The Huntington Library’s Volume of the Yongle Encyclopaedia (Yongle Dadian 湛樂댕듕): A Bibliographical and Historical Note

Duncan M. Campbell

The Find

During the summer of 2014, a large book in Chinese was brought to the attention of the Huntington Library’s archivist Li Wei Yang 楊立維. The volume, it soon became apparent, had been given to the Huntington Library almost half a century earlier, on 27 December 1968, by Mabel G. Whiting but, having been accessioned, had thereafter languished in the rich obscurity of the library’s backlog. A note tipped into the front cover of the book offered provenance: the volume had been ‘picked out of the debris’ of the library of the Hanlin Academy (Hanlin yuan 翰林院) by Mabel’s father, Joseph L. Whiting (1835–1906), and used to ‘barricade a window’ during the siege of the legations by the Boxers in Peking in 1900.[1] Following the lifting of the siege, when Whiting, a long-term Presbyterian missionary to China, was repatriated to the United States, the volume had accompanied him home.[2] What first caught the archivist’s eye was the fact that rather than having been printed, the book appeared to have been hand copied. Months of detailed work followed as Li Wei Yang sought to ascertain, through research and consultation, whether or not the book was authentically a missing volume of the famous Yongle Encyclopaedia (Yongle dadian 永樂大典), as its title claimed.[3] Once this process had been undertaken, the Huntington Library was able to announce the discovery on 16 October 2014. The volume has subsequently been on display in the library and has been digitised and made available online. Initial discussions were embarked upon about the volume’s possible repatriation to an institution in the People’s Republic of China.

Any book is always at once both a physical object subject to the vicissitudes of transmission, and the vessel of intellectual or literary content produced in a specific historical context, prone forever to shifting tides of understanding and interpretation.[4] In both respects, the story to be told about the Huntington Library’s volume of the Yongle Encyclopaedia is a remarkable one: respectively, a long-lost and magnificent volume of one of the world’s most grandiose attempts to capture under a single title the entirety of useful knowledge, and a chapter from one of the longest and most demanding of a Confucian canon that has now underpinned the trajectory of Chinese civilisation for more than two millennia and which, unexpectedly perhaps, seems likely to continue to do so. In what follows, I present both a bibliographical description of this rediscovered volume and a number of reflections on the history of the encyclopaedia.

The Book

As a book, the Yongle Encyclopaedia is a magnificent object. Remarkably, more than six hundred years after the work was first commissioned and four hundred after this copy of the Huntington Library’s volume was made, its properties are such as to continue to evoke both pride and wonderment.[5] The eminent Qing scholar and official Zhang Tingyu 張廷玉 (1672–1755), chief editor of the History of the Ming (Ming shi 明史), once remarked of it that ‘A handcopy, the strokes of the calligraphy are regular and correct, and the binding is exquisite. Both the paper and the ink give off the fragrance of antiquity’ (乃寫本字畫端楷裝飾工致紙墨皆發古香).[6]

We should first consider the physical properties of the book. All the volumes of the Yongle Encyclopaedia make use of two colours (red[7] for author names and book titles, punctuation [quandian 圈點],[8] marginal lines, borderlines, ‘elephant trunks’ [xiangbi 象鼻] and ‘fish-tails’ [yuwei 魚尾];[9] black for all text) and three point sizes (full-size for the main text [baiwen 白文 in Chinese; lit. white or unadulterated text]; half-size for the commentary and annotations [zhujie 註解], and a further reduced size for comments on the commentary, such as, particularly, ‘The rest of this commentary is the same as the commentary above’ 餘同前注疏). The volumes, measuring some 40.6cm in height by 25.4cm in width, were written on finest white Xuan paper 宣紙[10] in highest quality ink and bound in coarse yellow brocade.[11] The ‘Academy Style’ (guangeti 館閣體) calligraphy employed throughout the work (both original and copy) is said to be modelled on that of the early Ming 明 dynasty court calligrapher (and Hanlin Academician) Shen Du 沈度 (1357–1434) and was also known as the ‘Chancellery Style’ (taigeti 臺閣體); the Yongle 永樂 emperor seems to have had a particular affection for this man’s hand, saying of him that he was ‘… the Wang Xizhi 王羲之 of our dynasty’ — there being no higher compliment than to liken someone’s calligraphy to that of the greatest master of the art.[12]

Figure 1 The Yongle Encyclopaedia (Yongle dadian 永樂大典), inside front cover: handwritten note byJ.L. Whiting). RB 645576, The Huntington Library, San Marino, California

Figure 1
The Yongle Encyclopaedia (Yongle dadian 永樂大典), inside front cover: handwritten note by J.L. Whiting. RB 645576, The Huntington Library, San Marino, California.

When completed in 1408, having involved the efforts of some 2169 scholars working in the then capital of Nanking, the book totalled 22,877 fascicles (juan 卷), bound into 11,095 volumes (ce 冊), and stored on the shelves ten volumes to a box (tao 套). It included, in part or in whole, text taken from around 5500 titles. There remains considerable debate over whether or not it was ever intended that a copy of the work be printed. In any event, a fine copy (commonly referred to as the zhengben 正本 in the secondary Chinese literature) was made of the drafts of the text assembled. To this day, the fate of this copy remains a mystery, and the initial drafts of the work that this copy was based upon were probably consumed by flames in the 1440s. Perhaps this first or original copy was lost when the imperial archives burnt in 1644–45 as the capitals of Peking and then Nanking fell successively to the peasant army of the rebel Li Zicheng 李自成 (1606–45) and the invading Manchus as they replaced the Ming with their own Qing dynasty (1644–1911)? We are not even certain where the original was being held at the time of these cataclysmic events. What is seemingly beyond any doubt, however, is that throughout the entirety of the Qing dynasty nobody is known to have seen this original copy and not a single page of it, let alone a whole volume, has ever surfaced since. This remarkable fact suggests that the biblioclasm was both sudden and total, and was then followed by deliberate bureaucratic obfuscation on the part of those who had been responsible for the loss.[13] Fortunately, however, after a series of fires in the Forbidden City in Peking in the mid-sixteenth century, the Jiajing 嘉靖 emperor (1507–67; r. 1521–67) — the only emperor who himself ever made any use of the book — had ordered a copy be made of the original copy (usually referred to as the fuben 副本).[14] It is a volume (one of only four hundred or so that remain, mostly held in the National Library of China)[15] of this copy that resurfaced in the holdings of the Huntington Library, having been picked up out of the ashes after the siege of the legations by the Boxers by the American missionary Joseph Whiting, given to the library in 1968 by his daughter Mabel, and recognised for what it is by the Huntington Library’s archivist Li Wei Yang. The melancholy fate of this duplicate copy of the encyclopaedia, too, remains a resonant source of debate; the convenient and nationalistic attribution of the loss of all but about four per cent of the original encyclopaedia to the depredations of aggressive foreigners seems both simplistic and far from the truth of the matter.[16]

In Endymion Wilkinson’s estimation (22,937 juan of average 40 half-folio pages a juan, average 400 characters a page) in total, the compendium comprised around three hundred and seventy million characters.[17]

The volumes of the work most proximate in sequence to that discovered at the Huntington Library are juan 10,135 and 10,136 in the Bodleian Library at Oxford (consisting of historiographical texts) and juan 10,286 and 10,287 in the National Library of China (consisting of Taoist texts). Both are within the same rhyme (er zhi 二紙). Also within this rhyme are juan 10,115 and 10,116 (held at the School of Oriental and African Studies in London) and juan 10,309 and 10,310 (held in the National Library of China). Of interest is the fact that both the volumes noted above as being held in the National Library of China were returned to the library in 1951 from the Department of East Asian Studies of Leningrad University, indirectly, by way of a gift to the Ministry of Culture of the then newly established People’s Republic of China.[18]

The most recent discoveries of volumes of the encyclopaedia before that at the Huntington Library in 2014 were made in 1983 (in Shandong province), 1997 (in the library of Aberdeen University in Scotland), and, more recently, in Canada (in private hands, since acquired by the National Library of China).

Figure 2 Yongle Encyclopaedia, (leaf 1, recto), 40.6 cm (h) x 25.4 cm (w). RB 645576, The HuntingtonLibrary, San Marino, California.)

Figure 2
Yongle Encyclopaedia, (leaf 1, recto), 40.6 cm (h) x 25.4 cm (w). RB 645576, The Huntington Library, San Marino, California.

The Emperor

Of the Four Plagues of Books (Si e 四厄) traditionally much feared in China, water, insects, warfare, and fire, over time it was perhaps the last of these that proved the most destructive.[19] Book collectors in China (private and imperial) accepted as a fact of their lives a particular and dreadful moral hazard associated with assembling a large collection of books; the larger your collection, the greater the loss you would be held responsible for when fire struck, and they employed various means of staving off the inevitable: proximity to water for firefighting purposes, strict controls on access and hours of use (so as to avoid the need for dangerous sources of artificial light), word magic (how many of the libraries of China, both private and imperial, make use of the ‘water radical’ in their names?) and so on.

Imperial book collectors over the ages in China, however, seemed somewhat less prone to such moral qualms; they understood the size of their holdings to be an important token of their power and the moral quality of their rule.[20] This seems particularly the case with the Yongle emperor (1360–1424; r. 1402–24) (his reign title means ‘Perpetual Joy’), and his age has been characterised as one of ‘… economic growth, cultural regeneration, territorial expansion, and diplomatic glory’.[21] And yet in the case of the Yongle emperor, a number of additional considerations come immediately to mind. He was an emperor who had usursped the throne on the pretext of ‘Calming a Troubled Age’ (jingnan 靖難) (the phrase could perhaps be better understood as meaning ‘suppressing dissent’), and was certainly implicated in the death of his nephew, the ‘Establishing Civility’ (Jianwen 建文) emperor (1377–1402; r. 1398–1402), when the imperial palace in Nanking was set alight[22] (Jianwen’s short reign was then expunged from the historiographical record).[23] He then engaged in a post-victory purge of extraordinary and violent dimensions,[24] of which Shih-shan Henry Tsai, in his biography of the emperor, says ‘… was among the most brutal and barbarous political acts in Chinese history’.[25] When the previous emperor’s main advisor, Fang Xiaoru 方孝孺 (1357–1402) refused to acquiesce in the usurpation he was sliced in half at the waist and further punished with the extermination of his family to ten degrees of kinship (the only man so punished in Chinese history), while some 870 of his associates were executed.[26]

In a brutal age, the Yongle emperor proved a particularly brutal and ruthless man, and one seemingly also given to outbursts of sanctimonious Confucian moralising.[27] His reign is now understood to have served to intensify a distinct trend in late imperial Chinese political culture towards absolutism;[28] in the long term, also, his removal of the capital from Nanking to Peking in 1421 (at very great expense and for his own immediate political ends) was probably a grave mistake.[29]

The Encyclopaedia

In the case of the encyclopaedia, consideration of the personality of the emperor is of relevance. First, after all, he had allowed his reign title to serve as the book’s title. There were, of course, both proximate and more distant precedents for this procedure, and subsequent ones as well, as in the case during the Qing dynasty of the Kangxi Emperor’s Dictionary (Kangxi zidian 康熙字典, commissioned by the Kangxi emperor (1654–1722; r. 1661–1722) in 1710 and completed in 1716. But all the available sources point to the emperor’s intense and particular engagement in the project. He had rejected the product of an earlier exercise he had ordered his minister Xie Jin 解縉 (1369–1415) to undertake, entitled A Grand Compendium of Literary Sources (Wenxian dacheng 文獻大成), as being ‘insufficiently exhaustive’ (suo zuan shang duo wei bei 所簒尚多未備). In the eyes of many of the emperor’s contemporaries, and in the view of much subsequent scholarship, the fact that the emperor became so quickly engaged in large-scale scholarly projects of this type soon after his enthronement in such murky circumstances demanded that these projects be understood in the light of the possible extra-literary usages of such projects in the specific political circumstances of the time. To some extent or another, Yongle’s victory over his nephew represented the victory of the eunuch faction at court over the faction of the scholar-official; that is, power at court, increasingly, was in the hands of a very particular group of functionaries who were the personal servants of the emperor and not as bound by the conventions of the imperial bureaucracy as were their scholar-official contemporaries. In this context, the encyclopaedia projects can be understood to be a means of both persuading that scholar-official faction that he was a man devoted to learning, and of keeping otherwise disaffected scholars busy at their desks — not the first nor the last time that such a ruse was employed in Chinese history. The late-Ming scholar Sun Chengze 孫承澤 (1592–1676), in his wonderfully informative Record of Dreams of the Capital (Chun ming meng yu lu 春明夢餘錄), is very explicit in his view:

When the emperor initiated his ‘Calming a Troubled Age’ campaign, dissention was rife throughout the empire and so he made use of this scholarly project in order to dissipate obstacles in his path, this being his true intension at the time (至靖難之舉不平之氣遍於海宇文皇借文墨以銷壘塊此實係當日本意也).[30]

Further, a close reading of both the edict that resulted eventually in the Yongle Encyclopaedia (as recorded in the Veritable Records of the Ming (Ming shilu 明實錄), dated first month of the first year of his reign) and the ‘Preface’ that he wrote for it once it was completed (also recorded in the Veritable Records of the Ming and dated eleventh month of the fifth year) tend to confirm the impression of an emperor with explicit and ulterior motives. He argued that

The objects and affairs, both ancient and modern, of All-under-Heaven are recorded here and there in various books, such that their bulk is overwhelming, making them difficult to consult. We desire that the objects and events recorded in these tomes be gathered together and categorised, and arranged by means of rhyming category, in order to provide for ease of consultation and examination, as if one were plucking an object from one’s bag. I’ve often taken a look at the Assembled Jades of the House of Rhyme [Yunfu qunyu 韻府群玉] and the Confluence of Historical Resonances [Huixi shiyun 回溪史韻], and although affairs are comprehensively recorded here, the sources used in the compilation of these two books were too restricted, and thus their accounts of events are too sketchy. It is expected that you will undertake your work in accordance with my intention. Phrases found in all the books, from the very beginning of writing — the classics, the histories, the Masters, the literary collections, the works of the various philosophers, books of Astrology, local gazetteers, the Yin and Yang experts, of the physician and the fortune tellers, of Buddhist monk and Taoist priest, of the artisans and the artists — should be excerpted under a single title, with no thought given to the vast bulk that will result from this procedure [wu yan hao fan 毋厭浩繁].[31]

As emperor, he believed that there was an urgent need ‘to unify confusing systems and standardise government regulations and social customs’ (bi you yi tong zhi zhi zuo suo yi qi zheng zhi er tong feng su 必有一統之制作所以齊政治而同風俗). His ‘Preface’ manifests his pleasure that his orders had been carried out:

The product of the labour of this exhaustive process of compilation is a book that can satisfy all possible inquiries, such that searching for a word by means of its rhyme and examining affairs by means of this word, any reader can trace the trajectory of something from beginning to end, as easily as shooting a swan with one’s bow [ru she zhong hu 如射中鵠]. Nothing will remain hidden once you open up the pages of this book [kai juan er wu suo yin 開卷而無所隱].[32]

Figure 3 (leaf 2, recto). RB 645576, The Huntington Library, San Marino, California.

Figure 3
Yongle Encyclopaedia, (leaf 2, recto). RB 645576, The Huntington Library, San Marino, California.

Little wonder, then, that when the English-language version of Wikipedia announced in 2009 the posting of its three millionth article it did so in terms of having finally surpassed the scale of the Yongle Encyclopaedia as the world’s largest encyclopaedia.

Like all universal libraries, then, the Yongle Encyclopaedia is both slightly mad and vainglorious, and somewhat troubling of conception. Judging from the manner in which the emperor spoke about its purpose, the project of the book was not intended to generate, preserve, or disseminate knowledge, but rather to fix and consolidate existing knowledge and to make it more readily available (exclusively to the emperor personally and those closest to him) for the purposes of ruling All-under-Heaven.[33] In that sense, the usual English translation of its title employing the word ‘encyclopaedia’ seems at least doubly inappropriate; first, because it is not, in fact, an encyclopaedia as we commonly understand this term nowadays, and, second, because it would be inappropriate, through the connotations of the word ‘encyclopaedia’, to associate this work with, for instance, the European Enlightenment and the Encyclopédie, ou dictionnaire raisonné des sciences, des arts et des métiers.[34] The title should better be rendered Grand Compendium of Yongle or Yongle’s Grand Canon or some such.[35] In Chinese bibliographical terms, it is a ‘Book of Categories’ (leishu 類書). In essence, what the process of its composition involved was that copies were made of all the books then housed in the imperial library, the Pavilion of Literary Profundity (Wenyuan ge 文淵閣), supplemented by others acquired for the purpose of the project. These copied texts where then disaggregated, with their chapter divisions being retained but almost never whole titles being included, before being rearranged in accordance with the seventy-six rhyming categories the last (or sometime most important Chinese character) of that chapter title belonged, with a 60-fascicle index provided to give readers some idea of what had been included.

To us now, arranging an encyclopaedia by means of rhyming categories might seem an odd thing to do. This was not an innovation in the case of this encyclopaedia, however, and made perfect sense.[36] Chinese is not an alphabetic language, and in the traditional Chinese literary and intellectual world, the ability to write a good poem or effective prose was vital, rhyme being critical to both genres. A textual architecture arranged according to rhyme category provided readers of a book so arranged with the most effective search engine, and the rhyming dictionary was thus China’s true memory palace. The rhyming dictionary employed in this case, appropriately, was that compiled by the scholar Yue Shaofeng 樂紹鳳 (fl. 1355–80) and completed in 1375 on the orders of Yongle’s father, and which had been named after his reign title The Correct Rhymes of the Hongwu Era (Hongwu zhengyun 洪武正韻).[37]

What lends the encyclopaedia an element of distinction is the extent to which it cast its net widely, beyond the constricting definitions of the orthodox classical Chinese Confucian textual corpus to include items of popular and dramatic literature written in the vernacular. And neither is there any evidence that the emperor had his compilers engage in systematic censorship of the content of the encyclopaedia, for whatever reason.[38] Otherwise, the encyclopaedia seems a somewhat intellectually conservative and reductive project, to the extent to which it serves to reduce knowledge to information, text to extract, the philosophic (or religious) to the simply practical.[39] It was certainly immensely labour-intensive, involving the energies of many compilers over the course of the years 1405–07. Nominally under the general editorship of Yao Guangxiao 姚廣孝 (1335–1418) (also known by his clerical name Daoyan 道衍),[40] and Xie Jin, the bulk of the scholarly work was probably undertaken by men such as Hu Yan 胡儼 (1361–1443), Liu Jichi 劉季箎 (1363–1423), and, especially, Chen Ji 陳濟 (1363–1423). What is remarkable (and yet another element of mystery in the tale of the encyclopaedia) is that none of the official biographies of the scholars most involved in its compilation (all of which biographies were written after the dynasty had fallen) make any mention of their engagement in the project. A number of them were either executed (Xie Jin) or imprisoned (Chen Ji) soon after the completion of the encyclopaedia. From a twenty-first century perspective, perhaps, we can think about the Yongle Encyclopaedia as yet another instance of the frequently fatal embrace in China between knowledge and power.[41]

The Rites

The Huntington Library’s volume comprises juan 10,270 (28 folio pages) and 10,271 (21 folio pages) of the encyclopaedia, and contains the partial text of Chapter Six of the Li ji 禮記 — one of the five books that now constitute the Confucian canon.[42] Since its compilation in the Han dynasty (206 BCE–220CE), making use of text said to date from much earlier, the book had become an increasingly critical one in the context of orthodox Confucian state ideology. As Michael Nylan has argued,

Ritual was of inestimable importance to premodern society in China, for the single term ‘rites’ or ‘ritual’ (li 禮) denoted the full panoply of appropriate and thus mutually satisfying behaviors built upon emotional insights. These behaviors, expressed in dress, countenance, bodily posture, and verbal phrasing, where designed to strengthen communal bonds among the living, between the living and the dead, and with the gods.[43]

As Nylan also makes clear (p.171), this text (along with the other two texts that deal with the rites, the Rites of Zhou [Zhou li 周禮] and the Prescription [Yi li 儀禮]), ‘… more than any other of the Five Classics, inspired dramatic attempts to rethink and reorder Chinese social and political realities’. The particular chapter copied into the encyclopaedia, entitled ‘King Wen as Son and Heir’ (Wen wang shizi 文王世子), deals with the ritualistically complicated status of a child who was both son to the ruling emperor and his heir apparent, and thus, sometime in the future to be both emperor himself and ‘Son of Heaven’ (Tianzi 天子).

The first third of the chapter is missing from the Huntington Library’s volume. Missing also is the last short section of the text, consisting of between three and five folio pages. We may, therefore, conjecture that (a) this first section of the text (with its commentaries) was included in the no-longer-extant juan 10,269; and (b) that although for the bulk of the encyclopaedia the binding is two juan 卷 or fascicles to each ce 冊 or volume, examples do exist of volumes that have either one or three fascicles, and the extent of this text would suggest that this volume might well be an instance of the latter format. This second conjecture might explain why the volume had to rebound at some point in time once it had arrived in the US, as stated in L.J. Whiting’s letter pasted into the volume (‘The cover was put on in this country’); when he picked it up off the ground in 1900, both the front and back covers of this volume had been destroyed, it seems, during course of the siege of the legations, thus exposing to depredation both the first fascicle and the final pages of the volume.[44]

Item # 13 of the ‘Principles of Compilation’ (Fanli 凡例) for the encyclopaedia as a whole deals with the Confucian canon and its commentaries, as given below, in my translation:

The Book of Changes (Yijing 易經), the Book of Songs (Shijing 詩經), the Book of Documents (Shujing 書經, the Spring and Autumn Annals (Chunqiu< 春秋), the Rites of Zhou (Zhou li 周禮), and the Book of Rites (Li ji 禮記) all have attached to them prefaces, lists of contents, explanations of the filiation of the text through the various Confucian scholars, along with discussions of the overall meaning of work at hand. For present purposes, these have all be included under the character of the title of the particular classic (so that, for instance, such text attached to the Book of Changes is given under the character ‘I’ 易). The complete text of the various sections of these books is to be found, however, either under the first character of the section title or the most important character (…). As for the various commentaries to these texts, the choice has been to include, first, those written by the famous scholars of the Han, Tang, and Song dynasties (…), with later commentaries appended, included in chronological order. Occasionally the matters discussed in these texts involve famous objects that are of particular relevance to the maintenance of good order, and so these, too, have been included in the compendium in accordance with the rhyming category of their name. Of the Four Books, only in the case of the Great Learning (Daxue 大學) and the Mean (Zhongyong 中庸)has it proven difficult to disaggregate the text, and so they have been included in complete form (so that, for instance, the Great Learning is to be found under the character ‘learning’ (xue 學). The procedure adopted with both the Analects and the Book of Master Mencius (Mengzi 孟子) has followed this rule, and it has been also applied in the case of both the Five Classics and all the other books by the various philosophical masters.

The sections of the encyclopaedia covering the various canonical works dealing with the rites and ritual usages were the especial responsibility of the Hangzhou scholar Gao Deyang 高得暘 (1352–1420), one of the contributing editors of the overall project. In the estimation of the Qing dynasty scholar Dai Zhen 戴震 (1724–77), his work was done ‘… with particular care accorded the criteria of selection’ 去取尚為精審.

The Commentaries

The text of the chapter of the Book of Rites found in the Huntington Library’s volume is equipped with (overwhelmed by, perhaps one should say) commentary by a total of thirteen separate and named commentators.[46] These commentators can be grouped into three categories (and here I am particularly indebted to the preliminary analysis done of the text by Liu Bo of the National Library of China):

A. Eight commentaries (dating from the Han, Tang, and Song dynasties) that one would expect to see represented in most editions of the work, cited chronologically, as follows: Zheng Xuan 鄭玄 (127–200), Book of Rites Annotated (Li ji zhu 禮記注); (ii) Lu Deming 陸德明 (550?–630), An Explication of the Texts of the Confucian Classics: Phonetic and Semantic Glosses to the Book of Rites (Jingdian shiwen: Li ji yinyi 經典釋文·禮記音義); (iii) Kong Yingda 孔穎達 (574–648), Book of Rites: Correct Glosses (Li ji zhengyi 禮記正義); (iv) Zhu Xi 朱熹 (1130–1200), Classified Conversations of Master Zhu (Zhuzi yulei 朱子語類); (v) Wei Liaoweng 魏了翁 (1178–1237), Book of Rites: Essential Glosses (Li ji yaoyi 禮記要義); (vi) Wei Shi 衛湜 (fl. 1226), Book of Rites: Collected Explanations (Li ji jishuo 禮記集說); (vii) Chen Hao 陳澔 (1261–1341), Book of Rites: Collected Explanations (Li ji jishuo 禮記集說); (viii) Huang Zhen 黃震 (1213–81), Master Huang’s Daily Notes (Huangshi richao 黃氏日抄).

B. One commentary (dating from the Song dynasty) that is very rare: (i) Zhu Shen 朱申 (fl. 1050), Sentence by Sentence Interpretations of Every Section of the Book of Rites: Punctuated with Phonetic Glosses (Yindian li ji xiangjie jujie 音點禮記詳節句解).[47]

C. And four commentaries (also dating from the Song dynasty) that are unique to this volume: (i) Lü Zuqian 呂祖謙 (1137–81), Lü Bogong’s Accompanying Annotations: Punctuated with Phonetic Glosses (Lü Bogong yindian pangzhu 呂伯恭音點旁註) (juan 10270, 12a, 16a; juan 10271, 6a, 20a);[48] (ii) Chen Li 陳櫟 (1252–1334), Book of Rites: Collected Glosses and Detailed Explanations (Li ji jiyi xiangjie 禮記集義詳解) (juan 10270, 2a, 4b–5b, 12a, 13a, 16a–16b, 17a, 21a, 22b, 25b–26a; juan 10271, 1a–1b, 2a, 6a–6b, 7b, 9b–10a, 12b, 20a–21a);[49] (iii) Peng shi 彭氏 (?), Master Peng’s Illustrated Glosses and Annotations (Pengshi cuantu zhuyi 彭氏簒圖註義) (juan 10270, 3a, 9a–10a, 12b, 18a–18b, 23a, 26b; juan 10271, 1b–2a, 7b, 13a); (iv) Shi Jiongsun 史駉孫 (fl. 1324), Glosses to the Canon (Jing yi 經義) (juan 10270, 10a–11a).

In the case of this third category of commentary, that is text unique to this volume, the most important of the commentators was that of Lü Zuqian. Sadly, his four interventions recorded in this volume of the encyclopaedia are all simply (if importantly) phonetic glosses for the reading of particular characters.


Upon the Yongle emperor’s death in 1423, at Yumuchuan 榆木川 in the remote desert and on campaign against the Mongols one last time, the news was kept secret until the succession of his son could be ensured.[50] Encased in a tin coffin in an attempt to slow the rate of decomposition, food continued to be served to him in death as it had in life, as had been the case with the First Emperor of Qin. The emperor was later given both the temple name Grand Progenitor (Taizong 太宗) usually reserved for second emperors (in 1537 this was changed to Completing Emperor [Chengzu 成祖]), and the (outrageously exaggerated) posthumous name: The Literary Emperor who Opened up Heaven and Enlarged the Way, who was High-Minded and Brilliant and who Initiated our Present Good Fortune, who had both Sagely Martial Prowess and Divine Merit, and who was Purely Benevolent and Completely Filial (Qitian hongdao gaoming zhaoming shengwu shengong chunren zhixiao wen huangdi 啟天弘道高明肇運聖武神功純仁至孝文皇帝). His tomb in the hills just north of Peking, named Everlasting Tomb (Changling 長陵), is the largest of the thirteen imperial Ming tombs. A total of thirty palace women (including sixteen concubines) hung themselves in preparation for accompanying him to his grave. It is said that once his magnificent encyclopaedia was completed in 1408, the sixth year of his twenty-two-year-long reign, he never once consulted it. And yet, over the course of time since the end of the Ming dynasty, the various volumes of the copy of the compendium that were available to scholars (and of course access to it by scholars particularly after it’s relocation to the Hanlin Academy during the reign of the Yongzheng emperor (1678–1735; r. 1722–35)[51] much accelerated the disappearance of the encyclopaedia) has resulted in the ‘recovery’ (jiyi 輯佚) of between 500 and 600 books that had been otherwise lost to us and posterity. The legacy of the Yongle emperor has been a complex one; the Huntington Library may now lay a permanent and far-flung claim to a place within the textual dimensions of this legacy.

Figure 4 (leaf 3, recto). RB 645576, The Huntington Library, San Marino, California.

Figure 4
Yongle Encyclopaedia, (leaf 3, recto). RB 645576, The Huntington Library, San Marino, California.

A preliminary version of this paper was first presented, in part, as a lecture delivered on 8 January 2015, at the Huntington Library in San Marino, California, hosted by Dr Steve Hindle, W.M. Keck Foundation Director of Research. In working on both the lecture and this version of the paper, I am much indebted to Li Wei Yang, my fellow presenter on that evening and recently appointed Curator of Western American History with the Huntington Library, for his continued advice. Alan Jutzi, then Avery Chief Curator of Rare Books with the Huntington Library, too, as always, has been generous with advice and support. The report on the volume of the encyclopaedia discussed in this paper presented by Dr Liu Bo 劉波 of the National Library of China and which served to authenticate the volume is wonderfully informative; his conclusions inform this paper. I am also most grateful to the two anonymous readers of the paper; I have tried to respond to as many of their helpful suggestions as time and circumstances permit. Warm thanks are also due the editor of East Asian History, Benjamin Penny, and his able and long-standing editorial assistant Lindy Allen.

Duncan M. Campbell
Victoria University of Wellington

  1. Peter Fleming, The Siege at Peking (1959; rpt. Hong Kong: Oxford University Press, 1986) provides a very readable account of the event. E.D. Grinstead, ‘ “Yung-lo ta-tien”: An Unrecorded Volume,’ The British Museum Quarterly 26.1/2 (1962): 30–33 quotes a description of the events of the afternoon of 22 June 1900, from the unpublished diary of Captain Francis Garden Poole, the man in charge of the sector of the British Legation adjoining the Hanlin Library. On the loss of the Hanlin Library, see also Donald G. Davis, Jr and Cheng Huanwen, ‘The Destruction of a Great Library: China’s Loss Belongs to the World,’ American Libraries 28.9 (1997): 60–62.
  2. Whiting, about whom little more biographical detail seems presently available, appears to have known Chinese and thus been aware of the nature of the volume. Pasted onto the verso of the front cover he provides a handwritten bibliographical reference (‘The Great Cyclopedia of Yung Lo/ Vol. 10270/ Subject Sons./ Teaching the Heir Apparent’) and his own translations of a number of passages of the text of the volume, including the following: ‘Rites and Music are the essentials in teaching the Heir Apparent. Music to cultivate the inner man, rites (or rules of propriety) to polish the external conduct’.
  3. As argued below, I believe that the usual translation of the title of this book with the word ‘encyclopaedia’ is inappropriate. It is employed in this paper only because the bulk of the secondary English-language literature continues to label it as such. The secondary literature in Chinese on the history of this book is both voluminous, and, frequently, tendentious. An exception is Zhang Sheng 張昇, Research on the Circulation of the Yongle dadian and Attempts to Recover its Text (Yongle dadian liuchuan yu jiyi yanjiu 永樂大典流傳與輯佚研究) (Beijing: Beijing shifan daxue chubanshe, 2010). Zhang Sheng’s A Collection of Research Materials on the Yongle dadian (Yongle dadian yanjiu ziliao jikan 永樂大典研究資料輯刊) (Beijing: Beijing tushuguan chubanshe, 2005) is also invaluable. I am grateful to Caleb Hoyle for his timely supply of these two volumes.
  4. ‘For a book,’ D.F. McKenzie argues, ‘is never simply a remarkable object. Like every other technology it is invariably the product of human agency in complex and highly volatile contexts which a responsible scholarship must seek to recover if we are to understand better the creation and communication of meaning as the defining characteristic of human societies’, for which see his Bibliography and the Sociology of Texts (Cambridge: Cambridge University Press, 1999), p.4 (emphasis in the original). Timothy Brook reminds us of the equivocal status of the book in late imperial China in particular when he argues that they are at once ‘…objects through which information is stored and communicated’ and ‘social objects, bearing and transmitting messages about status’, for which see Timothy Brook, ‘Communications and Commerce,’ in eds Dennis Twitchett and Frederick Mote, The Cambridge History of China: Volume 8: The Ming Dynasty, 1368–1644, Part 2 (Cambridge: Cambridge University Press, 1998), p.666. See also, Duncan M. Campbell, ‘The Moral Status of the Book: Huang Zongxi in the Private Libraries of Late-Imperial China,’ East Asian History 32/33 (2006/2007): 1–24. I offer a caveat to what follows: much about the history of the book I discuss remains speculative and hotly disputed.
  5. The late-Ming dynasty (1368–1644) scholar Zhang Rulin 張汝霖 (1561–1625) may well have been one of the last people to have seen the original copy. His grandson, the essayist and historian Zhang Dai 張岱 (1597–?1684), tells us that he did so as he was putting the finishing touches to a massive encyclopaedia that he had been working on for more than thirty years (entitled My Mountain of Rhymes [Yun shan 韻山]). Shown a box of books containing thirty-odd volumes of the Yongle Encyclopaedia that had been secreted out of the Palace Library by a friend then working in the Ministry of Rites, he immediately put down his brush and sighed with admiration: ‘Goodness! How inexhaustible is this world of books. In my own paltry efforts I’ve been a bit like Jingwei, that bird of legend, trying to fill in the eastern sea one pebble at a time. By contrast with this book, what little I have to show for myself!’ (書囊無盡精衛銜石填海所得幾何), for which, see Xia Xianchun 夏咸淳 and Cheng Weirong 程維榮 eds, Dream Memories of Taoan: Dream Search for West Lake (Taoan mengyi: Xihu mengxun 陶庵夢憶: 西湖夢尋) (Shanghai: Shanghai guji chubanshe, 2001), p.98.
  6. For which, see Zhang Sheng, A Collection of Research Materials on the Yongle dadian, p.129.
  7. Red was a colour the use of which in writing was otherwise restricted to the emperor himself; its use in the case of the encyclopaedia served to confirm the extent to which it was, in no uncertain manner, an imperial production.
  8. The careful punctuation of both main text and commentary throughout the volumes of the encyclopaedia is unusual for texts of this type in a traditional context and thus of considerable interest; it is, perhaps, an indication of the extent to which the family origins of the ruling house of the newly established dynasty were such that their level of literacy was still somewhat rudimentary. The punctuation marks (ju 句) were impressed upon the paper with a round seal. A reduced-size flyleaf inserted into one of the fascicles of the work found in the Chester Beatty Library in Dublin lists two editors (of the six names listed as responsible for its copying) as being in charge of the punctuation: ‘Supervisors of Punctuation’ (Quandian jiansheng 圈點監生). For this information, see Shane McCausland, ‘Copying and Transmitting, Knowledge and Nonsense: From the Great Encyclopaedia to A Book from the Sky,’ in eds Nick Pearce and Jason Steuber, Original Intentions: Essays on Production, Reproduction, and Interpretation in the Arts of China (Gainesville: University Press of Florida, 2012), pp.236–63.
  9. These printer’s markers (broad lines to top and bottom of the ‘block heart’ (版心) of the ‘block face’ (版面), and a pair or more of sharp-angled spots at the upper and lower parts of the centre of the folio, respectively), in the case of the encyclopaedia all hand-drawn rather than printed, were vital in ensuring that the folds made to the folio pages were accurately done. For an illustration, see Tsien Tsuen-hsuin, Science and Civilisation in China: Volume 5: Chemistry and Chemical Technology: Part I: Paper and Printing (Cambridge: Cambridge University Press, 1985), pp.222–23.
  10. Tsien Tsuen-hsuin (in the work noted in the previous footnote) provides the definitive English-language treatment of traditional Chinese paper and its manufacture, for which see especially pp.38–64. Xuan paper is made from the bark of the blue sandalwood tree (Pteroceltis tartarinowii, Maxim.), in Chinese the qingtan 青檀.
  11. The colour chosen for the brocade covers, too, deliberately connoted the imperial provenance of the book.
  12. For both a portrait of this man (Fig.144), and a photograph of one of his ink stones (Fig. 145), see Craig Clunas and Jessica Harrison-Hall, eds, Ming: 50 Years that Changed China (Seattle: University of Washington Press, 2014), p.169. Shen Du’s calligraphy has a lineage that stretches back, through the early Ming dynasty calligrapher Song Ke 宋克 (1327–87), to the style developed by Zhao Mengfu 趙孟頫 (1254–1322), the scion of the ruling house of the Song dynasty (960–1279) who had taken up office under the Yuan (1271–1368). Shen’s calligraphic style also served as a model for a number of printing projects. On Shen Du, see Shane McCausland, ‘Copying and Transmitting, Knowledge and Nonsense: From the Great Encyclopedia to A Book from the Sky,’ in eds Nick Pearce and Jason Steuber, Original Intentions: Essays on Production, Reproduction, and Interpretation in the Arts of China, pp.256–58.
  13. Miao Quansun 繆荃孫 (1844–1919), the founding director of the library that was eventually to become the National Library of China, argues persuasively in his ‘An Investigation into the Yongle Encyclopaedia’ (Yongle dadian kao 永樂大典考), for which, see Zhang Sheng, A Collection of Research Materials on the Yongle dadian, pp. 243–56, that this original copy was lost in the disastrous fire in the Forbidden City in 1797 that destroyed completely the Palace of Celestial Purity (Qianqing gong 乾清宮), a palace that had also burnt to the ground during the Wanli era of the Ming (1572–1620).
  14. It appears that Zhang Chenshi 張忱石, in his Anecdotes about the Yongle Encyclopaedia (Yongle dadian shihua 永樂大典史話) (Beijing: Zhonghua shuju, 1986), is the originator of the (almost certainly apocryphal) story that the Jiajing emperor insisted that the original copy be buried with him. The long reign of this unpleasant man (so unpleasant was he that on one occasion his own concubines tried to strangle him) proved a disastrous one for the dynasty; during a period of almost twenty-five years, he refused to meet with his officials, relying instead upon the services of either incompetent and corrupt men such as Yan Song 嚴嵩 (1480–1567), or Taoist priests and eunuchs. John Dardess, in his Ming China, 1368–1644: A Concise History of a Resilient Empire (New York: Rowman & Littlefield, 2012), describes him as ‘… one of the most self-centered, self-indulgent, short-tempered, and humorless autocrats in the country’s history’ (p.49). He seems to have had a particular affinity for the Yongle emperor; the second character of the reign title he took, ‘Glorious Calm’ (Jiajing 嘉靖), is the same as the first character of the term that the Yongle emperor used as pretext for his usurpation, while, posthumously, the name given to the Jiajing emperor’s mausoleum, ‘Perpetual Tomb’ (Yongling 永陵), makes explicit reference to the reign title of his predecessor. Significantly in this respect, both men had not been in direct line to succeed to the throne. In the Jiajing emperor’s case, his insistence on promoting to imperial rank his own parents (rather than accepting posthumous adoption by the recently deceased Zhengde emperor (1491–1521; r. 1505– 21) as precedent required), led to a bitter dispute at court which lasted for over a decade and which is referred to as the ‘Great Rites Controversy’ (Da li yi 大禮議).
  15. Zhang Sheng, Research on the Circulation of the Yongle dadian and Attempts to Recover its Text, pp.239–54, lists the conditions and whereabouts of 418 volumes of the encyclopaedia.
  16. In his diary, the eminent Qing official Weng Tonghe 翁同龢 (1830–1904) provides a note that perhaps better reflects the truth of the slow loss of volumes of the duplicate copy of the encyclopaedia: ‘On the sixth day of seventh month of the eleventh year of the reign of the Xianfeng emperor (1861), I went to the ministry in search of a book, catching sight of the Yongle dadian as I did so. The book was housed in the Pavilion of Respect for the One (Jingyi ting 敬一亭) and had long languished in neglect. The pavilion was a three-bay one, with a throne sited in the middle bay, and twelve bookcases placed in each of the other bays. The encyclopaedia had pages lined in red and had been finely calligraphed, expertly bound. In size, it was about two chi 尺 tall, more than a chi wide. The volumes took up no more than half the shelf space; they were all covered by more than an inch of dust, and scattered in such disorder that one was unable to locate whatever it was that one wanted to get hold of (咸豐十一年七月初六日到署檢書見永樂大典是 書藏敬一亭久無人問矣亭屋三櫞中設寶座旁列書架十二大典本以朱絲界畫繕手工整高二尺許寬尺許在架者不及架之半塵積寸餘零落不能觸手矣). For which, see Chen Yijie 陳義傑, ed. Diary of Weng Tonghe (Weng Tonghe riji 翁同龢日記) (Beijing: Zhonghua shuju, 1997), p.127. An entry in this diary dated 1894 notes that only around 800 volumes of the encyclopaedia remain (剩八百餘本), Vol.5, p.2706.
  17. For which, see his fine entry on the compendium in his indispensable Chinese History: A New Manual (Cambridge: Harvard University Press, 2012), pp.958–59.
  18. The 62 volumes presently held by the Palace Museum in Taipei had been part of a collection of some 30,000 books that were held in safekeeping (and microfilmed) at the Library of Congress in Washington DC from 1941, by arrangement with the National Peiping Library, as it was then known. In 1965 this collection of books was transferred to the National Central Library in Taiwan for temporary custody until the Palace Museum building was completed. For an account of this footnote in the history of the Yongle Encyclopaedia, see Tsuen-hsuin Tsien, ‘How Chinese Rare Books Crossed the Pacific,’ in his Collected Writings on Chinese Culture (Hong Kong: The Chinese University Press, 2011), pp.236–39. The late Tsuen-hsuin Tsien ends his essay: ‘It is hoped that these rare materials, China’s national treasures from thousands of years and numerous dynasties, will eventually be returned to their original home’ (p.239). For a brief account of the two volumes of the encyclopaedia held by the Harvard-Yenching Library, see Shum Chun, ‘The Chinese Rare Books: An Overview,’ in Treasures of the Yenching: Seventy-Fifth Anniversary of the Harvard-Yenching Library: Exhibition Catalogue (Cambridge: Harvard-Yenching Library, 2003), pp.8–10.
  19. The Huntington Library’s volume of the encyclopaedia shows traces of having suffered from all four of these plagues.
  20. The Yongle emperor certainly well understood the dangers posed by fire. In the sixth year of his reign (1408) a fire in the Jiangshan Monastery 蔣山寺 appears to have completely destroyed the blocks carved to print the edition of the Buddhist Canon commissioned by his father and known as the Hongwu Southern Canon (Hongwu nanzang 洪武南藏), completed in 1398. On this, see Long Darui, ‘A Note on the Hongwu Nanzang, a Rare Edition of the Buddhist Canon,’ The East Asian Library Journal 9.2 (2000): 112–47. The emperor was later to insist both that this lost edition be recommissioned (to be known as the Yongle nanzang 永樂南藏) and that an edition also be prepared in Peking covering northern Buddhist textual traditions, known as the (Yongle beizang 永樂北藏). As I was usefully reminded by one of the anonymous readers of this paper, the Yongle emperor’s complicated relationship with his father, the Hongwu emperor (1328–98; r. 1368–98), finds expression in the extent to which, as in this case, he both continues and supersedes the compilation of a number of scholarly projects first commissioned by Hongwu.
  21. Perpetual Happiness: The Ming Emperor Yongle (Seattle: University of Washington Press, 2001), p.211.
  22. Rumours persisted throughout the Yongle emperor’s reign that the Jianwen emperor had escaped the palace dressed as a monk. Some have argued that emperor’s dispatching of the eunuch admiral Zheng He 鄭和 (1371–1433) on his celebrated expeditions overseas (another much-remarked upon but often misrepresented aspect of the Yongle era) was in part motivated by a search for the missing emperor.
  23. This historiographical fiction was effected by extending the reign of the Hongwu (‘August Martial Might’) emperor for a further four years after his death. The reign of the Jianwen emperor was only restored to the records in 1595.
  24. Timothy Brook, in his The Troubled Empire: China in the Yuan and Ming Dynasties (Cambridge: Harvard University Press, 2010), speaks of the persisting ‘stench of illegitimacy’ of Yongle’s reign.
  25. Perpetual Happiness: The Ming Emperor Yongle, p.71.
  26. Famously, after he had been sentenced to ‘death by a thousand cuts’ (lingchi 凌遲) by the emperor, Fang Xiaoru was said to have responded: ‘Let’s not talk about just nine generations; what about ten?’ (莫說九族十族何妨).
  27. Benjamin A. Elman, ‘ ‘‘Where is King Ch’eng?”: Civil Examinations and Confucian Ideology During the Early Ming, 1368–1415,’ T’oung Pao 19.1/2 (1993): 23–68, discusses the process whereby, in collaboration with a number of compliant Confucian scholars, the Yongle emperor sought to disguise the circumstances of his accession to the throne and to become recognised as a ‘sage-king’ (sheng wang 聖王).
  28. According to the late-Ming loyalist scholar Huang Zongxi 黃宗羲 (1610–95), after the rupture of the Yuan dynasty, ‘… nothing at all survived of the sympathetic, benevolent, and constructive government of the early sage kings’. As Timothy Brook makes clear in his discussion of this view, Huang’s comments were directed at the new Manchu rulers of the Chinese empire, but that (as was the case with both the Hongwu and Yongle emperors), ‘The charismatic Mongol khan having become a routine Chinese emperor, the potential nonetheless lingered for the Chinese emperor to claim the charisma of a Mongol khan, and to act without regard for the constitutional constraints of emperorship’, for which, see his The Troubled Empire: China in the Yuan and Ming Dynasties, p.81. As noted by Benjamin Elman, whereas during the Tang dynasty high officials had sat alongside the emperor at court, by the Ming and Qing dynasties, officials prostrated themselves before a seated emperor, for which, see his ‘‘‘Where is King Ch’eng?” z’, p.46.
  29. Huang Zongxi, one of traditional China’s most insightful political theorists, makes this argument powerfully in his Waiting For the Dawn: A Plan for the Prince (Mingyi dai fang lu 明夷待訪錄) from a variety of strategic, cultural, and economic perspectives: ‘It may be asked, ‘Why did the Northern Capital fall so quickly? What was the reason for it?’ I say it may have fallen for more than one reason, but since a mistake (shi suan 失算) was made in the original establishment of the capital, there was no hope of saving it’, for which, see Wm. Theodore de Bary, trans., Waiting for the Dawn: A Plan for the Prince (New York: Columbia University Press, 1993), p.122. In his celebrated letter to Huang after having read this work, Gu Yanwu 顧炎武 (1613–82), Huang’s fellow Ming loyalist, disagrees with this argument: ‘Mo-ling [Nanking] is only suited to serve a minor state’, for which, see Waiting for the Dawn, p.171.
  30. For which, see Sun Chengze, Record of Dreams of the Capital (Chun ming meng yu lu) (1883; rpt. Hong Kong: Longmen shudian, 1965), p.117.
  31. For which, see Zhang Sheng, A Collection of Research Materials on the Yongle dadian, p.12.
  32. For which, see ibid., p.17. To the emperor’s mind, this ease of use is in contrast to the difficulties previously encountered when trying to trace an idea or an event in the past, which he likened to ‘panning sand to uncover the gold, scouring the ocean’s depths for pearls’ (tao jin yu sha tan zhu yu hai 淘金於沙探珠於海).
  33. To my mind, it is significant in this respect that none of the recommendations made during the Ming that volumes of the encyclopaedia be published and made more widely available were ever realised. Of course, ruling the Ming was no mean feat. Craig Clunas neatly captures the dimensions of the task that faced the Yongle emperor: ‘… everything about Ming-period China was on a grander scale than that of its contemporaries … It had a greater land area, bigger cities (and more big cities), bigger armies, bigger ships, bigger palaces, bigger bells, more literate people, more religious professionals; and it produced more books, ceramic dishes, textiles and spears than any other state on earth at the time. It covered numerous ecological zones and varied environments, from the subtropical to the steppes, from the lush to the arid. Encompassing a population that was diverse in language and culture as well as in religion, the Ming empire required a highly sophisticated level of organization to hold it together as the largest polity on earth at that time,’ for which, see Craig Clunas and Jessica Harrison-Hall, eds, Ming: 50 Years that Changed China, p.25.
  34. I am grateful to my friend Ye Yang 葉揚 for insisting on this second point.
  35. The etymology of the Chinese character 典 ‘dian’, said to depict several books 冊 sitting upon a table, should be remembered: the volumes are there not to facilitate reading or consultation, but rather are placed there for the purposes of obeisance.
  36. For a brief history of the rhyming dictionary in China, see Endymion Wilkinson, Chinese History: A New Manual, pp.27–28.
  37. To arrange an encyclopaedia by means of rhyming categories was, in fact, a time-honoured practice. Fortunately, such was the nature of the language that rhyme was easily struck with the monosyllabic sounds of all Chinese characters falling into one of 20 rhyme categories, unevenly distributed among the four tones of the spoken language: ‘Level Tone’ (pingsheng 平聲), ‘Rising Tone’ (shangsheng 上聲), ‘Departing Tone’ (qusheng 去聲), and ‘Entering Tone’ (rueheng 入聲). Rhyme category thus provided the most effective of search engines; every educated Chinese person would know both the categories (and the headwords [yunmu 韻目] in Chinese) of these categories) and their sequence by heart. In the case of this volume, the last character of the title of its text means ‘son’ and is pronounced ‘zi’ (子); anyone looking for information on the topic of ‘son’, therefore, would go immediately to the second of the rhyming categories in the ‘Rising Tone’, under the category headword meaning ‘paper’ which is pronounced ‘zhi’ (紙), a close homophone.
  38. For a brief treatment of one instance of the Yongle emperor’s close engagement in the content of books being presented to him, see Ku Chieh-kang 顧頡剛 and L. Carrington Goodrich, ‘A Study of Literary Persecution During the Ming,’ Harvard Journal of Asiatic Studies 3.3/4 (1938): 245–311.
  39. R. Kent Guy makes a similar point in his The Emperor’s Four Treasuries: Scholars and the State in the Late Ch’ien-lung Era (Cambridge: Harvard University Press, 1987): ‘All imperial book collection projects, of course, had as one goal the collection of precedent and source material useful for the ruler. But the effort expended in earlier projects suggested … that earlier governments were concerned with the state of scholarship and recognised that the products of scholarly effort had a value in themselves. The Yung-lo Emperor’s rhyming reference guide, never actually printed, was clearly meant for narrower, more utilitarian ends. Thus, at the same time that private scholarship was becoming more independent and diverse, the government’s scholarly efforts were becoming more narrow and limited’ (p.16).
  40. Yao Guangxiao, whom the future Yongle emperor had first met at his mother’s funeral in 1382, became the emperor’s closest advisor. He had been responsible for the forging of the historiographical record to remove any mention of the four-year reign of the Jianwen emperor. Aside from his work on the Yongle Encyclopaedia, he also oversaw the composition in 1403 of both the Record of Obeying Heaven’s Command to Calm a Troubled Age (Fengtian jingnan ji 奉天 靖難記) and Genealogy of the Imperial Clan (Tianhuang yudie 天皇玉牒), both of which were works designed to confirm the Yongle emperor’s legitimacy. The inscriptions on a surviving portrait of the man, dressed in his monk’s robes and entitled ‘The True Visage of Lord Yao Guangxiao (named “Reverent and Pleasing”), Imperially Titled Preceptor and Posthumously Ennobled as Duke of Rongguo’ 敕封榮國恭靖公贈少師姚公廣孝真容, dated 1592 and 1601 respectively, are somewhat ambivalent; one speaks of ‘To serve as an official whilst wearing a monk’s robes,/ Is to be akin to a fly squatting upon an ice-pure face’ 染衣而官繩點冰顏, while the other picks up on this comment to say ‘He serves in office whilst wearing a black monk’s robe,/ His face showing no sign of shame’ 官爾緇衣無靦爾顏, for which, see Craig Clunas and Jessica Harrison-Hall, eds, Ming: 50 Years that Changed China, pp.80–81. Again, I am grateful to one of the readers of this paper for the discussion of one of these inscriptions.
  41. Yet another illustration of Yongle’s close involvement with (and supervision of) the scholarly world, is the fact that during the course of his reign, he oversaw a total of eight sessions of the highest level of the imperial examinations.
  42. For a succinct bibliographical treatment of this work, see Jeffrey K. Riegel, ‘Li Chi 禮記,’ in ed. Michael Loewe, Early Chinese Texts: A Bibliographical Guide (Berkeley: University of California, 1993), pp.293–97. More recently, Liu Yuchi and Luke Habberstad, ‘The Life of a Text: A Brief History of the Liji 禮記 (Rites Records) and Its Transmission,’ Journal of Chinese Literature and Culture 1–2 (2014): 289–308, present a judicious discussion of the complex textual history of the work. Of the section of the book found copied into the volume of the Yongle Encyclopaedia found in the Huntington Library, they note that in the ten-fold categorisation proposed in his Explanations of the Titles of Ancient Texts and a Method for Reading Them (Yaoji jieti ji qi dufa 要籍解題及其讀法), Liang Qichao 梁啟超 (1873–1929) allocated the chapter to that which constituted ‘Records of Specific Events’ (p.299). As they note, their conclusions about the text (that it is ‘a composite text made up of material from a wide range of ritual learning traditions’ (p.296), much of the content of which may have been composed during the Han dynasty) must remain preliminary, ‘particularly given the continual discovery and excavation of new texts’ (p.297). In Chinese, see also Qiao Xiuyan 喬秀岩, ‘Notes on Various Edition of the Book of Rites’ (Liji banben zashi 禮記版本雜識), Beijing daxue xuebao 5 (2006): 103–10.
  43. The Five ‘Confucian’ Classics (New Haven: Yale University Press, 2001), p.168. As Nylan points out (p.197), the Chinese character for ‘rites’ has not yet been found amongst the oracle bone inscriptions that remain to us today. She refers to the research of the great classical scholar Wang Guowei 王國維 (1877–1927) wherein he argued that the character derived from the character 蘴 feng (abundant), depicting two pieces of jade placed in a bronze vessel as an offering to the spirit ancestors residing in Heaven.
  44. As James Legge points out in his short introduction to his translation of this chapter of the Book of Rites, the Qianlong era (1735–96) editors, in the text of the work they include in the Four Treasuries project, divide the text into two sections, for which, see James Legge, trans., The Sacred Books of China: The Texts of Confucianism: Part III: The Li Ki, I-X (Oxford: Clarendon Press, 1885), p.22. Of the authorship of the work, Legge states: ‘No hint is given, nothing has been suggested …’.
  45. For which, see Chen Dengyuan’s 陳登元 magisterial treatment An Investigation of the Formation and Loss of Books Ancient and Modern (Gujin dianji jusan kao 古今典籍聚散考) (1936; rpt Shanghai: Shanghai shudian, 1983), p.256.
  46. For what remains the best short treatment of the role of the commentary in China, see Daniel K. Gardner, ‘Confucian Commentary and Chinese Intellectual History,’ Journal of Asian Studies 57.2 (1998): 397–422.
  47. The single extant copy of this work (published in Jianyang in Fujian Province during the early Yuan dynasty) is held by the National Central Library in Taipei. For a characterisation of the Classics produced in this printing centre during the Ming, see Lucille Chia, Printing for Profit: The Commercial Publishers of Jianyang, Fujian (11th–17th Centuries) (Cambridge: Harvard University Press, 2002), pp.221–26. The noted modern collector Fu Zengxiang 傅增湘 (1872–1949) tells us that, working from the volumes of the Yongle Encyclopaedia available to him, Dai Zhen collated reconstructions of both this work and the Pengshi cuantu zhuyi (3.iii below) but he seems never to have finalised the MS. The National Library of China in Peking holds a photographic version of the MS, taken in 1936. In his work collating this latter text, Dai Zhen is said to have made careful record of the both the fascicle and page number of the encyclopaedia from which the text had been recovered; without access to this photographic version of his text, we will not be able to ascertain whether or not Dai Zhen had access to the volume of the encyclopaedia discovered in the Huntington Library.
  48. For ease of reference to the Huntington Library’s volume of the encyclopaedia, in the case of these four unique commentaries, I append page references.
  49. The editors of the Four Treasuries note that this book was lost. It is important to remember of the Four Treasuries (Siku quanshu 四庫全書) project instituted by the Qianlong emperor in 1773) that: (1) it was explicitly designed to surpass in size that of the Yongle Encyclopaedia; and (2) the first step in the project was to make use of the volumes of the encyclopaedia to recover texts that had otherwise subsequently been lost. As discussed by R. Kent Guy, the Qianlong emperor, when presented with the 60-fascicle table of contents of the Yongle Encyclopaedia, declared himself unimpressed. To his mind, it was an exercise in ‘intellectual covetousness’ (tan duo wu de 貪多無得), little more than a reference book, whilst the use of rhyme to organise the book was distorting, for which see his The Emperor’s Four Treasuries, p.77.
  50. In the event, of course, his son, the Hongxi 洪熙 emperor (1378–1425; r. 1425), proved to have the shortest reign of the sixteen emperors of the Ming, during course of which he ordered the cancellation of any further maritime excursions, the destruction of the shipyards of Nanking (in the final years of his father’s reign, Zheng He’s expeditions had been suspended), and the return of the capital to Nanking. His first two actions could be said to have altered the shape of the modern world; the last command was never implemented.
  51. In a preface written in 1693 during the reign of the Kangxi emperor (1654–1722; r. 1661–1722), the scholar Xu Qianxue 徐乾學 (1631–94) tells us that the copy of the encyclopaedia housed in the Hanlin Academy was already missing volumes lost during the dynastic transition (dingge shi yi you yishi 鼎革時亦有佚失); Quan Zuwang 全祖望 (1705–55) notes that the removal of the encyclopaedia to the academy was occasioned by the need to make room on the shelves of the Historiographical Office for the (voluminous) Shengzu ren huandi shilu 聖祖仁皇帝 實錄, for both of which see, Zhang Sheng, Research on the Circulation of the Yongle dadian and Attempts to Recover its Text, pp.56–57. The reformist official Wen Tingshi 文廷式 (1856–1904) notes that he ‘borrowed more than three hundred volumes’ (借讀三百餘冊) of the encyclopaedia whilst he was working in Peking during the years 1885–86; he is now suspected of having failed to return more than a hundred of these volumes. The eminent bibliographer Ye Dehui 葉德輝 (1864–1927) reports having seen copies of the encyclopaedia at Wen’s house shortly after his death when they were being offered for sale by his descendants, for which see Zhang Sheng, p.79. Zhang Sheng’s account is the most convincing contemporary source on the fate of the encyclopaedia. After more than a decade of research on the topic, he arrives (in my translation) at the following five important and unimpeachable conclusions (p.111): (i) The second copy of the encyclopaedia was largely lost from the library of the Hanlin Academy during the years 1860 onwards; (ii) The difficulties to be overcome in seeking to track down further extant volumes of the work are compounded by the fact that they gravitated to numerous countries and (in some cases) into private hands; (iii) The final loss of the encyclopaedia did in fact take place during the Boxer Uprising when the Hanlin Academy was set alight, but by that time the library held only an estimated 10% of the original encyclopaedia, the other 90% having been removed by Chinese (guoren 國人). That is to say, the fire in the Hanlin Academy does not constitute the major disaster encountered by the encyclopaedia; (iv) As the volumes of the encyclopaedia started to find their way into the marketplace, they were seen by many people, and in some cases, acquired; these instances will doubtless provide new clues in the search for further extant copies of the work; and (v) In keeping with the extent to which discoveries of lost volumes of the encyclopaedia have been made over the years, I am fully confident that further discoveries will be made in the future.’ Zhang Sheng’s last point, in particular, has now of course been realised by Yang Li Wei’s recognition of what it was that The Huntington Library held in its vaults.