Introduction
In the late sixth century, an emissary from the Northern Qi dynasty (550–577 CE) visited Southern dynasty Chen (557–589 CE). As a southern host entertained this emissary, at one point the host jestingly remarked, while handling a betel nut (binlang 檳榔), that he had heard that someone in the north had recently been accused of committing an offense for chewing that mild stimulant. He asked if it was true that betel had been banned. The emissary replied that there had indeed been a case around 550 when chewing betel had been mentioned in a criminal charge, but betel had no more been forbidden than cosmetic facial powder (hu fen 胡粉) had been banned by the Han dynasty (202 BCE–220 CE) after Li Gu 李固 (94–147 CE) was unsuccessfully accused, among other things, of standing around looking elegant in facial powder when he should have been in mourning.Footnote 1 Betel chewing is typically associated with the remote south—during the Tang dynasty (618–907), for example, betel was among the local tribute presented by Annam 安南 (what is today northern Vietnam)Footnote 2—and betel might have still been considered somewhat exotic in north China in the sixth century. Correspondingly, dairy products (such as what was called lao 酪) were northern peculiarities that were considered outlandish in the south during the centuries of division after the Han.Footnote 3 The earlier Han dynasty facial powder might also at first glance appear to have been a parallel bit of northern exotica, because the character hu 胡 in hu fen normally indicates foreign northern origin, but in this case it is probably only a variant of a different character hu 餬 meaning something like “paste.”Footnote 4
The intriguing thing about this otherwise trivial anecdote—which is, furthermore, presumably no more reliably accurate than any other anciently recorded story—is that it demonstrates that even at a time when there were separate northern and southern dynasties within the geographic space we think of today as China (one of which—Northern Qi—had rulers who are conventionally classified as not Han ChineseFootnote 5), and at least somewhat different regional cultures, members of the elite from both states were nonetheless still able to communicate with each other fairly easily,Footnote 6 and could be expected to be familiar with even minor incidents from Han dynasty history. It could apparently simply be taken for granted that educated gentlemen representing entirely different governments—as well as all anticipated future readers of the written account of this story—would automatically recognize the passing reference to Li Gu's cosmetic powder, without need of explanation, even though the event happened more than four hundred years before the conversation narrated in this anecdote supposedly took place, and nearly six centuries before the text recording that conversation was compiled (and eight centuries before its oldest surviving source version). Although this is only one unimportant little story, it therefore nicely illustrates the kind of deeply shared historical memory that may have made the “reunification” of northern and southern geographic China after 589, under the Sui (581–618) and Tang dynasties, if not exactly “inevitable” (as some may too easily assume), then at least fairly natural.
Yet, it is also true that elites throughout East Asia in this era—extending well beyond China—shared a substantially common literate culture. For example, diplomatic exchanges between Sillan 新羅 (Korean) envoys and their Japanese hosts in 723 were conducted in part through the medium of written Chinese-language poetry (and possibly even orally in “vernacular Chinese”), with the Japanese also referring to Japan as “Hua” 華 in contrast to Sillan “Yi” 夷 (in Japanese, pronounced ka and i)—terms that, as we will see, had ancient pedigrees as quintessential designations meaning “Chinese” and “non-Chinese” (although also perhaps, more ambiguously, “civilized” versus “uncivilized”).Footnote 7 Even within China itself, moreover, the shared written language and high culture were closely tied to dynastic states and the relatively small literate elite that was qualified to administer them, which coexisted in perpetual “tension” with a much more numerous “private realm” of local and family interests and identities.Footnote 8 What, then, was the “China” that was supposedly “reunified” in 589?
The interrelated topics of Chinese identity, Han Chinese identity, and Chinese nationalism continue to provoke study and debate to the present day. Among the recent books published in English dealing with these issues, Critical Han Studies: The History, Representation, and Identity of China's Majority appeared in 2012. This volume is an edited collection drawing methodological inspiration from approaches such as Critical Race Theory and Whiteness Studies to analyze some of the complexities of Han identity—a purportedly “ethnic” Han identity that includes eight major “speech communities” and multiple different (sub-[?]) cultures and (sub-[?]) ethnicities.Footnote 9 By contrast, Nicholas Tackett's 2017 publication, The Origins of the Chinese Nation: Song China and the Forging of an East Asian World Order, argues that a consciousness of China as an ethnically Han Chinese nation-state had already emerged by as early as the eleventh century, during the Song dynasty (960–1279).Footnote 10 Meanwhile, Zhaoguang Ge's 2018 book, What is China?: Territory, Ethnicity, Culture and History, dismisses postmodern and Western theories that are critical of the nation-state, or that try to depict China as an “imagined community.” Ge also seems to simply assume the existence of the Han people as an unproblematic given, although he does explicitly acknowledge that “what is China” has changed over time and is pluralistic.Footnote 11
Indeed, most scholars now agree on the diversity of China's origins. Scholars in mainland China today, however, generally prefer to believe that China's diversity is contained within a “unified multi-ethnic framework” (duoyuan yiti geju 多元一體格局), or that, through a kind of “snowballing” process stretching across centuries, various different communities were absorbed into and assimilated around a common Han core.Footnote 12 Outside of mainland China, on the other hand, postcolonial and postmodern scholars (especially) are more inclined to question such assumptions. A particularly sharp line of disagreement seems to take shape over the issue of sinicization, or “transformation into Han” (Hanhua 漢化), whereby originally non-Chinese people supposedly become Chinese through assimilation. Chinese scholars generally take sinicization as an historical fact (and often assume that it is driven either by the compelling attractiveness, or more advanced level of development, of Chinese culture), while non-Chinese scholars tend to be more skeptical. An example is the case of the so-called New Qing History studies in the United States, which emphasize Manchu and other non-Han elements in the Qing dynasty (1644–1912) and deny that Manchu success as rulers of the Qing dynasty was due to sinicization.Footnote 13
Somewhat inadvertently, however, Zhaoguang Ge may have hit on a key to better understanding of Chinese identity when he observed that, despite undeniable periods of political division, “since ancient times China … has already been narrated by one ‘history’.”Footnote 14 A single continuous historical narrative is, indeed, what is typically presented in the standard textbooks. But history is a human construct, woven together from selected memories and documents into a comprehensible story, rather than some objectively absolute verifiable scientific fact. Once an historical narrative has been framed and articulated, however, that story can then become a kind of reality in itself. In combination with the hard political reality of living in the same (repeatedly recreated) centralized state, and having a shared written language and other institutions, historical memory may have helped consolidate a Chinese identity—one that brought together what may have originally been quite diverse peoples—that truly has been remarkably enduring. This essay will argue that a tenacious thread of historical memory was a major factor helping what we call Chinese identity to survive through an extended period of not merely political division, and ethnic, cultural, and linguistic diversity, but also when language, institutions, and material culture flowed both directions across the borders of what we think of today as China.
The Origins of a Chinese Identity
Before the Sui dynasty conquest of Southern dynasty Chen in 589, the approximate geographic space that we think of today as China Proper had actually only ever been unified into a single centralized country under the Qin (221–207 BCE), Han, and Western Jin (265–317 CE) dynasties, and not with stability for the entirety of even those periods. Although it remains popular to imagine a single continuous China existing since antiquity, and to assume a supposedly obvious Chinese ethno-racial identity, the towering modern intellectual Liang Qichao 梁啟超 (1873–1929) correctly recognized already a hundred years ago that even the original core Chinese identity emerged from a number of different population groups living in the Central Plain area, and that the Three Dynasties of classical antiquity—Xia, Shang, and Zhou—rather than being different dynastic houses ruling a unified China one after another in succession, may have originally been different groups of people, who for a time may have even coexisted separately.Footnote 15 Many of Liang's contemporaries did imagine the Chinese people to have a single cohesive ethno-racial identity, but even they tended to assume it was an “inclusive” race that had absorbed and assimilated various other peoples over the course of centuries.Footnote 16 Since Liang's time, a recognition of the multiple origins of Chinese civilization has arguably become the mainstream of serious scholarship.
The idea of an enduring unitary “China” has been challenged to such an extent that Edward H. Schafer once even flatly declared that the very concept of China “simply did not exist, except as an alien fiction.”Footnote 17 Yet, while it is true that “China” and “Chinese” are not Chinese language words, and have no exact Chinese language equivalents, and the geographic space we call China has not always been a single continuously unified state, nevertheless, few even among the skeptics would deny that a distinctive civilization did emerge in the Yellow River region of what is today north China over the course of the last millennium or two BCE. This civilizational identity was sometimes referred to broadly as Hua 華 or Xia 夏 (which were later often combined into HuaxiaFootnote 18), and its core state or states came to be called Zhongguo 中國 (the “Central Country/ies”).
This Hua/Xia civilization was conceived as contrasting with alien neighboring peoples, who were stereotypically labeled Rong 戎, Di 狄, Yi 夷, or Man 蠻. Even as these identities were crystallizing during the classical Zhou dynasty era (ca. 1045–256 BCE), however, interaction between the Hua/Xia and the alien Rong, Di, Man, and Yi peoples was sometimes extensive,Footnote 19 and a characteristic belief also emerged that any of these “non-Chinese” people who lived amidst the Hua/Xia and accepted Hua/Xia culture could potentially also become Hua/Xia (“Chinese”) themselves. Confidence in the exceptional attractive power of Chinese culture, and its ability to assimilate different population groups—sinicization—in fact became one of the more distinctive features of Chinese civilization.Footnote 20
Although a degree of subordination to the supreme Zhou dynasty king was commonly acknowledged by the various states existing in the geographic region we call China in antiquity, some of the more peripheral states also long remained rather marginal to the Zhongguo core. Around 860 BCE, for example, a ruler of the important southern state of Chu 楚 is supposed to have remarked: “We are Man and Yi [i.e., not Hua or Xia], and do not participate in the conferment of titles by Zhongguo” (我蠻夷也, 不與中國之號謚).Footnote 21 Even the northwestern Qin kingdom that forged the first version of the Chinese empire in 221 BCE “was originally a small and remote country, and all the Xia treated it as a guest, comparing it to the Rong and Di [i.e., non-Hua/Xia]” (秦始小國僻遠, 諸夏賓之, 比於戎翟).Footnote 22 After the Qin kingdom conquered the other Warring States and unified the Chinese world for the first time, there were still apparently some (non-Hua/Xia) Yi people living in eastern China (as it is geographically understood today) who were subsequently then “dispersed as subject households” (秦并六國, 其淮, 泗夷皆散爲民戶).Footnote 23
Qin imperial unity was brief, but, under the much longer-lasting Han dynasty, common imperial cultural, political, and economic institutions and networks, which were furthermore open and inclusive—exerting what Xu Zhuoyun 許倬雲 has described in currently fashionable (but not necessarily inappropriate) modern terminology as a “soft power” (ruan shili 軟實力) style attraction over all the peoples within their gravitational field—caused many inhabitants of the empire to increasingly converge on a shared Han imperial identity, becoming what would eventually literally be known as the Han people 漢人.Footnote 24 The Qin and Han empires were undoubtedly important stages in the solidification of a Chinese identity.
Ancient China has already been the subject of much scholarly attention, and the ancient origins of Chinese identity are not the focus of the present inquiry. Rather, we are primarily concerned here with how a sense of what might be called Chinese identity survived during the multi-ethnic, multi-state, and richly cosmopolitan centuries of division after the end of the Han dynasty, and was subsequently reconsolidated under the Sui and Tang dynasties. Unlike antiquity, the Age of Division is an epoch that has not yet been sufficiently studied, especially in English, and it was also a time during which China arguably confronted a serious identity crisis.Footnote 25 However, the growing consensus that ancient Chinese identity was, from the beginning, already a synthesis of different regional cultures and peoples may help make continuing ethno-cultural synthesis across the centuries of division after the Han dynasty somewhat easier to understand, and seem less radically exceptional. And, the enduring memory of antiquity and the Qin-Han imperial model throughout this period of fragmentation may have laid a template for future reintegration.
When the center failed to holdFootnote 26
Even after four centuries of Han dynasty imperial unity, many people in the geographic space we know as China still retained at least a kind of hyphenated identity. A third century CE text, for example, described the Di 氐 people living in what are now western Shaanxi, eastern Gansu, and northern Sichuan, as being familiar with the Chinese language and subject to Chinese administration, but speaking their own very different language among themselves, and still having their own native leaders.Footnote 27 A stone inscription erected in 367 by a Capital Protector (hu jun 護軍) at Pingyi 馮翊, northeast of modern Xi'an in Shaanxi, listed twelve different Yi peoples (夷類十二種) under his administration.Footnote 28 Modern estimates have placed the number of persons living inside geographic China but still retaining distinctly different, separately named, group identities during the centuries of division after the fall of the Han dynasty in the millions.Footnote 29 The so-called Five Hu 胡 (another generic label for northern non-Hua/Xia peoples) who overran north China in the fourth century tend to be the focus of most historical attention, but it has also been suggested that pre-Sinitic aboriginal peoples may still have constituted more than half of the total population of what is today south China during this period as well.Footnote 30 The dynastic history of the Sui reported, for example, that in the region of the ancient Chu kingdom (roughly modern Hubei) there were still many Man and Yi people. Those who dwelt among the Xia people 夏人 (i.e., what we call Chinese) were said to be no different from all the Hua 諸華 (also meaning what we call Chinese), but those who lived in remote mountain valleys still spoke incomprehensible foreign languages and had completely different cultures.Footnote 31
Whether these different groups should be considered “ethnicities” raises thorny theoretical questions about the definition of ethnicity which, for our purposes here, can be temporarily set aside. We merely need to invoke something like a concept of ethnicity, as Walter Pohl has written with regard to roughly contemporary western Eurasia, for “the simple reason … that ethnonyms figure so prominently in our sources.”Footnote 32 The primary sources are replete with references to separately named population groups in the geographic area of China during this period, which demand historical consideration.
The degrees of difference in language and culture between these separately named population groups could vary widely. The XianbeiFootnote 33 people, for example, who were prominent across the area of north China from the third through the sixth centuries, were speakers of a language that may have been related to later Mongolic, but was unrelated to Chinese.Footnote 34 The Xianbei themselves were then subdivided into multiple separately named subgroups, including, among others, the Tuoba 拓跋, the Murong 慕容, and the Tuyuhun 吐谷渾. Differences between these Xianbei subgroups may have sometimes been slight—the Tuyuhun, for example, were said to have been founded by a breakaway brother of a Murong leader whose personal name happened to be Tuyuhun, and whose following subsequently came to be named after him.Footnote 35 At the same time, at least in their larger incarnations, all of these groups typically also had mixed populations, and were not uniformly homogenous.
Identities could be surprisingly fluid, and there was considerable exchange and interaction between all of the groups throughout the region, including, of course, the Han or Hua/Xia (“Chinese”). It should also be understood that it is perfectly normal for an individual to simultaneously hold multiple identities, both at different levels—such as family, community, region, and state—and of different types—such as gender, social class, language or accent, occupation, ethnicity, and religious affiliation. It is also not uncommon for identities to be mixed—as a result of intermarriage, for example—and sometimes even appear mutually contradictory.
Over time, across much of Eurasia, the many small group identities that were so apparent in late antiquity tended to gradually disappear or consolidate into fewer, larger, and what at least in some cases are now more familiar proto-national identities, including the one we call “Chinese.” While a remembered claim to Xianbei descent still survives today among certain minority populations in China,Footnote 36 most of the group identities that were so conspicuous during the Age of Division have long since effectively vanished, and the descendants of those people (at least the ones living in geographic China) have mostly now become Han Chinese. However, during the Age of Division between the Han and Sui-Tang dynasties, among the dozens of kingdoms and empires that rose and fell in geographic China, a clear majority of those in the north had rulers with separately named group identities that are conventionally considered to have not (yet) been Han Chinese. And, relations between these non-Han rulers and their Han subjects could sometimes be strained. In 349, for example, there was a reactive mass slaughter of foreign-appearing persons (having protuberant noses and thick beards) in north China,Footnote 37 and it was alleged that in cavalry battles between the Xianbei-ruled Northern Zhou (557–581) and Northern Qi dynasties in the mid-sixth century, the Xianbei warriors “always drove Xia [i.e., Han Chinese] people [ahead of them] as fences of flesh” (周齊每以騎戰驅夏人爲肉籬).Footnote 38
Yet, despite a fair amount of reciprocal cultural exchange, and the conscious preservation of a sometimes haughty attitude of Xianbei superiority over their conquered Han subject population until nearly the end of the sixth century, most of the ostensibly non-Chinese ruling elites of this period had also, almost from the beginning, already acquired a great deal of what we can only call Chinese culture. The Chinese language enjoyed a near (though not quite total) regional monopoly of writing. Funerary practices in north China generally continued to follow Han dynasty and subsequent Wei and Jin dynasty traditions.Footnote 39 And, while separate administration of the Han and Hu populations in north China was initially common, and the old non-Chinese Xiongnu 匈奴 steppe nomadic title chanyu 單于 was long maintained alongside Chinese-style official titlesFootnote 40 (and an apparently new non-Chinese title, qaghan, was used by the Tuoba XianbeiFootnote 41), even the “non-Chinese” kingdoms and empires of fourth- through sixth-century north China were nonetheless to a large degree Chinese in style.Footnote 42 Indeed, they sometimes turned around and bestowed linguistically and institutionally Chinese titles on peoples and places that were more unambiguously not Chinese, such as on the rulers of the proto-Korean kingdom of Koguryŏ 高句麗 (alternatively spelled Goguryeo).Footnote 43 Much of the development of East Asian states in this early period was, in fact, associated with the adoption or acceptance of Chinese-style titles, up to and including the Chinese language title “king” (wang 王).Footnote 44 In the case of Japan, this eventually included the full range of supreme Chinese-language imperial titles.Footnote 45
As an example of border-crossing interactions in this era, in 372 one of the ethnically non-Han Chinese-ruled empires based in north China (Former Qin, 351–384, which was ruled by Di people) reportedly introduced Buddhism to Koguryŏ. Later that same year Koguryŏ established a Confucian academy, and in the following year Koguryŏ promulgated Chinese-style legal codes.Footnote 46 And, of course, Koguryŏ (as well as the other kingdoms on the Korean peninsula and Japan) also adapted the Chinese writing system and extensively utilized the Chinese written language itself.
Koguryŏ lay beyond the Chinese heartland in the northeast; in the northwest, also beyond the Chinese heartland, in a palace of the kingdom of Gaochang 高昌 (located near modern Turfan, in Xinjiang) in the late 500s there was reportedly a painting of a Duke of Lu 魯 asking Confucius about government, and education officials were established for three of the Confucian classics. But these Confucian classics were studied in a non-Chinese Hu language. As of 599 the ruler of Gaochang simultaneously held both the Chinese title “king” and the Turkic title eltäbär. And, although Gaochang's king attended the “reunified Chinese” Sui dynasty court in 609 and even accompanied a Sui invasion of Koguryŏ, Gaochang's adoption of Chinese cultural standards remained limited because it was, at that time, more directly subordinate to the Tiele 鐵勒 Türks.Footnote 47 Meanwhile, north China itself remained mostly under non-Han Chinese rule for nearly three centuries, from 304–581. All of this makes disentangling “China” from “not China” in this era surprisingly difficult.
Establishing legitimacy
Yet, there also was a firmly held ancient Chinese cultural assumption that there could only be a single legitimate supreme sovereign at any one time, who was the recipient of Heaven's Mandate to rule all Under Heaven. The arcane (and often self-serving) issue of determining dynastic legitimacy, therefore, becomes relevant to any exploration of Chinese identity, because recognition as the legitimate dynasty effectively anointed that particular state as the standard bearer for the whole heritage of China. Although shared cultural practices, independent of any state structure, may ultimately have become the key feature distinguishing Han from non-Han people,Footnote 48 in a study of premodern Chinese “national identity” Michael Ng-Quinn nonetheless explicitly took only the state into consideration, rather than ethnic or other social configurations.Footnote 49 Dynasties (that is, states), starting with Xia, Shang, and Zhou, formed the spine of premodern Chinese identity, and monopolize much of the narrative of Chinese history. Mark Elliott has even suggested that “the emergence of the Han as a single ethnic group was not so much the basis for Chinese unity as a consequence of it.”Footnote 50 Even in post-imperial modern times, John Fitzgerald argues that the Chinese nation, rather than simply being a naturally preexisting fact, has instead been repeatedly recreated and defined by the state. In China, Fitzgerald writes, “the appearance of a continuous history derives less from the preservation of a Chinese nation than from the ideal of the unitary state which transcends all state formations and is made identical with the idea of China itself.”Footnote 51
Precisely because there have been so many state formations in the geographic space we call China, however, determining which was the sole legitimate unitary state that constituted “China,” at any given time, becomes crucial. During the Age of Division after the Han dynasty, what came to be accepted standards for determining dynastic legitimacy included virtuous rule (worthy of Heaven's Mandate), actual control over the Central Plain heartland (or, better yet, unification of the whole of geographic China Proper), and patronage of traditional Chinese civilizational standards.Footnote 52 The mere fact of rule over the Chinese heartland could be enough to stake a claim to legitimacy. One fourth-century ethnic non-Han ruler (a Jie 羯), for example, was assured: “Since Your Majesty has wrapped up the two capitals [in north China], you are the Emperor of Zhongguo” (陛下既苞括二都, 爲中國帝王).Footnote 53 But legitimacy was more secure if it was widely recognized, and foreign recognition could be an especially strong confirmation of legitimacy. The same ethnic non-Han emperor of north China who introduced Buddhism to Koguryŏ in 372 (receiving tribute from the king of Koguryŏ in return), reportedly also received tribute from the rulers of sixty-two different states—including such offerings as the legendary “blood sweating horses” (汗血馬) from Ferghana (Dayuan 大宛).Footnote 54
At the same time, both the usual and the most unquestionable source of legitimacy was to inherit it from an established legitimate previous ruler; and claims to hereditary imperial descent (whether fabricated or authentic) were often the best assurance of recognition as true sovereign. Dynasties could be changed in premodern China, but never lightly or easily. Descent, more broadly, might also include ethnic descent, and there were indeed some explicit assertions made during (or, at least, later attributed to) this period that people with non-Han identities, such as Hu or Rong, could never be legitimate emperors. But such narrow ethnocentrism was flatly contradicted by Mandate of Heaven theory, which implied that whoever was worthy could be a legitimate ruler, and, just as ancient Zhou kings had legitimately replaced ancient Shang monarchs (even though Mencius once described King Wen of Zhou as “a man of the western Yi” [西夷之人也]Footnote 55), Hu or Rong people could arguably also now become perfectly legitimate emperors of what we call China.Footnote 56
In the end, though, claims to virtuous rule and the receipt of Heaven's Mandate could only be claims, and what was essential was to win acceptance of those claims by as wide an audience as possible. To overcome doubt, dynastic founders frequently invoked supernatural validation. After usurping the throne in a palace coup, for example, the first Sui emperor “feared the people were unsatisfied and spoke much of auspicious omens to dazzle them. Those he fabricated and advanced were beyond count” (恐黎元未愜, 多說符瑞以耀之. 其或造作而進者, 不可勝計).Footnote 57 It also became common to invoke theories concerning a proper dynastic place in the sequence of Five Phases (wu xing 五行) that supposedly naturally followed each other. After 491, for example, the Tuoba Xianbei rulers of the Northern Wei dynasty (386–534), in northern geographic China, claimed to represent the phase of “water,” legitimately succeeding the “metal” phase of the Western Jin dynasty, and thereby dismissing all of the intervening regimes (Eastern Jin and the Sixteen Kingdoms) as illegitimate.Footnote 58
There was also a literal seal of legitimacy, known as the “seal of state transmission” (chuan guo xi 傳國璽). This had supposedly been carved in jade on the order of the First Emperor of Qin, and then passed from Qin to Han, and from Han to Three Kingdoms Wei, then to Western Jin, and from 352–552 fell into the possession of the Southern dynasties, after which it was recovered by Northern Qi in the aftermath of a disastrous southern rebellion.Footnote 59 Whether or not there ever was an authentic actual seal, claims to such a seal could then be fabricated. For example, the Murong Xianbei ruler of the Former Yan dynasty (337–370), in northeastern geographic China, falsely claimed to have obtained the seal of state transmission when he proclaimed himself emperor of Zhongguo in 352.Footnote 60 In 446, the Northern Wei claimed to have recovered not one but two such seals from the ruins of a pagoda, one of which was inscribed as “the seal of state transmission that [Three Kingdoms] Wei received from Han” (魏所受漢傳國璽).Footnote 61
Much is sometimes made of ethnic non-Han claims to descent from legendary ancient Chinese cultural heroes, such as the Yellow Emperor, as a way of winning acceptance as legitimate rulers in this era.Footnote 62 In his grand history of the known world, however, Sima Qian 司馬遷 (145–ca. 90 BCE) had earlier traced nearly every “foreign” ethnic group he described—including both the Viets (Ch: Yue 越) in the south and the nomadic Xiongnu in the north—to similarly shared ancient Chinese ancestors.Footnote 63 In the Chinese imagination, all people (or at least their leaders) could be thought of as descending from common legendary ancestors,Footnote 64 and such claims made at this time on behalf of the Tuoba Xianbei and others were therefore maybe nothing really very remarkable. Nevertheless, the ability to imagine a shared legendary past undoubtedly did help facilitate forging a shared present and future reality as fellow subjects of the same state and civilization.
Rival Northern and Southern Chinas
The label “Han,” designating what we call ethnic Chinese, may have first begun to actually be applied during the Northern Wei dynasty, as a way of distinguishing the ruling Xianbei people from their conquered Han subjects. Shao-yun Yang has argued, however, that those same Xianbei may have nonetheless also identified themselves in a broader sense as Hua (Chinese), and as people of Zhongguo, defined in terms of civilization and geography rather than narrow ethnicity.Footnote 65 Satō Masaru 佐藤賢 speculates that the Tuoba Xianbei-ruled Northern Wei dynasty may have understood Zhongguo to include both the Central Plain of modern north China, which the Xianbei sometimes referred to as “Southern Xia” 南夏, and also the Xianbei homeland zone somewhat farther north, which they may have implicitly regarded as “Northern Xia” 北夏. The ethnic Han Chinese-ruled Southern dynasties, based in the Yangzi River drainage basin, could then be dismissed by the Northern Wei as merely a land of (non-Chinese/uncivilized) Yi and Di.Footnote 66
Conversely, of course, southerners sometimes dismissed the Northern Wei as a (non-Chinese/uncivilized) Hu regime, and claimed that the legitimate succession, and the Qin dynasty seal of state transmission, lay in the south.Footnote 67 Southern accounts condescendingly labeled northerners as “rope caitiffs” (suo lu 索虜)—a reference to the Tuoba Xianbei practice of wearing their hair in ropelike queues—while northern sources contemptuously called southerners “island Yi” (dao yi 島夷).Footnote 68 As political division stabilized somewhat into opposing Northern and Southern dynasties, both tried to depict themselves as being the sole legitimate line of (Chinese) dynastic succession.
During the Age of Division, the south long boasted a more flourishing Chinese-language literary culture. In the mid-sixth century it was claimed there were three times as many classic books in the south(east) as in the north (北于墳籍少于江東三分之一).Footnote 69 Even the greatest warlord of the northeast in the early sixth century, Gao Huan 高歡 (496–547), admitted that “gentlemen on the Central Plain view it [the Southern dynasty of his time] as where the orthodox calendar is located” (中原士大夫望之以爲正朔所在), that is, as the culturally legitimate dynasty.Footnote 70 The rulers of the Northern Wei dynasty initially issued orders in the Xianbei language rather than in Chinese, and, after becoming infected by Chinese customs, tried to preserve the Xianbei language by deliberately practicing it, calling it their “national language” (guo yu 國語).Footnote 71 After military disturbances and a split of the Northern Wei into opposing eastern and western regimes in 534, the warlord of the northeast, Gao Huan, still usually commanded his troops in the Xianbei language.Footnote 72 As late as 577, one northern emperor is explicitly described as speaking to officials in Xianbei.Footnote 73 In 549, a Wei edict ordered those Xianbei people who had taken Chinese surnames to revert to their old Xianbei names,Footnote 74 and in 554 the northwestern regime reorganized its military forces along the lines of what was purportedly (although it was probably really a new construct) the old Tuoba Xianbei tribal structure, extending Xianbei names to everyone who served in them.Footnote 75
But even an illiterate fourth-century ethnic Jie ruler of north China reportedly still had someone read the Dynastic History of the Former Han aloud to him, evincing an interest in Han dynasty history.Footnote 76 Even while the Northern Wei capital was still at its initial far northern location at Pingcheng 平城 (near modern Datong, in Shanxi), there were reportedly already copies of the Confucian Five Classics engraved on stone in the western outskirts of the city.Footnote 77 Despite the conscious preservation of a distinct Xianbei ruling identity in the north until nearly the end of the sixth century, the overwhelming majority of writing in both the Northern and Southern dynasties was always in the Chinese language. Even the emperor's speech in the Xianbei language in 577 that was mentioned in the preceding paragraph was actually a speech in praise of the exceptional Chinese-language writing abilities of a captured former enemy. Especially after a revival of Confucian learning in the north in the late fifth century, although north and south continued to take different approaches to scholarship, they ultimately “arrived at the same results from different directions” (殊方同致矣).Footnote 78
Throughout the Northern and Southern dynasties period there continued to be cross-border exchange, and the coming and going of people. For example, a member of an illustrious Southern dynasty family named Wang Su 王肅 (464–501) fled from a purge in the south to Northern Wei, where he helped to establish an official system “just like Zhongguo” (一如中國).Footnote 79 Beyond such interregional exchanges, moreover, there also clearly remained an enduring conception of a unitary Hua/Xia world based on the geographic memory of the Han dynasty empire.
The great Northern Wei geographer Li Daoyuan 酈道元 (d. 527), for example, included both northern and southern geographic China in his Annotated Classic of Rivers (Shui jing zhu 水經注), compiled around 520, suggesting that regardless of political division he still viewed the Chinese world as being roughly equivalent to the territory of the old Han dynasty.Footnote 80 As Andrew Chittick has demonstrated, even specifically local geographic writing during the Age of Division, while much concerned with documenting local customs and matters of local distinction, nonetheless overwhelmingly “shied away from defiant [regional] political posturing,” and was instead largely “devoted to cataloguing the evidence of classical and Han imperial culture in their home districts … . to show how local customs preserved and propagated a valid subset of an idealized imperial whole.”Footnote 81 Even advocates of decentralized rule and greater local autonomy (along the lines of ancient Zhou dynasty fengjian 封建) left unchallenged “the almost unquestioned ideal of a single legitimate sovereign ruling ‘All under Heaven’.”Footnote 82
Over the course of the Northern Wei dynasty, a number of the subordinate Hu tribes in the north were gradually transformed into ordinary tax-paying imperial subjects,Footnote 83 and, in the late fifth century, the emperor Xiaowen (r. 471–499) famously presided over a major sinicization program to deliberately make Northern Wei more of a Chinese-style state. He aspired to be remembered a thousand years later as being like the ancient (Chinese) Sage Kings Yao and Shun (千載之後, 我得髣像唐虞).Footnote 84 He “studied the ancient ways, followed the old rules, considered the former kings, and selected from their codes” (稽古, 率由舊則, 斟酌前王, 擇其令典).Footnote 85 Regarding his plan to relocate the capital south to the venerable ancient Chinese city of Luoyang, it was said: “The Yi and Luo [Rivers, in modern Henan] are a central area, which takes possession of all Under Heaven. Your Majesty is establishing imperial rule over Hua/Xia and making peace within the Nine Domains. When the people hear this they should celebrate greatly” (伊洛中區, 均天下所據, 陛下制御華夏, 輯平九服, 蒼生聞此, 應當大慶).Footnote 86 During the reign of this same ethnic Tuoba Xianbei emperor it is even believed that the first specifically Confucian temple ever erected in a capital city in Chinese history may have been constructed in 489.Footnote 87 As Katherine Tsiang observes, these sinicizing reforms may have been motivated “not in imitation of Chinese culture per se,” but rather primarily to recast the Northern Wei as heir to the legitimacy of antiquity and “as a successor to the great Han dynasty, invoking Han imperial precedent in its art, architecture, and court ritual.”Footnote 88
After the split of the Northern Wei into rival northwestern and northeastern regimes in 534, in 556 the northwestern regime (Western Wei, 534–557) implemented bureaucratic reforms based on an idealized vision of ancient Zhou dynasty institutions drawn from the Zhou li 周禮.Footnote 89 Although this proved to be a relatively short-lived and superficial archaism, the symbolic intent is nonetheless significant. The Zhou li was one of those classic texts, familiar to all members of the Chinese-language educated elite, which collectively formed an imaginary template that enabled the imperial model to “survive the demise of each of its incarnations” and replicate itself over and over again as a new (Chinese) dynasty.Footnote 90 Soon after these Zhou li-inspired reforms, there was a change of dynasties from Western Wei to Northern Zhou (557–581)—the new regime even invoking the ancient Zhou dynasty in its choice of dynastic names, and the regent who dominated the early years of Northern Zhou even explicitly comparing himself to the canonical figure of the ancient Duke of Zhou.Footnote 91
The second emperor of this Northern Zhou dynasty, Emperor Ming (r. 557–561), who was a XianbeiFootnote 92 with Chinese-language literary inclinations, compiled a genealogy (shi pu 世譜) in 500 juan 卷 stretching from Fuxi and Shennong in legendary Chinese antiquity down to the end of the Wei dynasty.Footnote 93 In other words, this Xianbei emperor was explicitly imagining the Xianbei-ruled Wei dynasties as lying in the direct line of legitimate Chinese historical succession. As Jack Chen has noted, this same emperor also wrote about listening to a song composed by the Han dynasty founder (“The Great Wind” 大風), implying his own imagined literary place in “a sort of orthodox lineage (zhengtong 正統) of sovereignty.”Footnote 94 This emperor's successor then notoriously purged Buddhism because it was foreign to China, unlike himself, reportedly even claiming (despite his Xianbei ancestry): “I am not one of the Five Hu” (朕非五胡).Footnote 95 In 579, Emperor Xuan of Northern Zhou (r. 578–580) and his court officials began dressing in the revived (supposedly Chinese-style) apparel of the Han dynasty and Three Kingdoms Wei, having previously apparently worn non-Chinese Hu-style clothing.Footnote 96
Imagining China
In 581, a palace coup brought a new dynasty, called Sui, to power in northern geographic China. Although its ruling Yang 楊 family had intermarried with the Xianbei elite and were culturally somewhat mixed, they are conventionally understood to have been of ethnic Han Chinese descent in the paternal line. Interestingly enough, the first Sui emperor's father had spent two extended periods of time (the first lasting for five years) living and serving in Southern dynasty Liang (502–557).Footnote 97 The family was therefore evidently capable of feeling at home in both the Northern and Southern dynasties. In early 581, even before he formally usurped the throne, the future first Sui emperor ordered persons who had previously been bestowed with Xianbei surnames to revert to Chinese names.Footnote 98 Soon thereafter, he ordered a reversion to the official standards of the Han and Three Kingdoms Wei dynasties.Footnote 99 That same year, he ordered a revision of the legal code, drawing upon precedents ranging from the Three Kingdoms Wei and Jin dynasties down to Southern dynasty Liang.Footnote 100 In 589, he successfully conquered the last Southern dynasty, and in 590 he ordered military men throughout the empire to be brought under regular civilian administration and compelled to register their households and begin farming “just like [ordinary] subjects” (一與民同), thereby helping to erase the distinction between Xianbei warriors and Han farmers and civilian officials that had until recently been so glaring.Footnote 101
This Sui dynasty “reunification of China” was in some ways a successful and permanent milestone. Never again would there be such a protracted Age of Division. Yet, the Sui dynasty itself only lasted two generations, and not only the south (conquered in 589) but also the northeast (which had been conquered by Northern Zhou in 577) long remained zones of military occupation facing at least potentially hostile local populations.Footnote 102 In the final defense of one strategic northeastern (Northern Qi dynasty) city against the Northern Zhou invasion, for example, it was reported that even young boys and girls threw bricks and stones from the tops of houses at their attackers.Footnote 103 People at the time did not necessarily welcome being “reunified” by military force. The Sui dynasty founder, moreover, did not always consistently support Confucian education, and portrayed himself as a Buddhist “Aśoka-like cakravartin sovereign,” and “Bodhisattva Son of Heaven” (pusa tianzi 菩薩天子), whose authority was more universal than a traditional Chinese-style monarch.Footnote 104
If the Sui dynasty reunified China, it was also in some ways more than merely what we think of today as China. Not content with reconquering what is now northern Vietnam, which had in fact been part of the old Han dynasty empire, in 605 the Sui also invaded Linyi 林邑 (Champa), in what is today central Vietnam—an utterly foreign land beyond the borders of any previous Chinese realm.Footnote 105 Sui armies also campaigned in the west, conquering and garrisoning the region that is now Qinghai.Footnote 106 And, most notoriously, the Sui launched a series of unsuccessful invasions of Koguryŏ, in what is today northern Korea and southeastern Manchuria. In 607, the second Sui emperor visited the Grand Qaghan of the Türks (Tujue 突厥) in his camp in what is now Inner Mongolia, where he discovered that an emissary from Koguryŏ was also present. The emissary from Koguryŏ was instructed to tell his king that he must come soon for an audience with the Sui emperor. When the king of Koguryŏ did not obey this Sui imperial summons, his insubordination supposedly instigated the massive series of Sui invasions.Footnote 107
Yet, if the Sui attempt to conquer northern Korea is an obvious example of the Sui dynasty not coinciding exactly with what we think of today as China, there is another sense in which they really were still operating within the field of traditional Chinese historical memory. The disastrous Sui invasions of Koguryŏ are conventionally blamed for the swift collapse of the Sui dynasty, and the founder of the subsequent Tang dynasty is said to have thereafter, more cautiously, considered not insisting that Koguryŏ acknowledge vassal status (何必受其臣).Footnote 108 As his contemporary, the dynastic historian of the Sui, pointedly observed, it had been a long time since the lands east of the Liao River (southeastern Manchuria and beyond) had actually been under imperial administration.Footnote 109 But the counterargument prevailed instead that not only had the putative Shang aristocrat Jizi 箕子 (Korean: Kija) supposedly founded a polity in the area of Koguryŏ in early Zhou dynasty times, and not only had it once been part of the Han dynasty empire, but, as a more general principle, “Zhongguo is to the Yi and Di as the sun is compared to the array of stars” (中國之於夷狄, 猶太陽之比列星).Footnote 110 That is, the Yi and Di (non-Chinese peoples like Koguryŏ) should properly be overshadowed (or, perhaps more precisely, be outshone by the radiance of, and illuminated) by Zhongguo. The Tang dynasty accordingly embarked on another series of attacks on Koguryŏ, which ultimately ended with a victory for Tang imperial armies operating in conjunction with Korean allies, and the unification of the Korean peninsula in 668—but then the swift expulsion of the Tang presence by 676, leaving Korea thereafter both unified and independent.
Imperial China was not a modern-style nation-state. Not only was there no concept of popular sovereignty, but the state was also not defined in ethno-national terms. To be sure, Nicolas Tackett has vigorously argued that an expectation “that ethnic, cultural, and political boundaries ought to align” had emerged already during the Song dynasty.Footnote 111 But even the later Ming dynasty (1368–1644) arguably remained a multi-ethnic empire outside of its Zhongguo core,Footnote 112 and the Mongol-ruled Yuan (1271–1368) and Manchu-ruled Qing dynasties were far from being ethnically defined Han Chinese nation-states. The Qing dynasty did, however, call itself Zhongguo (Manchu: Dulimbai gurun),Footnote 113 and effectively used the “discourse of Hua-yi and orthodox legitimacy (Ch. zhengtong), to construct a new identity” for itself as a “Chinese empire.”Footnote 114 Self-identification as Zhongguo (China) and Hua (Chinese) was always potentially separable from the somewhat narrower Han ethnicity, which itself had been an evolving synthesis from the beginning.
The existence of a repeatedly reconstituted centralized empire administered by officials sharing a common educational background (and expected to serve anywhere except their own native places) did generate a standardized imperial elite culture that eventually percolated down to and interpenetrated with non-elite popular culture. But even as late as the nineteenth century, Henrietta Harrison argues that the multiethnic, multicultural, and multilingual nature of the Qing dynasty meant that China was not, after all, an exception to the general proposition that nation-states are modern inventions.Footnote 115 In fact, the pluralistic legacy of the Qing dynasty arguably contributed to an alternative vision of a multiethnic modern Chinese national identity that continues today to challenge more narrowly ethnic Han Chinese nationalism.Footnote 116
If premodern China was not a modern-style nation-state, hyperbolic claims that the emperor literally ruled “all Under Heaven” as a truly universal sovereign were usually not taken entirely seriously either. Instead, the Chinese world was commonly conceived as an series of concentric zones, centering around the royal demesne (wang ji 王畿), and becoming progressively more alien, and untamed, with distance.Footnote 117 As one early Tang dynasty source put it, “Zhongguo is the root of all Under Heaven, and the Four Yi are like branches and leaves” (中國, 天下本根, 四夷猶枝葉也).Footnote 118 All the world might be interconnected, but beyond a certain point it was no longer Zhongguo. Thus, at the peak of Tang dynasty imperial power and expansive universalistic pretentions it was claimed that Tang authority reached all the way to Persia (波斯),Footnote 119 but actual administration of the far west was largely delegated to native vassal rulers.
In addition to the case of Koguryŏ described above, another interesting example involves the Tang handling of the Türks who had submitted after the Tang defeat of the Eastern Türk Qaghanate in 630. “Many gentlemen at court said” (朝士多言) that these Türk tribes should be divided up and scattered under Tang administration south of the Yellow River, where they should be put to work farming, so that “their customs would be changed, and a million Hu caitiffs could be transformed into Han [Chinese]” (變其風俗. 百萬胡虜, 可得化而爲漢).Footnote 120 This nicely illustrates the kind of open inclusivity that was often characteristic of early Chinese identity. Tang dynasty policy, for example, was that the children of submitted foreigners would become ordinary Tang subjects (凡內附後所生子即同百姓不得爲蕃戶也).Footnote 121 It may not have been entirely inconceivable during the Tang dynasty to imagine that even Türk nomads could potentially become Han Chinese.
But, of course, non-Han Chinese peoples might not wish to be assimilated, and forced assimilation is regarded by many observers today with something approaching horror. (Though a counter-argument could perhaps be made in favor of an ideal of tolerant inclusivity and a rejection of tribalism.) In any case, an alternative proposal that was suggested for dealing with the submitted Türks in 630 was that they should instead be driven far away, so that they would cease to be a threat, and the proposal that was actually adopted, instead of either of these other two, was to follow the old Han dynasty precedent in dealing with the Xiongnu and settle the Türk tribes intact on the northern frontier to serve as a defensive buffer. In other words, what the Tang dynasty actually did was to consciously follow an explicit Han dynasty model.
Tang dynasty China was in many ways quite different from Han dynasty China, and there were also some senses in which the early Tang, like the Sui, was more than what we might think of as merely “China.” After defeating the eastern Türks in 630, for example, the Tang emperor was famously hailed as “Heavenly Qaghan” (Tian kehan 天可汗) of the steppes as well as emperor of China, and it was asserted that Tang “control over the Four Yi began with this” (統制四夷, 自此始也).Footnote 122 At an extreme, it could even be argued that the Tang dynasty was not an old Han dynasty-style “Han Chinese” state, but rather another Tuoba Xianbei conquest empire, differing from the later Mongol and Manchu empires only in its degree of sinicization.Footnote 123 As we have seen, however, Chinese identity had been a synthesis from the start, which was perfectly capable of elastically expanding now to include new groups such as the Tuoba Xianbei.
The early Tang dynasty was famously cosmopolitan, and its emperors were at least partly of mixed descent. But the Tang was also based in geographic China, used the Chinese written language, openly laid claim to the heritage of Chinese history and civilization, and called itself Zhongguo. At least in the core regions of China Proper, the Tang also enjoyed the same kind of framework of shared cultural, political, and economic institutions and networks that Xu Zhuoyun has observed as promoting a common identity under the earlier Han dynasty. As Marc Abramson concludes, if the Tang began as “a self-consciously multi-ethnic empire,” it ended three centuries later as a much more thoroughly “mono-ethnic empire united by a significantly more homogenous Chinese culture.”Footnote 124 This greater degree of homogeneity may have then set the stage for the allegedly incipient proto-Han Chinese nationalism of the subsequent Song dynasty.
Rupert Emerson has identified some of the frequently reoccurring components of modern national identity formation as including a geographic homeland, a common language, a shared historical tradition, “and the intricate interconnectedness of state and nation.”Footnote 125 Imperial China was not a modern-style nation-state, but there may have been some similarities with regard to identity formation. The Qin and Han, and now the Sui and Tang empires, had effectively staked out the approximate territory of what we think of as geographic China Proper, and established the model of the Chinese imperial state. A single common written language enjoyed a near monopoly of writing (although there was considerable local variation in spoken language, and the same written language was also used by non-Chinese states and peoples, including Korea, Japan, and Vietnam; the linkage between language and state identity was not absolute). And, although the turmoil of the fourth century may have temporarily caused an identity crisis and a challenge to Chinese commonalities, the historical memory of antiquity and of the Han dynasty remained strong. As the anecdote rehearsed at the beginning of this article demonstrated, there was a widely shared historical memory among the educated elite in geographic China, even during times of prolonged political division and ethno-cultural diversity.
An astonishing eight of the total twenty-four Chinese dynastic histories were compiled within a single generation, either officially or unofficially, at the early Tang court.Footnote 126 The early Tang dynasty literally wrote (or, more precisely, edited) much of what we think of as Chinese history, and clearly identified it with the entirety of the geographic space we consider to be China Proper—for example, by including the ethnic non-Han regimes of the fourth-century north in the Jin shu (though not as the legitimate sovereign dynasties), and including both the separate lines of Northern and Southern dynasties. These histories were mostly edited and copied from older texts rather than being entirely new compositions, and, despite undoubted imperfections,Footnote 127 they form part of what is surely the world's finest and most complete premodern historical record. Indeed, to some extent it is the very fullness of China's historical record that creates the appearance of a permanently enduring “China.” Through the compilation of histories (and other activities), the Tang dynasty successfully depicted itself as heir to the whole of geographic China, and all of Chinese history stretching back into legendary antiquity.
The Tang located its capital near the site of the early Han dynasty capital, and revived the old Han dynasty name for it (Chang'an) in “a self-conscious act linking the T'ang to the last great Chinese empire of Han.”Footnote 128 The second Tang emperor (Taizong, r. 626–649) is said to have taken the Han dynasty founder as a model, although he even more especially preferred to associate himself with the memory of the Sage Kings of antiquity.Footnote 129 As Taizong allegedly professed, “what I like now is only the Dao of [the Sage Kings] Yao and Shun, and the teachings of the Duke of Zhou and Confucius” (朕今所好者, 惟在堯舜之道, 周孔之教).Footnote 130 After 657, as Howard Wechsler has pointed out, the Tang dynasty maintained regular sacrifices to the spirits of the ancient Sage Kings, the Zhou dynasty founders, and the first emperor of the Han dynasty (among others), asserting a Tang place in the legitimate line of Chinese political succession.Footnote 131
Grand historiographical debates over the precise sequence of legitimate dynasties (the so-called zhengtong lun 正統論) date mostly from the Song period, but the idea of an orthodox line of succession was older. For example, during the Han dynasty, Ban Biao 班彪 (3–54 CE) asserted that the Han imperial family (surnamed Liu 劉) was a continuation of the reign of the Sage King Yao 劉氏承堯之祚, and at the end of the Han dynasty Fu Gan 傅幹 (late second–early third centuries) concurred that the Han founder had inherited the rule of Yao (高祖襲唐之統), and credited the successful Later Han dynasty restoration (25–220 CE) as due to it being the “orthodox lineage of emperors” (帝皇之正統).Footnote 132
In terms of cultural rather than strictly political legitimacy, the most famous formulations of this orthodox line of succession idea were surely articulated in the late Tang dynasty by Han Yu 韓愈 (768–824). Of the Dao and the teachings of the ancient Sages, Han Yu wrote:
Yao transmitted them to Shun. Shun transmitted them to Yu. Yu transmitted them to Tang [the founder of the Shang dynasty]. Tang transmitted them to Wen and Wu [the iconic early kings of the Zhou dynasty]. Wen and Wu transmitted them to the Duke of Zhou and Confucius. They were written into books, and the people of Zhongguo preserved them for generations.
堯以是傳之舜. 舜以是傳之禹. 禹以是傳之湯. 湯以是傳之文武. 文武以是傳之周公孔子. 書 之 於册. 中國之人世守之.Footnote 133
Han Yu's storyline may have been rooted more in legendary imagination than in verifiable historical fact, the orthodox transmission of the Confucian Dao may have been later challenged by Qin Legalism and Buddhism, and imperial unity disrupted during the centuries of division after the fall of the Han dynasty, but (Chinese) civilization was imagined as a single line of direct descent, and confidently reclaimed as their heritage by people like Han Yu.
Conclusion
Continuing identification with antiquity and the Han imperial model may be an example of the approach to linear national history critiqued by Prasenjit Duara, through which “dispersed” historical narratives are appropriated and reorganized (often, perhaps, unconsciously) “according to present needs, thus revealing how the present shapes the past,” while, “by linking new events back to an animating story, this story is extended into the present which it seeks to shape.”Footnote 134 The early Tang compilation of dynastic histories reinforced a storyline that established the Tang dynasty as the legitimate heir to all of previous Chinese history. Such an editorial agenda did not necessarily preclude the possibility, however, of those histories remaining reasonably factually accurate, especially in contrast to the generally inadequate histories that are available from elsewhere in the world in the third to sixth centuries. And, if the Tang dynasty brought the compilation of history under state control and used it to “shape the past in such a way as to make it a guideline for the future,” it was not so much the newly edited histories of the relatively recent past as it was the idealized memories of remote antiquity and of the Han empire that were most reverently invoked.Footnote 135
Today, the official name of the country English speakers call China—both the People's Republic and the Republic—is “Central Hua” (Zhonghua 中華). A Mongol-era commentary to the Tang dynasty legal code had earlier defined “Central Hua,” however, essentially as civilization itself: the land, intimate with the teachings of kings, where “clothing and caps are dignified, customs are filial and fraternal, and [people are] disposed to propriety and justice, therefore we call it Central Hua, unlike the customs of the distant Yi and Di” (衣冠威儀, 習俗孝悌, 居身禮義, 故謂之中華, 非同遠夷狄之俗).Footnote 136 Rather than being a specific state bearing a permanently fixed name, premodern China was actually a multiplicity of separately named dynastic monarchies, sometimes ruled by differently named peoples such as the ancient Zhou, the Tuoba Xianbei, and the Manchus. Yet, despite all its complexity, diversity, and recurrent disruptions, Zhongguo is popularly imagined to have been a single more-or-less continuous identity reaching back thousands of years.
This study suggests that the Age of Division following the Han dynasty may have simultaneously been less of an aberration in Chinese history than is commonly assumed (the Tuoba Xianbei may not have been so very different from the later Manchus, for example, and what we think of as China has been a pluralistic synthesis from the beginning), and, at the same time, also really the single greatest period of crisis and challenge to the survival of a Chinese identity prior, at least, to modern times. Not only was the geographic space we call China politically divided, and not only was there much ethno-cultural and linguistic diversity, but also the borders between “China” and “not China” were porous and unclear. But the continuity provided by a single written language, standardized institutions, and remarkably vivid elite memories of the Han dynasty and antiquity all kept a sense of Chinese identity alive throughout four hundred years of division. After the Sui and Tang dynasties militarily restored the approximate contours of the old Han dynasty empire, it proved relatively easy for them to successfully depict themselves as heirs to a unitary China rooted in ancient memory
To suggest even a rough similarity to Benedict Anderson's influential depiction of modern nation-states as “imagined communities,” however, does not mean that such imagined identities—whether modern or premodern—are necessarily false, or mere phantasies. All human identities, after all, must be imagined, and they can be very powerful. The premodern sense of Chinese identity was nurtured through a prestigious body of shared texts and orally transmitted stories and legends, as well as through a broadly common language and culture, and it lived on in generations of human hearts and minds. There was no primordial and exclusive Chinese “race,” kingdoms and empires repeatedly rose and fell, the territory that those kingdoms and empires controlled expanded and contracted, and the details of culture evolved and changed over time, but a Chinese identity has proved enduring, even in an age of division.
