Summary: Language education in the Late-Qing Jirim League

The Jirim League, like the other Inner Mongolian Mongols, consisted of banners: administrative units established by the Qing rulers between 1634 and 1670. Most of the Jirim banners administered Khorchin Mongols, the others Gorlos, Dörbed, and Jalaid communities. ‘Banner chiefs’ (Mo. ǰasaγ) of local origin but imperially approved headed the banners, and the Military Governors of the “three eastern provinces” (Ma. dergi ilan golo) of Manchuria supervised them according to their geographic location.

The social and economic makeup of Manchuria and Inner Mongolia were in great flux in the period covered in the book, as large numbers of Chinese settlers were moving into the area. At the same time, Manchuria was a focal point of Russo-Japanese imperial rivalry (and in the 1904–1905, war). Both of these factors compelled the Qing authorities to act to assert their sovereignty and maintain order. Meanwhile, developments in Beijing also affected local policies in the Jirim League and Jilin Province.

At the height of their power in the seventeenth and eighteenth centuries, the Qing monarchs had aspired to a kind of siloing of populations into different status categories (bannermen, commoners or civilians, different categories of Mongols, and so on). The groups had different obligations and privileges and were supposed to be linguistically and culturally distinct. These communities were to the degree possible administered separately, in different languages and combinations of languages. The Manchu monarchy, in displays of control and cultural largesse directed at the politically most significant elite communities (including Chinese literati and Mongol nobles), produced books and erected stelae in different languages. In the minds of its rulers, the empire both needed and nurtured a range of linguistic competencies.

Over time, migration, economic integration, and demographic trends put stress on Qing institutions and administrative practices. In the Jirim League, as elsewhere in Inner Asia, new administrative units of the commanderies-and-counties (junxian 郡縣) type were needed to administer and extract taxes from the Chinese settlers. Similarly, the increasing number of legal cases involving individuals from different linguistic communities posed difficulties for officials used to working in only one language (90–94).

In this context, the “three eastern provinces” were restructured in 1907 to match those of China proper, and were headed by Governors and Governors-General. “The Mongolian affairs of the ten banners of the Jirim League” were placed under the authority of the Governor-General of Fengtian, in Mukden (92). As the Manchu government, under the influence of Chinese reformers, tried to transform the imperial state into a constitutional monarchy supported by a literate and informed citizenry, it encouraged Chinese literacy among the Mongols, at one point even disenfranchising those who could read only in Manchu or Mongolian (150). “The empire,” He writes, now “prioritized integration over diversity” (153).

However, these ambitions were simply not realistic. The Jirim Mongols could not be taught to speak and read Chinese overnight; and the attempt might dangerously alienate them. The provincial authorities set out to make citizens out of the Mongols by promoting reading in their own language, including by publishing Mongolian newspapers (169–70).

Manju monggo nikan ilan acangga šu-i tacibure hacin-i bithe | Manǰu mongγol kitad γurban neičetü udq-a-yin surγaqu ǰüil-ün bičig | Man-, Meng-, Han-, san wen hebi jiaokeshu 滿蒙漢三文合璧教科書, the rare trilingual textbook that is at the center of He’s study, encapsulates the tensions of the provincial authorities’ new policy. The textbook, first published in 1909 and subsequently revised, was translated from a Chinese textbook published by the Commercial Press in Shanghai for the empire’s new-style schools. The book presented Chinese culture and China Proper as the empire’s foundation, with little room for separate Manchu and Mongolian linguistic and cultural practices. The sanitized version of Chinese culture espoused in the textbook was paired with concepts and things considered useful in the modern world, such as the telephone, the world’s climate zones, and so on (175–76).

Using documents culled from an array of published and unpublished archival corpora, He discusses how new curricula, including classes in Mongolian and Manchu, were then implemented not only in the Jirim League, but also in areas under direct provincial administration in Jilin (Chapter 6). This discussion contains some precious information on how the Manchu language in particular was taught in the late Qing period (248–50). The trilingual textbook did not prove very popular as teaching matter, with some teachers even coming up with implausible excuses for not using it (236). Indeed, the Qing pedagogical paradigm proved quite tenacious, as many teachers apparently preferred to teach Mongolian and Manchu using bi- or trilingual pedagogical texts published in the eighteenth and nineteenth centuries. Even so, the trilingual textbook survived the Qing to be revised for use in the early Republic. Prominent educator Cai Yuanpei endorsed it, and its translator, Rongde, received a medal from President Yuan Shikai (267–68). The early Republic made continued attempts to educate the Mongols in Mongolian before losing control over them.

Ruling the Mongols of Manchuria consists of an introduction, conclusion, and six chapters: 1) “Kamcime: Ruling a Polyglot Empire” (on Qing multilingualism), 2) “The Linguistic Scene” (the language situation in the Jirim League), 3) “The Literacy Question” (the perceived urgency of literacy education and language reform in late-Qing China), 4) “Literate in What Language” (literacy education in the Jirim League and the role of Manchu in its linguistic configuration, as reflected in the trilingual textbook),Footnote ¹ 5) “Reimagining China and the World” (history and geography in the trilingual textbook), and 6) “Trilingual Practice in the Jirim League and Manchuria” (education in the League as well as in parts of Jilin Province following the publication and distribution of the textbook). Chapter 6 is the strongest, but Chapter 5 is also solid (its discussion of the textbook’s contents is a presentation of boilerplate late-Qing ideology that is unsurprising, but perhaps necessary for the argument).

In the book’s conclusion, He compares the linguistic order of the Qing Empire with that of the Ottomans, and brings the story of Mongolian education in Inner Mongolia up to near-contemporary times. The conclusion would have been stronger had she explored the ramifications of the different fates of the Ottoman and Qing empires in the twentieth century. Whereas the Ottoman Empire collapsed, the government of the People’s Republic of China is currently working to transform large parts of the old Qing Empire into a Chinese nation-state through linguistic and cultural assimilation, with great consequences for Mongolian-language education (not to mention other vulnerable minority nationalities). He’s statement in the introduction that “Mongolian language teaching is still considered essential by local communities and the central government in the 2020s” (20) ignores this important political reality.

Mistakes that do not affect the big picture

I do not want to lump all of the problems that I find in He’s book into the kind of laundry list of errors large and small that reviewers—in bad faith—sometimes use to obliquely discredit an author’s argument and dismiss their scholarly contributions. We all make mistakes and many do not matter in the grand scheme of things. Yet it is important to identify such mistakes.

In addition to quite a number of typos and linguistic infelicities,Footnote ² there are some problems in the treatment of foreign-language material. The names Ye Gaoshu, Shen Huaiyu, and Yu Yingshi are romanized contrary to the conventions adopted in the book; Shengjing is often referred to as such and not as “Mukden,” as the conventions adopted dictate; many simplified Chinese characters are used, again contrary to conventions; the PRC simplified character set is used inside the titles of Japanese articles and books (110); in several instances, yu 於 and shen 瀋 are used to write the surnames Yu 于 and Shen 沈; in several book titles, hui 匯 is used for hui 彙, and zhi 制 for zhi 製; bakši should read baksi (55); Gagai should read G’ag’ai (55); “Kanto” should read “Kantō” (109); Altai should read Alt’ai (< 알타이) (173); Zhigao 志高 should read Zhi-kuan 志寬 (190); goro should read golo (199); yanjiusuo jikan should read yanjiu ziliao jikan, unless it is the Chinese characters that are wrong (160); the titles of Japanese articles and books are not romanized; Russian titles are sometimes romanized, sometimes given in Cyrillic; Mongolian book titles and italicized words mix the Mostaert-Vladimirtsov system in its canonical form with a modified version without carons and with gh substituted for γ; Manchu-language forms are not given for the names of all Manchu individuals for whom they are known (e.g., I hin for Yixin, Guiliyang for Guiliang, and Žungde for Rongde 榮德);Footnote ³ and, finally, kamcime, a Manchu verb form that cannot modify nouns, is frequently used as such a modifier (e.g., “kamcime texts,” “kamcime style,” “kamcime system”), contrary to the recommendation that the editors of the Chicago Manual of Style make on their FAQ website regarding foreign-language phrases in English academic prose.

I turn now to problems with the actual contents of the book. Some very radical and unexpected statements are given without a citation, making them difficult to evaluate (e.g., that “the pronunciation and phonetic method” of Manchu are “similar to Indo-European languages” [185]). Some omissions risk confusing readers (e.g., both the banners of the Jirim League and Mongol bannermen are mentioned in the same paragraph [226, 238], but He does not explain that “bannermen” refers to members of the Eight Banners rather than the banners in Mongolia, and the book’s index points to pages that discuss the Jirim League banners as well as the bannermen of the Eight Banners under the single entry “banner system”).

I find the following factual errors and mistranslations, listed in the order in which they appear in the book: The reform of the Manchu script in the 1620s and 1630s was not the work of Hong Taiji and Dahai alone (56); there is no single “Chinese version” of the Nurhaci Veritable Records, but several (56); relatedly, it is thus incorrect to say that “the Chinese text” of the Records omitted the “detailed linguistic discussion on Manchu’s adaptation from Mongolian,” since only one Chinese version did so (57); the terms Jiu Manzhou dang 舊滿洲檔 and Manwen laodang (Manbun rōtō) 滿文老檔 do not refer to the same textual corpus (57); the pre-conquest Manchu archive does not contain documents “solely written in Manchu,” but also a substantial amount of text in Mongolian (57); bithesi does not mean “official translator” but something like “ranked clerk” (70);Footnote ⁴ the court-sponsored Manchu dictionary from 1708 was not “polyglot,” nor was it a “Manchu–Chinese dictionary,” but was in Manchu only (72); its title was not Han-i araha toktobuha manju gisun-i buleku bithe (my emphasis) (72); the Manchu–Mongol dictionary Han-i araha manju monggo gisun-i buleku bithe from 1717 did not carry the Chinese title Yuzhi Man-Meng hebi Qingwen jian 御製滿蒙合璧清文鍳, nor any Chinese title, but the Mongolian title Qaγan-i bičigsen manju mongγol-u gen-ü toli bičig, which He does not give (72); the court-sponsored trilingual Manchu–Mongol–Chinese dictionary did not carry Manzhou 滿洲 in its title, but Manzhu 滿珠 (in both the block-printed edition and the slightly later Siku quanshu manuscript) and this book was published not in 1771, but in 1780 (72); the compilation of the five-language version in the same series of dictionaries did not “dr[i]ve polyglot printing in the Qing dynasty to its peak,” since it was never printed (73); Ilarion Kalinovich Rossokhin’s translation of Qingwen qimeng 清文啓蒙 was not the “earliest Manchu grammar book in Europe” (105); Manbun shūin 滿文輯韻 was never published, but remained in manuscript (108); what Naitō Konan photographed in 1905 was a Qianlong-era re-edition of the pre-conquest Manchu archive, not simply the “Old Manchu Archive,” its later publication under that title notwithstanding (109); the bibliography gives no date or publication data for Qinding Shengjing tongzhi 欽定盛京通志 (119); qieyin 切音 “phonetic spelling” should not be translated as “Romanization” (136 et passim); in the translated sentence “The degree of difficulty of a language is directly proportional to the spread of education,” the terms should be reversed (144); four references to a 1934 Commercial Press edition of Li Jinxi’s 黎錦熙 Guoyuxue jiangyi 國語學講義 on pages 144–45 appear to actually refer to Li’s Guoyu yundong shigang 國語運動史綱, which is also cited elsewhere in the same chapter (I can find no edition of Guoyuxue jiangyi from 1934 in any library catalogue, and the earlier edition at my disposal does not contain the text cited, whereas the 1934 Commercial Press edition of Guoyu yundong shigang does contain the text, but on a different page range); if the extracts on these pages are from Guoyu yundong shigang, then the one straddling pages 144–45 should be presented as Li’s paraphrase, rather than a statement from a late-Qing government ministry (Li has no quotation marks around this passage, but does for other passages from the same statement); it is not correct to translate shengmu 聲母 “syllabic initial” as “consonant” and yunmu 韻母 “rhyme”as “vowel”; “August 4” should read “August 14” (169); Qianlong’s Manchu–Chinese dictionary does not date from 1771 but from 1772, for the preface is dated Qianlong 36/12/24, which fell in 1772, and some scholars have suggested that printing was probably finished in 1773 (175); the “contemporary Manchu” (Sibe?) word gisutun “telephone” is not “a combination of gisun (language) and tun (island)” but a so-called blend of gisu(n) and (soni)tun “machine,” a type of coinage frequent in Manchu going back to the eighteenth century (176); Rongde’s Manchu translation šang hai mederi hūdai sita bithe šuwaselara kuren for Shanghai shangwu yinshuguan 上海商務印書館 does not “completely [lose] the meaning of the original Chinese term” but represents a consistent practice (in this phrase, mederi does not modify hūdai sita to mean “maritime business affairs,” but constitutes the head of the noun phrase šang hai mederi, which is how Rongde refers to Shanghai, for which see footnotes 35 and 36 on page 204, where Shanghai is again translated as šang hai mederi, along with han keo angga for Hankou 漢口 [angga “mouth, pass” translates kou 口] and tai hū tenggin for Taihu 太湖 [tenggin “lake” translates hu 湖]). The title of the trilingual textbook is not an unconventional translation of the Chinese title (why would the Chinese title take precedence?), but is appropriate in light of the content of the book as well as earlier precedent (šu is short for gurun šu “national culture [or literature],” Rongde’s translation of the novel concept of guowen 國文; acambi is used instead of kamcimbi because Rongde is not referring to three scripts on the same page, but to three “literatures” being placed together, just as “three kinds of pronunciation” [ilan hacin-i mudan] are said to have been “placed together” [acaha] in the title of Qianlong’s Manchu–Mongol–Chinese dictionary from 1780, for which see above; [177–78]);Footnote ⁵ we are thus shown no evidence that Rongde’s Manchu text contains “mistranslations” that represent “careless mistakes,” nor any evidence that “for those who have a command of Manchu, Rongde’s writing may be unclear and weird” (177–78); the demonstration of the fanqie 反切 syllabic spelling method on pages 184–85 implies that gong 公 should be read *góng, which is incorrect (the example should be explained in reference to Middle Chinese); Jean-François Gerbillon did not publish Elementa linguæ tartaricæ in 1696—the book was written by Ferdinand Verbiest in China, probably in 1677–78, and published in Paris without his or Gerbillon’s involvement in 1687 and then again in 1696 (185); the translated quote from Charles de Harlez on page 186 contains several mistakes (the passage should be in present tense, not simple past; “Manchu people” should read “Tartar bands in Manchuria”; “reigning in China at the time” should read “currently rules over China”; “being the official language of the Beijing court” should read “it is one of the official languages of the Beijing court”; “used to sign treaties” should read “serves to compose the treaties”); no French or German dictionary of Manchu was published in the early nineteenth century (Stanislas Julien’s Manchu–French vocabulary from sometime before 1824 remained in manuscript and Hans Conon von der Gabelentz’s Manchu–German dictionary was published in 1864) (186); The Manchu golo “region, province” does not mean “tribe” (199 and 200); “twelve Mongolian initials” refers to twelve finals (236); and “red banner” should read “rainbow flag” (267).

I turn now to more interesting and more consequential issues, both of which concern the narrative and framing in Ruling the Mongols of Manchuria. One concerns Mongolian, with implications for our understanding of language education in the late-Qing Jirim League. The other concerns two Manchu terms that He uses to characterize the Qing language order and its transformation in the late nineteenth and early twentieth century.

The issue of the Mongolian language

Ruling the Mongols of Manchuria does not give the Mongolian language the attention it deserves. The names of the key institutions governing the Jirim League Mongols appear only in Chinese and English, including the Bureau of Mongolian Affairs of the Three Eastern Provinces, which plays such an important role in He’s book, even though the trilingual textbook He discusses gives the Mongolian name (J̌egün γurban muǰi-yin mongγol kereg-i sidkekü qoriy-a).Footnote ⁶ The names of the Mongolian newspapers likewise appear only in English and Chinese, yet I easily found the Mongolian-language title of the one I decided to look for in the Library of Congress’s catalog.

The neglect of the Mongolian language makes the narrative a little unbalanced and lacking in nuance. He Jiani writes that “Guo (國) and gurun had various meanings about [sic] nation and state in Chinese, Manchu, and Mongolian” (163),Footnote ⁷ but she never says which Mongolian word she has in mind. Ulus? Törö?

He engages at length with Qu Liusheng’s discussion of the Manchu neologisms found in the trilingual textbook (175–77), but says nothing about the book’s Mongolian vocabulary. Yet the problems that the textbook served to address—per the Qing official Xiliang’s preface, at least—concerned Mongolian-language use and the Mongolian–Chinese interface. Xiliang claims that aside from “Indian documents” (Ma. enetkek bithe, Mo. enedkeg-ün bičig; Ch. Fanwen 梵文; in reality, probably Tibetan), the Mongols did not study anything. With few individuals literate in both Chinese and Mongolian, not only the Chinese documents but “also the meaning of the Mongol documents has often become muddled” (Ma. monggo bithei gūnin jurgan be, inu farhūn farhūn ojorongge labdu, Mo. mongγol bičig-ün sanaγ-a jirum-i, basaču [sic] qarangγui qarangγui bolqui anu ülemǰi, Ch. 蒙文意義亦多昧昧). Thus “among the translated documents in each banner, half have coarse words and phonetic transcriptions mixed in; no wonder then that the meaning does not come across!” (Ma. gere gūsai ubaliyambuha bithe de, dulin muwa gisun be suwaliyaganjame, ubaliyambuha mudan bime jurgan be hafumbuhakūngge be aide wakašambikai, Mo. olan qosiγun-u orčiγuluγsan bičig-tür, dumdaγur büdügün üge-yi qoličalduγun, orčiγuluγsan ayalaγu büged ǰirum-i ese nebteregülügsen-i inu yaγun-dur buruγušaγaqu bolai, Ch. 無怪各旗繙譯之文半雜俚語、音譯而義未達也).Footnote ⁸ He Jiani correctly points out that Manchu was an “intermediate” language in this context (170), a bridge connecting Chinese and Mongolian (172). Therefore, what we really need to know is how Mongolian vocabulary was affected by the trilingual textbook. Were the Mongolian words for new concepts and objects simply calques on the Manchu?

The book’s neglect of Mongolian extends to the language’s social context, and He’s discussion of literacy education in the Jirim League suffers a bit as a result. He remarks that “spoken Mongolian … became an indispensable teaching language because it was the Jirim Mongols’ native language” (168). Spoken Mongolian here probably refers to the Khorchin dialect of modern Mongolian. But the modern Mongolian vernaculars differ substantially from the linguistically conservative so-called script Mongolian that was used and taught in the Qing period. Moreover, He mentions that the “colloquial newspapers” carried articles in vernacular Chinese with Mongolian translations (169), but were these Mongolian translations also vernacular? He does not mention that her actors were aware of Mongolian diglossia, but it seems to me that Xiliang was. Might not the “rude” or “coarse” words (büdügün üge) that Xiliang identified refer to Khorchin dialect colloquialisms? If so, the stratification of the Mongolian language clearly affected literacy in the Jirim League, and He’s actors knew it.

The Qing language order and language ideology

Ruling the Mongols of Manchuria brings up the language order in the Qing Empire, the court’s language ideology, and the way in which they changed in the late-Qing period, centering the discussion on two Manchu terms: kamcime “to place close together, to be together” and gurun-i gisun “the language of the state.” I will discuss these terms in the remainder of this review.

First, what can the term gurun-i gisun tell us about the Qing language order? He Jiani discusses this term in Chapter 1 (56–57) and then uses it repeatedly throughout the book. He’s source for this phrase is the Manchu version of the Nurhaci Veritable Records, which presents the Manchu ruler as insisting that the Mongolian script be used to write Manchu, thereby creating a written language for the nascent polity. He writes that after that point, gurun “country, state” became a “tag [?] for written Manchu,” by which I think she means that Manchu was specified to have been described as the gurun-i gisun of the Qing empire. “The meaning of gurun-i gisun expanded gradually from ‘the Manchu script’ to ‘the Manchu language’ including spoken and written forms,” she writes.

Setting aside the fact that gisun “language, word, phrase” to my knowledge never means “script,” the real problem is that the Nurhaci Veritable Records does not actually use the phrase gurun-i gisun, nor does Ruling the Mongols of Manchuria quote it from any other source. I have read only a small fraction of the Manchu books and documents that were produced in the Qing period, so I cannot say that none includes the term gurun-i gisun. I have read enough, however, to know that it played no significant part in ideologically charged discussions of language in Qing court publications, in administrative documents dealing with language issues, or in language pedagogy. Clearing up this matter will affect understanding of the Qing language order.

In the Nurhaci Veritable Records, the sequence (not the phrase!) gurun-i gisun occurs in the two contrasting phrases musei gurun-i gisun “the language of our state” and monggo gurun-i gisun “the language of the Mongol state.” The analogous structure of the two phrases serves to highlight the contrast between the language of our (Manchu) state and the language of (their) Mongol state. Both phrases are noun phrases (NP), because the head of the phrase (gisun) is a noun (N), which determines what functions the phrase can fill in a sentence. Inside each of the noun phrases, another noun phrase is nested: musei gurun and monggo gurun respectively. The relationship between these nested noun phrases to the larger noun phrase is one of subordination; the first noun phrase modifies the second. In both cases, subordination is marked by means of the genitive case particle -i (GEN; musei gurun-i and monggo gurun-i are genitive attributes). We can thus say that the phrases are structured like this: [[musei gurun-i] gisun] and [[monggo gurun-i] gisun], or [[NP_GEN] NP].

These diagrammatic representations show that just because the sequence gurun-i gisun occurs in the text, we cannot extract it from the text as one unit, as it does not form one unit there. By the same token, had the Veritable Records been talking about “toy car manufacturers,” or [[toy car] manufacturers], it would be wrong to say that it was using the term “car manufacturer,” that is, [car manufacturer]. Making such a claim implies that the Veritable Records was talking about a [toy [car manufacturer]], i.e., a non-functional car factory used for amusement.

We can understand the structure of these phrases in no other way. My reasoning goes as follows. If we assert that the Veritable Records uses the term gurun-i gisun, we have to eo ipso also assert that the phrases musei gurun-i gisun and monggo gurun-i gisun are structured like [[musei] gurun-i gisun] and [[monggo] gurun-i gisun]. I do not dare to say that these structures are impossible, but I am confident to say that they are very unlikely. The reason has to do with the phrase monggo gurun-i gisun, which, as I mentioned, the thrust of the passage demands be analogous in structure to that of musei gurun-i gisun.

The two attributes musei gurun and monggo gurun are both themselves noun phrases consisting of a head noun and its attribute. In musei gurun, the attribute’s subordination to the head is again marked by means of the genitive case particle -i affixed to a pronoun (Pro; muse means “we” and musei means “our”). In monggo gurun, subordination of the nominal attribute is not marked using the genitive case particle -i, but through juxtaposition.Footnote ⁹ And here we get to the crux of the matter. Never have I seen a noun phrase with the structure [[N] N_GEN N] in Manchu, in which the head is a noun phrase that contains a genitive attribute and has a modifier consisting of a single noun whose subordination to the head is marked through juxtaposition. In other words, I have only seen phrases such as [[alban-i _GEN] hūdai _GEN niyalma] “official merchant” ([[N_GEN] N_GEN N]) and [[nomhon morin-i _GEN] fiyenten] “Equestrian Section” ([[N N_GEN] N]), and the like. Again, I cannot say that noun phrases with the structure [[N] N_GEN N] do not exist in Manchu, but if they do, they are rare, and we could assume that such a structure occurs here only if some special circumstance favored such a reading. He Jiani shows no such circumstance. This reading must be discarded, so monggo gurun-i gisun must have the structure [[N N_GEN] N]. Ipso facto, musei gurun-i gisun has the structure [[Pro_GEN N_GEN] N]. Both can be described at a higher level of abstraction as [[NP_GEN] NP].

If more proof were needed, consider the official Mongolian translation of the phrase monggo gurun-i gisun in the Mongolian version of the Veritable Records: mongγol ulus-un kele “language (lit. tongue) of the Mongol country.”Footnote ¹⁰ The Mongolian phrase has a structure entirely analogous to its Manchu counterpart, and it must be interpreted as [[N N_GEN] N]. Phrases of this type cannot be read in any other way, according to one authoritative grammarian.Footnote ¹¹ We can confidently conclude that the phrase gurun-i gisun does not occur in the Nurhaci Veritable Records.

Once we strip He’s narrative of the term gurun-i gisun, the Qing language order appears in a somewhat different light. In He’s argument, the term serves as a foil for the Chinese guoyu 國語, which in the late Qing “represented a nascent Chinese nation being forged by Chinese nationalists.” By contrast, in the phrase “gurun-i gisun … gurun referred to the state founded by the Manchus and characterized by cultural pluralism” (44). By this logic, the many early-to-mid-Qing texts that use the Chinese phrase guoyu in reference to Manchu use guo in this sense, to refer to this polity not as a nation-state characterized by monolingualism and a culturally uniform citizenry, but as a pluralistic empire. I will not argue that gurun did not have these connotations (on this matter I defer to others), but it is quite telling that guoyu in reference to Manchu, as far as I know, never translates gurun-i gisun, but only manju gisun “Manchu language.” That is to say, the Chinese phrase is used in reference to the language of a particular community, not a language of a pluralistic state.

For this reason, I have generally preferred to translate guoyu as “dynastic language” rather than “national language” or “state language.” To my eyes, guo here refers to the quite small community consisting of the Manchu imperial house, supplemented by their most loyal servants, the Manchu aristocracy and the bannermen (at least sometimes including the members of the Chinese and Mongol Eight Banners).Footnote ¹² It was particularistic, not pluralistic.

This conclusion brings me to the second term used to characterize the Qing language order in Ruling the Mongols of Manchuria: kamcime. He Jiani borrows this term from Pamela Crossley, who used it in 1999 in reference to what she called the “simultaneous” expression of Qing emperorship in “multiple cultural frames” (26). Whatever utility this term might have had in Qing historiography in the late 1990s, we need not revive it now. Crossley (correctly or not) was talking about the public-facing aspect of the Qing monarchy, not language use within Qing education and institutions. Yet it is precisely in this latter area that Ruling the Mongols of Manchuria makes its most important contributions.

The late-Qing changes in language education in the Jirim League and Jilin Province might very well have developed largely from the tension between a pluralistic imperial polity and a nascent Chinese nation-state—I am not denying that. But it is also clear from He’s narrative and her sources that some of the problems that warranted educational reform had bothered Qing officials and emperors repeatedly for a long time, such as improper language use in official documents and substandard translations (see my discussion of Xiliang’s preface to the trilingual textbook above). As He also makes clear, when officials were able to advance Manchu and Mongolian language training, it often happened with the help of pedagogical texts that had been used since at least the eighteenth century. In its last decades, the Qing empire in Manchuria was to some degree engaged in a campaign similar to earlier policy initiatives.

The question is whether simultaneity helps us understand Qing language policy and practice. First, what exactly does “simultaneity” mean? He states that the term primarily refers to bi- or multilingual books with versions of the same text running in parallel columns (177–78). In this narrow sense, simultaneity is certainly a poor metaphor both for the public-facing, ideologically charged use of language by Qing state organs and, especially, for language use within the administrative apparatus itself.

In fact, comparatively few court-sponsored publications used the plurilingual interlinear format. We can exclude the bi- or multilingual dictionaries, since such dictionaries by their nature will present words in different languages on the same page. Besides, the definitions in these dictionaries, to my knowledge, in several cases remained in Manchu only, without translation into the other languages (the two Manchu–Mongol dictionaries contained the definitions in both languages). Excluding the dictionaries leaves a fairly limited corpus of texts. The most famous and most influential bilingual texts were certainly the Qianlong-period Manchu–Chinese versions of the Four Books and Five Classics, but many other books, even dating from the Qianlong period, were made in separate Manchu and Chinese versions, rather than one “simultaneous” version, and then sometimes bound together (Qianlong’s Ode to Mukden, Manzhou yuanliu kao 滿洲源流考, Kaiguo fanglüe 開國方略, etc.). Manchu translations of Chinese books were also published or written up without the original Chinese text (e.g., the 1655 Poetry Classic, historiographical works, etc.). I am prepared to wager that linguistically “simultaneous” works were largely a product of commercial and private publishing, not court-sponsored publishing.

More expansively, He also uses simultaneity to refer to plurilingual texts that are not presented interlineally, such as imperial diplomas and stelae. Such simultaneity certainly characterized public-facing emperorship in the Qianlong period, but not the Qing bureaucracy, which produced most official documents. I have seen clauses such as manju nikan hergen-i kamcime arafi “wrote jointly in Manchu and Chinese” in Qianlong-era administrative documents, probably referring to the production of texts that were bilingual in this second, looser sense of the word (something like jian xie 兼寫 in Chinese). Whatever the precise meaning, “simultaneity” was not always a good thing in these cases, with Qianlong reprimanding officials for having written palace memorials in two languages on the same sheet, rather than separately.

Indeed, more generally, the palace memorial system—one of the most important innovations in early to mid-Qing administrative practice—developed in part to avoid having to produce documents in several languages. Translations consumed time and energy. They could cause delays and, if not executed properly, confusion as well. In many situations, the state’s priority was to minimize the difficulties involved in administering a plurilingual empire, not to celebrate plurilingualism in “simultaneous” monuments and books.

New insights into the linguistic pluralism of the Qing Empire will, I think, come from studies of its institutions, practices, and training of personnel, rather than from discussions of ideologically charged pronouncements. And it is precisely in this area—language education in the late-Qing Jirim League and Manchuria—that He’s book makes its most important contributions.