Why study Manchu?

Why Should the Sinologue Study Manchu? (Or Mongolian?  Or Tibetan?  Or . . . ?)

Mark Elliott

Introduction: Hauer on the Jesuit Heritage

The title of this essay is a paraphrase of a famous article written almost seventy years ago by the eminent scholar Erich Hauer, best known for his compilation of one of the outstanding lexicographic aids for students of Manchu.  In a short essay called “Why the Sinologue Should Study Manchu,” which appeared in 1930, Hauer argued that the study of Manchu was undeservedly neglected and laid out three basic points that he hoped would contribute to its rehabilitation among scholars.[1]

Screen Shot 2013-01-18 at 08.14.27 - Version 2The first reason Hauer said people should study Manchu was for the purposes of historical linguistics.  Hauer informed the reader that Manchu was not yet a dead language, citing as proof, among other things, a 1925 newspaper he had received, called the Ice donjin afaha (“New tidings sheet”), published in Hailar.  It was not too late, Hauer said, to track down and record a “genuine Manchu pronunciation” from these last living speakers in Manchuria, although he warned that time was running out.  After enumerating the different Manchu grammars that had been written up to that point, Hauer underscored the importance of further such work, not least because most of the grammars he referred to were not only out-of-date, but were very hard to obtain.  This was his second reason.  He insisted on the need for more complete grammatical studies of Manchu that would include exercises, and on the need for good Manchu dictionaries in English, German, French, and Russian, as existing works were “neither complete nor reliable.”[2]  Moreover, he urged the wider dissemination of the Manchu translations of the Shu jing, Shi jing, Yi jing, Chun qiu, Zuo zhuan, and other classic texts, by calling for an international effort at transliteration and publication, an appeal that evidently fell on deaf ears.

Hauer’s third rationale for the study of Manchu – and the only one that really seems to have very much to do with his purported sinological audience, since the first two arguments would probably have been persuasive mainly to linguists – was that it was a useful tool in the study of Chinese history, language, and literature.  He spelled this out under five different rubrics.  The first two pertain to history, namely, that knowing Manchu allows one to correctly render Manchu personal and place names that have been “horribly mutilated” by their Chinese transliterations and that knowing these names can reveal their real meanings; the second two pertain more to Manchu’s usefulness as a “pony” in reading Chinese texts, since the Manchu translations of classic texts were expertly done by people perfectly familiar with their original meaning and with how best to express it in Manchu (he singles out the Manchu translation of the Peiwen yunfu as being especially helpful); and the fifth rubric he cites is that Manchu is not difficult to learn.  Here again he comes back to what for him appears to have been the main benefit of Manchu, which is that it “enables the student of Sinology to use the Manchu versions of the classics . . . in order to verify the meaning of the Chinese text.”[3]

All things considered, Hauer’s defense of the utility of Manchu was not too far removed from the Jesuit rationale provided by Father Amiot, whom he cites as saying that “the knowledge of Manchu would open a free entrance into the Chinese literature of all centuries: there is no good Chinese book which has not been translated into Manchu.”[4]  Though he admonishes Amiot for exaggerating the amount of the Chinese literary corpus that existed in Manchu, Hauer essentially agrees with him, declaring (with Laufer, whose authority he inserts here, too) that it is simply irresponsible of sinologists to ignore the Manchu translations available of many important works in making their own translations.  One can almost hear him stamping his feet in impatience.

Whatever one makes of Hauer’s arguments, it must be admitted that they do not appear to convinced a whole lot of people that taking the time to learn Manchu was really worth the effort – certainly not historians, which is the group I am mainly concerned about here.  The Jesuit heritage and its multilingual culture has passed.  Outside of Germany, about the only scholars who bothered were Japanese scholars, and their reasons for doing so hardly had much to do with Hauer’s.  In the twentieth century, only a few Chinese scholars troubled themselves to learn the language of their former imperial masters, and American sinology can scarcely be said to have produced large numbers of scholars fluent in all the languages of the Qing empire.[5]  By and large, this is the situation that has come down to us today, where scholars of Chinese don’t bother with other languages except, perhaps, Japanese.  Even if they wanted to, the number of places in this country one can undertake such study, never large, is today down to no more than two or three.

Why Learn Manchu?

Fletcher1So why, then, should the scholar of China bother to learn Manchu?  This is the question I raise in this paper.  But I broaden the question to include languages not mentioned by Hauer:  Mongolian, Tibetan, Turki, Russian, Japanese, Korean, even French, Latin, and English – any language other than Chinese that might be used in the study of Chinese history.  I do so deliberately because I believe that, whatever its special attractions, any really good argument supporting the use of Manchu ought to be equally valid when applied to other languages, too.  I use Manchu for my examples here only because I am most familiar with it and it has attracted the most attention. By looking briefly (and, admittedly, selectively) at the different reasons that have been put forward for learning Manchu, it should be possible to see in what ways they fall short of persuading the field as a whole that not just Japanese, but one of the other languages of the Qing, should be a requirement for all Ph.D.’s.  Because the truth is if we don’t do the language work in graduate school, few of us will have the time or inclination later in our careers to do it.

Let us imagine, then, a historian or graduate student in Chinese studies curious about the utility of Manchu, somewhat disappointed, as one imagines, after a reading of Hauer.  The next place he or she might look would then be one of the more recent statements of the usefulness of Manchu, such as the 1973 essay by Joseph Fletcher, with which most of you no doubt are familiar.  This essay begins straightforwardly enough:

Manchu sources have two main uses for historians of China.  They supply information that is unavailable in Chinese and, when both Manchu and Chinese versions of a given text exist, they provide controls for understanding the Chinese.[6]

This amounts essentially to a restatement of the position of Hauer and the Jesuits.  But reading on, it is clear that in other ways the approach here has evolved in some ways, the chief difference being that by the time Fletcher was writing, a fair amount of archival material in Manchu had been published, including the Manbun rôtô and the Jiu Manzhou dang.  There was also much more information available on unpublished archival collections, which Fletcher usefully summarized.  This permitted him to characterize more specifically the historical subjects for which Manchu might be useful:

. . . the history of the Manchus as a people, their policies for controlling the Mongols, their relations with the Ming dynasty, the administrative structure of their state, and the functioning of their empire in China.[7]

His bibliography of other kinds of semi-official or unofficial Manchu accounts also surpasses what is in Hauer, and is still worth consulting.

However tantalizing the last line of Fletcher’s essay may have looked to some readers (“What?  An Asian language you can learn in a year?!”),[8] it was not Fletcher’s aim to “sell” Manchu to the interested student in any explicit way.  In fact, the overall tone of the article tended to downplay the importance of Manchu and to emphasize the degree to which Manchu texts exist alongside Chinese ones.[9]  This may have reflected the long-standing belief among sinologists that whatever existed in Manchu was also available in Chinese.  As Fletcher himself noted, “almost all Manchu source material, even from the earliest period, was carried over in one form or another into Chinese.  For historians of the middle and late Ch’ing, Manchu records can be useful, but they are not necessary.”[10] Why learn Manchu, indeed!

The Discovery of the Archives

DSCN27662571 - Version 2Just five pages long, Fletcher’s essay was for a very long time the only statement in English of the potential uses of Manchu language study for the historian of China.  Only in 1985, with the appearance of an article by Beatrice Bartlett, would our curious graduate student find something else to consult.  Bartlett, who had spent more than a decade working in the archives and knew more about them than any other Western scholar, adopted a very different tone:

The received wisdom on the Manchu language archives of the Ch’ing dynasty has been that historians of the Ch’ing have little need either to learn Manchu or to use Manchu materials for research on historical subjects.   Learning Chinese will suffice, the argument goes, because the Ch’ing dyarchical principle guaranteed that everything written in Manchu was also recorded in Chinese.  Indeed, some Manchu specialists have averred that once the historian gets past 1644, Manchu materials merely provide opportunities to employ the many available dual texts to enhance knowledge of both languages.

Bartlett distanced herself from this view (Fletcher’s view) by describing her discovery in the First Historical Archives of significant Qing archival deposits that existed in Manchu only.[11]  These materials, which she went on to describe in some detail, led her to conclude that

In contrast with what has hitherto been thought, . . . many unique Manchu documents, never translated into Chinese, were produced in the middle and even the late Ch’ing.  [. . .]  Future Ch’ing historians may well find the study of Manchu worthwhile.[12]

Bartlett’s authoritative knowledge of the archives lent credibility to statements such as, “a run of Manchu archival records is more likely than the Chinese to be complete.”[13]  Her words offered a very concrete sort of encouragement to new people in the field, among them, myself.  The opening of the archives and the publication of this article were landmarks heralding a new level of awareness of the importance of Manchu for the history of late imperial China.[14]

Most recently, our student, now perhaps more inspired, might turn to the 1993 article by Pamela Crossley and Evelyn Rawski, “A Profile of the Manchu language in Ch’ing History.”  This is essentially an introduction to the different sorts of Manchu materials available to the historian, and perhaps fits more neatly in with a long lineage of Manchu bibliographic work, a tradition which features many illustrious names (including the names of some of those present today), but which need not detain us here.  At the same time, the essay’s emphasis on the opportunities opened up by a knowledge of Manchu puts it also in the genealogy we are tracing here, since it implicitly supplies an answer to the question, “Why should the sinologue study Manchu?”  Comments touching directly on this point emerge here and there, as when the authors note that,

Nineteenth-century Western scholars emphasized the significance of Manchu in the Ch’ing order, although after the fall of the dynasty some Ch’ing specialists questioned the necessity of acquiring literacy in the language.  In recent years access to Manchu materials has increased dramatically, and a reassessment of the importance of Manchu for Ch’ing research is in order.[15]

According to Crossley and Rawski, the study of Manchu matters for four main reasons:  Its position as the “imperial language” in administration; its symbolic importance as an instrument of “ideological and religious” manipulation; its role as a vernacular language, especially in Beijing; and its place in private scholarship in the Qing.[16]

As I see it, this brings us, as it were, to higher ground.  We have moved beyond the advantages that are typically cited, namely that Manchu either provides extra information, i.e., that it supplements Chinese sources, or that it provides exclusive information, i.e., that some things can only be studied using Manchu (these things understood to be limited to pre-conquest history and other topics for the specialist).  Crossley and Rawski acknowledge, of course, that knowing Manchu is useful for rectifying errors in Chinese texts, and they point out as well that it is vital for a broad range of important topics in Qing history (foreign relations, politics, military organization, the imperial court).  But they also show (perhaps unintentionally) that knowledge of Manchu can end up producing new types of problems and new lines of inquiry.  Let me explain what I mean, and why I think this points us in a new direction.

Escape from Sinocentrism

The facts of the matter are plain enough.  On the one hand, people are willing enough to believe in the usefulness of Manchu.  It does them no harm, and makes them appear open-minded enough to avoid being classed with the die-hard Sinocentrics, who are coming to earn a reputation that puts them in the same club as the “Orientalists.”  On the other hand, even with the appearance of Beatrice Bartlett’s article fourteen years ago, only a few historians have decided that Manchu is worth the trouble of learning themselves.  Interest in the language among graduate students is relatively small, though it may be growing slowly; yet this growth is hampered by the increasing scarcity of programs to train them.  So we have an apparently contradictory situation in which Manchu, though nearly universally acknowledged as important not just for the very early Qing, but for much or even most of Qing history, is still the province of a very few specialists in the world, whose numbers can be calculated easily enough if we figure that at least half of them are in this room right now.  Is there a way out of this dilemma?

Assuming we are in agreement that more, rather than less, scholarly activity using Manchu (or Mongolian, or Tibetan) sources is a good thing, and that it is desirable to break down the boundaries that have tended to marginalize the historical issues we work on, and keeping in mind the way that the academic world is structured, I see two paths to pursue.  One is institutional, the other intellectual.  Institutionally, we can make an effort to provide more opportunities for instruction in these languages, especially in the summer, since this is of greatest practical benefit to students at schools where they are not offered.  At programs where Manchu and the other Qing languages are offered, we can insist that students have at least one year of study under their belts before they begin their dissertation research.  And we can continue to insist that national organizations such as the AAS make greater efforts in their sponsorship of activities and publications related to the “Inner Asian” component of “China and Inner Asia” and to integrate these activities more energetically into broader educational and scholarly agenda.

None of this will happen, though, until our work is more fully invested intellectually – at which point it will probably begin to happen by itself.  It seems to me that people will act on the knowledge that Manchu is useful only when they see it usefully applied to problems they are interested in.  This is simply because life is short and we all must make choices.  Realistically, the sinologue will only study Manchu when there is something in it for him or her.

There is no way to guarantee this happening on anything approaching a universal basis, of course, but I can think of a couple of ways of increasing its likelihood.  One is to use Manchu to address exactly those problems, and showing how knowing Manchu can make a difference.  (This is exactly the reason that people learn Japanese.)  In effect, this is just the approach that Beatrice Bartlett has taken in her 1990 monograph, Monarchs and Ministers, where she demonstrates quite explicitly how even a little knowledge of Manchu can go a long way in elucidating the mysteries of Qing politics.[17]  We all know that Manchu is useful for military history.  What about economic history?  Intellectual history?  Art history?  Applying a knowledge of Manchu to what might be thought of as “mainstream” topics will go far to accomplish its integration into the curriculum and to convince our colleagues and students that Manchu is not just about Manchu studies, or frontier studies, or military history.  It is about “Chinese” studies – which must itself then be redefined.  In fact, this is already happening.

Which leads me to my second point, which is that the value of knowing Manchu – or any kind of linguistic knowledge relevant to the larger Chinese world – will be most powerfully demonstrated if we are able to use it to create a new framework for thinking about historical issues.  This gets us out of an old paradigm, in which knowing Manchu is either a specialization entirely unto itself, or a handmaiden to one’s “real” work, which is in the Chinese sources.  The marginality of Manchu is inevitable in the first instance and its supplementarity inevitable in the second, since “real” problems are in that case defined by those Chinese sources and those who use them.  Thus it seems to me that one of the chief points to be stressed in a discussion of using non-Chinese sources is that they enable us not just to frame familiar problems from different angles, but also to frame entirely new problems.

This is important because it provides an avenue of escape from the prison of Sinocentrism.  If examining European history from the point of view of the colonial has proved to be a key strategy to escaping Eurocentrism, then by the same logic examining Chinese history using the languages of people similarly marginalized (if not made subalterns in the same way), promises fresh perspectives.  It is up to us to create them.


To achieve this promise, we must do more than just preach to the choir (as I am doing here).  We must be more than “language jocks.”  We must find ways to transform various linguistic knowledges into intellectual currency that is both viable and valuable.  First, as R. Bin Wong and Kenneth Pomeranz are doing in their work on comparative Chinese economic history, we must find ways to make what we do understandable to our colleagues and students in related fields and disciplines using categories that are mutually relevant and meaningful, if we want more than a polite nod and a smile when we tell them how useful we have found these linguistic skills.  With a little creativity, just as a French historian can be made to care about China, so a Chinese historian can be made to care about Mongolia (a case in point might be the work of Tatsuo Nakami and Christopher Atwood on nationalism).

Second, in translating linguistic knowledge into heuristic knowledge, we must question the prevailing assumptions and presumptions of Sinocentric scholarship.[18]  Like Joseph Fletcher, we must think critically and theoretically as we search for new models and new modes of explanation that disentangle the warp as they integrate the weft.[19]  At the same time, in decentering privileged Chinese views, we must be careful to give them due weight, too, and avoid falling into the same sort of absolute thinking as those who hold that cultural interaction in the Sinic world is simply the story of one-way assimilation.  The success or failure of this endeavor will, I think, go far in determining the viability of new hypotheses on Qing hybridity and the importance of ethnicity in the creation of a modern Chinese nation.[20]

It will always be possible to research and write on Chinese history without using anything but Chinese sources.  Probably, this is how things will remain.  I am not under any illusions about that.  It is for this reason I would suggest that the most important thing those of us who use non-Chinese sources can do is to make those who do not more aware of the assumptions hidden in their sources.  If we can find a way to communicate the different “feel” provided by a Manchu or Tibetan document we can make those who rely exclusively on Chinese sources more conscious of the prejudices and special problems built right into the classical language they are written in.

If past experience is any guide, until we somehow demonstrate that more than just additional information is to be gained by learning another language, nothing is really going to change.  The sinologue will not learn Manchu or any other language of the Chinese world.  Through our work, we must strive to construct an original way of knowing the Chinese past and so reveal the flaws of older epistemological schemes.  We must show that behind the different information found in Manchu, Mongolian, Turki, or Tibetan sources lies a different world, an independent world as different, as valuable, and as real as the world the sinologue discovered the day that Chinese first began to make sense.

This paper was originally presented at the International Symposium on Non-Chinese Sources for Late Imperial Chinese History, 19-21 March 1998, at the University of California, Santa Barbara.  It appears here for the first time.  No attempt has been made to revise or update its contents.


[1] Erich Hauer, “Why the Sinologue Should Study Manchu,” Journal of the North China Branch of the Royal Asiatic Society 61 (1930), 156-164.

[2] Hauer, “Why the Sinologue,” 161.

[3] Hauer, “Why the Sinologue,” 164.  In this light, Joseph Fletcher wrote that the advantage of Manchu was that the “ambiguities of Chinese cannot be carried over into Manchu,” because each element of Chinese syntax and the meaning of each Chinese character was made explicit in the process of translation.  Joseph Fletcher, “Review of Walter Simon and Howard G.H. Nelson, Manchu Books in London: A Union Catalogue,” Harvard Journal of Asiatic Studies 41.2 (December 1981), 656.  While it is not hard to grasp the general point here, I confess to some uneasiness with respect to this conclusion, made even graver by contemporary theories of translation.

[4] Hauer, “Why the Sinologue,” 156-157.

[5] Indeed, I can think of only four:  Francis Cleaves, Joseph Fletcher, David Farquhar, and James Bosson.  I would be grateful to learn of others.  Tragically, Fletcher and Farquhar barely lived to fifty, preventing them from shaping the field as they undoubtedly would have were they still alive today.

[6] Joseph Fletcher, “Manchu Sources,” in Donald Leslie, Colin Mackerras, and Wang Gungwu, eds., Essays on the Sources for Chinese History (Canberra: ANU Press, 1973), 141.  Because it was not available to me at the time of this writing, I regret that I am not able to consider here another potentially relevant source, Denis Sinor’s Introduction to Manchu Studies (ACLS, 1963).

[7] Fletcher, “Manchu Sources,” 141.  I am not sure what Fletcher meant by “the functioning of their empire in China,” which seems to cover a lot of territory.

[8] The last line reads:  “Fortunately, it [Manchu] is an easy language, and for anyone who reads Chinese, as little as a year of study can unlock the vast store of Manchu sources.”  His reference here is to Hauer.  No doubt he had in mind Hauer’s note that Manchu was a “tiffin language,” i.e., “a language which could easily be mastered during the short time of a tiffin.”  Yet Hauer dismissed this as a “facetious exaggeration,” acknowledging only that Manchu was easier to learn than Chinese (Hauer, “Why the Sinologue,” 163), which is undoubtedly true.  My own experience leads me to the conclusion that it is misleading to expect (or to lead others to expect) that one year of study in fact permits very much more than an elementary understanding of the language, just sufficient to begin to turn to an analysis of primary source texts on one’s own.  (A later statement by Fletcher on the same subject in a 1981 review indicates that this may have been what he was trying to say.  See note 11 below.)   I would also add that, apart from a knowledge of Chinese, knowing Japanese can be of great help in learning Manchu.

[9] Imparted by such phrases as “there are personal accounts written in Chinese for which Manchu translations also exist” and “probably almost all of [these works] are available in Chinese versions.”  Fletcher, “Manchu Sources,” 143.

[10] Fletcher, “Manchu Sources,” 145.  He did allow, however, that Manchu sources were “essential” for the history of the early Ch’ing, i.e., the first half of the seventeenth century.

[11]In fact, a hint of things to come was dropped by Fletcher himself in a 1981 review article in the Harvard Journal of Asiatic Studies, by which time what he had seen of Bartlett’s work caused him to change his tune.  Though it was still unclear what secrets the Manchu archives held that were not in the Chinese materials, Bartlett’s research had apparently convinced him that the utility of Manchu was not limited to the pre-1644 period.  Fletcher wrote: “A Ch’ing scholar who wants to do first-class work in the archives must, from now on, learn Manchu and routinely compare the Manchu and Chinese sources for their topics of research.”  Yet, he added, “Very few scholars of Ch’ing history presently possess this capacity.”  Fletcher, “Review of Walter Simon,”  653-656.

[12] Beatrice Bartlett, “Books of Revelations: The Importance of the Manchu Language Archival Record Books for Research on Ch’ing History,” Late Imperial China 6.2 (December 1985), 25-26.

[13] Bartlett, “Books of Revelations,” 33.

[14] The author’s modest contribution to this rising awareness is an article published in Japanese, “Dai’ichi rekishi tôankanzô naikaku to kyûchû Manbun tôan no gaijutsu,” Tôhôgaku 85 (January 1993).

[15] Pamela Kyle Crossley and Evelyn S. Rawski, “A Profile of the Manchu Language in Ch’ing History,” Harvard Journal of Asiatic Studies 53.1 (June 1993), 64.

[16] Crossley and Rawski, “A Profile of the Manchu Language,” 100-102.

[17]Beatrice Bartlett, Monarchs and Ministers (Berkeley: University of California Press, 1990).  By the same token, one can also think of instances where a little knowledge of Manchu has proved a dangerous thing.

[18] For two recent examples of this sort of thinking, see Nicola Di Cosmo, “Ancient Inner Asian Nomads: Their Economic Basis and Its Significance in Chinese History,” Journal of Asian Studies 53.4 (November 1994), 1092-1126, and James Millward, “New Perspectives on the Qing Frontier,” in Gail Hershatter, et al., Remapping China: Fissures in Historical Terrain (Stanford: Stanford University Press, 1996), 113-129.

[19] See R. Bin Wong, “China and World History,” Late Imperial China 6.2 (December 1985), 10-11.

[20] For an outline of these trends, see Evelyn S. Rawski, “Reenvisioning the Qing: The Significance of the Qing Period in Chinese History,” Journal of Asian Studies 55.4 (November 1996), 829-850.


Get every new post delivered to your Inbox

Join other followers: