this time demonstrate that there was an ideal type of nation commonly held by all authors: the nation is a group of people who share common language, cultural and physical traits, and are descendants of a physically, linguistically and culturally homogeneous proto-group. Thus the field researchers of that time tried to find the physical, linguistic, and cultural proto-types of the Qiang, and therefore could delineate the ethnic boundary of the people. Even though they failed to find a normative Qiang culture, their attempts to do so, and the data they recorded in these quests, have reinforced the concept of the Qiang nationality both for the Han and the natives.

In contrast to ethnographers who could only give a vague concept of the Qiang through describing the native culture, linguists provided a much more clearly defined human category of the people through linguistic investigation and classifications. Thanks to the contributions of many linguists, among whom Wen Yu was the first Chinese scholar in this field, the Qiang language found a place as a distinct member in the Tibeto-Burman family of the Sino-Tibetan language phylum. For many scholars, not only did the specific language provide a “scientific” way to identify the Qiang people, but it also delineated the hierarchical structure of linguistic relationships (Qiang language ® Tibeto-Burman family ® Sino-Tibetan phylum) and encouraged them to connect the Qiang with wider human categories historically as well as culturally. Therefore, while the Qiang people are believed to be a unique ethnic group, on a higher level they are also part of a collective nation, Di-Qiang minzu, corresponding to a specific linguistic group. And on the highest level, Di-Qiang minzu are part of the super nation, Zhonghua minzu.

The Making of Qiang History

Nevertheless, ethnographers and linguists did not create the ethnic category of the Di-Qiang minzu; it was the invention of historians. As I have mentioned, the Han Chinese have a long and abundant historical memory concerning the Qiang. Therefore, in the eyes of Han Chinese, the history of the Qiang has always been much more attractive than the ethnography of the people. In 1905, a young revolutionary, Liu Shipei 劉師培, published a book entitled Zhongguo Minzu Zhi 中國民族志 (History of the Chinese Race), intended to awaken the Chinese nation; the Chinese nation was equivalent in his point of view with the Han race (Liu 1905). The Di-Qiang zhongzu 氐羌種族 (Di-Qiang race) that appeared in this book were a branch of Tibetans whose descendants had been partially absorbed into the Han race (Liu 1905: 8-9). After the establishment of the Republic, historical studies of the Chinese nation and nationalities came to be a major part of the project of nation-building. In these works, the Qiang or Di-Qiang minzu represent a large group of people whose offspring are now scattered among the Han, the Tibetans, and many southwestern non-Han peoples (Lin H. 1934; Lü S. 1934; Lü Z. 1948). Nevertheless, during this period the Qiang or Di-Qiang minzu seemed to be a historical concept; very few scholars connected them with the contemporary people living in the upper Min River Valley.[5]

In 1949, the Chinese Communist Party came to power and established a new regime called the People’s Republic of China (PRC). Like its predecessor, the new regime asserted authority over all nationality areas and stressed the unity of the Chinese nation. But unlike the former republic, the new regime had more positive policies toward nationality issues. At first, the equality and autonomous rights of all nationalities were guaranteed by the Constitution of the PRC. Subsequently, identification of nationalities was carried out over all of China. Until 1979, a total of 55 minority nationalities had been formally recognized by the state. Along the old Qiang belt, the identification or distinction of nationalities basically followed the Chinese concepts of Qiang, Fan, and Yi (Figure 3.6). Most of the people in this belt who had been labeled as fan now were classified as Zang zu (Tibetans), and became the eastern-most portion of a total of 4,593,330 Zang zu. The people at the southern end of this belt who had been labeled Yi (to avoid its barbaric hint the character 夷 was now replaced by a homophonous character 彝) were identified as the Yi zu and became the northern-most portion of a total of 6,572,173 Yi people. The Qiang, a nationality not found elsewhere, comprise only 198,252 people. Even though the number of Qiang was small, because of the wealth of materials on the Qiang in Chinese archives, through historical imagination it was possible to connect the Zang, the Yi, and many southwestern Di/Qiang-connected nationalities with the Han Chinese.

In this new environment, the history of the Qiang nationality became a focus of historical attention. During the 1970s and 1980s, many leading scholars, such as Gu Jiegang 顧頡剛 and Ma Changshou 馬長壽, engaged in this study. The result was a series of reconstructed histories that not only connect all the Qiang from the Shang dynasty down to the present, but also link a wide range of northwestern and southwestern nationalities, through their Di-Qiang connections, with the Han (Gu 1981; Ma 1984; Ran, Li, and Zhou 1984). A folk story concerning a war between the natives and the Geji people (Geji ren 戈基人), adopted from the Qiang in the Wenchuan and Lixian areas, was interpreted by scholars as evidence for the Qiang’s southward migration (Lin X. 1983; Ma 1984; Ran, Li, and Zhou 1984). This folk evidence further connects the Qiang today in northwestern Sichuan with the lineal history of the Qiang or Di-Qiang minzu reconstructed from China’s historical memories. Thus it became common knowledge among the literati of the Han and of many nationalities that the Qiang were the descendants of the legendary Yan Di 炎帝 (the Emperor Yan) and Dayu, and, though once a very populous and powerful nationality, were scattered over various places after being defeated.

Identifying the Emperor Yan of the Jiang 姜 clan with the ancestor of the Qiang or Di-Qiang minzu is especially crucial.[6] In Chinese legendary history the status of the Emperor Yan was twofold: according to some sources he was an ancient leader rivaling the Emperor Huang for leadership (SJ 1.1; DWSJ), but another source states that the two rival emperors were actually brothers (GY 10). The Emperor Huang (Huang Di) was commemorated by the Han Chinese as their common ancestor in the nation-building scheme during the late Qing and early Republican periods. At the same time, Emperor Yan, the defeated brother of Emperor Huang, has been collectively reconstructed in national history (minzu shi 民族史) as the ancestor of many non-Han nationalities, especially those that could be connected to the Di-Qiang minzu. Thus the term Yan Huang zisun 炎黃子孫, literally the descendants of the Emperor Yan and Emperor Huang, has become a synonym for the Zhonghua minzu including all Han and non-Han peoples; it is a historical metaphor confirming the brotherhood of all nations/nationalities. It is also a historical metaphor distinguishing offspring of the conqueror from those of the conquered. This explains why even though the Qiang today are a minor population its history is crucial for Chinese historians.

History and Ethnography of the Qiang Nationality from the Native Viewpoint

The Qiang today are not only a non-Han category in the eyes of the Han Chinese, but also an ethnic group in their own right, since they identify themselves as Qiang zu (the Qiang nationality) and believe that they share a common language, culture, and history. This development occurred mainly after 1949.

It is difficult to estimate the extent to which the administration of the KMT government, and the scholarship of foreign and Chinese scholars in the first half of the century, contributed to the formation of the ethnic self-awareness of the Qiang. However, during my own fieldwork, I have been told by many old people that they had not heard about the Qiang and had not thought of themselves as such until the Han Chinese said they were in and after the 1950s. Obviously, during the late Qing and early Republic, the ethnonym Qiang min was merely an official terminology recorded in writing documents; in daily life, they were labeled by the people upstream as erh and by the people downstream as tshep, sbie or ferh, and by the Han Chinese, manzi (barbarians). Further, while they were young, the autonym rma (or local variants) ¾ a native notion of “our people”¾ denoted only the people of their own valley, small or large. Needless to say, no one had heard anything about the “descendants of Israelites”.

With the coming of the People’s Republic, the situation altered radically. Besides investigating and classifying nationalities, the new government introduced a series of policies toward minority nationalities. These policies have varied from time to time and certain projected goals have never been achieved, but the prohibition of discrimination and economic subsidies for minority nationalities are attractive enough to make minority status welcome. Thus in the Qiang areas, not only previous “barbarians” (manzi) were willing to accept minority status with the assigned label Qiang, but also many people who had previously declared themselves as Han now made every effort to find their Qiang connections in order to acquire this non-Han status. Against this background, more and more people in the upper Min River Valley and Beichuan region became member of the Qiang nationality. In addition, after being taught about their “past” in China's social/historical memory through school education, they began to select, interpret or invent their own cultures and history.

Qiang Language, Writing, and Culture

For the Qiang people, the most serious obstacle in building their common identity is the lack of a native writing system and a common language. Starting from the mid-1980s, after receiving state approval, Qiang literati began to create their own writing system (using a Romanized alphabet) and to compile a dictionary of the Qiang language, based on the standardized Qugu 曲谷 dialect. Since 1994, more teachers who have learned the standard Qiang language and the newly invented writing system have been sent to mountain villages of all places. Even though it is not true that all Qiang people can communicate with each other with the standardized native language, it is now common sense among the people that they are speaking the “same language.” This common sense is created by linguists’ fieldwork and simplified publications.

Besides the common language and writing, the highlights of the newly selected or invented Qiang culture are the new year festival of the Qiang calendar, the salong (guozhuang 鍋庄 in Chinese) singing dance, the national dress, and a folk story concerning the Qiang-Ge war. In the 1910s, Torrance noticed that the Qiang in the Wenchuan area celebrated the festival of the Ox-king (niuwanghui 牛王會) in the first day of the tenth month of the lunar year. However, believing that the Qiang were descendants of Israelites, he suggested that the day is actually a ‘new year festival of the Qiang calendar’ instead of the Sinicized festival of the Ox-king (1920: 29-35). It is questionable whether the people Torrance studied had the custom of a new year festival as their tradition. First, elder aborigines I met told me that they celebrated a festival of the Ox-king, but the most cheerful day in a year was the Chinese new year, the first day of the lunar calendar. In some less Sinicized villages people say directly that the festival of the Ox-king is a Han custom that did not previously exist. Furthermore, in the natives’ concept of time, according to the native language, there are only two periods in a year, one equivalent to spring and summer, the other, autumn and winter; their concept of months was borrowed from the Han Chinese, and so is that of the first day of a new year. Nevertheless, in 1987, the autonomous government of Aba Prefecture officially pronounced the first day of the tenth month to be the new year festival of the Qiang calendar. From that year on, four Qiang counties─Maoxian, Wenchuan, Lixian and Beichuan─have taken turns hosting the official activities of the day for all the Qiang people. Young people enjoy the opportunity to sing and dance with their friends from different localities, but the elders are not interested. They even complained that it is an insult to impose the festival of the Ox on the Qiang people.

The most important activity in the new year festival of the Qiang calendar as well as in other celebratory events is the salong singing dances ¾ a kind of group dance that is commonly believed to be shared between the Tibetans and the Qiang in Aba Prefecture. Nevertheless, natives in some regions admit that they did not have this tradition until young people brought it back from towns in recent years. In some fortress-villages (in northern Maoxian and southern Songpan), while young people dance salong, elders perform nisa instead ¾ another kind of group dance with more of a ritual or religious sense. It is noteworthy that very few people, if any, know the meaning of salong songs even in villages that had practiced this tradition. In these villages, elders still remember that, before the standardized salong was introduced to the people, the salong dances varied from village to village, and from valley to valley. Since 1989, under the support of the autonomous government of Aba Prefecture, both the standardized salong of the Qiang and that of the rGyalrong Tibetans were gradually introduced to all the people. Salong dances now are widely practiced in every official or private celebration; it is not only a symbol of the Qiang culture but also a symbol of unification of the rGyalrong Tibetans and the Qiang in Aba Prefecture.

While performing salong, women wear their traditional national dress of very fine, delicate, colorful, and varied embroidery patterns striking contrast to the impression the Qiang’s dress made upon early investigators, which was always described as dull, plain and ethnically indistinguishable (Torrance 1937: 39-41; Graham 1958: 20-21; Jiaoyubu Meng Zhang Jiaoyu Si 1943: 23; Hu 1944: 39). Furthermore, early investigators noticed that the practice of embroidery seemed to be flourishing in some more Sinicized or Tibetanized Qiang areas; thus they suspected that the Qiang might have borrowed the art of embroidery from their neighboring Han or Tibetans (Hu 1944: 40; Graham 1958: 21). Now embroidery is commonly believed and proclaimed to be a national hallmark of the Qiang nationality. Therefore, by 1996, Wenchuan was assigned by the state the name of Zhongguo Minjian Yishu Qiangxiu zhi Xiang 中國民間藝術羌繡之鄉, “Home of the Qiang’s embroidery of Chinese folk art” (Bianji weiyuanhui 1997). I have mentioned elsewhere that the changes of Qiang women’s dress in this century illustrated a process of fashioning minority ethnic identity (Wang 1998).

The Qiang’s Folk History

For many scholars studying Qiang culture, the most interesting character is probably duangong 端公 ¾ exorcist, medicine man, and storyteller. During the Cultural Revolution, duangong and the religion they practiced almost disappeared. But, in the 1980s, as the sense of ethnic identity among China’s nationalities sharply increased, the story of the Qiang-Ge war adopted from duangong oracles and the related white stone worship have become a symbol of the Qiang culture. The story has been reinterpreted by Han scholars to relate the Qiang in northwest China during the Han dynasty to the people in the upper Min River Valley and other southwestern regions today; thus a lineal history of the Qiang or Di-Qiang minzu was completed. The Qiang literati are willing to accept from the story that they were once a strong and warlike people, and, most important, are the mainstream exemplars of the Di-Qiang minzu today including their populous neighbors ¾ the Zang (Tibetans) and the Yi. Therefore, the story of the Qiang-Ge war and the subsequent Qiang “history” have become a crucial part of recently published work by Qiang literati. However, for most villagers in mountain fortresses, since they could neither learn what duangong sang in the past, nor acquire the story from books in Chinese today, the story of the Qiang-Ge war is basically foreign. Meanwhile, the history has been simplified in oral tradition as “the Qiang used to be a strong and warlike people but dispersed in all directions after being defeated” (Wang 1997b: 142). Therefore it is a common belief among the people that, beyond the Qiang in Aba Prefecture and Beichuan areas, there are, or must be, some sub-branches scattered in different places around the whole country, or even around the world.

While the story of the Qiang-Ge war has been adopted by Han scholars to construct a lineal history of the Qiang, the same story has also been reinterpreted or rewritten by natives to fit their own interests and historical mentality. That is, as the story said, after the war the god Mubita, or the winning hero Paigo, dispatched Qiang soldiers to different valleys to establish their own fortress villages (QZGSJ: 26-27). In one new and popular version of the story (Luo Shize 1983; Ran, Li, and Zhou 1984: 209), the nine places Paigo’s sons (nine brothers) had ruled included all the Qiang autonomous counties and major towns today (Songpan, Maoxian, Wenchuan, Lixian, Mianchi, Beichuan), and the place (Heisui 黑水) its inhabitants are believed to be the Qiang according to Qiang literati’s linguistic knowledge, and the places (Niangziguan 娘子關, Guanxian 灌縣) that are believed to have been Qiang areas according to Qiang literati’s historical knowledge. I have mentioned elsewhere that “brotherhood stories” ¾ which attribute origins of closely bound families, lineages, or villages in an area to the simultaneous arrival to the region and subsequent dispersion of brothers ¾ were, and in some areas still are, a kind of history in natives’ minds (Wang 1999). By telling these histories, the brotherhood among all families, lineages, and villages in an area could be strengthened. Thus, we find two kinds of historical mentality in the story of the Qiang-Ge war and its interpretations. On the one hand, the story as part of a lineal history connects the Qiang and other Di-Qiang nationalities with the Han through which ethnic relations of majority-versus-minority, or of superor-versus-inferor, are revealed. On the other hand, the story as part of a structural history of the coming and separation of brothers connects all the Qiang people in different locations; and, in this historical scenario, all descendants of “the brothers” are equal.

By telling the story of the Qiang-Ge war and providing a historical interpretation, the Qiang literati built up their self-image as the strongest opponents of the Han. Meanwhile, by acquiring other ancestors in Chinese historical documents, a historical role as the Han’s brother and savior has also been constructed. These ancestors include the legendary Emperor Yan 炎帝, the Great Yu 大禹, and Li Bing 李冰 of the Qin dynasty, as well as the heroine Fan Lihua 樊梨花 of the Tang dynasty. The Emperor Yan in the Chinese legend was the brother of Emperor Huang, and also the inventor of agriculture. Both Great Yu and Li Bing were known as heroes who harnessed the rivers and saved people from flood and drought. The heroine Fan Lihua, daughter of a non-Chinese warlord, helped her husband, a Chinese general in the Tang dynasty, to defeat the invasion of the Xifan (western barbarians), namely, the Tibetans. All these legendary or historical characters are recognized, for various reasons, by the Qiang literati their ancestors. The Great Yu, the founder of the Hsia dynasty, has been most often commemorated by the educated Qiang people; there is even a good deal of propaganda and academic debate among Qiang intellectuals in Wenchuan and Beichuan over the birthplace of the sage-emperor (Wang 1997b: 144-45, 152).

In the natives’ construction of their own culture and history the Qiang literati have played an important role. Thus, a few remarks on the general backgrounds and characteristics of these people is in order. The Qiang literati in this chapter refer to natives who have acquired knowledge by learning Chinese language and writing and therefore became teachers, public servants of autonomous governments, or holders of other administrative or academic positions in various levels of government. Their privileged social position comes from their role as intermediaries between their own people and the state or the Han Chinese. The emergence of these people as a social class is quite new. During the Qing and Republican periods, even though a few prosperous native families might have the opportunity to send their young to study Chinese, this did not create a group of literati confident enough to declare a native identity; nor could they hold a prestigious position in Chinese society. A “learned barbarian” was still a “barbarian.” Therefore, it was not until the state ensured the equality of all nationalities and provided minority nationalities with economic subsidies, educational opportunities, as well as leniency in the number of children allowed, that the Qiang became a popular identity. And, it was not until state education extended deep into all mountain villages and educated young were able to serve in autonomous governments or schools that the Qiang literati became spokesmen for their people.

Nevertheless, the history of the Qiang in Chinese social memory was basically unknown to natives in this area before this century; thus, only natives well versed in Chinese are able to select suitable pasts from Chinese archives to construct their glorious history. Moreover, mountain dwellers in this area traditionally identified themselves with people in a relatively narrow world (a few villages or valleys), as manifested in the traditional concept of rma. Therefore, only natives who are aware of their minority-of-minority ethnic status through coming into contact with the outside world feel the need to glorify their Qiang identity. Finally, the Qiang nationality is presumably an inseparable part of the Chinese nation and the communist state. Consequently, only educated natives serving in the state or party system and acquainted with all these ideologies are in a position to carefully establish and propagate their Qiang identity, and at the same time avoid threatening the state or the unity of the nation.

Conclusions

From the Late Han to the Qing period, the concept of the Qiang as a western ethnic boundary of the ancient Chinese, first referring to a huge range of people in the west, gradually came to denote only the non-Chinese people living in the upper Min River Valley and Beichuan area. This process was the result of more and more western “barbarians” becoming Chinese or being classified by the Chinese into new ethnic categories. The history of the Qiang is segnificant, serving as a bridge between many minority nationalities and the Han in the past and the present, reconstructed mainly by Han intellectuals in this century as part of a nation-building scheme. And I have described the native movement of constructing a common language, writing, culture, and history of the Qiang nationality, as well as the role of native literati in this movement as receivers, interpreters, and propagators of their own history and culture.

In conclusion, it is important to elucidate some theoretical aspects of national identity formation and the “history” involved. In the study of history of nations, explanations for the formation of a nation usually take one of two forms: “how did the past create the present,” or, “how did the present create the past?” The past in the former explanation always involves “what really happened,” and those in the second, “the one of selection, reconstruction, or imagination.” It is undeniable that the single coherent past and the standardized explanation about the origin of a group almost always result from political-economic power oppositions. But it is also noteworthy that deconstructing the “history” people created through historical imagination raises the question: is it the anthropologists’ or the historians’ business to tell people under study that the history they believe is wrong (Tonkin, McDonald, and Chapman 1989: 9)?

The “fact-versus-fiction” also exists in the historical studies of Chinese nation or nationalities. Most Chinese historians hold the present-in-the-past explanation about national histories, of which all nationalities in China are believed to be old and together form an inseparable national unit. Western scholars, however, influenced by the trend of emphasizing the imaginary or inventive nature of history and traditions in the formation of nations or nationalities, now mostly favor the past-in-the-present explanation. The Chinese nation and its nationalities are thus an entirely modern phenomenon produced under the impetus of nationalism, and the related history is viewed as a modern historical narrative.

My own opinion falls in between because the meaning of the history of the Qiang is twofold: it is a history of a minority nationality, and also a history of the Chinese in respect to boundary formation and changes. In the former sense, my arguments demonstrate that the formation of the Qiang nationality involved selected, invented, or imagined aspects of history and culture. In the latter sense, if we consider the history of the Qiang as a process of formation, expansion and change of Chinese boundaries (and the Tibetan boundary as well), this chapter illustrates how the past created the present, and underlines the continuity of this history. However, the most important past that created the present Han-Qiang relations is obviously not what really happened on the Qiang side but how Chinese thought about their ethnic boundaries through the concept of the Qiang. In other words, what has remained as a part of history is not the Qiang nationality but a central-versus-periphery dynamic of human relations.

Scholars in Chinese studies have noticed the progression of Chinese thought on national identity from the traditionally cultural and hierarchical ideas of civilization and territory (space) to the modern concept of ethnicity (or nation) based on racial differences (Thierry 1989; Crossley 1990; Gledney 1991). The core issue in these studies is the traditional notion of Chinese nationality represented by the identity of Zhongguo ren 中國人 which implies civilized people living in the Central Kingdom. However, the sense of the people in the Central Kingdom was not a self-sustained notion but one that needed to be supplemented by alienating boundaries; through describing the alien and uncivilized nature of those on the periphery, and commemorating military acts toward them, the oneness of the people within these boundaries was confirmed (Wang 1997a: 289-319). The Qiang was a Chinese ethnic boundary of this kind. But, most significantly, it was the oldest, least stable, and most nebulous boundary between the Han and non-Han people. This explains why, although now a minor population, the history of the Qiang is crucial to the Han and other nationalities as well as a force in strengthening the historical unity of the nation ¾ Zhonghua minzu.

Acknowledgement

I wish to thank the Chiang Ching-kuo Foundation for International Scholarly Exchange for a contribution to my field expenses in 1995-1997. I also thank National Science Council in Taiwan for the support of a project on archival research on China’s southwestern minorities in 1997-1998.

References

Anderson, Benedict. 1991. Imagined Communities. Rev. edition. London: Verso.

Barth, Fredrik. 1969. ‘Introduction.’ In Ethnic Groups and Boundaries, ed. by Fredrik Barth, 9-38. London: George Allen & Unwin.

Bentley, Carter G. 1987. ‘Ethnicity and Practice.’ Comparative Study of Society and History 29.1: 24-55.

Bianji weiyuanhui 編輯委員會. 1997. ‘Chunhou de minzu fengqing’ 醇厚的民族風情 (Glorious National Culture), in Daxiongmao de Guxiang: Zhongguo Wenchuan 大熊貓的故鄉: 中國汶川 (Mysterious Panda Hometown: Wenchuan China). Wenchuan: Wenchuanxian Renmin Zhengfu.

Comaroff, John and Jean Comaroff. 1992. Ethnography and Historical Imagination. Boulder: Westview.

Coser, Lewis A. 1992. ‘Introduction: Maurice Halbwachs.’ In On Collective Memory, ed. & trans. by Lewis A. Coser. Chicago: The University of Chicago Press.

Crossley, Pamela. 1990. ‘Thinking about Ethnicity in Early Modern China.’ Late Imperial China 11, no. 1: 1-35.

QZGSJ Qiangzu Gushi Ji 羌族故事集 (A Collection of Stories among the Qiang Nationality). Sichuan sheng Aba Zangzu Qiangzu zizhizhou wenhuaju Press 四川省阿壩藏族羌族自治州文化局, 1989.

de Crespigny, R. 1984. Northern Frontiers: The Policies and Strategy of the Later Han Empire. Canberra: Australian National University Press.

Duara, Prasenjit. 1995. Rescuing History from the Nation: Questioning Narratives of Modern China. Chicago: The University of Chicago Press.

DWSJ Diwang Shiji 帝王世紀 (A Geneology of Emperors and Kings). Authored by Huangfu Mi of the Jin dynasty 晉皇甫謐著.

Fentress, James and Chris Wickham. 1992. Social Memory. Oxford: Blackwell Publishers.

Geertz, Clifford. 1963. ‘The Integrative Revolution: Primordial Sentiments and Civil politics in the New States.’ In Old Societies and New States: The Quest for Modernity in Asia and Africa, ed. by Clifford Geertz, 105-57. New York: Free Press.

Gledney, Dru. 1991. Muslim Chinese: Ethnic Nationalism in the People‘s Republic. Cambridge: Council of East Asian Studies, Harvard University.

Graham, David Crockett. 1958. The Customs and Religion of the Ch‘iang (Qiang). City of Washington: The Smithsonian Institution.

GY Guo Yu 國語 (Dialogues of States) . In Sibu Congkan 四部叢刊 (Collectanea of books in the four bibliographic categories). Shanghai : Shangwu Press, 1937.

Halbwachs, Maurice. 1992. On Collective Memory, ed. & trans. by Lewis A. Coser. Chicago: The University of Chicago Press.

Hobsbawm, Eric and Terence Ranger, ed. 1983. The Invention of Tradition. Cambridge: Cambridge University Press.

Hu Jianmin 胡鑑民. 1944. ‘Qiangzu di xinyang yu xiwei’ 羌族的信仰與習為 (Beliefs and practices of the Qiang nationality). Bianjiang Yanjiu Luncung 邊疆研究論叢.

Jiaoyubu Meng Zang Jiaoyu Si 教育部蒙藏教育司, ed. 1943. Chuanxi Diaocha Ji 川西調查記 (A Report on an investigation in Western Sichuan). Jiaoyubu Mong Zhang Jiaoyu Si Press.

Keyes, Charles F. 1981. ‘The Dialectics of Ethnic Change.’ In Ethnic Change, ed. by Charles F. Keyes, 4-30. Seattle: University of Washington Press.

Gu Jiegang 顧頡剛. 1980. ‘Cung guji zhong tansuo woguo de xibu minzu: Qiangzu 從古籍中探索我國的西部民族: 羌族 (Exploring western people of our country from ancient archives: the Qiang nationality), Shehui kexue zhanxian 社會科學戰線 1: 117-152。

Lü Zhengyu 呂振羽. 1951 (1948). Zhongguo Minzu Jianshi 中國民族簡史 (Brief History of Chinese Nationalities), Beijing: Sanlian shudian 三聯書店.

Lü Simian 呂思勉. 1934. Zhongguo Minzu Shi 中國民族史 (History of Chinese Nationalities), Shanghai: Shijie Press. 世界書局.

LFTZ Lifan Ting Zhi 理番廳志 (Lifan Ting Gazetteer). Authored by Wu Gengmei of the Qing dynasty 清吳羹梅, 1866.

Lin Xiang 林向. 1983. ‘Qiang Ge dazhan de lishi fenxi’ 羌戈大戰的歷史分析 (A Historical Analysis of the great Qiang-Ge war). Sichuan Daxue Xuebao Congkan 四川大學學報叢刊 20: 8-16.

Lin Huixiang 林惠祥. 1936. Zhongguo Minzu Shi 中國民族史 (History of Chinese Nationality), Shanghai: Shangwu Press 商務印書館.

Liu Shipei 劉師培. 1962 (1905). Zhongguo Minzu Zhi 中國民族志 (History of Chinese Race), Taipei: Zhongguo Minzu Xuehui 中國民族學會.

Luo Shize 羅世澤. 1983. ‘Qiang Ge Dazhan.’ 羌戈大戰 (The Great Qiang-Ge Wars). In Mujiezhu yu Douanzhu 木姐珠與斗安珠 (Mujiezhu and Douanzhu)., ed. by Lo Shize 羅世澤 and Shi Fongchun 時逢春. Chengdu: Sichuan Minzu Press 四川民族出版社.

Ma Changshou 馬長壽. 1984. Di yu Qiang 氐與羌 (The Di and The Qiang), Shanghai: Shanghai Renmin Press 上海人民出版社.

Macherras, Colin. 1994. China‘s Minorities: Integration and Modernization in the Twentieth Century. Oxford: Oxford University Press.

MZZ Maozhou Zhi 茂州志 (Maozhou Gazetteer). 清楊迦懌等修 (ed. by Yang Jiaduo of the Qing dynasty), 1831.

MSL Ming Shilu fu Jiaokan Ji 明實錄附校刊記 (Authentic Records of the Ming dynasty with collating notes). Taipei: Institute of History and Philology, Academia Sinica 中央研究院歷史語言研究所, n.d.

Ran Guangrong, Li Shaoming & Zhou Xiyin 冉光榮、李紹明、周錫銀. 1984. Qiangzu Shi 羌族史 (History of the Qiang Nationality). Chengdu: Sichuan Renmin Press 四川人民出版社.

Ren Naiqiang 任乃強. 1984. Qiangzu Yuanliu Tancuo 羌族源流探索 (A Study on the Origins and Spread of the Qiang Nationality), Chungqing: Chungqing press 重慶出版社.

Roosens, Eugene E. 1989. Creating Ethnicity: The Process of Ethnogenesis. London: Sage Publications.

Schwartz, Barry. 1990. ‘The Reconstruction of Abraham Lincoln.’ In Collective Remembering, ed. by David Middleton & Derek Edwards. London: Sage Publications.

Shen Sung-chiao 沈松僑. 1997 ‘Wo yi wo xie jian Xuanyuan: Huang Di shenhua yü wan Qing de guozu jiangou’ 我以我血薦軒轅: 黃帝神話與晚清的國族建構 (The Myth of Huang-ti, Yellow Emperor, and the Construction of Chinese Nationhood in Late Qing), Taiwan Shehui Yenjiu Jikan 台灣社會研究季刊 28: 1-77.

Silverman, Marilyn & P. H. Gulliver. 1992. Approaching the Past: Historical Anthropology through Irish Case Studies. New York: Columbia University Press.

Smith, Anthony D. 1987. The Ethnic Origins of Nations. New York: Basil Blackwell.

SJ Shi Ji 史記 (Records of History), authored by Sima Qian 司馬遷 of the Han dynasty. Beijing: Zhonghua Press, 1959.

Thierry, Frrancois. 1989. ‘Empire and Minority in China.’ In Minority Peoples in the Age of Nation-States. Ed. Gerard Chaliand. London: Pluto Press.

Thompson, Richared H. 1989. Theories of Ethnicity: A Critical Appraisal. New York: Greenwood Press.

Tonkin, Elizabeth, Maryon McDonald and Malcolm Chapman ed. 1989. History and Ethnicity. London: Routledge.

Torrance, Thomas. 1920. The History, Customs and Religion of the Ch‘iang (Qiang). Shanghai: Shanghai Mercury Press.

1937. China‘s First Missionaries: Ancient Israelites. London: Thynne & Co. LTD.

Van den Berghe, Pierre L. 1981. The Ethnic Phenomenon. New York, Elsevier.

Wang, Ming-ke. 1992. The Ch‘iang (Qiang) of Ancient China through the Han dynasty: Ecological Frontiers and Ethnic Boundaries. Ph.D. diss. Cambridge, MA: Harvard University.

------. 1997a Huaxia Bianyuan: Lishi Jiyi yu Zuqun Rentong 華夏邊緣: 歷史記憶與族群認同 (On Chinese Borderlands: Historical Memory and Ethnic Identity). Taipei: Yunchen Press 允晨文化公司.

------. 1997b ‘Hanzu bianyuan de Qiangzu jiyi yu Qiangzu benzhi’ 漢族邊緣的羌族記憶與羌族本質 (The Memory and Ethnicity of the Qiang on the Han‘s Borderland). In Cong Zhoubian Kan Hanren de Shehui yu Wenhua 從周邊看漢人的社會與文化 (A View of Chinese Society and Culture from the Periphery). Ed. by Ying-kuei Huang and Chuen-rong Yeh. Pp. 129-165. Taipei: Institute of Ethnology, Academia Sinica.

------. 1998 ‘Qiangzu funü fushi: Yige minzuhua guocheng de lizi’ 羌族婦女服飾：一個民族化過程的例子 (Women’s Dress of the Ch’iang: A Case of Fashioning Ethnic Identity). In The Bulletin of the Institute of History and Philology, Academia Sinica 中央研究院歷史語言研究所集刊 69.4: 841-885. Taipei: Institute of History and Philology, Academia Sinica.

------. 1999 ‘Genji lishi: Qiangzu de dixong gushi’ 根基歷史: 羌族的弟兄故事 (Primordial History: the Brotherhood Stories of the Qiang). In Shijian, Lishi yu Jiyi 時間、歷史與記憶 (Time, History and Memory). Ed. by Ying-kuei Huang. Pp. 283-341. Taipei: Institute of Ethnology, Academia Sinica.

Wen Yu. 1941. ‘Chuanxi Qiangyu de chubu fengxi’ 川西羌語的初步分析 (A Preliminary Analysis of the Qiang linguage). Studia Serica 華西大學中國文化研究所集刊 2: 58-90.

[1] Before receiving the ethnic label Qiangzu, an ethnic autonym rma or its sound-alikes have represented the widest range concept of “our people” among natives include no more than people of a few mountain villages or inhabitants in a small valley. While the natives become Qiang zu, rma and its sound-alikes are generally used by native litrati, ignoring the variant pronunciation of the word, in Mandarin as erma.

[2] For Tibetanized people in this period I mean the people who became Lama Buddhists and whose leaders declared their Tufan or bod (Tibetan) origins.

[3] In some areas, there are sbie or ferh, instead of tshep, refering the barbarians upstream.

[4] During the Warring States period (450-221 B.C.), an ethnonym Di-Qiang 氐羌 appeared in Chinese documents. It appeared mainly in the works of philosophers as a name for some legendary western people known only by hearsay. During the Qin (221-206 B.C.) and the Former Han (202 B.C.-A.D. 7) dynasties, as knowledge about the western peoples increased greatly, Chinese split the concept of Di-Qiang into two human categories, the Di and the Qiang, indicating two groups of culturally distinguished people in the Lungxi area. For details on the conceptual changes concerning the ethnic labels, Di-Qiang, Di and the Qiang, in ancient Chinese mind’s eye, see Wang 1992: 116-20; 1997a: 236-43.

[5] As Wen Yu stated in his article in 1941: “We would not know that there were still Qiang people in this territory if we did not come to western Sichuan. A race that has long appeared in ancient documents has not been extinguished through countless migrations and mixtures (with others) in thousands of years, and no one in the world even knows this. It is most regretful to our academic circle.” (Wen Yu 1941).

[6] The Jiang clan was an important political and military force with which the Zhou allied to conquer the Shang, and also the clan from which the Zhou kings frequently took their wives. It has been widely assumed that the Jiang clan was a branch of, or identical with, the Qiang; The phonetic and graphic similarities between the surname Jiang 姜 and the ethnic label Qiang 羌 led scholars to believe that these two words are identical etymologically (Pulleyblank 1983). I have argued against this theory in my doctoral dissertation (1992: 105-108) and other work (1997: 231-33).