NOTES

ABBREVIATIONS

INTRODUCTION

1 Hanzu are the members of the Han minzu.

2 See Anderson 1972; Blake 1981; J. Watson 1988; R. Watson 1988; Honig 1992a; Choi 1995; Constable 1996; Chen 1999; Hansen 2005; and Vasantkumar 2012.

3 See Eberhard 1962; Fei 1992; Duara 1995; Dikötter 1996; Ebrey 1996; Xu 1999; Harrison 2000; Leibold 2007; Elliott 2012; and Joniak-Lüthi 2014.

4 These socionyms are collective identity labels conceived by some Han to describe other Han. They are capitalized as a reminder that they are to be read as emic identity labels and not as analytical terms.

5 I would like to thank Ellen Hertz for her insightful comments on the question of relationality.

6 In localized contexts, Han-ness/Chinese-ness was likely intertwined with other identities, such as “the cultured ones”; “farmers, not herders”; or “having bound feet.” In late nineteenth-century (and also twentieth-century) Guangdong, bargaining over who was more and who was less Han/Chinese could be as fierce a debate among Han/Chinese as the differentiation between Chinese and “barbarians” (Blake 1981; Constable 1996; Leong 1997).

7 I use biopolitical in the meaning discussed by Foucault, for instance in Discipline and Punish (1991).

8 For a discussion of the anti-Manchu rhetoric, see Crossley 1997 (189–201).

9 For detailed analyses of this process, see Duara 1995; Chow 2001; Harrison 2001; and Leibold 2006. See also Rhoads 2000 for an analysis that draws on the perspective of Manchu-Han relations.

10 See Elliott 2012.

11 Minzu is not an indigenous category of identity sensu stricto but rather was imported from Japan in the late nineteenth century.

12 The Han minzu is to a certain degree a new form conceived to regulate a much older and fragmented attachment that has been perpetuated in a complex interplay between imperial and local identity processes.

13 An individual can only have one minzu (even if her or his parents are of two different minzu) and one nationality (as Chinese). Double citizenship is not permitted in China.

14 In overseas communities, Chinese-ness has different associations that tend to focus on common ancestry and shared cultural identity. Political and scholarly elites in mainland China, however, present the overseas Chinese as a transnational extension of the Chinese nation (Barabantseva 2010).

15 Putonghua (lit., “common speech”) is the sole national language in the People’s Republic of China. It is based on the vocabulary of the northern Han languages, Beijing pronunciation, and grammar as used in literary works written in baihua, a colloquial script. Putonghua is the standard language; it is what is referred to in English as Mandarin Chinese.

1. NARRATING “THE HAN”

1 See Gladney 1991 (72–73) for some illustrative quotes from Fei’s book. Xu Jieshun and the Han Studies Center he established at the Guangxi University for Nationalities played a crucial role in relaunching Han studies as a discipline in mainland China during the late 1980s.

2 Compare James Leibold’s 2010 study of Han supremacism on the Chinese Internet and his contribution in Critical Han Studies (Leibold 2012).

3 References to earlier dynastic periods in this book are offered to demonstrate that some processes in late imperial China clearly originate in much earlier periods.

4 I would like to thank Jonathan Lipman for his helpful comments on the meaning of Semuren and Semuguan (Officials of Various Categories). Although some scholars in China argue that semu means “colored eyes”—for example, Weng 2001—Lipman follows Frederick W. Mote, convincingly arguing that the usage of semu in Yuan-period sources indicates no such meaning. See also Lipman 1997 (33).

5 Among others, see Eberhard 1962 (18–30); Ebrey 1996 (23); and Harrison 2001 (30–31).

6 Sun, n.d. (1–6).

7 In Sun’s lectures, Han and Chinese often appear as interchangeable terms.

8 See Ramsey 1987 (8–11); Dikötter 1996 (250–51); Harrison 2000; Chow 2001; and Leibold 2006.

9 For examples, see Faure and Siu 1995; Zhang 2001; Hansen 2005; Whyte 2010; Carrico 2012; Joniak-Lüthi 2013.

10 However, Duara (1993) argues that the spatial idea of the Han community inhabiting a concrete country was not an invention of twentieth-century nationalism but had already been proposed by part of the scholarly elite during the Jin invasion in the twelfth century CE. Furthermore, the classification of Han subjects by the Yuan dynasty was spatial too. Northern Han/Chinese, together with Khitans, Jurchens, and other former subjects of the Jin dynasty, became classified as Han People. This was a category distinct from the Southern People, which included both Han/Chinese and other subjects of the defeated Song dynasty (see Elliott 2012). One can also argue that there was a spatial link between Han-ness and a sedentary lifestyle. Han were “fixed” spatially by cities, fields, towns, and markets, in contrast to the fluidity of the pasturelands. I am grateful to James Leibold for pointing out that association.

11 Compare Weng 2001 for examples of these terms used in historical records.

12 Yet there were certain limits to this openness. See Dikötter 1992 (10–17) and 1997 (20–21) for a discussion of the racial discourse of exclusion with regard to “African slaves” in China. Kang Youwei, on the other hand, argued that even “African slaves” could be “improved” through change of dietary habits, intermarriage, and migration (Dikötter 1992, 89).

13 It is necessary to pay attention to the agency of non-Han/Chinese in assuming or rejecting the Han identity, a point ignored in earlier acculturation and sinicization theses that presented Chinese culture as an overwhelming power that assimilated everything in its way (compare Crossley 1990).

14 For examples, see Blake 1981 (7–16, 87); Choi 1995 (104–22); Hayes 1995 (90–92); and Ching 1996 (58).

15 Until the seventeenth century CE, access to imperial power structures was largely restricted to the Han/Chinese through the system of imperial examinations. Wolfram Eberhard (1962, 18–30) emphasizes that non-Han were permitted a certain quota for the imperial examinations only from the seventeenth century onward, under the Manchu. Drawing on the reality of imperial examinations, he argues that upward social mobility was not as easy in China as many authors maintain. Indeed, large groups were excluded. While gentry and farmers were admitted, merchants were admitted only much later, and even then they were limited to 0.3 percent of the total quota. Sons of criminals (even if they came from the two top classes), monks, or non-Han were also excluded. Moreover, until the seventh century CE no free competition was possible, as candidates had to be recommended by local authorities according to a quota system. At certain times, however, a degree could be purchased. Periods of war also opened paths for upward social mobility.

16 See Blake 1981; Ching 1996; M. Cohen 1996 (1962); Segawa 1996; Leong 1997; and Harrison 2001.

17 For examples, see Fei 1980 on Chuanqing who “lost” the memory of being Han; Rhoads 2000 (278–79) on Han who joined Manchu banners and later registered as Manchu; Tapp 1995 and 2002 on complex identity constellations in mixed Han-Miao families; and Hansen 2005 on minzu switches of the Hanzu in present-day Xishuangbanna/Sipsongpanna.

18 Weng Dujian (2001, 61) quotes Discourses of the States (Guoyu), compiled probably in the fifth and fourth centuries BCE, which reports that “Rong and Di . . . are like beasts [ruo qinshou ye].”

19 However, Tamara Chin (2012, 133–34) emphasizes that there was no single “Chinese” worldview. Although the classical philosophers of the fifth to third centuries BCE generally distinguished between those within and outside of the Central States according to their adherence to Zhou ritual code, the cultural superiority of the Zhou was not always assumed.

20 On Han culture, see Leibold 2007 (23). Leibold demonstrates that even among Han-dynasty officials, the belief in the transforming power of Han culture and the emperor’s way (dao) was not shared by all. Some officials believed that some barbarians (such as the Xiongnu) could never be transformed into cultural beings; the only option was to keep them at distance and tame them by appeasement.

21 There are also some double-syllabic Han names such as Sima or Zhuge. They are, however, very rare.

22 For more details and examples of names, see Eberhard 1942; Fan 1961; and Müller 1980.

23 Xing originally meant a kin group of blood-relatives, a lineage. Another similar term, shi, referred to patriarchal clans of the nobles. During the Qin and Han dynasties, the two terms fused into one: xingshi, a surname group (Yuan and Zhang 2002, 3).

24 Among the Han, children inherit ancestral home place, or a place of origin along the patriline (jiguan), after their fathers. The idea of jiguan is very similar to the Swiss concept of Heimatort (patrilineal ancestral home place). People with the same family name, or branches within large family-name groups, share the same Heimatort (which is in most cases different from birthplace). For a detailed discussion of the concept of home place, see chapter 3.

25 James Watson (1993) isolates nine stages: ritual wailing to announce death, wearing hempen garb and other symbols of mourning, the ritualized bathing of the corpse, the transfer of goods (through the medium of fire) to the deceased, the preparation of a soul tablet with the name written in characters, the use of copper and silver coins in ritualized contexts, high-pitched piping and percussion to mark transitions in rites accompanied by a procession, sealing the corpse in a wooden coffin, and expelling the corpse from the community.

26 A Chinese expression that refers to times of great social disorder reads, “Fathers do not act like fathers, and sons do not act like sons” (Fu bu fu, zi bu zi) (Fei 1992, 128).

27 It is believed that Confucius taught that a correct mind follows from proper behavior: “When funerals are conducted in accordance with the rites, and sacrifices to remote ancestors are given devoutly, the morality of the people will naturally reach its peak” (Lunyu 1994, 6).

28 The oldest characters found on oracle bones date from the Shang dynasty (sixteenth to eleventh century BCE).

29 From the sixteenth century CE onward, bureaucrats and merchants also shared a common spoken language, the Officials’ Speech (Guanhua) (Yuan 1983; Ramsey 1987). Even if not a standard language—pronunciation differed depending on the speakers; and southern speakers were proud of their southern accents, which retained older distinctions from the Tang period lost in the North—Guanhua likely functioned as an important marker of belonging to these social classes (Ramsey 1987, 5).

30 Although the category of Boat People is organized to a considerable degree around the occupational distinction to agriculturalists, not all Boat People have actually engaged in fishing. Some have also been bamboo-basket weavers and farmers (Ye 1995, 83–84).

31 The Hakka (Kejia in standard Chinese, lit., “Guest People”) believe themselves to be—and historians partly confirm their self-representation—descendants of Han/Chinese who migrated from north-central China before the fourth century CE and settled in southeast China by the fourteenth century (Constable 1996, 7–15). The Hokkien (or Fujian People) are descendants of migrants from Fujian; they speak southern Min languages (dialects of Quanzhou, Xiamen/Amoy, or Chaozhou). Punti (Bendi in standard Chinese, lit., “local,” “native”) claim the status of “nativity” in Guangdong.

32 Cole (1982) reports that the Fallen People (Duomin) from Shaoxing Prefecture in Zhejiang, also referred to as “beggar households” (gaihu), were, like the Boat People, emancipated in 1723 from legal discrimination. Still, they were barred from engaging in farming and commerce in some parts of Shaoxing as late as 1945. Naquin and Rawski (1987, 148), as well as Eberhard (1962, 18), argue that the low status of the Fallen People is not only related to their occupations (hairdressers, bridal attendants, matchmakers, prostitutes, opera singers, banquet attendants, and fortune tellers, among others) but also perhaps to their origin among non-Han people, who, in the course of Han expansion, lost their livelihoods and came to specialize in these hereditary professions.

33 Interestingly, the imperial Qing understanding of China and the Chinese is similar to what these terms nominally represent today. In 1755 the emperor Qianlong pronounced, “There exists a view of China (Zhongxia), according to which non-Han people cannot become China’s subjects and their land cannot be integrated into the territory of China. This does not represent our dynasty’s understanding of China, but is instead that of the earlier Han, Tang, Song, and Ming dynasties” (Zhao 2006, 4).

34 Pamela Crossley (personal communication, 2014) emphasizes that while genealogies were an important part of Nurgaci’s (1559–1626) state building, they were first used in an ethnicized way to demonstrate Manchu identity in the mid-eighteenth century. Part of Nurgaci’s state building consisted of forcing subjects to write down their genealogies—which were earlier mostly committed to memory by designated members of the lineage, as paper was rare and expensive—for purposes of organizing companies within the Eight Banners. However, although committing genealogies to writing is an important historical threshold, genealogization as such does not begin with Nurgaci’s rule. Likewise, ethnicization of Manchu identity does not automatically begin with genealogization. For instance, imperial genealogies (yudie) were already being compiled in the mid-seventeenth century. Yet, the yudie were the exclusive preserve of the Aisin Gioro zongshi (clan) and thus not particularly relevant to Manchu identity generally. The yudie trace ancestry back to the time of Giocangga (in the case of the central line, back to Mongke Temur). This was important in establishing rights within the banners and the stipendiary bureaucracy of the Qing but had little to do with emphasizing ethnic boundaries. See also Crossley 1990a, 1999.

35 For an excellent study of Manchu-Han relations in late premodern and early modern China, see Rhoads 2000.

36 Sun, n.d. (1–6). Others, like Wang Jingwei or Liang Qichao, added a shared habitat, national spirit, and writing system (Harrell 2001, 29; Harrison 2001, 104).

37 Sun, n.d. (31).

38 Tracing ancestry back to mythical ancestors like Shennong (also called Yandi) and the Yellow Emperor (Huangdi) was already established by Sima Qian’s time and was still present in the genealogies from the Tang and Song periods (Ebrey 1996, 26–28). Revolutionaries reinvented the Yellow Emperor as the ancestor of all Han rather than of individual lineages. For Peking Man, see Sautman 2001.

39 Chinese revolutionaries and reformists tried to avoid making European history “the standard reference in temporality” by adopting the birthdays of either Confucius or the Yellow Emperor as the beginnings of Chinese chronology (Chow 2001, 63–65). However, judged by today’s awareness of these temporalities among the Han in the mainland China and among the overseas Chinese, the revolutionaries’ efforts did not have a lasting success. See also Ramsey 1987 (8–11); Dikötter 1992, 1996 (250–51); Duara 1995; Harrison 2000; Chow 2001; and Leibold 2006.

40 Although the Han used the same written language throughout the empire, spoken languages differed greatly. Some of the northern and southern languages were and remain mutually unintelligible.

41 Harrison 2001 (115).

42 On the Minzu Classification Project, see Wong 1979; Fei 1980; Heberer 1984; Harrell 1995; Huang 1995b; Gladney 1998; Wang, Zhang and Hu 1998; Joniak and Lüthi 2001; and Mullaney 2011.

43 The last of the currently fifty-six minzu, the Jinuo, was officially recognized in 1979. Since then, no new minzu have been officially recognized, but some groups have been given semiofficial status as distinct ren (people) within a larger minzu.

44 Stalin’s definition of nation echoes Marx and Engels: modern nations are ethnocultural and linguistic communities with their own states, while nationalities are ethnocultural and linguistic groups that did not develop into full nations because they lacked their own state. Marx and Engels also argued that “national communities incapable of forming national states are hindering the development of the progressive centralization and uniformation of humanity, and must therefore assimilate to more ‘vital’ and ‘energetic’ nations capable of forming national states with democracy ‘as compensation’” (Nimni 1995, 72). Inspired by both Marx and Stalin, Mao envisaged the new Chinese nation as a composite of the Han, constituting its core, and the nationalities located at the margins of the national imagery.

45 Stalin chose a dual path of citizenship and nationality: one was Soviet (citizenship) and at the same time, for instance, Ukrainian or Russian (nationality). Stalin’s solution to the “national question” was also implemented by Mao in China: one is simultaneously Chinese (citizenship) and Han (minzu), or Chinese and Tibetan, and so on. Although Mao imported many Soviet political ideas, he never copied the Soviet state model as a union of republics. China was to become a multiethnic but unitary state.

46 Although this model was adopted as Marxist in China, for Asia (especially for India and China) Marx and Engels originally conceived a distinct “Asiatic mode of production” that posited Asian societies as doomed to stagnation and lacking the driving power that would push their development forward. Marx and Engels interpreted the activity of colonial powers in China as modern capitalist institutions making the country develop the “proper” way (Wicker 1974, 3). Understandably, the Asiatic mode was a cause of consternation in the East, and it was eventually deposed from Marxist theory at a conference in Leningrad in 1931 (Wicker 1974, 127). Still, Marx was not consistent on this point, and in the introduction to A Critique of Political Economy he wrote. “Asiatic, ancient, feudal, and modern bourgeois modes of production can be designated as progressive epochs in the economic formation of society” (quoted in Tong 1989, 196).

47 Morgan’s model reached the Chinese Communists through Engels’s interpretation of his ideas in Der Ursprung der Familie, des Privateigenthums und des Staats (http://marxwirklichstudieren.files.wordpress.com/2012/11/engels-ursprung-der-familie-usw.pdf). Engels also based his analysis on Marx’s notes on Morgan.

48 Morgan linked savagery with a “Punalua family” (group marriage). The early stage of barbarism was connected with matriarchal family structure, the middle period with both matriarchal and patriarchal structures, and the late with patriarchal family structure. The final stage of civilization was marked by monogamous marriage (Morgan 1976, 33–34).

49 In the 1990s, the label “living fossil” was successfully promoted as a slogan to attract Han tourists searching for exotic marital customs.

50 Similar to the early twentieth-century revolutionaries, Mao conceptualized the Han as a community that had existed uninterrupted since time immemorial and that had developed in an evolutionary way from a primitive classless society to a feudal class society and, later, to a semicolonial one. In Mao’s version of history, after thousands of years of classless society China evolved four thousand years ago into a phase of slavery, and during the Zhou and Qin dynasties (eleventh through third century BCE) into feudalism. Afterward, it developed only slowly, with the period of feudalism lasting almost three thousand years. After the Opium Wars, China transformed gradually into a semicolony and then into a semifeudal society (Mao 1968–69, 354–88).

51 For popular contemporary markers of Han-ness, see chapter 2.

52 In one case, Han minzu belonging is a matter of choice and not descent only, namely when a child of mixed non-Han and Han parents decides at the age of eighteen which minzu she wants to belong to, her mother’s or her father’s.

53 In fact, there is still a substantial population with an “unidentified minzu status” (wei shibie de minzu); most of them live in Guizhou. This population without a minzu status numbered 734,438 in the year 2000 (ZRTN 2002, 106) and 640,101 in 2010 (National Bureau of Statistics of China, www.stats.gov.cn).

54 However, signifiers like “mixed blood” (hunxue; indicating a person of a mixed-minzu descent) or “half Yi, half Han” (Stevan Harrell, personal communication, 2014) are also used as self-denominations in contexts where the information may be important. Mixed descent may be deliberately concealed in other contexts. Significantly, these individual negotiations have no impact on the official classification of the persons who use them.

55 Compare Ann Anagnost’s (1997) “nation as narration” and Homi Bhabha’s (1990) notion of nation as a narrative of historical progress.

56 Duara (1995) brilliantly calls for “rescuing history from the nation” by attending to the appropriation of history by nationalisms.

57 Compare Harrell’s (1996a) argument about the role of ethnic leaders in creating teleological narratives of national and ethnic unfolding.

2. CONTEMPORARY NARRATIVES OF HAN-NESS

1 For analysis of these processes in contemporary China, see Gladney 1994; Schein 2000; and Blum 2001. For the pre-1949 period, see Duara 1995; Dikötter 1996; Harrison 2001; Chow 2001; Leibold 2006, 2012; Elliott 2012; and Giersch 2012.

2 Some markers could also be called stereotypes, but this term does not as clearly convey the act of marking, which is the focus here.

3 Chu (n.d.) demonstrates that Tibetan, Uyghur, but also Mongol minzu are the most often represented “minorities” in elementary-level textbooks.

4 But see Blum 2001 (57–59) for how Han-ness also differs from whiteness in several important respects.

5 Kaiser (1997, 11) compellingly argues that a nation is a “mass based community of belonging and interest, whose members share a backward looking sense of common genealogical and geographic roots, as well as a forward-looking sense of destiny.”

6 By “High Communist Period,” I am referring to James Scott’s (1998) definition of Authoritarian High Modernism.

7 Stevan Harrell (personal communication, 2014) points out that, in quite a contrast, non-Han such as the Yi stereotype “the Han” as having no family feelings or family loyalty. Similarly, Uyghur individuals often contrast their complex system of family obligations and filial piety with “the Han.” Especially the fact that Han children often live away from their parents in cities and thus do not fulfill the filial obligation to take care of parents has been popularly discussed.

8 The three terms Hanyu (Han language), Zhongwen (Chinese language), and Putonghua (Common Speech) are customarily rendered in English as “standard Chinese.” I use them interchangeably in this book. Interestingly, parallel to the tension between Han-ness and Chinese-ness, a tension in linguistic nomenclature exists in the overlapping use of the terms “Han language” and “Chinese language.”

9 Hansen (2005, 22) observes that apart from being a lingua franca, Chinese is also a “language of success.”

10 Since the Central Committee of the Communist Party endorsed the popularization of the new national language (Putonghua) in 1954, it has been promoted as the only official language. This despite the fact that about 30 percent of the Hanzu speak languages that are mutually unintelligible with Putonghua, and almost all non-Han minzu have distinct languages. Although the difficulty of mastering the Han language makes the idea of multiple national languages attractive, the dominant position that the Han language and education in the Han language grant the Hanzu leaves little space to entertain such ideas in the current political constellation. On how the national language standard was negotiated, see Yuan 1983 and Ramsey 1987.

11 Although the markers about clothing were collected between 2002 and 2003, they relate directly to the later Han clothing movement (Hanfu yundong) (Leibold 2010; Carrico 2012). The goal of the Han clothing movement, which had momentum in the late 2000s, was to “revive” the “traditional” Han silk robes in contrast to the “foreign” Qing robes. It was also intended to equip the Han with some attributes of ethnic uniqueness and particularity, qualities most popularly associated with Chinese “minorities.”

12 Lin Yutang in My Country and My People offers a prominent example of how Han-ness/Chinese-ness can be biologized: “[The] cultural homogeneity [of the Han/Chinese] sometimes makes us forget that racial differences, differences of blood, do exist within the country. At close range the abstract notion of a Chinaman disappears and breaks up into a picture of a variety of races, different in their stature, temperament and mental make-up” (Lin 1998, 17–18).

13 Non-Han, particularly those from rural areas and autonomous regions, receive for instance extra points at the university entry exams.

14 This argument does not relate to non-Han peoples, such as the Manchu or Mongols, who have assumed Han identity in the 1950s as a protection against discrimination and persecution. When the Maoist era concluded, these people and their descendants began to openly manifest their previously concealed identities, often changing their minzu status.

15 At the time of research, Han female fashions in the Lugu Lake region included self-embroidered shoes, blouses, and trousers. Some of these items were also exhibited for sale on the village streets.

16 It is possible that some of these markers and associations reached villagers during political education meetings, particularly the long public meetings of the Cultural Revolution, which many villagers remembered. It is difficult to assess what influence newspapers or migration had on transmission of these markers. Newspapers may have had some influence on the literate Han. Migration and word of mouth may have had some influence as well, but they do not explain why younger generations of school attendees who had not yet left the village were so articulate about these markers.

17 This is particularly visible in the control of private educational institutions. For instance, private kindergartens or language schools run by Uyghur individuals in Xinjiang are subject to much more control and institutional discrimination than other Uyghur-run enterprises.

18 Because Zuosuo is located within the Liangshan Yi Autonomous Prefecture, the junior high offers Yi (Nuosu) language classes. These were mandatory for all non-Han students in 1999 and voluntary for the Han, a Yi-language instructor reported.

19 Some Nuosu settled on the Zuosuo plain in a government program that provided them with land and timber for house construction. At the time of research, some of those who had originally moved had already returned to the mountain hamlets where they had lived before.

20 The Chinese state message is somewhat ambiguous on this point. “Minority” languages are constitutionally guaranteed the right to be used as a means of communication and education. At the same time, in practice they are rarely the languages of instruction beyond the third elementary grade. But there are exceptions. For instance, in southern Xinjiang in 2011–12, Uyghur was the language of instruction in the majority of primary Uyghur schools and in some secondary schools. Following the pressure to switch to “bilingual education” (shuangyu jiaoyu), circa 2010 Uyghur schools began turning to a “bilingual” curriculum where standard Chinese would be the language of instruction from third grade on. Given the lack of qualified teachers, however, this transformation has been occurring very slowly. There are also different models for the transition from a native language to standard Chinese as the language of instruction in non-Han-dominated regions, even if standard Chinese usually begins to replace other ethnic languages beginning in the third grade. While in some schools the change of language is abrupt, in others it occurs gradually. In yet other schools, some subjects are taught in Chinese while others remain taught in Uyghur far beyond the third grade, for instance due to the lack of Chinese-speaking teachers.

21 Compare Gladney 1991, 2004; Schein 2000; and Harrell 2001.

3. TOPOGRAPHIES OF IDENTITY

1 In chapter 4, I discuss other boundaries and attachments that emerged in my research data.

2 Ho 1966; Naquin and Rawski 1987; Watson 1988; Fei 1992; Honig 1992a; Goodman 1995; Leong 1997; Gladney 1998; Zhang 2001; Yi 2002; and Liu Xiaochun 2003, among others.

3 Jia translates as “family,” “household,” “home,” “people who share the same surname,” or simply “surname”; gu refers to services in which sacrifices are offered and also means the “old,” “former,” or “ancient.” Xiang indicates a “village,” “native place,” or “hometown.” Another popular term is laojia (old home), which tends to refer to birthplace or ancestral home place. Goodman (1995, 2, 4) also mentions other related terms such as sangzi (native place) or yuanji (ancestral home, native place).

4 Zuji, where zu means “ancestors” and ji means “record” or “registry,” is a term similar to jiguan.

5 Jiguan has to be registered, for instance, when checking in at a hotel.

6 This seems to indicate that Fei Xiaotong conceives of Han-ness as a kind of shared, racial descent.

7 Territorial bonds, like kinship bonds, come into play in economic life (Fei 1980, 95). Ho Ping-ti (1966, 120) also points out that common native place, together with kinship ties, used to be the most fundamental basis for voluntary association.

8 Contemporarily, clear specialization appears to exist too. For instance, Yi (2002, 111) discusses specialization among food sellers in Shanghai where Cantonese sell fish porridge, Shaoxing People specialize in fermented dried vegetables, Subei People market dough twists fried in sesame oil, and Ningbo People sell dumplings made of glutinous rice.

9 In late Qing, there were also guilds (hang) in Guangzhou, which gathered local merchants and were organized on the basis of specific services or products. However, unlike the Shanghai guilds, they did not have a strong influence on public life in Guangzhou (Rhoads 1974, 103). As merchants were not granted any political voice, the guilds were traditionally not involved in politics. This changed as the Qing weakened and the revolution neared.

10 That people are today migrating around the country does not mean that the household registration system has been abandoned. The transfer of registration from rural to urban areas is as difficult for rural migrants as it was under Mao. Migrant workers’ vulnerability continues in the cities. Although the need for changes in household registration has been recognized, including by central authorities, changes have so far been limited (Chan and Buckingham 2008).

11 Other possible terms with which to discuss Han place-based attachments are tautotopic identity (place-based identity) or topocentric koinosis, that is, a sense of commonality based around a place. I would like to thank Jonathan Lipman and Stevan Harrell for brainstorming about this terminology. Ultimately, however, the final choice of home-place identity was mine. I bear the responsibility for any conceptual slippages and inconsistencies.

12 I discuss the importance of state education in establishing minzu labels as viable categories of identification in my master’s thesis, “‘Since they have never been to school, they do not know who they are’” (University of Bern). The title is a quote from a government official in Zuosuo, Sichuan, who reflected on some villagers’ lack of awareness about the officially endorsed minzu classification.

13 Honig (1992a) thoroughly studies the processes through which Subei identity was created and stigmatized.

14 Jiangnan refers literally to the region south of the Yangzi River. South of the Yellow River would be henan in Chinese, but this is precisely the term that the interviewee wanted to bend.

15 Self-denominations put forth by the respondents from Subei included the following: Person from South of the River, here meaning the Yangzi River (Jiangnanren); from Xuzhou (Xuzhou de); Hanzu; Shali Person (Shaliren); Local (Lata); Huai’an Person (Huai’anren); Qihai Person (Qihairen); Person from North of the Yangzi River (Jiangbeiren); Taizhou Person (Taizhouren); and Shanghainese. No respondents referred directly to the term Subei in their self-denominations. Self-identifiers used by respondents from Henan included the following: (Person from) Central Plains (Zhongyuan), Zhengzhou Person (Zhengzhouren), Chinese (Zhongguoren); and Henan Person (Henanren).

16 For examples, see Chen 1999; Lin 2001; Ding 2002; Ma 2002; Liu 2002; and Liu Xiaochun 2003.

17 Hakka and Kejia are two different pronunciations of the same socionym, Guest People. Hakka is its pronunciation in the Hakka and Yue (Cantonese) languages, while Kejia is the standard Chinese pronunciation. Although these two pronunciations are most commonly used in China and abroad, other pronunciations include Khe-ka in the Minnan language (as used in Fujian) and in Taiwan, Khelang. Hakka from Sichuan are locally called Guangdong People (Guangdongren) (Stevan Harrell, personal communication, 2014).

18 Blake (1981) and M. Cohen (1996) argue that it was language and not migrant-native discourse that played a central role in the delineation of “us” and “them” in Guangdong and Guangxi where the Hakka settled.

19 This also seems to be one of the motivations of Ma Rong’s idea to “depoliticize” minzu categorizations in China (Ma 2014). He suggests that “the minorities” should stop imagining themselves as political collectives and instead focus on a self-understanding based on cultural differences among Chinese citizens. These citizens, Ma argues, should enjoy equal protection from the state and equal rights on the basis of being Chinese. In light of the Chinese government’s inability to guarantee equal rights today (e.g., in terms of language use or forms of representation in school textbooks), and the rather difficult task of convincing “the minorities” to abandon “emotionality” and focus instead on culture and socioeconomics, it is difficult to grasp who Ma’s call addresses, how it should be implemented, and by whom.

20 “Origin” must here be in quotation marks, as it is as likely to be invented as inherited.

21 A similar phenomenon was observed by Siegel (1989) in the Central African Copperbelt.

4. OTHERING, EXCLUSION, AND DISCRIMINATION

1 In chapter 2, to emphasize their marking function in processes of identity, I referred to tedian as markers. Here I wish to emphasize their stereotypical quality and collective reproduction.

2 In localized communities, other paradigms may be important. Compare for instance localized markers of distinction in Zuosuo in chapter 2.

3 These adjectives are quoted directly from my research interviews.

4 See also Hanson’s (1998) historical analysis of the discourse of difference between the North and South based on the notion of qi.

5 On the historical notions of culture (wenhua), see Fairbank 1968; Bauer 1980; and Watson and Rawski 1988.

6 Though southern provinces also have large agricultural outputs, in the perception of my informants they did not function as “agricultural provinces.” Instead, southern China was unanimously associated with developed industry and commerce.

7 Though southern provinces also have large rural populations, many of the informants nonetheless seemed to link agriculture with backwardness and the North.

8 For historical and linguistic explorations of regional distinctions and regional characteristics, see Moser 1985.

9 See Elliott 2012 for a description of how this influenced the idea of Han-ness and Chinese-ness.

10 On the idea of peasants as the embodiment of the premodern, and city people (shimin) as synonymous with modern, self-determining citizens, see Day 2013.

11 The Jing Youzi socionym was used by some Beijing People too. They argued that it positively connotes being experienced and having a broad knowledge of the world. The immigrants from Shandong and Inner Mongolia said instead that it reflected the slippery and cunning character of Beijing People.

12 See Müller 1980; Heberer 1989; and Thierry 1989.

13 In some of the southern provinces, however, the gentry were more urbanized. The southern gentry began shifting their residence to market towns and cities in the sixteenth century CE. However, at the same time wealthy city dwellers began to invest in agricultural land (Elvin 1977, 459–60).

14 See Cohen 1993 for a discussion of the politics of peasant categorization. However, Cohen (1993, 154) points out that the notion of countryside as a locus of feudalism and superstition was not shared by all. In particular, social anthropologists and sociologists with fieldwork experience in rural China did not subscribe to this reading of the countryside.

15 But Chan and Zhang (1999, 830) point out that long-term undocumented rural migrants, the so-called black households, did exist in cities prior to 1978.

16 For an excellent account of rural/urban inequality, see Whyte 2010.

17 See Malkki 1992; Feld and Basso 1996; Paasi 1996; Lovell 1998; and Navaro-Yashin 2012.

18 In China’s multiethnic borderlands, the issue of who is “native” to an area (are Han settlers “native” to Yunnan, for instance?) is overtly political (Hansen 2005). On the role of “nativity” in ethnic tourism in China, see Swain 2001. Zhang (2001) explores the construction of migrant “strangeness” in Beijing. I discuss the question of belonging to a place and the making of Han places in Xinjiang in Joniak-Lüthi 2014.

19 Southern Jiangsu was referred to briefly as Jiangnan (Region South of the Yangzi River) and Sunan (Southern Jiangsu). Together with immigrants from Guangdong and Subei, Jiangnan natives have comprised some of the largest migrant communities in Shanghai from the mid-nineteenth century on. As Honig (1992b, 239) argues, “Which of these areas one hailed from was critical in shaping work opportunities, residential patterns, cultural activities, and social status. Hierarchy was structured largely according to local origins: the elite was composed primarily of people from Guangdong and Jiangnan, the unskilled service sector staffed mostly by migrants from Subei.”

20 In contrast to this prevailing view, some interviewees from Subei argued that Subei People are “industrious,” “nimble,” “quick,” “hardworking,” “reliable,” and “earnest.” Further, they believed that they have “willpower” and “perseverance” and that they are “kind” and “good-hearted.”

21 Ma (2002, 54–58) reports having seen notices posted in shops and restaurants warning, “Henanese are forbidden to enter” (Henanren bude ru nei) or “We do not employ Henan People” (“Bu zhao Henanren” or “Bu yao Henande”).

22 Anhui People were also associated in Shanghai with the category of rural strangeness, but only by a couple of my informants. Additionally, the terms used to characterize Anhui People were much milder than those used for Subei People.

23 Other identifications, for instance with work unit, once of a critical importance, were not mentioned.

24 See Blake 1981; Choi 1995; Hayes 1995; Liu 1995; Ching 1996; Constable 1996; and Leong 1997.

5. FRAGMENTED IDENTITIES, THE HAN MINZU, AND ETHNICITY

1 I do not imply here that the minzu scale is located above the “regional differentiation” scale or that it encompasses it; on the contrary. Because most of my informants claimed that identification with home place is more central to them than identification with the Han minzu, one could argue that these scales are parallel or that the minzu scale is actually secondary, including historically so, to that based on the notion of home place.

2 Compare for instance this quote from Weng 2001 (3): “At the times of Xia, Shang, and Zhou [twenty-first century BCE–256 BCE], each minzu of our country [woguo ge minzu] and their mutual relations experienced great progress and development.”

3 My thanks to Stevan Harrell for suggesting this parallel.

4 See Rowe 1984 (247); Moser 1985; Watson 1988 (133); and Cole 1996 (161).

5 The term zu, a component of minzu, connotes territorial lineage. This makes it related to ren or home-place identities. As a term that relates to kinship, it is also somewhat similar to jia. However, it is virtually never used as a suffix in contemporary Chinese in ways similar to ren and jia. For instance, while it would be possible to say women Zhangjia (lit., “we people of the Zhang family”), it is not possible to say women Zhangzu. This is because zu as a suffix was politicized as a corruption of minzu.

6 For a detailed analysis of this process, see Mullaney 2011.

7 Yet we should not forget that the label was arbitrarily imposed onto some local groups and denied to others during the process of minzu classification. For examples, see Fei 1980.

8 I would like to thank Steve Harrell for his insightful comments on the spatiality of ren.

9 Before the Minzu Classification Project, some non-Han in the Southwest were also referred to in the Chinese language as jia, as in Minjia or Yijia.

10 On the Danmin, see Anderson 1972; on the Duomin, see Cole 1982. As “people,” the term min is used, for instance, in combination with some minzu identifiers like Huimin (Hui People) and in designations like Nongmin (Peasants).

11 But see Carrico’s (2012) study in Guangdong, where he found people calling for the independence of Guangdong from the North. Still, I understand this to be a call for the revision of power relationships with Northerners rather than for independence from Han-ness.

12 But compare Hansen (2005, 40), who reports on Han cadres in Sipsongpanna registering their children as “minorities.” Also, children of mixed couples (Han and non-Han) have the right to choose between the minzu affiliations of their parents. Often, these youth decide to identify as a “minority,” as this status has certain advantages (e.g., permission to legally have more children, priority access to positions in local government, and extra points on university entrance exams).

13 This may be different in other cases. For example, the Mosuoren of the Lugu Lake region were granted recognition as a distinct people (ren) within the Naxi minzu (in Yunnan) and Mongol minzu (in Sichuan) to which they were assigned during the Minzu Classification Project. This ren identity is likely less negotiable and individual-dependent than the ren identities of Han in eastern China.

14 But see Barabantseva 2012, which analyzes the discourse of overseas Chinese minorities and the ways in which the Chinese government reinvents the Chinese minzu categorizations in overseas Chinese communities.

15 A similar project of nesting minzu identities within the Chinese national identity has been less successful in relational terms. When the Chinese national identity is challenged, or when the scale of interaction becomes international, a similar “disappearance” of minzu identities in Chinese-ness does not occur, with the exception of Han-ness, which is Chinese-ness itself. The nesting efforts of the Chinese state have been the least effective with transnational minzu such as the Tibetans, Uyghur, or Dai. While Han-ness merges into Chinese-ness at the international scale of interaction, these minzu identities do not disappear in Chinese-ness but remain to a significant degree distinct. Hence, while most ren, min, and jia identities of the Hanzu have been successfully nested in the Han minzu, the parallel nesting of minzu identities in the Chinese identity has been much less effective and is still very much in progress.

16 The notion of nested identities (Brewer 1993, 1999; Calhoun 1994; Medrano and Gutiérrez 2001) is similar to how the functioning of multiple minzu and non-minzu identities is conceptualized in this book, namely as compatible because each fulfills a different function. However, I argue here that it is crucial to pay close attention to how identities become “nested.” Further, my research data do not support discussing nested identities in terms of higher- and lower-order identities or higher- and lower-level identifications. Instead, identities appear more parallel. They may be verticalized and hierarchized for certain purposes and by specific actors, but not in analytical terms.

EPILOGUE

1 Barabantseva 2012 describes Chinese attempts to export these categorizations.

2 I am grateful to James Leibold for this metaphor.

3 This distinction was formulated by the late Karl Lo, University of Washington librarian (Stevan Harrell, personal communication, 2014).