Was There an Administrative Revolution?

Abstract

This essay takes a close look at Maura Dykstra's monograph Uncertainty in the Empire of Routine (Harvard Asia Center, 2022). It analyzes the book's multitude of problems, such as its flawed conception, numerous factual blunders, failure to engage existing scholarship, problematic choice of primary sources, and dubious citation practices. Most significantly, this essay aims to provide ample evidence to demonstrate how the book systematically misrepresents the majority of its primary sources to support an untenable thesis. It argues that the book's central claims are ungrounded in evidence.

Maura Dykstra's book Uncertainty in the Empire of Routine is an ambitious foray into the Qing Empire's information management and the making of Qing state archives. Unfortunately, as a historical monograph, the book fails to meet basic academic standards. Failing to engage most of the relevant historiography and filled with misinformation, the book demonstrates a poor command of its subject matter. The author bases her arguments on questionably chosen primary sources without critiquing them or explaining her strategies in using them, and she exacerbates this problem with a citation method that makes tracking her sources unnecessarily difficult. There are at least a dozen places where the citations do not match the content of the book.¹ As a result, the book is conceptually, methodologically, and factually unsound.

Moreover, the author systematically misrepresents her primary sources by mistranslating texts, exaggerating them, taking them out of context, and embellishing them with non-existent information and details. The majority of primary sources in this book are misinterpreted such that they support the author's untenable claims, while evidence undercutting these claims is ignored. The author also makes many claims throughout the book that are not supported by any sources. Remarkably, the book contains hundreds of errors.² We all make mistakes, and any single error in this book, in isolation, would be embarrassing but excusable. Yet the scale and seriousness of the problems here far exceed those of any other academic monograph that I have read, necessitating a close look at many cited sources.

This review unfolds in four parts. First, I introduce the book's thesis and its major claims. Second, I provide a detailed assessment of the book's problems, including its primary conceptual flaw, significant factual errors, failure to engage relevant scholarship, problematic choice of primary sources, and dubious citation practices. Third, I focus on the author's systematic misrepresentation of primary sources. Fourth, I dissect Chapter 3—the book's central chapter—to demonstrate the extent and depth of its problems and why the book's claims are unsound. I end with an overall appraisal of the book.

Thesis, Storyline, and Claims

By combing through what appears to be a random collection of tedious and mundane excerpts from imperial regulations, court orders, administrative boilerplate, and legal advice with an “information-centered approach” (1), the author claims that an “administrative revolution” (3) occurred between the Manchu conquest in the mid-seventeenth century and the peak of the Qing dynasty's power in the mid-eighteenth century.

For Dykstra, the Qing “administrative revolution” was the culmination of a hundred years of increasing and tightening regulations imposed upon administrators. At the center of it was routine information reporting and archive-making. According to Dykstra, preoccupied with disciplining local officials and stamping out bureaucratic malfeasance, the Qing court constantly implemented ever-stricter regulations to standardize reporting procedures and record keeping, so that the central state could learn more about various aspects of territorial governance. Citing the expansion of these regulations, the author argues that routine paperwork at various levels of the government increased exponentially over time. This enabled the central state to chase down corruption and identify malfeasance by tallying statistics and cross-checking reports. At the same time, this administrative revolution produced vast quantities of records in government archives—the remnants of which became the Qing archives that historians rely on today. In Dykstra's words, this “administrative revolution” produced a “new information/data ecology” and a “radical epistemological shift” that in turn remade the Qing state and its archives (3). As a result, the Qing emerges as an “empire of routine.”

Despite its “tectonic effects,” the book describes this revolution as “profound and yet subtle” (4). It was “unexpected, unintended, and heretofore overlooked” (3). It has “remained invisible to both actors and historians” and was “imperceptible even to the men at the center of the state” (4). In other words, over the past two centuries, nobody, including the emperors and ministers who authored the revolution and the generations of historians who have studied them, ever noticed it—until now.

The revolution proved to be a double-edged sword: a hundred years of “quotidian efforts to produce greater bureaucratic accountability” buried the Qing state in a mountain of information, especially “information about malfeasance in the territories.” As the court received ever-more information, it became aware of problems that might have previously gone unnoticed. This instilled a sense of paranoia and paralysis; the former because problems were increasingly legible, the latter owing to billowing piles of paperwork. In Dykstra's words, “The more the Qing central court sought to root out corruption, the more corruption it found. By the close of the eighteenth century, the helm of the Qing state became convinced that it was in the throes of decline” (3–4). The Qing emperors and ministers who “unknowingly authored” a successful revolution in information gathering became its unwitting victims.

Problems and Overall Assessment

Dykstra lays out her ambitious claims in the first few pages of the book. A few pages on, things start to go awry. In a section in the prologue called “Vagaries of the Qing Archive in the Republican Period,” the author begins by stating that “after the fall of the Qing, the imperial archives from the palace fell under the jurisdiction of the Republic of China's Ministry of Education,” and then “in 1916, the records were transferred to the care of a newly established Museum of History in Beijing” (xxvi). This is surprising because between 1912 and 1924 the Manchu court still occupied the Forbidden City: How could the Republican government transfer the imperial archives that belonged to the rump Qing authorities away from the palace?

It turns out the dramatic story of the dispersion of Qing imperial archives that Dykstra tells is not the story of all the imperial archives; rather, she has recounted the famous “eight thousand sacks incident” (baqian madai shijian 八千麻袋事件) and mistaken the early Republican-period peregrination of these sacks of Neige daku 內閣大庫 (Grand Secretariat Archives) documents, which ended up in the Academia Sinica, for the entire Qing imperial archives. As a result, she further misunderstands the different provenances of the archives in Taipei's National Palace Museum (mostly consisting of palace and Junjichu 軍機處 memorials) and the Neige daku documents at Academia Sinica, which she wrongly suggests were split from the same source into these two separate collections after they arrived in Taiwan. Adding to this confusion, she erroneously claims that another portion of these materials “made its way back to Beijing” where it formed the basis of the collection of over ten million documents at the First Historical Archives (FHA), the vast majority of which had in fact been in Beijing all along. For a book that purports to tell “the double stranded tale of the development of the Qing bureaucracy and its archive” (xviiii), a fundamental misunderstanding of the provenance of the three existing archives of Qing central administration bodes ill.³

Dykstra's confusion extends beyond the provenance of the Qing archives to her sources. In Chapter 5, for instance, Dykstra cites five memorials and sources them all to the Neige daku (207–223). But a close examination shows they are confidential memorials from the National Palace Museum's database, not routine memorials from the Neige daku. Has the author simply mislabeled them in a rush? One would hope so, but the citation for a different document that actually comes from the Neige daku sources it to a “database,” which she titles Zhongyang yanjiuyuan lishi yuyan yanjiusuo cang mingqing shiliao 中央研究院歷史語言研究所藏明清史料 (Ming-Qing Historical Sources from the Institute of History and Philology, Academia Sinica)—a “database” that I have never heard of (221). On checking the link (http://catalog.digitalarchives.tw) provided in the bibliography, I found the website Diancang Taiwan (Important Collections in Taiwan), which offers a unified catalogue of digital collections across many of Taiwan’s museums and research institutions, including postwar sources, but I could not find the database in question. Looking at the actual document, I realized the mistake: the cited Neige daku document appears as a photocopy of the original, printed on stationary with the letterhead “Zhongyang yanjiuyuan lishi yuyan yanjiusuo cang mingqing shiliao.” In a purported history of the archives, Dykstra has mistakenly invented an archival database based on a letterhead.

These errors make one wonder whether Dykstra really understands the different types of documents being produced by different central offices and how they were managed. During the Qing, a distinction existed between a routine channel of information that centered on the Grand Secretariat—the central bureaucratic office that received and processed regular information from provinces and ministries for the throne to review—and the more confidential, personal, and non-routine channel of information between high officials and the emperor through “palace memorials” (zouzhe 奏摺), which were processed by the Grand Council after its creation during the Yongzheng reign. In contrast to the Grand Secretariat, the central organ of the bureaucracy, the Grand Council served as the emperor's personal deliberative council, a part of the Inner Court. These two channels of information produced different archival holdings.

The book shows no evidence of awareness of the existence of these two distinct flows of information, both of which connected the capital with the provinces. The Qing court, unlike its Ming predecessors, relied heavily on the non-routine channel of confidential palace memorials sent directly to the throne to acquire and triangulate crucial information about important affairs across its vast territories and to monitor and control its provincial officials. But Dykstra never mentions “palace memorials” or “confidential memorials,” nor does she make the distinction between routine (tiben 題本) and palace memorials (zouzhe 奏摺). There are a couple of places in the book where Dkystra showcases how much the emperors cared about memorials (6, 229), but she never specifies which kind. Did emperors really care that much about routine memorials that contained lists of names of capital-sentence criminals or summary reports on the sanctions of county-level officials for mishandling legal cases? Or, alternatively, did they care more about the outcome of battles, sudden rebellions, or major corruption charges against high-ranking officials and ministers, all of which would show up on their desk in the palace memorials? In any event, shouldn't the functions of the crucial palace memorial system be discussed in a book about the Qing's information management and bureaucratic control?

In several places the author betrays a limited knowledge of the central information-processing organs of the Qing. In Chapter 3, she claims that there were “Three Personnel Offices (analogous to the Three Legal Offices)” in Beijing, and says they were the Ministry of Personnel (Libu 吏部), the Censorate (Duchayuan 都察院), and “the Henan Circuit section of the Personnel Office of the Grand Secretariat” (111). The problem is that the Grand Secretariat never had a personnel office, let alone a Henan Circuit underneath it.⁴ In Chapter 4, Dykstra first mistranslates the Neige (Grand Secretariat) as the Grand Council (161) and then, in discussing the Autumn Assizes, mistakes the Nine Ministers (Jiuqing 九卿) who presided over the Autumn Assizes for the Grand Council, incorrectly claiming that the Grand Council was in charge of reviewing death sentences (164). In fact, the Nine Ministers were composed of the presidents of the six ministries, the Censorate, the Court of Review (Dalishi 大理寺), and the Office of Transmission (Tongzhengsi 通政司); they had nothing to do with the Grand Council. These mistakes are akin to a book about the US government stating that the White House reviews legal cases submitted to the Supreme Court.

These shortcomings of the book are related to a lack of engagement with existing scholarship. Several important monographs on these subjects exist, any of which would have clarified the division of administrative functions, proper flow of documents, responsibility for oversight, etc. But Dykstra does not appear to have meaningfully engaged them. For example, this book does not cite Chuang Chi-fa's authoritative study on the Qing palace memorial system.⁵ While the bibliography includes Silas Wu's 1970 monograph on the same topic, it never appears to be consulted in the chapters.⁶ The book quotes only a few minor details from Beatrice Bartlett's classic study of the Grand Council, and the author appears to be unaware of its core messages.⁷ Despite acknowledging Philip Kuhn's Soulstealers to be a “seminal work” in the study of Qing bureaucracy (21), the book ignores its analysis of the conflict between the routine and non-routine aspects of the Qing administration. Nor does Dykstra respond to Kuhn's argument that “the imperial effort to achieve closer control over bureaucrats had to reach resolutely beyond routine procedures.”⁸ Instead, Dykstra arrives at the opposite conclusion, emphasizing only the routine and alleging that the Qing was “an empire of routine.”

Indeed, Dykstra's bibliography (titled “Works Cited”) is incredibly thin. Although the Yongzheng Emperor's reforms are central to the book's thesis, Dykstra omits reference to Madeleine Zelin's seminal study on the topic.⁹ Ignoring T'ung-tsu Ch’ü's classic monograph on Qing local government, she ventures to tease out the functions of information reporting by local administrations.¹⁰ Taking no account of Lei Rongguang and Yao Leye's introduction to the myriad types of Qing official documents, she aims to create a new understanding of “the role that texts played in mediating the governing structures of the empire” (5).¹¹ Without reference to Qin Guojing's two authoritative studies of Ming and Qing archives and record-keeping practices, Dykstra bravely accuses previous historians of misunderstanding the Qing archives.¹² Despite placing particular emphasis on the Qing court's efforts to control territorial officials, Dykstra cites no works on the Qing bureaucratic supervision system (jiancha zhidu 監察制度). While focusing attention on routine administration of judicial matters, Dykstra never mentions Matthew Sommer's analysis of either the Qing central legal archives or the tradition of legal commentaries.¹³ This dismissive approach to some of the most important and relevant scholarship is deeply troubling. As a result, the book not only suffers from conceptual flaws and factual blunders but also fails to clearly demonstrate how it alters our view of the Qing administration and the nature of the Qing archives.

Dykstra also neglects the scholarship on the “New Qing History” and Qing borderland studies that have transformed our understanding of Qing administration over the past thirty years.¹⁴ This is unfortunate, because ethnic tensions between the Manchus and the Han have been shown to have spurred many of the important administrative reforms that the Qing adopted after the mid-seventeenth century. In addition, the book only concerns what happened in China proper and fails to acknowledge that Qing administration and information management in frontier regions such as Mongolia, Manchuria, Xinjiang, and Tibet were radically different. Only by ignoring the Qing Empire's administrative plurality does Dykstra come up with statements like: “by the end of the eighteenth century the routinization of the bureaucracy exposed the territorial administrations in even the farthest reaches of the empire to the scrutiny of the central state” (187).

Dykstra's choice of primary sources is equally problematic. Despite the claim that she is “writing a history using the archives of the Qing central court” (xviiii), the book's primary sources are the published Shilu 實錄 (Veritable Records) and the Huidian 會典 (Imperial Compendia of Regulations), plus a few legal and administrative handbooks. A rough count reveals the Shilu cited 130 times, the Huidian (sometimes the Qianlong and sometimes the Guangxu editions) 74 times, and various legal handbooks 57 times. The archival sources that do appear are of minor importance. They include 18 documents from the First Historical Archives (of which 9 come from the same ten-day period in the same unspecified series),¹⁵ 13 documents from the Baixian (巴縣, Ba County) Archives, 5 palace memorials from Taipei's National Palace Museum, and 1 document from the Neige daku database at Academia Sinica. Almost all the book's non-archival sources are accessible in online databases. Although Dykstra claims to have written a history about the making of Qing archives, this is not an archive-based study.

The author never justifies the book's reliance on the Shilu and the Huidian or explains her strategies in using them. Qing scholars know that these are problematic sources—they were compiled by the Qing court to serve political purposes. The Shilu are the official chronicles of the various reigns composed by court historians, while the Huidian are compilations of administrative regulations and precedents. They contain heavily edited and sanitized materials. Historians treat them with suspicion, especially when they can refer to better sources, namely, the Qing archival documents that these compilations are based upon. It is ironic that Dykstra turns to heavily-edited and already published sources to reconstruct the history of Qing archive-making.¹⁶

Finally, the author makes tracing her primary sources more difficult than it should be by not citing page numbers or even volume numbers for either the Shilu or the Huidian, both of which are enormous compilations. Nor does she cite the particular physical editions or specific online databases used for any of her primary sources. Instead, she only cites the dates for contents in the Shilu and section subheadings for the Huidian. This new practice, she claims, is “to make the links to original sources more durable in our new age of access (and lack of access) to Qing sources in both print and digital formats” (xv), even though she is aware that “scholars tracing my citations in their original sources may have to spend a bit more time poring over the table of contents” (xvi). It is of course tautologically true that if you do not cite the edition, volume, or page number of a source then it will not be superseded by new editions; it also defeats the purpose of citations. It appears that since Dykstra relies almost exclusively on searches in digital databases to do research, she fails to see the value of citational conventions.

The Core Problem: Systematic Misrepresentation of Primary Sources

A greater concern than those outlined above is the fact that this book systematically misrepresents its primary sources in ways that make otherwise untenable claims seem plausible. A comparison of the original texts and Dykstra's representations of them, including all citations from the Shilu and legal handbooks, plus a portion of those from the Huidian, reveals hundreds of errors. The book contains twenty-seven block quotations from Qing-era sources; twenty-five of them are either mistranslated or misinterpreted—often both.¹⁷ Most of the book's shorter quotations and paraphrasing of textual evidence are also problematic. The Conclusion cites a total of eleven primary sources, for example, and apart from two from the First Historical Archives that I could not check, all include misrepresentations, some egregious. To be sure, not every error is equally significant, and any one of these errors, taken individually, would be a minor issue. But when errors are so numerous, and when most err in support of the book's thesis, they become a systematic problem.

I identified five problems with the author's treatment of sources. 1) Misidentification. Dykstra often misunderstands the structure of Qing official documents and so misidentifies the imperial institutions producing them. For example, she sometimes mistakes the words of a minister for those of the emperor (229, 231–232) or vice versa (202). 2) Mistranslations and inaccurate paraphrasing. Many translations and paraphrases in this book significantly distort the meaning of the original texts, sometimes even contradicting them. 3) Decontextualization. By omitting crucial information from the original texts, Dykstra often takes quotations out of context, rendering their meaning pliable and then misleadingly recontextualizing them for her purpose. Relatedly, she also uses sources in anachronistic ways. 4) Exaggeration. Dykstra frequently exaggerates the significance of a source, making minor administrative tweaks—or the lack thereof—appear to be dramatic reforms, and making regulations with a specific, narrow application appear as if they were applied more generally. 5) Embellishing. Dykstra habitually embellishes her translation or paraphrasing of texts with information not in the original. By doing so, she alters the meaning of the original texts to fit her story. It is often her embellishments and alterations, not the information in the original sources, that constitute crucial evidence for her claims.

As evidence, I will first provide two unrelated incidents taken from the book that show how Dykstra handles sources. In the next section, I illustrate how the book's central chapter is built on a string of misrepresented sources.

Example One

The first stand-alone example comes in the introduction, where Dykstra cites a 1774 edict “commanding two capital offices to update and disseminate their statutes” (13). Dykstra suggests that the following is the Qianlong Emperor's commentary on how imperial regulations (defined as those in the Huidian) emerged from ministerial ones:

Each of the ministries is a general hub for the cases from each of the provincial territories. The normal generation of imperially approved regulations (事例) is largely the result of deliberations on the advantages and disadvantages of adapting practice to certain places and times. These things cannot but be compiled into ministerial precedents (不得不纂為則例) so that everyone in both the capital and the provinces may know and follow them. (13)

This text comes from the Qianlong shilu; here is the original with the cited section underlined.

諭。御史陳朝礎奏請修內閣都察院則例一摺｡殊可不必｡各部為直省案件總匯｡其常行事例｡多有因地因時。斟酌損益者｡不得不纂為則例｡俾內外知所適從 … 至內閣固為絲綸重地｡然收發章疏｡繙本票簽｡及承辦誥敕寶冊等事｡並有一定章程｡祇須遵循罔斁｡遇有改簽事件｡則係朕親閱本章｡折衷酌定｡特降諭旨｡皆非閣臣所能參與｡又豈或有成例可稽｡若都察院雖風紀攸司｡而事非繁劇｡如監禮糾儀。稽察巡查。奏派諸務｡悉係奉行成憲｡并無庸臨事權衡｡是閣務院規｡均不過恪守舊章｡非若六部比擬例案｡必須互證兼資者可比｡又何必附纂例之故套｡而為無益之虛文乎。 (QLSL 39.7.21, vol. 963)¹⁸

By omitting most of the text, Dykstra distorts its meaning and intent. The preceding sentence—the first sentence of the edict—tells us that this edict is the Qianlong Emperor’s response to Censor Chen Chaochu’s memorial, in which Chen requested the compilation of zeli (ministerial regulations) for the Grand Secretariat and for the Censorate for the first time. The emperor resolutely rejected this proposal. Yet Dykstra represents the emperor as supporting Chen's recommendation, as if Qianlong is literally “commanding two capital offices to update and disseminate their statutes.” Nothing could be further from the truth.

In the rest of this source, the Qianlong Emperor in fact reasons that compiling standard regulations for the two offices in question, as Chen recommends, is not necessary. On the one hand, Qianlong notes that the Grand Secretariat, as the central office for processing routine memorials, reports, and imperial documents, follows time-honored practices with little change in regulations over time. On the other hand, the Censorate, while crucial in supervising and disciplining officials, does not have as many affairs as the Six Ministries, and therefore also just follows old rules. That is to say, this source is entirely at odds with Dykstra's claim that the central government had been increasing regulation of routine reporting and tightening discipline in a way that amounted to an “administrative revolution.”

While wholly distorting the meaning of the translated section, Dykstra also mistranslates a key common term in Qing official documents. Here the word shili, which literally means “cases and examples,” has nothing to do with “imperial regulations” defined in the Huidian; the Qianlong Emperor uses the term here as a common generic term meaning “regulations for administrative affairs.” This quote should be translated as follows: “Each ministry is a hub for cases from the provinces. Many regulations for oft-repeated administrative affairs are deliberated upon, added, or subtracted, adapting to time and place. That is why we have to compile ministry regulations. By having them, everyone in and out of the capital can know what to follow.” In other words, it explains why the zeli is necessary for the six central ministries: because ministerial regulations often change, the provinces need stable guidelines from these ministries when administering related affairs. There is nothing surprising or new here. The emperor is simply commenting on the day-to-day functioning of the bureaucracy. To the extent he is making any recommendation, it is not to tighten or update regulations, but to keep doing business as usual.

Example Two

A second example comes from Chapter 4, where Dykstra uses one passage from the Qianlong Shilu to make claims about Beijing's increasing demand for information from the provinces and the resulting proliferation of paperwork during the Qianlong reign. Dykstra writes:

The sheer volume of reports demanded by the Beijing ministries soon grew so large that a new genre of reports began to proliferate: reports on reports. These documents were known as “summary memorials (huiti 彙題).” The idea of requiring provinces to submit a summary memorial about their cases from each year was first suggested in 1744 by Henan Office Investigating Censor Peng Zhaozhu (b. 1699). He reasoned that “every provincial administration affects the lives and customs of the people. Although affairs of serious import are reported [to the throne] and resolved via memorial, it would be fitting to produce a summary memorial of them at each year's end.” Peng proposed to make provincial offices responsible for compiling information about the yearly tallies for certain cases in their administration, for ease of perusal in Beijing. (156)

Here is the original text for Censor Peng's proposed policy, of which Dykstra has only translated the underlined sentence:

河南道御史彭肇洙奏。各省關繫民生風俗重大之事｡雖經題結｡仍宜年終彙奏一次｡如戶部。則曰計今歲某省旱｡某省澇｡某省旱澇不為災｡某省豐｡某省歉｡某省豐歉居半｡如刑部。則曰計今歲某省強盜案件若干｡姦淫案件若干｡鹽梟案件若干｡謀故殺案件若干｡鬥殺案件若干｡干犯名義案件若干｡於封印日。簡明彙奏｡總計天下水旱豐歉姦詐盜偽之數｡歲歲考之｡則各督撫平日留心地方｡教養斯民｡有無成效之處｡可見一斑｡褒獎切責｡功罪不至相掩｡得旨。著照所請行｡該部知道｡ (QLSL 9.12.1.4, vol. 230)

Dykstra has misrepresented Peng's proposal in multiple ways. First, Dykstra confuses huiti 彚題 (routine summary memorials) and huizou 彙奏 (summary reports submitted through palace memorials). Dykstra uses this text as evidence for the emergence of huiti, which she claims to be “a new genre of documents.” However, in the original text, Peng uses the term huizou and makes no mention of huiti at all. What Peng actually suggests here is that a summary of major affairs should be submitted to the throne through the non-routine palace memorials (奏摺), although they had already been concluded in routine memorials (雖經題結). Undergirding this suggestion was the perceived importance of the palace memorials: for routine administrative matters to gain the emperor's attention, they would have to be presented through the palace memorials. Dykstra's confusion here stems from her apparent obliviousness to the distinctions between the tiben 題本 and zouzhe 奏摺, the two distinct documentary channels at the center of the Qing state. Moreover, it is unclear why Dykstra considers huiti as a new genre of documents, as the terms and practices of summary memorials (both huiti and huizou) appeared as early as the Ming, and there are numerous references to such reports in the Qing Shilu prior to the Qianlong reign.

Second, Dykstra interprets the policy that Peng proposed as commanding provincial administrations to compile information and produce summary memorials.¹⁹ Building on this interpretation, Dykstra goes on to claim that provincial governments functioned like “clearing houses for all manner of information” (157). But reading the original text in full shows that the policy was not about reporting from provincial administrations at all. Instead, it was the Ministry of Finance and the Ministry of Punishment who were to produce these summary reports. These two central ministries were commanded to sort empire-wide information by province in their reports, so that the emperor and his aides could compare the performance of different provincial administrations; the provincial administrations themselves are not even mentioned in it. As a matter of fact, in a different section of Chapter 4, Dykstra quotes Peng's policy again in block and interprets it correctly as evidence of the central ministries compiling “empire-wide statistics for perusal by the heads of state”²⁰ (160). It is puzzling how she could interpret Peng's policy in contradictory ways in different sections of the same chapter to make claims about new reporting practices in both provincial administrations and central ministries.

Dykstra not only misstates the provincial administrations’ reporting functions but also invents a non-existent type of document. She states in a matter-of-fact fashion that there emerged a “massive end-of-year reporting ritual: by the second half of the eighteenth century, [provincial] governors were required to submit a dozen summary reports in a single mega-memorial as part of the sealing up of the yamen and its archive at the end of each year” (157). No citation is provided for the claim, nor is there any discussion elsewhere of what this so-called “mega-memorial” from the provinces might refer to. No historians or archivists have ever discovered any of such “mega-memorials.” In fact, as Dykstra acknowledges, no such “mega-memorials” could be located in the First Historical Archives. Instead of reconsidering her own analysis or consulting the FHA for more reliable information, she claims that “the packets of summary memorials submitted by each province to the throne have been compromised by recent reorganization and digitization efforts by the First Historical Archives in Beijing.” She further states that the “critical information” about those reports “has been lost forever to historians” (157) without providing any proof to support such accusations. In sum, these “mega-memorials” seem to exist only in Dykstra's imagination; she offers no proof that the paperwork that provincial administrations submitted to the central ministries had indeed proliferated during the Qianlong reign.

Overall, the discussion above demonstrates how Dykstra, on the basis of one misconstrued passage in the Shilu, makes a series of inaccurate and contradictory claims, invents a non-existent type of documents and reporting practices, accuses archivists of wrongdoing without proof, and presents a distorted picture of the Qing administration. The treatment of this source is but one example of how Dykstra frequently makes large and incorrect assertions based on limited and misused evidence.

Anatomy of A Problematic Chapter

While source problems pervade the book, space does not permit detailed critique of each chapter. I thus analyze Chapter 3, “Imperial Routines in the Local Archive: Synchronization and Scrutiny in the Yongzheng Era,” which is central to Dykstra's primary claims, to illustrate how Dykstra consistently misrepresents sources to build her story.

In this chapter, Dykstra argues that the Yongzheng Emperor, in order to combat bureaucratic inertia and information problems, engineered key reforms. These, so the argument goes, created new documentary practices that led to the making of state archives and a “new data ecology” for the Qing. This “archival turn,” we are told, went on to drown the Qing state in paperwork. The question here is whether Dykstra has provided convincing evidence to show that such revolutionary changes actually occurred.

Censorial Duties in the Provinces

Dykstra identifies two major approaches to information problems during the Yongzheng reign. The first approach is what she calls “censorial duties in the provinces.” She begins by providing a rough account of how personnel documents—along with a host of registers about local fiscal and judicial conditions—were passed from the county magistrate, through the provincial bureaucracy, to Beijing during the so-called “great audit” (daji, 大計, triennial personnel reviews of territorial officials). She claims this documentary stream allowed provincial governments to absorb censorial duties in supervising and evaluating officials. This claim is unobjectionable. But can we attribute this approach to the Yongzheng reign (108)?

Unfortunately, the answer is no. Here Dykstra relies heavily on Pan Biaocan’s 潘杓燦 Weixinbian 未信編 (An Unreliable Treatise) as evidence to reconstruct this hierarchical documentary stream. It is an odd choice, for Pan’s book was composed in the 1670s and printed in 1684, so the practice must have already been in operation half a century before the Yongzheng reign. It is unclear why Dykstra attributes this historical phenomenon to the Yongzheng reign and calls it a Yongzheng approach. Later I will show how Dysktra uses Pan's treatise in other surprising ways.

Relatedly, when did the provincial administration begin to absorb censorial duties? Previous scholars have pointed out that this phenomenon happened after the Qing abolished the xun'an 巡按 system in 1661. (The xun'an was a Ming censorial institution that would rotate roaming censors dispatched from Beijing to inspect and assess the performance of officials in the provinces but remain independent of the provincial governments.) As soon as the xun'an was abolished, its responsibilities were taken over by provincial administrations.²¹ The xun'an system features nowhere in the book, but its abolition is rather important to the evolution of the Qing administration and its struggle to control local officials—a central theme of this book. Without an independent censorial institution in the provinces, provincial governors would become too powerful. Debate over how to curb their power and whether to restore the xun'an continued into the Yongzheng reign before being resolved.

And the Yongzheng Emperor's solution to the problem? It was not to standardize documentary streams of routine reporting, but rather to expand the coverage of the non-routine, secretive palace memorial system—a system with which Dykstra is apparently unfamiliar. The Yongzheng Emperor gave financial and judicial commissioners (in addition to governors-general, provincial governors, and the emperor's confidants) the prerogative to send secretive palace memorials directly to the throne, so that each of the high officials in the provincial capitals could report on their colleagues.

Archival Conquest of Central Ministries

The second approach Dykstra identifies is the Yongzheng Emperor's insistence that offices be held accountable for paperwork. Although this sounds rather mundane, Dyktra calls it the Yongzheng Emperor's “administrative stroke of genius” (104). According to her, this constituted Yongzheng's “archival conquest,” a transformation that proceeded from the central bureaucracy out to the local governments.

First, for changes in the central ministries in Beijing, Dykstra provides a total of three pieces of evidence, and all three are misrepresented. To avoid excessive detail, I will only analyze the first piece of evidence. This piece is an edict issued in the first year of the Yongzheng reign (1723) from the Shilu, an edict Dykstra considers “the first stone in the foundation of this new approach to bureaucratic discipline through documentary fidelity.” Here is Dykstra's translation:

An edict to the yamen of every metropolitan ministry and office regarding every single dossier and every single case in every one of the [capital] yamen … Stored dossiers must not only be sealed—when opened up for examination and review they cannot leave the hands of the responsible scribe. Any theft, any misplaced document, any changed characters might give rise to manifold malfeasances. And furthermore [in case of such actions] there is no basis for discovering [the malfeasance] and punishing it! Henceforth upon the occasion of the transfer of any administrators working in the ministries, all of the case dossiers for which he was responsible (both those he inherited and those he created) will be audited and transferred. Each individual [the outgoing and the incoming officials] will sign a pledge, read it aloud before his superiors, and append it to a dossier in the archive. (116–117)

Dykstra further explains that “the introduction of a pledge to be made on the transfer of office made each individual responsible for the actions documented in the archive compiled under his tenure” (117, my emphasis).

Neither Dykstra's translation of the original text from the Shilu nor her interpretation stand up to scrutiny. The underlined sections are those translated by Dykstra:

諭部院各衙門｡凡一應衙門卷案｡ 各有典守之官稽查引據。全以舊檔為憑｡此固一代之典冊。六官之掌故。不得視為具文也｡ 收貯卷案｡封禁雖嚴｡而翻閱查對。不能脫書吏之手｡盜取文移。改易字跡。百弊叢生｡莫可究詰｡嗣後司官遷轉｡將所掌卷案｡新舊交盤｡各具甘結說堂存案。如有疏失換易等弊｡一經發覺｡與受同罪｡爾各部院衙門。急宜查核清楚｡設法封貯｡永杜弊端。不得因卷案浩繁。畏難退沮｡其交盤事例｡爾部院諸臣。公同確議具奏｡尋議。各衙門卷案。俱應呈堂用印收貯｡遇有查閱｡滿漢司官。親身驗看｡陞轉之日｡出結交代。并請揀發專寫檔案筆帖式｡三年無誤保送補用｡從之。(Yongzheng shilu 1.3.6, vol. 5)

Her translation includes significant details that are not in the original. First, pay attention to the “every,” “every single,” “all,” and “each individual” in her translation, only one of which appears in the original. By adding these words, Dykstra changes the tone of the text, making a matter-of-fact regulatory tweak sound like an urgent, passionate command befitting the start of a “conquest.” Second, there is nothing in the original about “all of the case dossiers for which he was responsible (both those he inherited and those he created) will be audited and transferred.” Which word suggests “audit” here? Furthermore, there is nothing indicating that each individual will read his pledge aloud “before his superiors.” Which characters point to the presence of a superior?

Moreover, Dykstra misreads a string of sentences in the middle of the passage, making her translation contradict the meaning of the original. Her translation reads: “Stored dossiers must not only be sealed—when opened up for examination and review they cannot leave the hands of the responsible scribe. Any theft, any misplaced document, any changed characters might give rise to manifold malfeasances.”

A more accurate translation of the original should read: “Although [we] strictly closed off the archived case files, ministries have been relying on scribes to read and check them. This [the scribes’ access to documents] leads to theft and manipulation of documents. It has resulted in numerous ill effects and becomes impossible to investigate.”

In other words, this new policy did not intend to solve the problem of official malfeasance, as Dykstra claims, but rather scribes' unhindered access to archived documents. This reflects a common feature of late imperial Chinese governance, namely, deeply ingrained distrust of runners, scribes, and clerks and blaming them for all kinds of administrative problems.²² The proposed solution was two-fold. On the one hand, the officials had to take responsibility for the integrity of these ministry archives and sign pledges during a transfer of office. On the other hand—as shown in the latter part of the original that Dykstra does not quote—the existing documents should be sealed off and kept off limits from scribes. When they are needed, the Manchu and Han officials should personally read and verify them. Why would Dykstra focus on a pledge during office transfers? Moreover, as I will demonstrate, the requirement of a pledge from officials during office transfers was nothing new in the eighteenth century; it was by no means the Yongzheng Emperor's “administrative stroke of genius.”

Finally, and most importantly, why did the emperor initiate this new regulation? Was it intended to make “each individual responsible for the actions documented in the archive compiled under his tenure”? The secret lies in the ellipsis in her translation. This omitted part (which I have italicized in the original text) says that evidence of administrative precedents for future reference comes from the case files in the ministry archives. That is to say, preservation and integrity of these archives was not intended to make officials responsible for the actions documented therein, but to serve as references for administering future affairs and for compiling regulations such as those in the zeli (ministry regulations) and huidian (imperial compendia). By omitting crucial information in the original text, Dykstra distorts the meaning and purpose of this policy to make it fit her story.

The other two pieces of evidence that Dykstra has provided to illustrate Yongzheng's “archival conquest” of the central ministries are also mispresented.²³ All three pieces have been taken out of context, mistranslated, and embellished, and their significance has been exaggerated. They were, in fact, all minor procedural tweaks with no intrinsic connections among them; but Dykstra states that the supposed patterns demonstrated in them were repeated several times and in multiple fields over the Yongzheng era. And yet she provides no further citations. In which fields did these patterns repeat themselves? How do we know? Where are the sources? And what were the patterns? What happened to ministries other than the ones mentioned above? All these issues remain unaddressed.

Archival Turn in the Provincial and Local Administrations

Now let's turn to the alleged “archival creep” in local administrations. I examine three pieces of evidence and then evaluate Dykstra's claims.

To prove what she calls “documentary proliferation” in local administrations, Dykstra provides a 1723 regulation as an example that “some new policies simply demanded documentation of personnel or the quotidian functions of territorial yamen.” This regulation, according to Dykstra, required that “personnel hired in the gubernatorial yamen be documented with pledges from their superior officials” (121).

But the original text says something quite different. This edict from the Yongzheng Emperor concerns only the private secretaries of the provincial governors (a few dozen individuals in each province at most, known as mubin 幕賓), only a tiny subsection of “personnel hired in the gubernatorial yamen,” which could amount to thousands of people. Does the production of a register of a few dozen people's names constitute documentary proliferation for a province? Dykstra here takes a regulation about a specific category of people and vastly expands its meaning. Indeed, Dykstra provides no other evidence showing that more information was requested beyond the private secretaries or that more information was requested about governmental personnel below the provincial level. Finally, the original text never mentions “pledges from their superior officials.” Dykstra has simply added it.

To buttress her argument about local documentary proliferation, Dykstra directly quotes a 1725 regulation: “Every time a banditry case is encountered, on the one hand, it should be reported to the supervising official, and on the other hand, notice should be sent to the neighboring jurisdiction and the patrolling officers” (122). But what appears as a direct quote here is nowhere near the original. The original reads, “When trying to subpoena suspects across provinces, the official in charge shall, on the one hand, provide a detailed report and request his provincial governor to send communications and, on the other hand, dispatch runners to inform the local officials in that neighboring province, requesting them to send runners to collaborate in arresting [the suspects]” (凡隔省關提人犯。承問官一面詳請督撫移咨。一面差人關會隔省該地方官添差協緝).²⁴ Nothing in the original text suggests this is about reporting every banditry case.²⁵ How could anyone translating this sentence mistake it to be about “every time a banditry case is encountered”? Moreover, the purpose of the edict is not to acquire information, but to deal with the specific problem of overlapping jurisdictions in the context of subpoenas. Here again Dykstra has mistranslated a passage in a way that supports her claims.

A third major component of the Yongzheng “archival turn” in local administration, Dykstra states, regards the transfer of offices. The Yongzheng reforms demanded “archival accountability” for magistrates, who were required to sign pledges vouching for archival integrity. Moreover, “the process of transfer entailed a much broader examination and reckoning of the office than ever before” (127). To substantiate these points, she provides two extracts from the Qinban zhouxian shiyi 欽頒州縣事宜 (Instructions for Magistrates Published by Imperial Order) from the Yongzheng reign, which instructed newly appointed magistrates to closely inspect every aspect of administrative matters before taking over from their predecessor. (The first extract is seriously mistranslated, while the second contains minor mistakes.) Dykstra claims that these instructions prove that after the Yongzheng reforms, “transfer was no longer only about reckoning revenue”; the incoming magistrates were now expected to audit and inspect matters such as granaries, public works, criminal cases, and orders from superiors, whereas the outgoing magistrates were to provide documentation for verification—contributing to proliferation of documents. To her credit, Dykstra provides a transfer audit from Baxian's local archives from 1786 to showcase the complexity of such documentation.

But was this really a new phenomenon, a result of the Yongzheng “archival turn”? A simple glance at Huang Liuhong's 黃六鴻 Fuhui quanshu 福惠全書 (A Complete Book Concerning Happiness and Benevolence)—arguably the most important handbook for magistrates during the Qing—invalidates Dykstra's claim. It is worth quoting a substantial section from Huang, who wrote in the 1690s during the Kangxi reign:

The transfer of office concerns not only one affair, and it requires many kinds of registers, each which traces respective matters. The registers that need to be made [for transfer] are the following: a register of household and tax grain, a register of miscellaneous taxes, a register of granaries, a register of postal stations and their staff and horses, a register of orders from ministries and superiors, and a register of prisoners in the jail, a register of confiscated goods in the warehouse, and a register of sale prices for salt confiscated from smugglers … After making these transfer registers, the outgoing magistrate audits each item in them. After that, he signs a pledge stating that there are no more incomplete tax collections and keep the pledge [along with the registers]. Then in the company of assistants, he presents in person all these registers made by the various county departments, two copies for each, along with the various archived documents, to the incoming magistrate [or an acting one]. The latter must inspect closely and make sure they contain no errors. And after that, on the front page of these registers, the new magistrate writes “received” and signs off, while the old magistrate writes “yielded” and signs off too. Each keeps a copy (of each register) for proof. The new magistrate then writes a pledge to give to the old magistrate. After that, they each provide pledges to their various superiors, stating that all the archival documents had been transferred. At this point, the old magistrate may finally depart at ease.²⁶

Dykstra is certainly familiar with Huang's book and has quoted it substantially (72, 79). It is hard to believe that she could have failed to notice Huang's chapters on office-transfers. The paragraph above shows that the things Dykstra claims to be new reforms of the Yongzheng reign—archival integrity as demonstrated by the officials’ pledges, the “documentary proliferation” of registers for many aspects of the magistrates’ administrative affairs (taxes, judicial records, granaries, postal stations, confiscated goods, etc.), and the rigor with which transferred documents should be audited—had already been in practice during the Kangxi reign, if not earlier. Only by ignoring this crucial piece of evidence from a well-known source could Dykstra attribute these long-standing practices to the Yongzheng reforms. Moreover, the Qinban zhouxian shiyi from the Yongzheng Reign that Dykstra cites here is not a set of new regulations but rather a very brief digest (only one volume) of existing effective practices. In fact, Dykstra provides no evidence showing the Yongzheng court issued any radically new regulations regarding office transfers.

Finally, to emphasize the revolutionary effects of the Yongzheng “archival turn” on the transfer of offices in local administration, Dykstra resorts to artistic liberty:

The handing over of the registers, accounts, and property of the yamen entailed an opportunity and a responsibility to pore over every facet of the administration. Every debt, every kernel of grain stored, every purchase or sale of grain, every expense passed down by the provincial offices, every type of fund passed up by subordinate yamen, every surcharge, every tax scheme, every conversion rate, every physical office, every street, every bridge, every horse, and every postal station were subject to scrutiny. The deadlines related to cases slated for central review were inherited and had to be accounted for, lest the incoming magistrate begin his tenure by taking the fall for wrongdoings committed by the previous occupant of his office. To make the transition as simple as possible, by the nineteenth century, manuals encouraged outgoing magistrates to refrain from permitting too many new cases, to wrap up open files, to clear out their jails, and to destroy everything that did not have to be preserved (132).

This is one of the most puzzling paragraphs in the book. First, note to the word “every,” which is repeated fourteen times. Did magistrates really have the opportunity to inspect every kernel of grain? Every horse? Every street? Was it even within the magistrate's responsibilities to inspect the streets? Moreover, Dykstra never considers another significant question: did Beijing—or even provincial administrations for that matter—know anything about the contents in these transfer registers? Did all these registers compiled for the transfers ever end up in the central or provincial governments? Or did they just remain in the localities? The latter is most likely. If that was the case, how did this help the central state to gather information about the local administration?

Second, Dykstra asserts that, as a result of the over-rigorous demand for archival integrity, by the nineteenth century manuals encouraged outgoing magistrates to destroy everything that did not have to be preserved. The search for archival integrity, according to her, produced the totally opposite result. But if we examine the citation, a major problem emerges. The only source that Dykstra cites to prove this point again comes from Pan Biaocan’s An Unreliable Treatise, a book printed in 1684—not a nineteenth-century text. If outgoing magistrates were already burning their records in the seventeenth century, how could it be a consequence of Yongzheng’s supposed “archival turn” in the early eighteenth century?

What is more troubling is that Pan’s treatise is not an esoteric source, but a crucial one cited frequently in the book. Recall that Dykstra has used a section of this Kangxi-era book to demonstrate an alleged Yongzheng era approach related to the documentary streams during the “great audit.” Moreover, Dykstra bases Chapter 2 on Pan’s treatise, which discusses Kangxi-era changes. She even provides a short biography of Pan and correctly identifies him as a seventeenth-century author.²⁷ Astoundingly, the section of Pan's treatise that Dykstra cites to make claims for the nineteenth century is the very passage quoted in the prologue (xxiii)—the first block quote in the entire book—to demonstrate that archive burning was a time-honored practice that magistrates used to prevent authority-bearing documents from ending up in the wrong hands. In both cases, she even provides verbatim footnotes introducing the source (xxiii, 132). I am not suggesting that Dykstra knowingly cites a seventeenth-century material to prove a claim about a nineteenth century development for which there are no other sources. But her use of the same Kangxi-era treatise to illustrate three different phenomena that supposedly took place in three different centuries is truly creative.

After reviewing all the evidence analyzed above, I remain unconvinced that the Yongzheng Emperor had launched an “archival conquest” because it is not borne out by the sources. Consequently, the alleged “new information ecology” is not grounded in evidence. This dissection of Chapter 3 serves as a case study of how Dykstra builds major claims almost entirely on misrepresented sources.²⁸ I have gone through many sources in every chapter, finding similar distortions and consistent misuse of historical materials.²⁹ In fact, very few primary sources in this book are interpreted accurately.

Conclusion: The Danger of Stretching the Evidence to Make a Dramatic Story

Although I may have tried the reader's patience, I have aimed to put forth enough evidence to demonstrate that what Dykstra calls the “administrative revolution” of the early Qing was not the accumulation of subtle yet powerful reforms, but rather the imaginings of an author who has misread, mistranslated, and misinterpreted not merely a few, but the majority of the primary sources in the book. None of the author's major claims—Yongzheng's archival conquest, the multiplying and tightening of administrative regulations, the radical increase of local information accessible at the center, or her oddly labeled “case-ification” of administrative affairs—withstand scrutiny. Instead, the evidence shows regulatory tweaks, administrative advice, and bureaucratic minutiae that Dykstra randomly cobbles together to create a story that is groundless, self-contradictory, and ultimately untrue.

For the past two centuries, a consensus has existed among observers and historians alike that a crisis of corruption as a symptom of profound socioeconomic and demographic changes had begun to undermine the power of the Qing in the late eighteenth century. Challenging this long-held view, Dykstra's book suggests that, perhaps, this crisis of corruption was only a new awareness of corruption that had been there all along, made possible by a “new information ecology” and the so-called “administrative revolution.” As the people at the helm of the Qing state learned more about local administrations, they became increasingly convinced of the system's flaws. A crisis of competence was thus merely a crisis of confidence. But since Dykstra has failed to prove that the “administrative revolution” even happened, or that the high-Qing information ecology was in any substantive way “new,” this radical reinterpretation of the crisis of corruption is no more than an illusion.

Dykstra's Conclusion provides a timely lesson for our field. In response to the disillusionment about routine, she claims, a desire for the extraordinary began to capture the imagination of Qianlong-era officials. To illustrate this point, she quotes a 1740 edict, which discusses the problems that emerged as a result of provincial governors’ pursuit of extraordinary achievements. Dykstra has taken this edict out of context, wrongly translated most of it, and misinterpreted it.³⁰ But the first part deserves attention. Here is my translation:

Recently, provincial governors and governors-general often assume that performing basic responsibilities is not sufficient for them to catch the attention of the emperor or the public, so they extend their imagination to experiment with significant reforms. If their programs achieve any little success, they can request recognition of merits … People like these, occupying important offices, hope to use their achievements to show their proactive attitude. If they succeed, they get merits; if they fail, they do not get reprimanded. But in terms of their basic responsibilities such as promoting agriculture, collecting taxes, storing provisions, famine relief, etc., they never do them well. (QLSL 5.7.22.2, vol. 123)

An analogy can be made to the Chinese history field today. When we judge a work of history, or teach students to write papers or theses, do we value and reward only those offering grand and sensational stories? Or do we still care about the basic standards of historical research, such as diligently studying one's sources, faithfully representing them, respectfully engaging with existing scholarly works, backing claims with evidence, and trying to be as factually accurate as possible? In this age of misinformation, if we allow our desire for the extraordinary to run unchecked, we will soon face a full blown “crisis of competence.”