Yongle Dadian, juan 10,110-10,112 (Chester Beatty Library, C 1757)

One physical volume, three juan, out of the roughly 400 that survive from an encyclopedia that once ran to 11,095. It is not the 1408 original commissioned by the Yongle Emperor, which no longer exists in any form; it is part of a hand copy made in 1562-1567 on the order of the Jiajing Emperor, the only later emperor known to have actually used the book.

This particular volume, Chester Beatty Library C 1757, spent about three centuries in the Hanlin Academy library in Peking until Boxer forces set the Academy alight on 23 June 1900 during the siege of the Legations. It was taken from the ruins by a British customs official present at the siege, kept by his family for more than half a century, and sold to the mining magnate and collector Alfred Chester Beatty in London in September 1954 for 120 pounds sterling, together with a sibling volume. It has been in Dublin, on public view, ever since.

The corpus it belongs to is still being pieced back together: volumes continue to surface at auction and in private hands into the 2020s, and Chinese institutions are actively buying them back.

One volume out of eleven thousand

The Yongle Dadian was compiled between 1403 and 1408 on the order of the Yongle Emperor of the Ming dynasty: a state attempt to copy out, in full or in part, essentially every book then held in the imperial library, arranged not alphabetically but by rhyme category, so that a reader could search by the sound of the last character of a term. Completed, it ran to 22,877 juan (chapters), bound into 11,095 physical volumes (ce), and comprised, on one scholarly estimate, around 370 million characters. Nothing else made in the pre-modern world approached its scale.

That original fine copy no longer exists. What survives, in scattered pieces across more than thirty collections in about eight countries, is a small fraction of a duplicate copy made a century and a half later. This record is about one physical volume of that duplicate: Chester Beatty Library C 1757, containing juan 10,110 to 10,112, now in Dublin. It is not the 1408 original. No volume anywhere claiming to be the 1408 original has been authenticated; the object described here, like every other surviving Yongle Dadian volume in every institution that holds one, is understood by its own holding institution's cataloguing to date from 1562-1567.

Why it was made

The Yongle Emperor's own preface, recorded in the Ming Veritable Records, states the ambition plainly: to gather the scattered contents of every book, ancient and modern, into one arrangement, so that a reader could trace anything as easily as shooting a swan with one's bow, and that nothing will remain hidden once you open up the pages of this book. He had already rejected one attempt, an earlier compilation by Xie Jin called A Grand Compendium of Literary Sources, as insufficiently exhaustive, and ordered it done again on a far larger scale.

The scholar Duncan Campbell has argued, citing the contemporary Ming writer Sun Chengze, that the project cannot be read only as a scholarly ambition. The Yongle Emperor had seized the throne from his nephew, the Jianwen Emperor, in a campaign he called 'calming a troubled age', and followed the coup with a purge in which his predecessor's chief advisor was executed and some 870 of his associates put to death. Sun Chengze's near-contemporary account holds that the emperor used the compilation project, in part, to occupy and pacify a scholar-official class he had reason to distrust, and to present himself as a patron of learning rather than a usurper. This is Campbell's argument, made in a citable peer-reviewed publication; it is not universally repeated in general accounts of the encyclopedia, which more often present the project simply as an act of Confucian cultural patronage.

How it was made

Some 2,169 scholars worked on the project under the nominal general editorship of the monk and advisor Yao Guangxiao, known by his clerical name Daoyan, and the Hanlin academician Xie Jin, with the bulk of the actual editorial work carried out by men including Hu Yan, Liu Jichi and Chen Ji. Existing books held in the imperial library, the Pavilion of Literary Profundity, were copied out, broken into their constituent passages, and reassembled under the rhyme category of the last character of the relevant chapter heading, using a rhyme dictionary called the Correct Rhymes of the Hongwu Era, compiled in 1375 on the order of Yongle's father. A 60-fascicle index was provided as a finding aid.

Neither Xie Jin nor Chen Ji fared well afterward: Xie Jin was executed and Chen Ji imprisoned, in the years following the encyclopedia's completion, for reasons connected to court politics rather than to the project itself. No individual scribe who worked on the later 1562-1567 copy, the version this volume belongs to, is identified in any source consulted for this record.

The original that vanished

Nobody living today has seen the 1408 fine copy, and not one page of it has been identified since the fall of the Ming dynasty. The most likely explanation is that it was lost or destroyed during the fighting of 1644-45, when Peking and then Nanking fell first to the rebel army of Li Zicheng and then to the invading Manchu forces that founded the Qing dynasty; but its exact location and fate at that moment are not recorded, and Duncan Campbell notes that the completeness and suddenness of its disappearance, without even bureaucratic trace, points to deliberate concealment of the loss by those responsible for it, as much as to the loss itself.

A second, more colourful explanation circulates: that the Jiajing Emperor, who alone among later emperors is known to have actually consulted the book, ordered the original buried with him when he died in December 1566. This claim originates with the writer Zhang Chenshi in his Anecdotes about the Yongle Encyclopaedia, and Campbell describes it as almost certainly apocryphal. No excavation or documentary evidence supports it. It is included here as a proposal attributed to a named source, not as an established fact.

What is certain is the copy that was made instead. After fires in the Forbidden City in the mid-16th century threatened the imperial libraries, the Jiajing Emperor ordered a full duplicate, the fuben, transcribed from the original. It took from 1562 to 1567 to complete, finishing only months after the emperor's own death. Every volume known to survive today, at every institution that holds one, is part of this Jiajing-Longqing duplicate. The Chester Beatty volume is dated by its own holding institution, in the Wikimedia Commons record of the leaf reproduced here, to exactly this window: 1562-1567.

Two catastrophes, and how the loss is apportioned

The duplicate copy did not fare much better than the original. By 1875 it had already been reduced from around 11,000 volumes to about 5,000, a loss generally attributed to the Anglo-French occupation of Peking during the Second Opium War in 1860, when British and French soldiers are recorded to have taken sections of the manuscript as souvenirs alongside destruction and fire. By 1894, on the numbers most commonly cited, only around 800 volumes remained in the Hanlin Academy library in Peking. Fewer than 400 volumes survive worldwide today, across all institutions.

The 1900 event most associated with the encyclopedia's destruction is the burning of the Hanlin Academy on 23 June, during the siege of the Foreign Legations in the Boxer Rebellion. Troops under General Dong Fuxiang set fire to the Academy buildings, immediately adjacent to the British Legation, apparently hoping the wind would carry the flames into the Legation compound itself. A mixed force of British and American marines and Imperial Maritime Customs volunteers, together with the journalist George Ernest Morrison and others, breached the Academy wall and fought the fire; contemporary accounts record that volumes of the encyclopedia had already been used by combatants on both sides to build barricades. Once the fire was under control, several of those present took surviving volumes away as personal souvenirs, including the man who later held the volume described in this record.

Two framings of this loss coexist in the scholarship, and this record gives both rather than choosing between them. The framing more common in general and Chinese-language accounts holds the 1860 and 1900 foreign interventions chiefly responsible for the destruction and dispersal of the corpus. Duncan Campbell, in the peer-reviewed source cited throughout this record, argues that this is, in his words, a 'convenient and nationalistic attribution' that is 'simplistic and far from the truth of the matter': on his reading, the numbers already show the bulk of the loss, from roughly 11,000 volumes down to around 800, had occurred by 1894, well before the Boxer Rebellion, through the ordinary neglect, mismanagement, and internal pilferage that afflicted large imperial book collections across the Qing dynasty generally. The 1900 fire and its aftermath, on this view, accounts for only the final and smallest fraction of a loss that was already close to total.

This volume: from the ruins to a customs officer

The provenance of Chester Beatty Library C 1757, together with its sibling volume C 1756 (juan 803-806), has been traced by the bibliographer David Helliwell to James Russell Brazier, who at the time of the siege was Chief Secretary of the Inspectorate of the Imperial Maritime Customs, and who was present in the Legation Quarter throughout. Helliwell states plainly that 'the provenance of all three volumes' obtained by Brazier during the siege 'can thus be traced to the Siege' of 1900. Brazier gave one further volume he had obtained, juan 11,907, to his alma mater, the University of Aberdeen, in 1922.

A red paper name-slip still kept with this volume bears characters read as 'bai lai xi', understood to be a Chinese transliteration of Brazier's own name. The volume's silk-covered tao, or protective wrapper, along with its sibling's, is of a construction and labelling that suggests it was made in Peking, consistent with having remained in China, in Brazier family hands, for some time after 1900 rather than being immediately rebound in Britain.

Fifty-four years, then a phone call

The two volumes passed to Brazier's son, William Russell Brazier, born in Aberdeen in 1897 and known in later correspondence as 'Col. Brazier'. In September 1954 he brought both volumes to the British Museum's Department of Oriental Printed Books and Manuscripts, intending to sell them to the dealers Kegan Paul & Co. for 100 pounds.

What happened next is recorded in a letter, preserved in the Chester Beatty Library Archives, written the same day by H.C. Lowe, a curator in that department and also an outside advisor to Chester Beatty: 'Col. Brazier walked into the Dept. this morning when by luck I was in Mac's room; he was just on his way to K. and P., to sell those two vols of the Chinese Encyclopedia for 100 pounds. I said I could offer him a better price at 110 pounds, he said he would feel happier at 120, so at this price we closed; he was happy, and SO WAS I.' A second letter, dated the following day, records that Chester Beatty's secretary, John Wooderson, collected the two volumes from the British Museum at noon on 22 September 1954.

120 pounds in September 1954 sits, by a rough and disclosed method only, somewhere in the range of three to four months' pay for an average UK worker at the time: the Bank of England's 'Three Centuries of Data' series gives average UK weekly earnings of 5.60 pounds in 1950 and 8.90 pounds in 1959, and a straight-line interpolation between those two published figures puts 1954 earnings at somewhere around 7 pounds a week. That is a period-appropriate order of magnitude only, not a precise contemporary figure, and it should be read as such.

The collector and the library that holds it now

Alfred Chester Beatty was born in New York in 1875, made a fortune as a mining engineer and consulting geologist across the American West and Mexico, and was known in the business press as the 'King of Copper'. He built his personal library of Asian, Middle Eastern and Western manuscripts, papyri and miniatures over decades in London, then, in May 1950, at the age of 75, moved himself and his collection to Dublin, a relocation that reportedly caused consternation in London and delight in Ireland. In 1957 he became the first person granted honorary Irish citizenship, a distinction created for him under the Irish Nationality and Citizenship Act of 1956, and which he called the proudest moment of his life.

He bought this volume, with its sibling, in London three years before that citizenship was conferred, while still assembling the collection that would become a public bequest. On his death in Monte Carlo in January 1968, he was given a state funeral at St Patrick's Cathedral in Dublin, the first private citizen and the first person born outside Ireland to be so honoured. His library and collection passed to the Irish state, and now form the Chester Beatty Library, a public museum on the grounds of Dublin Castle since 2000, and European Museum of the Year in 2002.

What the volume looks like

Volumes of this edition are written by hand in two colours of ink, red for author names, book titles, punctuation marks and the ruled borders and running heads, black for the body text, in three graduated sizes of the formal 'Academy Style' court calligraphy associated with the early Ming calligrapher Shen Du. They are copied onto fine white Xuan paper and bound, in the standard format, in boards covered with coarse yellow silk brocade; the standard page measures approximately 40.6 by 25.4 centimetres.

Chester Beatty Library C 1757 shows condition problems consistent with a hard survival: water damage and mould spread from the spine towards the outer edge of the leaves, suggesting to Helliwell that the volume was at some point stored or exposed while standing on its lower edge and doused with water, plausibly during the firefighting of 1900 itself. The yellow silk covers appear to be original to the volume, though the supporting boards beneath them have become limp. No formal conservation report for this specific volume was located in the sources consulted for this record.

A pattern, repeated at other institutions

This volume's path, a volume grabbed from the Hanlin ruins by a foreign national present at the siege, kept in a family for decades, then sold or given to an institution, recurs closely enough across the surviving corpus that it amounts to a documented pattern rather than an isolated story. Lancelot Giles, son of the sinologist Herbert Giles and himself present at the siege, wrote in his own diary that after the Hanlin library was 'almost entirely destroyed', he 'secured volume 13,345 for myself, merely as a specimen'; that volume is now at the Library of Congress. He kept a second volume, juan 16,343-16,344, which his father presented to Cambridge University Library on his behalf in April 1901, where it remains today, catalogued as manuscript FH.20.71.

An American missionary, Joseph L. Whiting, picked a volume, juan 10,270-10,271, out of the same debris and used it to barricade a window during the siege; it travelled home with him, was donated to the Huntington Library in California by his daughter in 1968, and was not recognised for what it was until 2014. All three men were present in the besieged Legation Quarter in the same weeks as James Russell Brazier.

Piecing the corpus back together

Interest in relocating and reuniting the scattered volumes is not new. The librarian Yuan Tongli conducted an early census of known surviving volumes, completed in 1928, and corresponded directly with private owners, including the collector Wilfred Merton, offering photostat exchanges from the National Library of Peiping's own holdings in return for copies of volumes in private hands; that correspondence survives in the Chester Beatty Library's own archives, concerning a different volume the library later acquired. In 1951 two volumes were returned to the National Library of China from the Department of East Asian Studies at Leningrad University, by way of a gift to the new government's Ministry of Culture.

The search continues into the present decade. Volumes have surfaced in Shandong province in 1983, at the University of Aberdeen in 1997, in Canada in private hands more recently, and, most recently documented, at a French auction house in May 2020, when two further volumes, covering juan 2,268-2,269 and 7,391-7,392, were bought privately by the collector Jin Liang because Chinese public institutions cannot bid at foreign auction; after clearing export approvals and customs, they were formally donated to the Hangzhou branch of the China National Archives of Publications and Culture in July 2026. In January 2026 the National Library of China Publishing House unveiled a new facsimile edition of the Jiajing-Longqing copy, reproduced on handmade Xuan paper with high-precision colour printing, intended, in the words of its project director Xu Haiyan, to be 'the closest we can get to touching what was lost'. The Library of Congress separately completed digitisation of its own 41 volumes in 2023, publishing them online with no known copyright restrictions.

What is still unresolved

The fate of the 1408 original remains genuinely unknown, and the burial legend attached to it is flagged in the scholarship itself as unproven. The precise custody of this particular volume between 1900 and 1954, within the Brazier family, is not documented beyond the fact that both father and son at different times possessed it. No source consulted for this record states which rhyme category or which underlying text juan 10,110-10,112 actually preserves, a gap a specialist reader would expect this record to close and cannot.

Even the basic count of what survives is not fully settled: the National Library of China's own holdings are given as 221 volumes in one 2023 source and as considerably more in looser recent summaries, a discrepancy this record has not been able to resolve, and one made more unstable by the fact that volumes are still actively being repatriated. What is settled is smaller and more solid: this one volume, its condition, and the names of the men who carried it out of Peking in 1900, sold it in a phone call in 1954, and bought it.