Hostname: page-component-546b4f848f-hhr79 Total loading time: 0 Render date: 2023-06-02T10:11:37.146Z Has data issue: false Feature Flags: { "useRatesEcommerce": true } hasContentIssue false

“Our Roots Are the Same”: Hegemony and Power in Narratives of Chinese Linguistic Antiquity, 1900–1949

Published online by Cambridge University Press:  21 October 2022

Gina Anne Tam*
History, Trinity University, San Antonio, Texas, USA
Corresponding author. Email:
Rights & Permissions[Opens in a new window]


Since the beginning of the twentieth century, a frequent claim among speakers of local Chinese languages (called fangyan in Chinese) is that their native languages preserve the language of antiquity better than the Beijing-based national language, Mandarin. This paper explores the origin of these claims and probes their significance in the making of the Han ethnoracial collective identity. I argue that claims of linguistic proximity to the imagined ancient origins of Chinese civilization represent a form of “hegemonic Han-ness”—an idealized form of the Han collective identity that was both internally hegemonic, in that it was meant to supersede other expressions of Han-ness, and externally hegemonic, in that it was meant to uphold the superiority of the Han people over other ethnoracial groups. From Zhang Taiyan, whose work provided a model for drawing linguistic connections between contemporary local languages and the language spoken at the dawn of Chinese civilization, to local gazetteer authors, who used linguistic data to prove their mother tongues directly had preserved the language of antiquity without being adulterated by the languages of non-Han peoples, this paper explores how various groups drew upon the cultural power of an idealized Han-centered past to challenge the authority afforded to the national language by the state.

Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (, which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited.
© The Author(s), 2022. Published by Cambridge University Press on behalf of the Society for the Comparative Study of Society and History


I am often told that Cantonese is the oldest Chinese language. In casual conversations, interlocutors tell me that Cantonese is much closer to the language of the Tang dynasty (618 CE–907 CE) than the current Chinese national language, Putonghua (Mandarin in English). Read the poems, study the rime tables, they insist, and I will see that the linguistic origin of the Chinese language came not from Beijing but from the southern coasts. As this common refrain is articulated by South China Morning Post editorialist Alex Lo, “Cantonese has a much longer and venerable lineage than Putonghua (普通话) [Mandarin].… The ancient canonical Chinese texts predated Putonghua, but not necessarily Cantonese.”Footnote 1

Cantonese is not the only local Chinese language to which such claims about linguistic antiquity are applied. Enthusiasts make similar claims about local languages across China, including those of Hubei, Fujian, and Sichuan, and this list is by no means exhaustive.Footnote 2 Today, while the exact natures of these popular arguments vary slightly, they share several attributes.Footnote 3 First, they are made by speakers of Hanyu fangyan (漢語方言), a term usually translated as “Chinese dialects,” in which the “Chinese” refers not to the national geobody (Zhongguo 中國) but the Han ethnoracial identity, an identity often referred to in English as “ethnically Chinese.”Footnote 4 Second, these claims are almost always comparative. Not only are these languages described as old, but they are purported to have a “more venerable” connection to antiquity than do other Chinese languages, especially the national language. Third, and most importantly, these claims never concern only the particularities of historical phonologies: To be the inheritors of the language that hews closest to the origin of a civilization is to similarly inherit a rightful claim to represent its history. It is cultural legacies, not morphemes, that peddlers of these narratives seek to reclaim.

These claims reveal a profound contradiction at the heart of the Han identity as it was reimagined within the context of Chinese nation-building. As one of the world’s largest collective identities, it is deemed by the government of the People’s Republic of China (PRC) the nation’s majority minzu (民族), a term we normally translate as “ethnic group.” With such a designation, the Han identity is presumed to have the same inherent unity as other similarly designated collectives.Footnote 5 But among those who identify as Han, there is remarkable diversity, including innumerable unique cultural practices and several distinct, mutually unintelligible, language groups.Footnote 6 Today, this contradiction is often masked by a veil of unifying rhetoric, as various actors, including the PRC state, emphasize certain characteristics as representative of Han identity—a singular linear Han history, philosophical traditions like Confucianism, or holidays like Chinese New Year.Footnote 7

Chief among these characteristics is language. Since the early twentieth century, the Chinese national tongue has been upheld as a core attribute of the Han identity’s rhetorical unity. Based on the language commonly spoken in Beijing, it was named the national language (Guoyu 國語) in the 1920s and, in the 1950s, adopted and renamed by the PRC government “the common language” (Putonghua). Deemed today the representative language of both the Chinese nation and its majority ethnoracial group, Mandarin serves as a blunt instrument of power in two ways. On one hand, it represents the superiority of Han people over other ethnic groups. It is forced upon and used to measure the relative “assimilation” and “civilization” of indigenous ethnicities like Tibetans, Uighurs, and Mongols, making it a powerful tool in their marginalization.Footnote 8 On the other hand, Mandarin is normalized as the titular “Chinese language” and given near-universal institutional support, making it a powerful mechanism in the diminishment of other non-Mandarin Sinitic languages.Footnote 9

Certainly, efforts to enforce a standard oral Chinese language, and rhetorical pushback against those efforts that relied upon harnessing the cultural power of antiquity, pre-date modern nation-building. In the early Qing dynasty, emperors tried to force the country’s educated official class to learn a standard northern-based oral language; speakers of southern fangyan balked, emphasizing their own languages’ phonological proximity to esteemed Tang-dynasty texts to justify their opposition.Footnote 10 But in the context of the twentieth century, these claims that certain fangyan are more ancient than Mandarin reveal efforts to challenge the prevailing presumption that state sponsorship matters more than historic legacies in defining the Han ethnoracial identity. The popularity of these claims questioning the hegemony of the national standard raises fundamental questions: How do particular traits of Han empirical diversity become part of, representative of, or erased from the term’s rhetorical unity and why? And more importantly, how do these competing expressions of hegemonic identity influence, take away from, or reinforce the relative power of the Han identity within the nation as a whole?

To answer these questions, it is helpful to look at theoretical frameworks outside of China. In particular, I take inspiration from studies on hegemony within collective identity, the best known of which is Raewyn Connell’s “hegemonic masculinity.” Taking aim at the presumed uniformity of the category “male,” Connell highlights the coexistence of multiple masculinities, defining hegemonic masculinity as dominant expressions of masculinity that hold discursive and institutional power.Footnote 11 As Connell reminds us, hegemonic masculinity is both “internally” and “externally” hegemonic—it is hegemonic over femininity and other genders, but also over non-dominant expressions of masculinity. Since Connell coined the term, others have extended her ideas to better understand intra-group power dynamics. One application is in studies of “hegemonic whiteness,” which explore the shifting, contextual, boundaries of “idealized modes of whiteness” through expressions such as foodways, popular culture, leisure, and, significant for this paper, language.Footnote 12 These studies remind us that there were and are multiple expressions of whiteness that co-exist with a presumed rhetorical unity, and in the maintenance of that rhetorical unity, certain expressions gain hegemony over others. More importantly, expressions of whiteness that gain ideal status do so because they buttress white supremacy over nonwhite groups. Internally hegemonic constructions, in other words, become internally hegemonic because they best uphold external hegemony.

These studies also underline the coexistence of competing hegemonic expressions of one identity. Take Nelson Flores’s study of bilingual education in the United States.Footnote 13 Flores argues that in the wake of the Civil Rights Era, antiracist groups pushed for a deemphasis of monolingualism in formal education, a relic of white supremacist colonialism that marginalized nonwhite bilingual students. Yet the benefits of the 1968 Bilingual Education Act designed to address these inequities were largely reaped by white Americans who learned a second language rather than bilingual Latino students, who continued to experience marginalization. As a result, Flores argues, two visions of hegemonic whiteness—bilingual and monolingual—emerged. And though their coexistence reveals an intra-white power struggle, they both nonetheless furthered the same outcome: the reinforcement of external hegemonic whiteness.

There are, of course, significant differences between these examples and that of Han-ness. For one, these studies focus on identities that are universally externally hegemonic, including whiteness, masculinity, or heterosexuality. Han-ness is not. Because Han people were subjects of imperial violence in the modern period, their power was historically, and remains still, significantly more contingent, especially outside Han-majority countries. Nonetheless, these studies offer an insightful lens through which to examine the significance of battles over representation within the Han ethnoracial group. They encourage us to move beyond merely noting the existence of diversity within the Han identity and focus instead on the issues of “hierarchy and power,” in Kevin Carrico’s words, inherent to its modern construction.Footnote 14

This paper draws upon these studies’ hegemonic identities to probe how proximity to linguistic antiquity became a competing trait of “hegemonic Han-ness”—defined as an expression of Han-ness that is discursively idealized and hegemonic—that challenged the hegemony of the national language in the early twentieth century. It does so by tracing how speakers of non-Mandarin Sinitic languages claimed they preserved the nation’s ancient history better than the state-sponsored national language. I argue that these counternarrative claims had a dual effect. First, they offered a competing vision of hegemonic Han-ness, one in which connection to the imagined origins of Chinese civilization proffered a more powerful claim to representing the nation than state sponsorship. Second, because these counternarrative claims relied upon a national history that was frequently used to validate the ethnoracial superiority of the Han people, they also reinforced, and were reinforced by, external Han hegemony.

This paper explores the relationship between language and hegemonic Han-ness in two sections. I begin in the late Qing, when a faction of Chinese elites, faced with the daunting task of constructing a new nation able to withstand the threat of imperialism, sought to reclaim political and cultural power for the country’s Han majority. Focusing on prominent revolutionary Zhang Taiyan, I show, first, how Zhang upheld an idealized antiquity as the source of Han ethnoracial supremacy and inspiration for an idealized Chinese nation, and second, how his research in philology laid the groundwork for later generations to center contemporary fangyan in the search for the origin of that antiquity. This section ends by tracing the impact of Zhang’s research on continuing discussions about language and Han-ness in the face of the state-sponsored promotion of a national language. The second section examines how speakers of non-Mandarin fangyan directly and indirectly capitalized on Zhang Taiyan’s research to claim a uniquely direct connection between their own local languages and an imagined Chinese antiquity after being marginalized by state language policy. In so doing, I show how local elites’ emphasis on linguistic proximity to antiquity challenged the hegemony of the national language but also, simultaneously, reinforced the broader project of Han supremacy.

National Language and Linguistic Antiquity

The question of Han hegemony in the late-Qing dynasty (1644–1911) was complicated.Footnote 15 Han people did constitute the Qing’s population majority and most of its intellectual, political, and economic elite. The Qing’s imperial examination system—itself the replication of long-established Chinese institution—exalted the philosophical and historical traditions of Chinese dynasties past. Most people, if they learned to read, learned Sinitic script, which served as the common written language for most of the empire’s proceedings. And yet, the Qing imperial family was not Han, but ethnically Manchu. It was the Qing rulers who had the power to dictate policy and decide whether to institutionalize Chinese cultural touchstones through education and civil service, and they could also install Manchu men in prominent government and military positions.

The problem of Han hegemony came to the forefront in the late nineteenth century, as the empire’s intellectual class became convinced that Manchu leadership was steering the Qing toward collapse. The empire’s defeat in multiple wars, culminating in the loss of territory, infrastructure, funds, and global power, led Chinese intellectuals to propose increasingly radical reforms. These practical proposals also supported a deeper kind of creative construction: new imaginings of what a modern Chinese nation-state, prepared to enter and survive in a world of equally sovereign nation-states, might look like. And while there were certainly debates among reformers about the role of Manchus within this newly imagined nation, few questioned that its cultural and political institutions would be culturally Chinese.Footnote 16

There similarly seemed to be little doubt that any standardized Chinese national language would be a Sinitic language. Faced with the fact that Han people spoke multiple, mutually unintelligible languages, many reformers—looking both at earlier Qing policies that required officials to learn an official pronunciation (guanhua 官話) based on northern speech, as well as examples of places like Japan and France that based their national standard on the common language of their respective capitals—suggested instituting the majority language of Beijing as the national standard. Others felt that Beijing was a particularly poor linguistic representative of a nation ruled by Han people because the city was too linguistically, culturally and politically associated with the Manchus. More than that, they feared that choosing just one existing language to represent the collective whole would alienate most of the nation’s Han people from the nationalizing project. Any standard language, these dissenters believed, should represent the nation’s Han majority in its entirety.

A prominent advocate of this latter position was Zhang Taiyan, one of early twentieth-century China’s most influential thinkers. Zhang is notoriously difficult for scholars to categorize, in part because his work was broad, multivalent, and complex. While famed as an anti-Manchu revolutionary, he was simultaneously a prolific scholar of classical Chinese history and philology, an ardent advocate for the restoration of an idealized Chinese past, and a strident critic of global imperialism and the modernity that imperialism defined.Footnote 17 But the difficulty also emanates from scholars’ own overly simplistic analytical categories. Scholars of late-Qing China have a propensity to “take for granted” a series of dichotomies in our categorization of key figures—traditional or modern, revolutionary or conservative.Footnote 18 But in Zhang’s work these frameworks were often more complementary than oppositional. Rather than presume that Zhang was simply a revolutionary because he advocated for the end of Qing rule, or a conservative because he advocated for the revival of a traditional past, we ought to, as Viren Murthy argues, see these dual positions as integrated parts of a coherent ideology: a desire to “reviv[e] the living nucleus of tradition” in the name of an alternative modernity grounded in the rejection of Western imperialism.Footnote 19

Zhang’s philosophical, political, and cultural contributions to late-Qing intellectual history are too numerous to recount here. One of his most influential ideas, though, was that the China nation was, and should be, defined by the history and culture of the Han historic race (lishi minzu 歷史民族).Footnote 20 His advocacy began after the Boxer Uprising (1900), when Zhang, formerly a reformer, began to call for the Qing’s collapse. He reinforced this political position with a Han ethnoracial nationalism built upon virulent anti-Manchuism, emphasizing that the stark embodied differences between the ruling Manchus and the majority Han population required a revolutionary overthrow. In a 1901 essay, “On the Correct Hatred of the Manchus,” Zhang called the Manchus “a different zu (族 ethnoracial lineage),” “mean and debased,” and ignorant of “governmental affairs [as well as] agriculture and commerce.”Footnote 21 As such, Zhang reasoned, they were unfit for political power. Mobilizing the logic of Social Darwinism, he warned that a nation governed by the Manchus, a less-evolved race, was in peril. Han political hegemony was the “key” to the nation’s evolutionary survival.”Footnote 22

After suffering a particularly traumatic prison stay from 1903–1906, Zhang’s anti-Manchu rhetoric underwent a notable shift. Whereas before 1903 he focused on how the Han people were the more evolved group, Zhang soon began criticizing the idea of historical evolution altogether.Footnote 23 Instead, he emphasized, a true revolution for the Han people would not be achieved by advocating for modernity for modernity’s sake, but rather, by “restoring” the political systems and power dynamics of antiquity. And for him, that meant the restoration of the “Chinese race … Chinese states and counties … and Chinese political power” (Zhongguo zhi zhongzu … Zhongguo zhi zhouzun … Zhongguo zhi zhengquan 中國之種族…中國之州郡…中國之政權).Footnote 24 By stressing restoration over evolution, Zhang reconciled his imagined Han-centered nation with a broader commitment to anti-imperialism in all its forms.Footnote 25

Zhang supported his goal of cultural restoration with a reinvented national history.Footnote 26 While Zhang was not as prolific a historian as other fellow restoration advocates, he frequently used historical narrative to advocate for the political dominance of the Han ethnoracial lineage.Footnote 27 His national history began with the mythical Yellow Emperor, the first ancestor of the Chinese people, and the civilization his descendants, the “Xia clan” (Zhongxia zhi shizu (中夏之氏族), built in the “Central Plains” (Zhongyuan (中原)).Footnote 28 Zhang’s history continued with a sequence of achievements by historical figures who represented his idealized past, from Zhou dynasty philosophers to Ming restorationists.Footnote 29 And while Zhang hardly celebrated all descendants of the Xia clan, non-Han groups were always portrayed as fundamentally different and often as inferior. In addition to his dehumanization of Manchu people, Zhang also referred to Mongols, Tibetans, and Turkic Muslims as “less evolved” or “less than human.”Footnote 30 His later writings shifted, targeting culture rather than biology as the basis for this imagined hierarchy; in some cases, he even maintained the possibility that non-Han groups might eventually be assimilated. Yet these shifts did not alter Zhang’s fundamental conclusion that Han people were historically and presently entitled to political and cultural hegemony. In other words, Zhang was consistent in his insistence that history proved the Han ethnoracial lineage distinct, and while he did not always imagine the boundaries between Han and non-Han peoples as impenetrable, he did always uphold that the very existence of those boundaries ought to translate into unequal access to political power.Footnote 31

Zhang Taiyan’s rhetorical shift from evolution to restoration coincided with his earnest pursuit of philological research. Innovatively combining the methods of philologists from the High Qing with those of scholars in Meiji Japan and Europe, Zhang penned a series of studies focused upon uncovering the origin of Sinitic languages and script and tracing their evolution.Footnote 32 One of his most extensive works on this subject was New Fangyan (新方言Xin Fangyan), named in reference to the first century BC compendium of regional phrases, Fangyan (方言) by Yang Xiong.Footnote 33 This work, first published as a collection of articles between 1907–1908, traced the etymology of several hundred contemporary regional phrases back to the Han dynasty, in particular to the third-century BCE dictionary Erya (爾雅), Xu Shen’s second-century dictionary of characters, Shuowen Jiezi (說文解字), and the aforementioned Fangyan. Put simply, Zhang’s primary goal was to recover the phonological and semantic connections between contemporary phrases and characters from antiquity, using this to both uncover the origins of modern expressions and their evolution through history.

Zhang’s methodologies were not new; he drew heavily on the work of High Qing philologists.Footnote 34 Scholars like Dai Zhen, Duan Yucai, and Qian Daxin, men who dedicated their lives to understanding the true meanings of the texts they took as philosophical and moral authorities, predicated their study of the classics on the presumption that understanding the Sinitic script used to write those texts required knowledge of the historic languages upon which that script was based.Footnote 35 As such, they held that a reconstruction of the dialectical process by which both a character’s phonology and its meaning evolved and “stretched” over time could uncover each character’s “original meaning” (本義) and unlock the texts’ true substance. This contention, Zhang explained in his introduction, inspired his own methodology.Footnote 36 At the same time, he connected his work to his own political convictions in a way that was specifically modern. In his introduction, Zhang criticized his intellectual predecessors for ignoring the value of contemporary fangyan, which he believed acted as “fossils,” in the words of Lin Shaoyang, that preserved key information about the “origin of language” and the Chinese ancient civilization that he saw as a model for his restorative politics.Footnote 37 As he explained in a later work, “The spirit (性) of the Zhongxia are preserved in their [spoken] languages,” a quote that shows how his philological research had clear implications for his modern nationalism.Footnote 38

Zhang’s work emphasized the evolution of the “phonology of antiquity” (guyin 古音) into present-day fangyan. Footnote 39 Yet he did not view all fangyan as equal descendants. For Zhang, the fangyan that had strongest connection to earlier periods of Chinese history were spoken in the south.Footnote 40 In a 1904 addendum to his Book of Urgency, Zhang laid out why, in his words, “only the southern pronunciation(s) evolved perfectly.”Footnote 41 In the centuries between the reign of the Yellow Emperor and the Zhou dynasty (1046 BC–221 BC), Zhang wrote, the heart of the Xia clan moved from the Central Plains to the southern state of Chu, near present-day Hubei and Hunan.Footnote 42 As he explained, “The language from the Spring and Autumn period stretched from the Song dynasty (960–1279) until today. But its roots remained in the state of Chu.”Footnote 43 As for those populations who remained in the north, their languages had “broken” with the language of antiquity by mixing with the non-Han polities of the borderlands.Footnote 44 By focusing on linguistic mixing with non-Han groups, Zhang implicitly connected his diachronic study of language to his beliefs about Han superiority, a connection that an afterword to New Fangyan written by Liu Shipei made explicit: “As the Hu, Jie, Di, and Qiang moved to live in Zhongxia, there were foreign sounds [in our language]. Then after the chaos of the Mongol [invasion] and Jianzhou [the Manchu invasion of 1408], our customs were replaced … if we want to expel foreign language from our Xia sounds, this book must be our forerunner.”Footnote 45 Liu’s afterword directly tied Zhang’s philological work to his broader ideas about history and politics, helping to provide a basis for an intra-Han hierarchy that hinged on proximity to an imagined antiquity.

The intra-Han hierarchy Zhang implied was not significant only in the abstract—it also had concrete applications. While he was writing New Fangyan, other reformers had begun earnest discussions about Chinese language and script reform. Zhang played a prominent role in these debates, often pushing back against proposals that, in his mind, were antithetical to the kind of Chinese society he sought to restore. He saved his most ardent vitriol for script reform.Footnote 46 Blasting increasingly popular proposals among some Chinese reformers to replace China’s script with a Romanized system, Zhang claimed that the nation’s rich history was evidenced in Chinese characters.Footnote 47 He railed, “Malaysians to the south and the Mongols to the north had phonetic scripts—was their culture truly superior to China’s?”Footnote 48 Zhang similarly opposed efforts to promote a standardized prose based upon the “congruence of speech and writing” (言文一致), and instead touted the value of teaching various genres of literature, including vernacular prose and classical forms,Footnote 49 and denounced proposals to replace Chinese languages altogether with Esperanto.Footnote 50

But the clearest application of New Fangyan was in regard to standard oral language reform. Zhang was not entirely opposed to instituting a national standard. Rather, in writings that nodded to his work in New Fangyan, Zhang opposed the mainstream proposal to base the national language the majority language in Beijing, arguing instead that a unified national pronunciation should reflect the imagined past he praised. While he upheld the contemporary languages spoken in a central region near Hubei as the “most correct,” he was wary of simply picking an existing language as the national standard, warning “north and south, [all languages] have divergences.”Footnote 51 Instead, Zhang advocated for a new invention: a language reconstructed from philological research to mirror the “correct language” of antiquity. As for the majority language of Beijing, he was resolute: “The correct language is that of Xia and Chu … it cannot be the language of Beijing.”Footnote 52

After the Qing fell, Zhang continued to play a prominent role in China’s intellectual life until his death in 1936. Yet his late-Qing writings remained the most impactful. On the topic of language, Zhang’s writings took on a life of their own. In policy-making circles, his ideas served as a cornerstone of debates over the new Republic’s language policy. Within the scholarly realm, New Fangyan played an influential role in the burgeoning fields of historical linguistics and dialectology. But most importantly, Zhang’s idea about language, national culture, and ethnoracial purity filtered down, cementing in the national imagination the idea that a Han-centered antiquity should serve as the mainstay of the new nation.

The Lasting Legacies of Zhang Taiyan

The 1911 revolution imbued Zhang’s ideas about how language bestowed political and cultural power with new meaning. The new Republican Ministry of Education, helmed by future Peking University President Cai Yuanpei, immediately began spearheading a series of aggressive reforms aimed at modernizing the country, including language standardization. Within this context, reformers who shared Zhang’s disdain for the influence non-Han groups had on China’s languages, and for that reason opposed making the majority language of Beijing the standard, suddenly had the opportunity to influence policy. Zhang’s philological research helped them fortify their positions.

In 1913, competing visions of a national language came to a head at a Ministry of Education conference convened to determine a national oral standard.Footnote 53 While the conference was contentious, ultimately the seventy or so men in attendance agreed on an invented language that took the majority language of Beijing as its base but, in an attempt to make their language more inclusive, also included phonological characteristics from other fangyan regions.Footnote 54 Within these deliberations, Zhang’s influence was clear. This radical construction that, in Jeffrey Weng’s words, “reflected the nature of the imagined future society it was meant to serve,” resembled Zhang’s suggestion of an invented national standard that brought together core aspects of diverse fangyan to reconstruct the language of Chinese antiquity.Footnote 55 The attendees even included several of Zhang’s students, who wielded notable influence promoting his vision.Footnote 56

Yet, within a few years, this national language engineered to represent the Han identity in its entirety faced backlash. Criticisms generally focused upon practicality: the new standard lacked a significant body of native speakers to effectively teach it. Once these challenges became clear, reformers who had once promoted the 1913 idealized constructed language began to distance themselves from it and advocate instead for alternatives that would facilitate fast, efficient promulgation.Footnote 57 In 1925, a group of reformers proposed a national phonology based solely on the majority language in Beijing, a proposal that was quickly adopted by the KMT-led Ministry of Education after they had secured control over the Republic’s central government. It has remained in place with few significant changes to this day.Footnote 58

In some ways, the institutionalization of the majority language in Beijing represented a repudiation of the Han ethnoracial supremacy of the late Qing. This was, in part, because the revolution had neutralized the power of Han supremacy in key ways. After the revolution, those who advocated a Han ethnonationalist state had to come to terms with the very real choice of either integrating non-Han groups into the national body or seeing the nation’s borders substantially shrink—a prospect that, in the age of Western imperialism, many feared would signal global weakness. As a fully realized Han ethnonationalist state seemed no longer tenable, calls for the restoration of the greatness of Han civilization were quietly replaced with celebrations of China as a multi-ethnic community of equals. The early Republic even adopted a five-color flag symbolizing “five races as one family” to signal this shift from a nationalism grounded in ethnicity and culture to a more civic understanding of citizenship.Footnote 59

But these realities did not mean that Han ethnonationalism, or a national language policy to support it, disappeared after 1911. After all, in Benno Weiner’s words, the discourses of civic nationalism were “not as convincing or powerful as those that preached a more limited nationhood based on arguments of common culture and descent.Footnote 60 As such, post-revolution nationalists, rather than continue to directly advocate for Han superiority or abandon Han ethnonationalism altogether, adjusted: they began to cloak Han supremacy in civic nationalism by wordlessly treating Han-ness as the default, presumed embodiment of Chinese national identity.Footnote 61 We can see this in several ways. Widely circulated literature exploring radical changes in Chinese national construction often conflated “Chinese” and “Han.” The title personage of Lu Xun’s famed “True Story of Ah Q,” meant to be a “representation of national character” or even “China itself,” was clearly, though never explicitly, coded as Han, allowing Lu to equate China’s national citizenry with Han-ness without ever actually saying so.Footnote 62 Similarly, school textbooks taught children that the history of China was equivalent to the history of the descendants of the Yellow Emperor—a narrative created and promoted by late-Qing Han nationalists but reimagined for Republican schoolchildren as national rather than Han history.Footnote 63

A Beijing-based national language was well-suited to support Han supremacy in this new guise. As a language spoken by Han people that did not claim its legitimacy from its connection to antiquity, the choice shifted focus away from ethnoracial purity and toward contemporary political power—Beijing was, after all, not the site of the Han ethnoracial lineage’s supposed origins, but the current seat of government of a Han-majority nation. More concretely, the national language reinforced external Han supremacy through its explicit ties to state power. Whereas late-Qing claims of Han supremacy were galvanized by a lack of political power, the revolution diffused much of their violent racialized vitriol by placing power officially in the hands of Han elites. This made the state a more trustworthy messenger for Chinese nationalism, and the capital more associated with modern Chinese power than Manchu autocracy. A national language based on Beijing phonology could thus continue the work of Han supremacy: it was supported by a state that now had the legitimacy to articulate a clear, coherent vision of what it meant to be Chinese, one that took Han-ness as its basis.

Yet while the national language project upheld the ethnoracial hierarchies for which late-Qing reformers advocated, it also institutionalized an intra-Han hierarchy. Large swaths of the citizenry were suddenly living in a country in which only one expression of Han-ness had institutional support, and it was not their own.Footnote 64 This loss was particularly personal for much of China’s intellectual class. Many of China’s most famous writers, thinkers, and academics came from southern provinces, which meant that they learned their own country’s national language later in life, if at all.Footnote 65

This feeling of collective loss among non-Mandarin fangyan speakers fueled a counternarrative about the role the national language could, or should, play in the new nation. Some compartmentalized, treating the national language as simply a practical communicative tool.Footnote 66 Others held that all fangyan were, at their core, languages of the Chinese nation by virtue of being Han cultural practices; even if only the national language had state support, other fangyan remained critical to the making of a national culture.Footnote 67 Still others emphasized that the national language could yet be improved to better reflect the diversity of the Han majority. Historian Rong Zhaozu, for instance, explained that for the national language to be “unified, promulgated, expanded, and improved,” reformers needed to consider the historical significance of other fangyan. Praising Zhang Taiyan by name, Rong credited his work for “using the present to explain the past.”Footnote 68 Others went further. In a robust debate in the late 1930s and 1940s among leftist writers about the proper model for a written language based upon a spoken vernacular, prominent writers such as Chen Boda and Ke Zhongping argued forcefully that non-Mandarin fangyan were critical pieces of national heritage. In the words of Wang Hui, these debates emphasized that fangyan were not and should not only be “resources for the formation of local identity, but aid in the formation of national identity.”Footnote 69

Yet perhaps the most common reaction to this loss of political power among speakers of non-Mandarin fangyan was to find alternative outlets to research, promote, and validate the value of those languages. One such outlet was the growing field of Chinese linguistics. Beginning in the 1920s, prominent academics began to clamor for a more systemic study of Sinitic languages, arguing that they were critical to understanding the nation’s culture. In the following decades, dozens of scholars published detailed surveys of contemporary fangyan, as well as new, scientific studies of historical phonologies of Chinese antiquity.Footnote 70

These studies frequently cited Zhang Taiyan’s New Fangyan. Take, for instance, A Chinese Syllabary Pronounced According to the Dialect of Canton, published in 1941 by Cantonese scholar Wong Shik-Ling.Footnote 71 Wong’s goal was to provide a scientific rendering of the Yue phonology—a southern fangyan region that includes present-day Cantonese. But in his introductory material, in which he praises Zhang Taiyan for his “meticulous” study of ancient phonology and local languages, Wong articulated stakes beyond rendering accurate data. In particular, he juxtaposed the publication of his book with the national language movement in order to explain the national importance of preserving and studying the Yue fangyan. “When we use Chinese characters, we cannot neglect the Yue fangyan … when we read aloud essays, in particular the rhyming verses written after the Wei and the Jin (AD 317–420) dynasties, using the Yue language makes it rhythmic and sonorous … [it] has best preserved the characteristics of these texts.”Footnote 72 With this statement, Wong argued that evidence of Han superiority—grounded in historical cultural achievements or the longevity of Sinitic script—was directly preserved in his own language. When judged by the extent to which a language embodied and retained that evidence, Wong concluded, the Yue fangyan was clearly more valuable than the national language.

By 1941, when Wong published his Chinese Syllabary, it was undeniable that the majority language spoken in Beijing had institutionalized power to represent Han people and the Chinese nation. But Wong’s work and others like it make clear that there were discursive strategies to challenge institutional support as the only or even primary source of power, and in wielding these strategies these men often cited Zhang Taiyan’s work. They consistently argued that the political power arising from historical lineage was absolute; it predated and overrode any power emanating from governmental authority. In their efforts, Zhang’s work provided inspiration for using phonological data to prove their language’s connections to that historical lineage and claim the power to which they believed they were therefore entitled. In other words, Wong’s work shows what Zhang inspired: a blueprint for challenging Beijing’s linguistic power by drawing on alternative wellsprings of power.

Claiming Hegemonic Han-ness in Local Gazetteers

The last section focused on a historical rupture. One language had been chosen as the national standard, while the rest of China’s Han majority were left with a well-established narrative of ethnoracial superiority but no institutional recognition of the significance of their own languages to that superiority. The last section also hinted at the cascading effects of that rupture, showing how the national language project created an impetus for Han speakers of non-Mandarin fangyan to explore new ways of regaining cultural power. The purpose of this final section is to explore how these strategies were wielded at the local level. In so doing, this section shows not only the evolution of narratives of Han supremacy as the national language project took hold, but also how communities whose culture remained quite distinct from that of the capital found meaning for themselves in the nationalizing project.

To explore how narratives of hegemonic Han-ness evolved beyond high-level academic and political conversations, this section focuses on local gazetteers (fangzhi).Footnote 73 These compendiums of a locality, composed by local elites, were lengthy, informational tomes meant to promote a locality’s uniqueness and, often, superiority.Footnote 74 As a genre, they have a long history, systematically produced since at least the Song dynasty. Sometimes comprising thousands of pages, they usually included the same basic information: topographical features, prominent lineages, local economic production, historical relics, and exemplary persons. Gazetteers also served an important historical role. As Maybo Ching notes, gazetteer compilers saw their task as “creat[ing] a chain that connected their localities to the nation,” allowing local elites to communicate with the capital and shape policy.Footnote 75 Their impact was thus not just political, but also cultural: they highlighted how a locality’s ecology, economy, and culture were integral to empire’s cohesive integrity. These portraits thus served as a critical space for local elites to both express local pride and ensure they had a voice in the creation of empire-wide narratives.

Given this historic role, twentieth-century gazetteers are excellent sources for examining how localities absorbed, interpreted, and shaped nationally circulated narratives.Footnote 76 This includes the discursive devices that reinforced Han superiority discussed above. Authors often treated the Han ethnoracial group as the unspoken standard representative of both the nation and their individual localities, as Han peoples’ biographies, histories, and accomplishments comprised the bulk of the gazetteers’ information. Sometimes references to Han superiority were overt. As compared to earlier volumes, the twentieth century saw a sharp increase in local histories that explicitly placed the local population within the lineage of the Yellow Emperor,Footnote 77 described as the ancestor of the “exceptional Chinese ethnicity” (youxiu de Zhonghua minzu 優秀的中華民族).Footnote 78 So too was there an uptick in histories of violent invading outsiders that emphasized either the grit and resilience of the local Han population or espoused the inferiority of those who were “not descendants of the Yellow Emperor.”Footnote 79

In some ways, therefore, Republican-period gazetteers mirrored other contemporaneous national histories. What was unique was how gazetteers interlaced well-known narratives with local anecdotes to demonstrate the locality’s role in the preservation and representation of those national histories. This was done in several ways: some included sections on historical architecture, others on local heroes or folk tales. But one of their primary strategies to demonstrate the locality’s historical importance to the nation was with a laudatory tribute to the local language.

While sections on local language, labeled fangyan, yuyan [language 語言], or tuyin [vernacular pronunciation 土音], did not appear in the majority of twentieth-century gazetteers, they were fairly common.Footnote 80 Their content was highly variable. They varied in length, from just a paragraph to dozens of pages, and in content, with some focusing on vocabulary and idioms, others on phonology. There was also variation in the kinds of expertise the authors cited. Some authors simply provided a curated list of unique local vocabulary, whereas others described the local tongue by citing recent scholarship, or in rare cases, conducting their own fangyan surveys.

Yet these sections had some attributes in common. It was typical to begin the section with a short narrative explaining why Han people spoke so many different languages. Authors from Ba county in Sichuan and Songming county in Yunnan, for example, explained how post-Han dynasty southern migration led to the “fracturing” of “local customs” and languages.Footnote 81 Authors from Xinfeng county in Henan offered a simpler explanation: “As our ancestors moved and spread around, they lost their original pronunciations (benyin 本音).”Footnote 82 Yet amidst these varying chronicles, none I encountered questioned the core premise of Han historic unity. Authors from Shaotong county in Yunnan claimed that all Han languages shared a root, as “each ethnoracial lineage [zu] has their own language”Footnote 83; another wrote, “Our linguistic roots are the same.”Footnote 84 The gazetteer from Wenxi county in Henan was more specific: “Our [Han] minzu has one branch, our language and script have one origin.”Footnote 85

A second generalization we can make is that most gazetteer authors described their local language by referencing historic texts. The typical format consisted of a list of local verbiages compared to their corollaries in historic dictionaries and character compendiums.Footnote 86 These sections usually began with common words—Mi county in Henan began with the words for father and motherFootnote 87—but some placed local particularities at the forefront, such as Jiande county, which began with words for coal and eggplant.Footnote 88 In these comparative lists, most referents were the aforementioned Han-dynasty texts Shuowen Jiezi, Erya, and Fangyan, but other texts appeared frequently as well. Yichun county in Jiangxi explained the origin of their local vocabulary in the Daoist philosophical text Zhuangzi (莊子) and the narrative history Zuo Zhuan (左傳Commentary of Zuo) from the Eastern Zhou (771–256 BC).Footnote 89 Taicang county in Jiangsu listed examples from the philosophical classic Mencius and poetry from the Han and Tang dynasties.Footnote 90 Chixi county in Guangdong cited all of the above and the fifth-century Ballad of Mulan. Footnote 91

This method of describing linguistic histories by comparing contemporary vocabulary to earlier character compendiums appeared in pre-twentieth-century gazetteers as well. As Hirata Shoji argues, language sections began appearing in the Ming and Qing dynasties, and they frequently touted the relationship between the local language and antiquity in a way that exuded regional pride.Footnote 92 So, too, was there a long history of exacting cultural capital through associations with classical texts, since these texts had long afforded power to those who could claim connection to them.Footnote 93 Yet in the context of Chinese nation-building, this familiar method supported new discourses. In a poetic rhythm of “X is said as Y in our local language,” Republican-period gazetteer authors treated millennia-old linguistic texts not simply as cultural authorities, but representations of a Han-centered national history intricately tied to language. By connecting their locality to these texts vis-à-vis linguistic lineage, authors leveraged the texts’ symbolic power to aggrandize their locality’s standing in a nation where connection to an idealized antiquity translated into access to the power of Han hegemony.Footnote 94

Connections between vocabulary and canonical texts were not the only way gazetteer authors drew connections to an invented past. Others focused on phonology. These narrations were also referential, the most common reference being the “phonology of antiquity” (古音).Footnote 95 Other times, the authors compared initials, finals, or tones in their language with data from historic rime tables, such as the authors of the Xining county gazetteer in Guangdong, who noted that their fangyan had the “opening tone” for characters that also carried that tone in the Guangyun, a tenth-century rime table.Footnote 96 Other gazetteers referenced particular geographic spaces. Zhongyuan, or Central Plains, was a common referent—Xining county near Guangzhou, Liling county in Hunan, and Yi county in Anhui all claimed that their phonology was either “based in” or “had elements of” the phonology of Zhongzhou or Zhongyuan.Footnote 97 And, importantly, these strategies were not mutually exclusive. The Ningxiang county gazetteer compared one character’s local phonology to the Guangyun, and another character to the fourteenth-century Zhongyuan yinyun. Footnote 98 Clearly, authors sought to associate their local languages with historic referents, all of which held power as part of the national canon.Footnote 99

Though diverse, many of these strategies were inspired by Zhang Taiyan’s use of “the present to explain the past.” I found sixteen gazetteers that cited him outright.Footnote 100 Sometimes these references were superficial, as authors cited Zhang’s explanations of China’s linguistic diversity or referenced his map of fangyan regions. Other times, gazetteer authors proclaimed that Zhang’s specific argument in New Fangyan could be used to prove significance of their own language. In the gazetteer for Linqu county in Shandong, for instance, the authors cited Zhang’s claim that any national representative of the Han needed to take “fangyan as its base” because only through local languages could citizens find the “origin” of guyin. Take, they argued, their own fangyan—reading Tang dynasty poetry aloud in Linqu fangyan revealed traces of guyin. Footnote 101

Not all gazetteers emphasized proximity to linguistic antiquity. The authors of the Guangshao county in Shandong offered instead a praise-laced narrative of the Chinese national language movement, excoriating those who sought to “accommodate fangyan” thereby “blocking the progress of national language unification.”Footnote 102 Yet even in instances like these, hegemonic Han-ness emerged in other contexts. In this same gazetteer, the authors praised how Guangshao county honored the memory of Confucius, native son of Shandong, by ritualistically singing a remembrance song, reproduced in the gazetteer’s pages. This example should remind us that proximity to linguistic antiquity is one of many expressions of hegemonic Han-ness. It is not one singular narrative, but competing, overlapping, and contextual discursive strategies.

Ultimately, it is clear that references to an imagined past are ubiquitous in Republican-period gazetteers, bringing into sharp relief the power that lay in a connection to antiquity. It is also clear that the contexts and contents that defined antiquity were flexible, diverse, and sometimes contradictory. But in the context of hegemonic Han-ness, the multivalence of antiquity gave it strength. With blurred borders and an ever-shifting core, antiquity could easily be slotted into a diverse array of unique local narratives, and the power it bestowed could be defended by a diverse range of actors. In a word, the ambiguity of antiquity as a referent made the existence of competing claims to represent it possible.

Detailed Claims from the South

The previous section narrated how Republican-period gazetteers described local languages using references to antiquity in order to underscore claims to represent hegemonic Han-ness. Its goal was to show the sheer prevalence of such claims and how, unlike earlier discussions of language in gazetteers, authors from the Republican period took influence from twentieth-century narratives about race and power. This final section moves from breadth to depth, taking two gazetteers as case studies to excavate the kinds of logic, evidence, and strategies local authors used to situate their own languages’ histories—and the power they thought those histories ought to rightfully confer—within national narratives.

The first comes from Dabu county in Guangdong near the city of Meizhou. Dabu was, and is today, majority Hakka, a group that migrated to the south of China after the Song dynasty. Published in 1943, the gazetteer focused heavily on the Hakka and their relationship to the Han ethnoracial identity.Footnote 103 The reason for this was likely related to a long local history of tumult. In the nineteenth century, a rise in violent clashes between local Hakka and Cantonese populations in Guangdong, as well as the increasing resonance of ideas about ethnoracial purity among local elites, fueled discrimination against the Hakkas, derisively referred to by Cantonese speakers as an “outside zu.Footnote 104 In response, prominent Hakka elites funded a deluge of scholarship to prove that the Hakka had a greater proximity to Han antiquity than did their neighbors.Footnote 105 Such scholarship featured prominently in the Dabu county gazetteer. The authors cited these instances of discrimination, repudiating the “mean and shallow” people who claimed that the Hakka were not part of the “Han race” (漢種 Hanzhong). To the contrary, the authors proclaimed, Hakka were actually the best representatives of the Han ethnoracial lineage. The gazetteer extolled Hakka accomplishments, citing their business acumen, their industriousness, and the sophistication of their education system. Ultimately, the authors concluded, citing the words of Yale geographer Ellsworth Huntington, the Hakka among the Han people were the “cream atop milk.”Footnote 106

To reinforce this narrative of Hakka elitism, the authors devoted a long section to proving how the Hakka had best preserved the lineage of the Xia clan. And here, language served as the primary evidence. According to the authors, other Chinese languages had strayed far from the language of antiquity. Phonological data proved that Cantonese speakers had moved south from Zhongzhou to Guangdong province during the third century; with distance and time, their language deviated from the language of antiquity. And yet, after the fourteenth century, northern languages became influenced by non-Han groups, meaning that the language spoken in the original location of Zhongzhou had similarly mutated. Only the Hakka avoided both of these fates, having moved south from Zhongzhou within those intervening centuries. As the gazetteer explained, though the Hakka “could not avoid” mixing with their southern neighbors over time, their language “remained close to the language of Zhongzhou.”Footnote 107 Indeed, they continued, the Hakka were the only community to have “handed down the language of antiquity” in an “enduring, long-lasting” way.Footnote 108

Importantly, the authors reminded us, this was not simply about language: the fact that the Hakka preserved the language of antiquity also meant they preserved its customs. For proof of this connection, they cited Zhang Taiyan. In a later addendum to New Fangyan, Zhang claimed that the root of the Hakka language exuded the “hat and belt” culture—a reference to Sima Qian’s second-century sweeping historical tome “Records of the Grand Historian,” in which he uses the term to denote a kind of Chinese civility directly opposed to the culture of non-Chinese groups—rather than that of foreigners. In the Dabu county gazetteer, the authors used this “hat and belt” reference to prove that the Hakka language had a connection not just to antiquity in general, but an antiquity marked by civility. In a word, their language proved a connection to a culture that was distinct because of its superiority over other indigenous groups.

Our second case study comes from Liling county in Hunan. According to the authors, Liling county had a uniquely long, continuous history. First settled in the Eastern Han (AD 25–220), the county’s residents oversaw all of the important moments in the “evolution” of the Chinese nation: from the “authoritarianism” of the imperial period—a direct reference to Manchu rule—to the “democracy” of the present, from the “overthrow” of outsiders (yizu 異族) and the creation of a Republic.Footnote 109 This emphasis on a national evolution specifically measured by the defeat and expulsion of non-Han peoples—and the anti-Manchu rhetoric inspired by the kind Zhang Taiyan popularized—appeared throughout the manuscript, framing sections on government, customs, and importantly for our purposes, language. In particular, the language section praised the people of Liling county for their ability to withstand historical change and outside influence. To prove the stability of their language, the authors offered an unprecedent amount of evidence. First, they presented original phonological research in a series of charts comparing their fangyan’s phonology to the tenth-century rime table Guangyun. The authors then offered a philological examination of the origins of local verbiage to the Han-dynasty dictionaries Fangyan and Shuowen Jiezi. To the authors, the data, meticulously recorded over dozens of pages, definitively showed that the people of Liling had preserved the language of antiquity.

They presented their data according to a familiar model, notable only in their sheer quantity, but their explanation of their language’s connection to antiquity was unique. They began with the region’s geography. Situated at the crossroads of several fangyan regions, the people to the north, east, and south all spoke distinct languages. Only at the center of the county did people speak the unchanged language of antiquity. “Because of the mountains and rivers, the veins of the ancient linguistic system in our language are still detectable,” the authors wrote.Footnote 110 This connection with antiquity, they claimed, facilitated their ability to both absorb and communicate with neighboring groups. The authors explained, “In the middle of our village is the pure local language. Even though people from the Min and Yue fangyan families moved to our location, within a generation or two, they would completely be absorbed into our local language. Even when guests come to our village and have not changed their language, when they are out in our society they can adapt to the local language. [For those who live] to the west of the village … because their language’s roots are the same, there are no gaps in communication.”Footnote 111

Liling county’s language embodied an ideal that strikingly resembled Zhang Taiyan’s ideas about a national standard: a language that preserved the historic roots of antiquity that all fangyan regions, to varying degrees, contained. Liling’s gazetteer authors thus made claims that were at once common among other gazetteers—that their language preserved ancient linguistic roots—and strikingly unique, in that they proved that preservation not just with historic texts, but also with the mutual intelligibility of their language with other fangyan regions.

There are clear similarities in the strategies the authors from Dabu and Liling used to emphasize their county’s connection to antiquity. In both, authors drew upon well-known markers of antiquity. The Guangyun and the Shuowen were not just texts; they were evidence of a diachronic history of the language of the Xia clan. Zhongyuan was not just a place; it marked the heart of Chinese civilization. In both, their lengthy section on language was a Republican-period addition, the topic not appearing in Qing-period gazetteers from either place. And finally, both gazetteers combined phonological data with narratives about migration. The Dabu gazetteer emphasized the Hakka’s migration from Zhongzhou, while the authors of the Liling gazetteer emphasized migration history to refute the presumption that their language’s tether to antiquity was subject to deterioration.

I chose these examples because I found them notable in the quantity of their data and the uniqueness of their narratives. Yet they are not the only gazetteers to examine local language with this much detail. When we include other similar gazetteers, we can draw one final generalization: they overwhelmingly came from the south. Though the peoples of Hainan, Guangdong, and Hunan all spoke different languages, they all cited Zhang Taiyan to prove that “only the south[ern] pronunciation had evolved perfectly.” It is impossible to know the impact Zhang Taiyan had on regional notions of Han-ness, especially since beliefs about the connection between southern languages and antiquity predated his work. Nonetheless, Zhang was remembered and exalted in these texts as using philological evidence to prove the preservation of Han antiquity—the source of Han superiority—in southern regions. In a word, his work seems to have helped buttress a particular and spacialized response to the national language project.Footnote 112

It is often difficult to trace the origins of a story or a myth. Sometimes narratives gain institutional backing, and we find them in textbooks, state policies, or public communications. Other times they acquire popular resonance, and we can see them in popular media or hear them via word of mouth. When it comes to the narrative of linguistic antiquity as a marker of hegemonic Han-ness, it seems to have been spread through popular reinterpretations of academic studies that, when taken together, created a unified field of understanding about Han-ness, language, and history across China. While we cannot definitively prove a clear link between these gazetteers and common retellings of similar stories today, these gazetteers reveal how late-Qing ideas about language and Han-ness, many of them inspired by Zhang Taiyan, were wielded by everyday people to battle national language hegemony and claim a competing trait of hegemonic Han-ness for their own mother tongues.

Conclusion: Hegemonic Han-ness in Comparison

This paper has probed why proximity to linguistic antiquity became a competing construction of hegemonic Han-ness that challenged the hegemony of the Chinese national language in the early twentieth century. It has shown, first, how the turmoil of the late Qing fueled an explicitly Han ethnocentric nationalism centered on the exaltation of Chinese antiquity, and second, how that nationalism supported claims by speakers of non-Mandarin local languages that the proximity of their language to antiquity made them better representatives of Han-ness than speakers of the national language. Moreover, once the national language debate concluded, officially precluding other fangyan from gaining institutional recognition, local elites latched onto a counternarrative about how languages defined Han-ness: that in the making of the Han ethnoracial identity, it was a language’s proximity to antiquity, not its institutionalization by the state, that mattered most.

This counternarrative among speakers of non-Mandarin fangyan, I argue, had two effects. First, they offered space for those who did not speak the national language to grasp power for themselves in a nation that defined itself as Han. This counternarrative, in other words, broke apart the notion of a top-down, unified Han nationalism, instead revealing a power struggle among those who identified as Han over the ability to define its boundaries. Second, and more importantly, these claims nonetheless reinforced the very basis of Han supremacy in the aggregate. By contending that the value of any Chinese language was rooted in its purity and lack of adulteration by outsiders, this counternarrative devalued non-Han groups in the making of the nation and racialized them as both other and inferior. In the end, these counter-discourses joined state discourses in directly or indirectly establishing that, however Han-ness was defined, its external hegemony was central to imagining what the Chinese nation was and should be.

This power struggles born of these competing discourses have clear resonance today. As was true a century ago, Chinese language policy remains both emblematic of and a tool for reinforcing external Han hegemony.Footnote 113 After 1949, while the new ruling Chinese Communist Party harshly denounced what they called the “Han chauvinism” of their predecessors, their purported celebration of indigenous groups never amounted to a dismantling of Han power in bureaucratic, legal, or cultural structures. Even during the high socialist period, defined by a broad effort to rebuild society into one defined by economic justice, central and local leaders silently upheld Han-ness as emblematic of the progressive values of new China while they simultaneously violently forced non-Han peoples into assimilation.Footnote 114 As a result, the remarkable societal changes of the Maoist period did little to shake Han hegemony. Indeed, Han dominance has only become more absolute.Footnote 115 Today, economic disparities between Han and non-Han regions are stark. Schools throughout the PRC continue to equate Han history with Chinese history much as they did a century ago, whereas space for learning non-Han histories is rapidly disappearing.Footnote 116 And within this context of increasingly totalizing Han supremacy, language often serves as a powerful tool for Han dominance. In majority non-Han regions, Mandarin education is frequently touted as a tool for promoting “ethnic harmony,” and the achievement, labor potential, and political loyalty of non-Han people are often judged according to their Mandarin proficiency. Perhaps the bleakest example is how, since 2017, the CCP government has seized and detained countless ethnic Uighurs—a Muslim indigenous group that lives in the northwest province of Xinjiang—in extrajudicial camps designed to “re-educate” them and root out so-called “dangerous” activities; within these camps, forced Mandarin education is a prominent part of daily life, alongside “thought education,” forced confessions, and bodily torture.Footnote 117

At the same time, Mandarin’s hegemony over other fangyan has expanded.Footnote 118 Today it is the only Chinese language regularly taught in schools,Footnote 119 and by far the most common language used in news media, television, radio, and public service announcements. It, and it alone, is deemed the “symbol of the Chinese nation” in the PRC constitution. It is also afforded institutional support in more subtle ways. In 2019, a popular video-based social media app called Douyin began sending warning messages to users who posted videos in Cantonese, asking them “please use Mandarin.” When pressed, Douyin’s owner, Bytedance, responded that their aim was not to censor Cantonese; they simply lacked the infrastructure to effectively moderate Cantonese content.Footnote 120

These current power dynamics raise an important question: Are earlier counternarratives about the proximity of non-Mandarin fangyan to linguistic antiquity still connected to hegemonic Han-ness today? This question would be impossible to sufficiently answer in the space here, but two recent protests offer a glimpse of existing connections. In 2010, protests broke out in the Guangzhou over the city’s government’s decision to gradually replace the already shrinking selection of Cantonese television offerings with shows in Putonghua.Footnote 121 After a series of marches, sit-ins, and sympathetic op-eds highlighting the historical value of Cantonese because of its “long history,” the government conceded, maintaining the existing Cantonese offerings.Footnote 122 In comparison, in 2020, protestors marched in Inner Mongolia against a government program to replace Mongolian language education with Putonghua. These protests, however, were not met with the same permissiveness, and resulted instead in thousands of arrests and no concessions.Footnote 123 Indeed, generally speaking, non-Han peoples such as Tibetans or Uighurs who publicly advocate for the preservation of their linguistic rights tend to face scrutiny, censorship, and state punishment.Footnote 124

There are many reasons why the state might respond more sympathetically to protests for Cantonese preservation than similar demands for Mongolian, which have nothing to do with ethnoracial identity.Footnote 125 For one, in 2010 the tolerance for public dissent was greater, and in fact since the seeming success of those protests Cantonese television programming has been quietly disappearing. Yet I propose that these distinct and unequal responses in part reflect the central government’s conceptual reliance upon Han supremacy to ensure the consent and support of the majority Han population. Given that the state has only increased its insistence on the hegemony of Mandarin in the past several decades, claims of proximity to Han antiquity remain a viable way for speakers of non-Mandarin local languages to garner cultural authority. As such, these speakers and their protests theoretically represent a challenge to state power. Yet when Cantonese speakers claim their language is older or more venerable than the national language, their claims rely on a historical narrative that reveres purity of lineage and denigrates non-Han groups—a narrative that ultimately reinforces the Han hegemony the state relies upon. In this way, while these fangyan-lauding counternarratives challenge the state’s insistence that the state alone speaks for the Han identity, they simultaneously aid the state in upholding the cultural dominance of the Han ethnoracial majority. Put bluntly, both protests sought to challenge the hegemony of Mandarin, but only one challenged the hegemony of Han-ness. Given that the state seems to see the latter as a greater threat than the former, it seems logical that counternarratives about linguistic antiquity still afford their adherents a certain amount of power.

This history of hegemonic Han-ness is not only relevant to understanding China today; it also has comparative potential. To return to Flores’s example of the United States, after the Bilingual Education movement, educational and bureaucratic institutions were redesigned to uphold bilingualism only insofar as it supported external hegemonic whiteness. What is unique about China is that there is less leeway for diverse expressions of Han-ness in today’s PRC than there is for whiteness in the United States. Yet the Cantonese protest reveals that, sometimes, the PRC government must weigh external hegemonic Han-ness against internal hegemonic Han-ness; in those cases, the state may sacrifice its claim to define the latter so as to reinforce the former. These counternarratives about the importance of local languages due to their proximity to antiquity are therefore still instructive in how Han people leverage external Han hegemony to fight a state-defined internal Han hegemony. The state response, on the other hand, implies that the PRC sees challenges to external Han hegemony as the greater threat.

Ultimately, both examples reveal how power relationships between hegemonic and marginalized groups intersect with dominant and marginalized expressions within hegemonic groups. The example of China shows us how those power relations play out when state institutions are not only expressly committed to upholding the hegemony of one group, but also one singular expression of it. It allows us to see how powerful institutions give weight to these levers of control, a story that resonates across the world.


I’d like to thank Alexander Statman, Jeffrey Weng, Ulug Kuzuoglu, Mårten Söderblom Saarela, Jenny Gavacs, Y. Yvon Wang, Agnieszka Joniak-Lüthi, Alan Baumler, and the thoughtful, anonymous CSSH reviewers for their extensive and empathetic comments. Thanks also to the University of Texas Library for access to their extensive materials, my colleagues in the Trinity University History Department for their support, and Wing Sun Tam, Matcha, and Callie for their unending encouragement and love. All translations heretofore are my own unless otherwise noted.


1 Alex Lo, “Why Cantonese Is a Real Language in Hong Kong,” South China Morning Post, 4 Feb. 2014.

2 Kejia fangyan he wenzi: Zhongyuan yu Hanyu de ‘huohuashi,’” Kejiaren zazhi, 25 July 2017; Putonghua he Guhanyu de chaju you duoda? Fayin he na zhong fangyan bijiao jiejin ne?” Meiri Toutiao, 23 Sept. 2017,

3 Many linguists today agree that southern fangyan better accord to reconstructions of ancient and archaic Chinese than northern fangyan, including Mandarin. See Ho, Dah-An, “Chinese Dialects,” Sun, Chaofen and William, S. Y. Yang, eds., The Oxford Handbook of Chinese Linguistics (Oxford: Oxford University Press, 2015), 154 Google Scholar; Ballard, W. L., “Aspects of the Linguistic History of South China,” Asian Perspectives 24, 2 (1981): 163–85Google Scholar.

4 While “dialect” is the most common translation for fangyan, many scholars find this problematic. See Mair, Victor, “What Is a Chinese ‘Dialect/Topolect’? Reflections on Some Key Sino-English Linguistic Terms,” Sino-Platonic Papers 29 (Sept. 1991): 129 Google Scholar. I will either use “non-Mandarin Sinitic language” or “local language,” or leave fangyan untranslated.

5 Both the translation of and categorization of the term “Han” are hotly debated among scholars. For explanations of these debates, see Joniak-Lüthi, Agnieszka, The Han: China’s Diverse Majority (Seattle: University of Washington Press, 2015)Google Scholar; Agnieszka Joniak-Lüthi, “The Han Minzu, Fragmented Identities, and Ethnicity,” Journal of Asian Studies 72, 4 (2013): 849–71; Mullaney, Thomas, “Critical Han Studies: Introduction and Prolegomenon,” in Vanden Bussche, Eric, Mullaney, Thomas, Leibold, James, and Gros, Stephane, eds., Critical Han Studies: The History, Representation, and Identity of China’s Majority (Berkeley: University of California Press, 2012)Google Scholar, 1–16. This paper will use the term “Han” untranslated.

6 Tam, Gina Anne, Dialect and Nationalism in China, 1860–1960 (Cambridge: Cambridge University Press, 2020)CrossRefGoogle Scholar.

7 Joniak-Lüthi, Han; Mullaney, “Introduction,” 2–4.

8 Li, Jin and Moore, Danièle, “Multilingualism, Identities and Language Hegemony: A Case Study of Five Ethnic Minority Students in China,” International Journal of Bias, Identity and Diversities in Education 2, 2 (2017): 4256 CrossRefGoogle Scholar; Roche, Gerald, “Articulating Language Oppression: Colonialism, Coloniality and the Erasure of Tibet’s Minority Languages,” Patterns of Prejudice 53, 5 (2019): 487514 CrossRefGoogle Scholar.

9 Ding, Sheng and Saunders, Robert, “Talking up China: An Analysis of China’s Rising Cultural Power and Global Promotion of the Chinese Language,” East Asia 23, 2 (2006): 333 CrossRefGoogle Scholar.

10 Shoji, Hirata, Wenhua zhidu he Hanyu shi (Beijing: Peking University Press, 2016), 247–56Google Scholar.

11 Connell, R. W., Masculinities (Berkeley: University of California Press, 1995)Google Scholar; R. W. Connell and J. W. Messerschmidt, “Hegemonic Masculinity: Rethinking the Concept,” Gender & Society 19, 6 (2005): 829–59.

12 Hughey, Matthew, “The (Dis)similarities of White Racial Identities: The Conceptual Framework of ‘Hegemonic Whiteness,’Ethnic and Racial Studies 33, 8 (2010): 1289–309CrossRefGoogle Scholar; Jennifer Guglielmo, “Introduction: White Lies, Dark Truths,” in Jennifer Guglielmo and Salvatore Salerno, eds., Are Italians White? How Race Is Made in America. (New York: Routledge, 2003), 1–16.

13 Flores, Nelson, “A Tale of Two Visions: Hegemonic Whiteness and Bilingual Education,” Educational Policy 30, 1 (2016): 1338 CrossRefGoogle Scholar.

14 Carrico, Kevin, “Recentering China: The Cantonese in and beyond the Han,” in Vanden Bussche, Eric et al., eds., Critical Han Studies: The History, Representation, and Identity of China’s Majority (Berkeley: University of California Press, 2012), 2344; 2425 Google Scholar.

15 The term “Han” was variably used to describe a people beginning in the Northern Wei (AD 386–534) but gained its contemporary connotations in the Ming dynasty (1368–1644) and absorbed modern ethnoracial significances in the Qing. Mark Elliott, “Hushuo 胡说: The Northern Other and the Naming of the Han Chinese,” in Eric Vanden Bussche et al., eds., Critical Han Studies: The History, Representation, and Identity of China’s Majority (Berkeley: University of California Press, 2012), 173–90; Chow, Kai-Wing, “Narrating Nation, Race, and National Culture: Imagining the Hanzu Identity in Modern China,” in Chow, Kai-Wing, Doak, Kevin, and Fu, Poshek, eds., Constructing Nationhood in Modern East Asia (Ann Arbor: University of Michigan Press, 2010), 4883 Google Scholar; James Leibold, “Searching for Han: Early Twentieth-Century Narratives of Chinese Origins and Development,” in Eric Vanden Bussche et al., eds., Critical Han Studies: The History, Representation, and Identity of China’s Majority (Berkeley: University of California Press, 2012), 210–33.

16 Zarrow, Peter, After Empire: The Conceptual Transformation of the Chinese State (1885–1924) (Stanford: Stanford University Press, 2012), 147–72.Google Scholar

17 Kenji, Shimada, Pioneer of the Chinese Revolution: Zhang Binglin and Confucianism, Joshua Fogel trans. (Stanford: Stanford University Press, 1990)Google Scholar; Tang Zijun, Zhang Taiyan nianpu changbian (Taibei: Taiwan shangwu yinshuguan, 1979); Laitinen, Kauko, Chinese Nationalism in the Late Qing Dynasty: Zhang Binglin as Anti-Manchu Propagandist (London: Curzon, 1990)Google Scholar; Young-tsu, Wong, Search for Modern Nationalism: Zhang Binglin and Revolutionary China, 1869–1939 (Oxford: Oxford University Press, 1989)Google Scholar; Viren Murthy, The Political Philosophy of Zhang Taiyan: The Resistance of Consciousness (Leiden: Brill, 2011).

18 Murthy, Political Philosophy, 3.

19 Yamada Keiji, quoted in Murthy, Political Philosophy, 15; Shimada, Pioneer of the Chinese Revolution, 21.

20 Fogel, Joshua, “Race and Class in Chinese Historiography: Divergent Interpretations of Zhang Bing-Lin and Anti-Manchuism in the 1911 Revolution,” Modern China 3, 3 (1977): 346–75CrossRefGoogle Scholar; Onogawa Hidema, “Zhang Binglin de painan sixiang,” Dalu Zazhi 44, 3 (1972): 155–76; Kai-Wing Chow, “Imagining Boundaries of Blood: Zhang Binglin and the Invention of the Han “Race” in Modern China,” in Frank Dikötter, ed., The Construction of Racial Identities in China and Japan (Hong Kong: Hong Kong University Press, 1997), 34–52; Frank Dikötter, The Discourse of Race in Modern China (Stanford: Stanford University Press, 1992).

21 There is much debate over how to translate Zhang’s terms. Zhang usually referred to Han as a “zu,” which scholars often claim historically translated to lineage or clan, but by the late Qing was better translated as ethnicity. Some scholars, such as Agnieszka Joniak-Lüthi and Kai-Wing Chow, maintain that Zhang’s emphasis on blood lineages and racial essentialism make “racial lineage” or “race-lineage” more appropriate. To make matters more complicated, Zhang sometimes used “Hanzhong” (種) instead of Hanzu, replacing zu with a character usually used to describe racial groups like yellow or white. Given that his rhetoric on non-Han indigenous groups was often essentializing, hierarchical, and embodied—characteristics scholars of race would see as constitutional of racialized rhetoric—I use “ethnoracial lineage” for Hanzu. Schneider, Julia, Nation and Ethnicity: Chinese Discourses on History, Historiography, and Nationalism (1900s–1920s) (Leiden: Brill, 2017), 198202 CrossRefGoogle Scholar; Chow, “Imagining Boundaries”; Dikötter, Discourse of Race; Joseph Esherick, “How the Qing Became China,” in Joseph Esherick, Hasan Kayalı, and Eric Van Young, eds., Empire to Nation-State: Historical Perspectives on the Making of the Modern World (New York: Rowan and Littlefield, 2006), 229–59; 245.

22 Leibold, “Searching for Han,” 213; Dikotter, Frank, “Racial Discourse in China: Continuities and Permutations,” in Dikötter, Frank, ed., The Construction of Racial Identities in China and Japan (Hong Kong: Hong Kong University Press, 1997), 13 Google Scholar.

23 As one of the organizers of the Asian Brotherhood Association, which focused on global struggles against Western imperialism, Zhang often framed his anti-Manchuism within a broader commitment to transnational anti-imperialism. Lin Shaoyang, “Zhang Taiyan ‘zizhu’ de lianya sixiang: Minbao shiqi Zhang Taiyan yu Riben zaoqi zuoyi yundong” (Zhang Taiyan’s Pan-Asianism: The Context of Japanese Early Leftist Movement, Pan-Asianism, the Anglo-Japanese Alliance and the Overseas Indian Independent Movement) Zhang Taiyan sheng ping yu xue shu (2016): 969–1015 (Lin Shaoyang’s translation).

24 Zhang Taiyan, “Geming daode shuo,” in Zhang Taiyan quanji: di yi ban (Shanghai: Shanghai Renmin Chubanshe, 1982[1906]), 292. Given that the essay flatly dehumanizes Jurchens, Mongols, and Manchus, “Chinese race” meant Han, though his emphasis on ethnoracialized power served a broader thesis of anti-imperialism. See Christopher Peacock, “Intersecting Nations, Diverging Discourses: The Fraught Encounter of Chinese and Tibetan Literatures in the Modern Era” (PhD diss., Columbia University, 2020), 47. The translation is the same as in Murthy, Political Philosophy, 83.

25 This tone shift is used by scholars such as Kauko Laitinen and Wong Young-tsu to argue that Zhang’s rhetoric was not racist but rather “misunderstood” propaganda; his true goals, they claim, were anti-imperialistic rather than racist and “political rather than genocidal.” Wong Young-tsu, Search for Modern Nationalism, 147–48; and Beyond Confucian China: The Rival Discourses of Kang Youwei and Zhang Binglin (New York: Routledge, 2010), 55. Yet, as Julia Schneider posits, Zhang continued to link “different identifications of zhongzu to superiority and inferiority” despite his rhetorical shift after 1906. This, she writes, is “racist or racialist … in the very modern sense of racism being equaled with a discrimination of people legitimized by the assumption of humans being divided into ‘separate and exclusive biological entities,’ some of which are superior to others, irrespective of whether the division lines between these entities are penetrable or not” (Nation and Ethnicity, 202). In my view, whether practical or ideological in its impetus, or genocidal or political in its intentions, global racism is defined by essentialization and dehumanization of other groups in the pursuit of power. This definition of racism is consistent with Zhang’s rhetoric, and not incompatible with his anti-imperialist project.

26 Chow, “Imagining Boundaries,” 47.

27 Hon, Tze-ki, Revolution as Restoration: Guocui Xuebao and China’s Path to Modernity (Leiden: Brill, 2013), 6 CrossRefGoogle Scholar; Liu, Lydia, Translingual Practice: Literature, National Culture, and Translated Modernity (Stanford: Stanford University Press, 1995)CrossRefGoogle Scholar.

28 Taiyan, Zhang (Zhang Jiang), “Xin fangyan,” Guocui Xuebao 4, 6 (1908): 4755, 52Google Scholar.

29 Schneider, Nation and Ethnicity, 170–208.

30 Taiyan, Zhang, “Zhonghua Minguo Jie (Explaining the Republic of China),” Par Cassel, trans., Stockholm Journal of East Asian Studies 8 (1997[1907]): 1540 Google Scholar.

31 Peacock, “Intersecting Nations,” 31–35.

32 Zhang took inspiration from Fredrich Max Müller, who argued that the human tendency to describe the unknown through metaphor resulted in a “disease of language.” According to Lin Shaoyang, Zhang viewed the evolution of Chinese written language similarly, in which a “crisis of representation” emerged as characters stretched beyond their original meaning. See Shaoyang, Lin, “Xixue xiangyu zhong de Zhang Taiyan ‘yinshen’ gainian xinjie: yu qi ‘wen’ lun, yuyan sixiang de guanxi,” Hangzhou shifan daxue xuebao 2 (Mar. 2020): 6078 Google Scholar. See also Kobayashi Takeshi, Zhang Taiyan yu Mingzhi sichao (Shanghai: Shanghai renmin chubanshe, 2018).

33 Taiyan, Zhang (Zhang Jiang), “Xin fangyan zixuGuocui Xuebao 3, 9 (1907) 5869; 58Google Scholar. The entirety of Xin Fangyan, including its afterwords and addendums, can also be read in Zhang Taiyan, Zhang Taiyan quanji, vol. 4 (Shanghai: Shanghai renmin chubanshe, 2018), 1–129.

34 Elman, Benjamin, From Philosophy to Philology: Intellectual and Social Aspects of Change in Late Imperial China (Cambridge: Harvard University Press, 1984), 89110 Google Scholar; Sela, Ori, China’s Philological Turn (New York: Columbia University Press, 2018)Google Scholar. Qian Daxin in particular stressed how fangyan could help scholars trace oral language change. Behr, Wolfgang, “Language Change in Premodern China: Notes on Its Perception and Impact on the Idea of a ‘Constant Way,’” in Helwig Schmidt-Glintzer, Achim Mittag, and Jörn Rüsen, eds., Historical Truth, Historical Criticism, and Ideology: Chinese Historiography and Historical Culture from a New Comparative Perspective (Leiden: Brill, 2005), 28 Google Scholar.

35 Not all scholars agree that mid-Qing scholarship constituted diachronic linguistic study. See Behr, “Language Change”; and Benjamin Elman’s review, T’oung Pao 96 (2010): 232–34.

36 Zhang, “Xin fangyan zixu,” 58.

37 Zhang, “Xin fangyan zixu,” 58–59; Shaoyang, Lin, Dingge yiwen: Qingji geming yu Zhang Taiyan fugu de xinwenhua yundong (Revolution by Means of Culture: The Late Qing Revolution and Zhang Taiyan from 1900 to 1911) (Shanghai: Shanghai renmin chubanshe, 2018), 122 Google Scholar; Elisabeth Kaske, Politics of Language in Chinese Education, 1895–1919 (Leiden: Brill, 2008), 362–66.

38 Taiyan, Zhang. “Guoyuxue caochuang xu” (Introduction to the creation of national language studies) Repr. in Guogu 3 (1919): 1112 Google Scholar.

39 Zhang, “Xin Fangyan zixu,” 59.

40 Rui, Wang, Ziguo Zixin—Zhang Taiyan yu Zhongguo chuantong sixiang de gengsheng (Shanghai: Shangwu chubanshe, 2019)Google Scholar. Zhang was not the first to argue this; for centuries, southern scholars emphasized that their mother tongues best preserved the oral language of past dynasties. According to Hirata Shoji, these arguments are rooted in the imperial examination system, as success on the exam required detailed knowledge of Tang-dynasty rime tables, which, generally speaking, better accorded to oral languages spoken in the south. The use of these same rime tables from the Song to the Qing reinforced the belief that their phonological categories—and the southern fangyan that consistently matched them—as having scholarly value. Hirata Shoji, Wenhua zhidu, 254.

41 Zhang Taiyan, “Qiushu: Chongding ben—fangyan” (Book of urgency: revised edition—fangyan) in Qiushu: Chukeben, Chongdingben (Shanghai: Zhongxi shuju, 2012), 176.

42 Kaske, Politics of Language, 357–58; Fenghua, Sun, Zhang Taiyan, Lian Huang minzu wenhua sixiang zhi bijiao (Jiuzhou Press, 2013), 127–30Google Scholar; Schneider, Nation and Ethnicity, 169–70; Lin, Dingge, 128; Yoshida, Kauru, “Zhang Taiyan yu ‘Xia yin,’Zhongguo xiandai wenxue yanjiu congkan 1 (2018): 167–77, 171–73Google Scholar.

43 Zhang, “Qiushu,” 173–74.

44 Ibid., 176; Zhang Taiyan, “Chongkan <Guyun biao zhun> xu” (Introduction to the revised edition of the <standard of ancient phonology>). Taiyan wenlu chubian (Shanghai: Shanghai Renmin Chubanshe, 2014[1915]), 209.

45 Liu Shipei, “Xin fangyan houxu,” in Zhang Taiyan, Xin Fangyan, 113–16.

46 Yurou, Zhong, Chinese Grammatology (New York: Columbia University Press, 2019), 3034 Google Scholar.

47 Zhang was not opposed to a phonetic system altogether, just the Roman alphabet. Indeed, he invented his own phonetic system based on seal script and rooted in the Shuowen. Kuzuoglu, Ulug, “Codes of Modernity: Infrastructures of Language and Chinese Script in an Age of Global Information Revolution” (PhD diss., Columbia University, 2018), 119–21Google Scholar.

48 Zhang Taiyan (Zhang Jiang), “Bo Zhongguo yong Wanguo Xinyu shuo” (On disputing the use of Esperanto in China), Minbao 21 (1907): 52.

49 Kaske, Politics of Language, 378–82.

50 Zhang, “Bo Zhongguo,” 49–72.

51 Ibid., 340.

52 Zhang made this speech sometime between 1907 and 1910. “Zhongguo wenhua de genyuan he jindai xuewen de fada” (The origin of Chinese culture and the development of modern knowledge), Zhang Taiyan quanji, vol. 14 (Shanghai: Shanghai renmin chubanshe, 2018), 74–79, 75.

53 Oral standardization was only one part of language reform; there were also debates about how and whether to standardize script and prose. See DeFrancis, John, Nationalism and Language Reform in China (Princeton: Princeton University Press, 1950)Google Scholar; Zhong, Chinese Grammatology; Li Jinxi, Guoyu Yundong Shigang (Shanghai: Shangwu Yinshuguan, 1934).

54 Jeffrey, Weng, “What Is Mandarin? The Social Project of Language Standardization in Early Republican China,” Journal of Asian Studies 77, 3 (2018): 611–33Google Scholar.

55 Ibid., 611.

56 Kaske, Politics of Language, 371–72.

57 Weng, “What Is Mandarin?” 621–25.

58 Li Jinxi, Guoyu Yundong Shigang, 55–56.

59 Esherick, “How the Qing Became China,” 247.

60 Weiner, Benno, The Chinese Revolution on the Tibetan Frontier (Ithaca: Cornell University Press, 2020), 14 Google Scholar.

61 In some cases, Han supremacy was more direct. Matthew Wong Foreman, “Racial Modernity in Republican China, 1927–1937,” Asian Ethnicity (2020): 377–97.

62 Peacock, “Intersecting Nations,” 54.

63 Zarrow, Peter, Educating China: Knowledge, Society and Textbooks in a Modernizing World, 1902–1937 (Cambridge: Cambridge University Press, 2015)CrossRefGoogle Scholar; James Leibold, “Competing Narratives of Racial Unity in Republican China: From the Yellow Emperor to Peking Man,” Modern China 32, 2 (2006): 181–220; Zarrow, After Empire, 212–26.

64 Weng, 624; Zhu Linggong, ed., Guoyu wenti taolun ji (A discussion of the problems of the national language) (Shanghai: Zhongguo shuju, 1921); Li Jinxi, “Guoyu san sa gang ji Guoyin zhi wu da wenti” (Three grand proposals for the national language and five big problems with national pronunciation) Shishi xinbao, 14 Oct. 1920: 1.

65 As one example, see Mareshi, Saito, “Liang Qichao’s Consciousness of Language,” in Fogel, Joshua A., ed., The Role of Japan in Liang Qichao’s Introduction of Modern Western Civilization to China (Berkeley: Institute of East Asian Studies, University of California, 2004), 247–71Google Scholar.

66 Yuen Ren Chao, Yuen Ren Chao, Chinese Linguist, Phonologist, Composer and Author. Interview by Rosemary Levenson. Berkeley: Regional Oral History Office, Bancroft Library, University of California, 1977, 81.

67 Zuoren, Zhou, “Guoyu gaizao de yijian” (Opinions on the improvement of the national language) Dongfang Zazhi 17 (1936): 715 Google Scholar.

68 Zhaozu, Rong, “Zhengji fangyan de wojian” (My opinions on collecting fangyan), Beijing daxue rikan 35 (1923): 14, 2Google Scholar.

69 Hui, Wang, The Politics of Imagining Asia, Huters, Theodore, ed. (Cambridge: Harvard University Press, 2011), 106 Google Scholar.

70 Changpei, Luo, “Zhongyuan yinyun shenglei kao” (Study of the initials in the Zhongyuan yinyun), Zhongyang lishi yuyan yanjiuyuan jikan 2, 2 (1932): 423–40Google Scholar; Dong Tonghe, “Shanggu yinyun biaogao” (Draft tables for ancient Chinese phonology), Zhongyang lishi yuyan yanjiuyuan jikan 18 (1948): 1–249.

71 Wong Shik-ling, Yue yun yin hui (A Chinese Syllabary Pronounced According to the Dialect of Canton) (Zhonghua Shuju, 1940).

72 Ibid., 2.

73 For this section, I consulted the Zhongguo fangzhi ku database. I used keyword searches for the terms fangyan, tuyin, guyin, Yellow Emperor, Hanzu, and Hanzhong, as well as others, to locate post-1911 gazetteers on relevant topics. After exhausting keyword searches, to ensure I captured geographic patterns, I read several gazetteers from various localities at random, privileging provinces that were less represented in keyword searches. I also looked at samplings from the Qing dynasty to trace change over time. Altogether, I consulted approximately 130 gazetteers.

74 “Xiuzhi shi lie gaiyao” (Guidelines for compiling local gazetteers) Zhejiang minzheng rikan 7 (14 Jan. 1930): 25; “Difangzhi shu zouzhi banfa” (Methods for compiling local gazetteers) Zhongyang ribao 2 (15 May 1944): 5; Chenzhi Wang, “Chinese Local Gazetteers: Evolution, Institutionalization, and Digitization,” Journal of East Asian Libraries 149 (Oct. 2009): 45–54, 45–47.

75 Dennis, Joseph, Writing, Publishing, and Reading Local Gazetteers in Imperial China, 1100–1700 (Cambridge: Harvard University Press, 2015), 3Google Scholar; May-bo Ching, “Classifying Peoples: Ethnic Politics in Late-Qing Native-Place Textbooks and Gazetteers,” in Tze-ki Hon and Robert Culp, eds., The Politics of Historical Production in Late Qing and Republican China (Leiden: Brill, 2007), 55–78, 58.

76 For references to nation-building goals, see Ren Yaoxian and Zhang Guishu, Fushan xianzhi (Fushan county gazetteer), 42 vols. (1935), 1; Chen Zhisun and Yang Sicheng, Songming xianzhi (Songming county gazetteer), 38 vols. (1945), 660; Chongxiu zhenyuan xianzhi (Newly edited Zhenyuan county gazetteer), 19 vols. (1935), 70; Zhu Shide, Junlian xianzhi (Junlian county gazetteer), 7 vols. (1948), 123; Deng Zhanxian, Xuxiu Wusu xianzhi (New edition Wusu county gazetteer), 2 vols. (1921), 99; Na Ruzhen and Jiang Shifang Zhenkang xianzhi (Zhenkang county gazetteer) (1936), 260. See also Ren Liao and Chen Tingting, “Mingqing yu minguo shiqi Guizhou fangzhi Zhong minsu shuxie bianhua yanjiu,” Hebei beifang xueyuan xuekan 33, 6 (2017): 24–28, 26.

77 Examples include Wang Weiliang and Liao Liyuan, Mingxi xianzhi (Mingxi county gazetteer), 15 vols. (1943), 133–256; Zan, Huang and Ruzhen, Zhu, Yangshan xianzhi (Yangshan county gazetteer), 18 vols. (1938), 397411 Google Scholar; and Zhang Han and Qiu Fu, Shangkang xianzhi (Shangkang county gazetteer), 36 vols. (1939), 985–88. In Qing dynasty gazetteers, the Yellow Emperor only appeared as a story cited in the Shiji (史記), or as the namesake of monuments. For one example, see Gu Renji and Shen Chengguo, Shangkang xianzhi (Shangkang county gazetteer) (1761), 3326–28.

78 Chen Bojia and Li Chengjun, Chongxiu Runan xianzhi (Refurbished Runan county gazetteer), 22 vols. (1938), 1186.

79 Gan Renjun, Chongde xian xin zhi (New Chongde county gazetteer) 18 vols. (n.d.), 1.

80 Li Lan, Deng Yawen, and He Ling found around 10 percent of post-1911 gazetteers included a language section: “Difangzhi zhong yuyan cailiao jilu qingkuang gaishu” (A summary of the compilation of linguistic materials in gazetteers) Journal of Changshu Institute of Technology (Philosophy & Social Sciences) 1 (2019): 102–9, 103.

81 Zhihong, Zhu and Chu, Xiang, Baxian zhi (Gazetteer for Ba county), 23 vols. (1939), 801–2Google Scholar; Chen Zhisun and Yang Sicheng, Songming xianzhi (Songming county gazetteer), 38 vols. (1945), 522.

82 Xu Zhaoyu and Yang Zhaotai, Xinfeng xianzhi (Xinfeng county gazetteer), 10 vols. (1929), 81.

83 Fu Tingquan, Shaotong xianzhi, 388–89.

84 Long Yunchang and Zhou Zhongyu, Xinzuan Yunnan Tongzhi (New Yunnan gazetteer), 266 vols. (1949), 6664.

85 Baozi, Yu and Yitian, Yang, Wenxi xianzhi (Wenxi county gazetteer), 25 vols. (1919), 147 Google Scholar.

86 Yimi, Yu and Shi, Bao, Wuhu xianzhi (Wuhu county gazetteer), 60 vols. (1919), 120 Google Scholar; Wang Zushi, Taicang zhouzhi (Taicang prefecture gazetteer), 28 vols. (1919), 105.

87 The format mimics Yang Xiong’s Fangyan. Wang Zhongxiu and Yan Fengwu, Mi xianzhi, (Mi county gazetteer), 20 vols. (1924), 241.

88 Riao, Xia and Ren, Wang, Jiande xianzhi (Jiande county gazetteer), 15 vols. (1919), 176–78Google Scholar.

89 Zuan, Xie and Yuxian, Su, Yichun xianzhi (Yichun county gazetteer), 24 vols. (1940), 9991001 Google Scholar.

90 Wang Zushi, Taicang zhouzhi (Taicang county gazetteer),101–14. See also Jiang, Qian and Yugui, Fan, Jianning xianzhi (Jianning county gazetteer), 28 vols. (1919), 212 Google Scholar.

91 Dalu, Wang and Jixi, Lai, Chixi xianzhi (Chixi county gazetteer), 8 vols. (1920), 153 Google Scholar.

92 Hirata sees a direct connection between these gazetteer sections and southern intellectuals who protested using guanhua to teach canonical texts since “their own languages were direct descendants” of Tang dynasty rime dictionaries (Hirata Shoji, Wenhua zhidu, 254–55).

93 The history of extracting power from ownership of or connection with ancient texts dates to at least the Song dynasty. See Baldanza, Kathleen, Ming China and Vietnam: Negotiating Borders in Early Modern Asia (Cambridge: Cambridge University Press, 2016)CrossRefGoogle Scholar.

94 Zhang Taiyan, “Zhonghua Minguo Jie,” 1907.

95 I counted thirty-seven that used guyin as a referent.

96 Tianrui, He and Xiu, Gui, Xining xianzhi (Xining county gazetteer), 34 vols. (1937), 168 Google Scholar; Zhou Zhenlin and Liu Zongxiang, Ningxiang xianzhi (Ningxiang county gazetteer) (1941), 1838.

97 In some cases, this referred to the “Zhongzhou zhengyin,” or “Zhongyuan yayin,” which implied the language spoken during later dynasties in the northern areas. See Jiang, Qian and Yugui, Fan, Jianning xianzhi (Jianning county gazetteer), 28 vols. (1919), 212 Google Scholar; and Xu Ganxiu and Wang Guoxian, Qiongshan xianzhi (Qiongshan county gazetteer), 28 vols. (1918), 117. But in others, it referred to the language of Zhongyuan in earlier time periods, such as the “Tang dynasty”; Zhongyuan. Xining Xianzhi (1937), 182. Other times, which time period’s Zhongyuan was referred to remained unclear; Hu Luoling, “Yi su xiao ji” (A small record of the Yi vernacular), in Wu Kejun and Cheng Shoubao, Yixian si zhi (Fourth Yi county gazetteer), 16 vols. (1923), 102; Kun, Chen and Qian, Liu, Liling xianzhi (Liling county gazetteer), 10 vols. (1948), 1771 Google Scholar.

98 Chuansha Xianzhi (Gazetteer for Chuansha county), 24 vols. (1936), 1028–35.

99 Hirata Shoji emphasized that these competing claims over which language denoted zhengyin had existed since at least the Yuan dynasty and increased in the Qing amidst imperial efforts to enforce Guanhua as the official standard. Hirata Shoji, Wenhua zhidu, 247–50.

100 Anhui tongzhi (1934); Dinghai xianzhi (1924); Yaoan xianzhi (1948); Xianshan xianzhi (1926); Xinxiu Fenshun xianzhi (1943); Chongxiu Nanchuan xianzhi (1926); Xinzou Yunnan tongzhi (1949); Tai xianzhi gao (n.d.); Fengtian tongzhi (1934); Linqi xuzhi (1935); Ba xianzhi (1939); Liling xianzhi (1948); Zhaotong xianzhi (1924); Shiping xianzhi (1938); Gaoyi xianzhi (1933); and Qingyuan xianzhi (1937).

101 Junying, Zhou and Qianqian, Liu, Linqu xuzhi (Supplemental Linqu county gazetteer), 22 vols. (1935), 858 Google Scholar.

102 Laifeng, Pan and Yinshan, Wang, Xuxiu Guangshao xianzhi (Re-edited Guangshao county gazetteer), 28 vols. (1935), 453 Google Scholar.

103 The Hakka are notably absent in earlier Dabu county gazetteers, such as Zhang Hongen, Dabu Xianzhi (Dabu county gazetteer) (1873).

104 Constable, Nicole, Guest People: Hakka Identity at Home and Abroad (Seattle: University of Washington Press, 1996), 4147 Google Scholar.

105 Xianglin, Luo, Kejia yanjiu daolun, Repr. (Shanghai: Shanghai chubanshe, 1992), 79 Google Scholar.

106 Liu Zhichao and Xiu Wenting, Dabu xianzhi (Dabu county gazetteer), 39 vols. (1943), 1076.

107 Ibid.

108 Ibid., 1328–29.

109 Kun, Chen and Qian, Liu, Liling xianzhi (Liling county gazetteer), 10 vols. (1948), 1Google Scholar.

110 Ibid., 1769.

111 They claim their language belongs to the Guanhua fangyan region. Chen and Liu, Liling xianzhi, 1771.

112 Other examples include Fengsheng, Xu and Ruzhen, Zhu, Qingyuan Xianzhi (Qingyuan county gazetteer), 20 vols. (1937)Google Scholar; Dinfen, Liang and Renchang, Ding. Panyuxian xuzhi (Panyu county gazetteer), 44 vols. (1931)Google Scholar; Yicheng, Zhang, Liang Guanxi Yangjiang zhi (Yangjiang county gazetteer), 39 vols. (1925)Google Scholar; Xu, and Wang, , Qiongshan Xianzhi (Qiongshan county gazetteer) 28 vols. (1917)Google Scholar; Xu and Yang, Xinfeng Xianzhi (Xinfeng county gazetteer); Pu Youwen and Wang Guohui, Danxianzhi (Dan county gazetteer), 18 vols. (1934).

113 Roche, Gerald Lexical Necropolitics: The Raciolinguistics of Language Oppression on the Tibetan Margins of Chineseness,” Language & Communication 76 (2021): 111–20CrossRefGoogle Scholar.

114 Weiner, Chinese Revolution, 161–80. Dru Gladney, “Internal Colonialism and the Uyghur Nationality: Chinese Nationalism and Its Subaltern Subjects,” Cahiers d’études sur la Méditerranée orientale et le monde turco-iranien 25 (1998): 47–64.

115 Tuttle, Gray, “China’s Race Problem,” Foreign Affairs 94 (2015): 39–46, 39 Google Scholar.

116 Wang, Zheng, Never Forget National Humiliation: Historical Memory in Chinese Politics and Foreign Relations (New York: Columbia University Press, 2014)Google Scholar.

117 Çaksu, Ali, “Islamophobia, Chinese Style: Total Internment of Uyghur Muslims by the People’s Republic of China,” Islamophobia Studies Journal 5, 2 (2020): 175–98, 180CrossRefGoogle Scholar.

118 Tam, Dialect and Nationalism in China, 186–218.

119 Wang, Hui and Zhongrui, Yuan, “The Promotion of Putonghua (Mandarin Chinese): An Overview,” In Yuming, Li and Wei, Li, eds., The Language Situation in China (Berlin: De Grueyter, 2013), 2739 Google Scholar.

120 Masha Borak, “China’s Version of TikTok Suspends Users for Speaking Cantonese,” South China Morning Post, 3 Apr. 2020,

121 Tania Branigan, “Protestors Gather in Guangzhou to Protect Cantonese Language,” Guardian, 25 July 2010.

122 Xiaoling, Zhang, and Guo, Zhenzhi, “Hegemony and Counter-Hegemony: The Politics of Dialects in TV Programs in China,” Chinese Journal of Communication 5, 3 (2012): 300–15Google Scholar.

123 Shin Watanabe, “China Arrests 130 Ethnic Mongolians over Language Policy Protests,” Nikkei Asia, 15 Sept. 2020,

124 Chris Buckley, “A Tibetan Tried to Save His Language: China Handed Him a Five-Year Sentence,” New York Times, 22 May 2018,

125 Li, Yao, “A Zero-Sum Game? Repression and Protest in China,” Government and Opposition 54, 2 (2019): 309–35Google Scholar.