The Beauvais Missal

A missal written for the use of Beauvais Cathedral at the end of the thirteenth century, 309 leaves of parchment, the third volume of a set of three. It was given to the cathedral in September 1356 by a canon, Robert de Hangest, in exchange for a Mass said for him every year on 3 November, and it stayed in that library for about 450 years.

It no longer exists as a book. In October 1942 the New York dealer Philip Duschnes bought it out of William Randolph Hearst's collection for USD 1,000, through Gimbel Brothers, and had single leaves on the market within a month at USD 25 to USD 40 each. Otto F. Ege of Cleveland, who had published a defence of book-breaking in 1938, distributed leaves as number 15 in forty boxed sets of Fifty Original Leaves from Medieval Manuscripts. The leaf carrying Hangest's donation inscription was lost in the process: the page recording why the book existed was destroyed in the act of selling it.

This record exists because biblioclasm is a provenance subject and almost nobody treats it as one. The chain here is unusually complete for a medieval manuscript, and the destructive act sits in the middle of it, fully documented, carried out by named professionals for a return that can be calculated. Lisa Fagin Davis has located 122 of the 309 leaves. One of them turned up framed on a wall at an estate sale in Waterville, Maine, and was bought for USD 75.

It was the third volume of three, and nobody has ever found the other two

The Beauvais Missal was a summer volume: the Mass texts and the chant for the feasts falling between the week after Easter and the end of the liturgical year, written for the use of Beauvais Cathedral in the last decades of the thirteenth century. Complete, it ran to 309 leaves of parchment, about 290 by 204 mm, ruled in two columns of fifteen or twenty-one lines, or ten staves of music, with the text block about 195 by 135 mm.

It was one book of a set of three. The Latin donation inscription calls it istam terciam partem missalis, this third part of the missal, and the French registers of Beauvais record the gift as a missal in three volumes. No trace of the first two has ever been identified. That is the first thing a reader should hold on to: two thirds of this object were already lost before anybody thought to cut up the third.

A canon bought a yearly Mass with it, and the page recording the transaction was the first thing destroyed

Robert de Hangest, a canon of Beauvais Cathedral, gave the missal to the cathedral in September 1356, reserving its use for his own lifetime, on condition that his death be commemorated every year on 3 November. He died that November. The transaction is a purchase in the only currency that mattered to him: a book against an obit in perpetuity.

The inscription recording it stood on the blank recto of the first leaf of the calendar. It read, in the transcription made for the 1926 auction catalogue, Dominus Robertus de hangestis quondam canonicus Beluacii legauit Istam terciam partem missalis Ecclesie Beluacii pro obitu suo in eadem perpetuum singulis annis faciendo tercia die Novembris. That leaf no longer exists. It was lost when the manuscript was broken up, which means the document establishing why the book was in the cathedral at all was destroyed in the act of turning the book into merchandise. Everything now known about the donation is known from copies made by other people.

Peter Kidd has pointed out the obvious difficulty with calling Hangest the patron. The missal is dated by art historians to within a decade either side of 1290 and Hangest died in 1356; he is unlikely to have been old enough to commission it. Kidd notes a relative, Guillaume de Hangest, Tresorier de France, who founded a chapel in the Beguine house at Beauvais in 1294, and asks whether he commissioned the book for a younger kinsman. That is a question, not an answer, and it is recorded here as one.

Four hundred and fifty years in one library, then the Revolution

The missal stayed at Beauvais and is identified in successive cathedral inventories. A seventeenth-century catalogue lists it as Missale Roberti Hangesto; en beau velin et belles miniatures. Then the cathedral library was dispersed, as most French ecclesiastical collections were, at or soon after the Revolution.

What happened next is the one genuine hole in the chain, and the scholar who reconstructed the rest of it says so plainly: this is the only undocumented part of the missal's provenance. A group of Beauvais manuscripts had reached Lyon by 1833. Nobody has shown how.

Lyon, a new binding, and four generations of one family

The book surfaces in the collection of Didier Petit de Meurville of Lyon, and appears in his sale of 1843 as lot 354. By then it had been rebound and gilt by the Lyon binder Bruyere, with the title Missel de Bauvais au pointille tooled on the lower edge of the bookblock; the binding description in 1843 matches the one published in 1897, which is how the binding is dated to Petit's ownership. When it was last intact it was in Russian leather with brass catches and clasps and metal studs along the board edges.

Henri Auguste Broelemann, a commercial broker in Lyon, bought it in 1843. It then passed through four generations of one family: to his son Emile-Thierry, to his son Arthur-Auguste, to his sister Albertine, and to her daughter Blanche Bontoux, Madame Etienne Mallet. The Broelemann books were catalogued privately by Arthur-Auguste in 1897, which is the document that fixes the collection.

4 May 1926: lot 161, and a bidder who was not bidding for himself

Blanche Bontoux sold the missal at Sotheby's in London on 4 May 1926, lot 161. It made GBP 970 to the dealer William Permain.

Permain was not the buyer. Peter Kidd established that he was acting as agent for William Randolph Hearst, who took the manuscript to the United States. This is the single most important correction in the early modern chain, and it was made by a provenance scholar working through catalogues, not by the auction house or by any of the institutions now holding the pieces. Named on the ticket: the dealer. Named in the record: nobody, for eighty-nine years.

The GBP 970 is reported as one figure. No buyer's premium is separately recorded, and none is stated in the sources consulted here, so the sum should be read as the price at which the lot was knocked down.

October 1942: one thousand dollars, through a department store

Hearst kept the missal for sixteen years and sold it in October 1942, through Gimbel Brothers in New York, for USD 1,000. Hearst's collections were being liquidated through a department store because that is how the Hearst dispersal was run; a thirteenth-century French missal moved through the same channel as furniture.

The buyer was Philip C. Duschnes, a New York dealer. Less than a month after the sale he began offering single leaves at USD 25 to USD 40 apiece, the price depending on how much gold was on the particular page. Broken Books puts the commercial logic in one sentence: Duschnes needed to sell forty of the cheaper leaves to break even, and he had more than three hundred.

The arithmetic is worth doing because nobody ever does it. At USD 25 to USD 40 a leaf, 309 leaves are worth between about USD 7,700 and USD 12,400 against an outlay of USD 1,000. That range is DERIVED here from the leaf prices and the leaf count, not quoted from any source, and it assumes every leaf sells, which no dealer achieves. Even discounted heavily, the return on breaking the book was somewhere between seven and twelve times the return on selling it whole. That ratio, and not vandalism or ignorance, is the entire explanation of biblioclasm.

Number 15 in the portfolio, and the man who wrote a defence of book-breaking

Duschnes sold many of the leaves to Otto F. Ege of Cleveland, a professor of art history who was also a dealer, and a friend and business associate. Leaves of the Beauvais Missal are best known as number 15 in Ege's boxed sets, 'Fifty Original Leaves from Medieval Manuscripts', of which forty were issued, each containing one leaf from each of fifty different manuscripts.

Ege did not do this quietly. In March 1938 he published an essay called 'I Am a Biblioclast' in the hobby magazine Avocations. It opens: 'For more than twenty-five years I have been one of those strange, eccentric, book-tearers.' He notes that William Blades and Holbrook Jackson each devoted a chapter to condemning mutilators of books, and that Andrew Lang classified them, adding: 'I find I am the aesthetic ghoul of the book world.'

He then sets out a code of practice, which begins with a rule he did not apply to this book: never to take apart a museum piece or a unique copy if it is complete. And he offers the justification the trade has used ever since: 'Surely to allow a thousand people to have and to hold an original manuscript leaf, and to get the thrill and understanding that comes only from actual and frequent contact with these art heritages, is justification enough for the scattering of fragments. Few, indeed, can hope to own a complete manuscript book; hundreds, however, may own a leaf.'

The argument is not stupid and it should be stated at full strength, because it is the argument that won. What it omits is that a leaf out of a missal cannot be read as a missal: the sequence, the quire structure, the calendar and the inscription that told you whose book it was are properties of the codex, and they are exactly what the sale destroys.

What the leaves show, and what a connoisseur can still say about the hand

The decoration runs to initials of one to seven lines throughout in gold and colours, with tendril-like marginal extensions ending in leafy finials, many incorporating serpents, dragons or birds, with small dogs and other animals. The art historian Alison Stones has affiliated the book with the cluster known as the Cholet Group, attributed to Amiens or Paris in the last decades of the thirteenth century, comparing the curling hair, the deeply modelled drapery and the delicately drawn eyes with a glossed Psalter of that group, Bibliotheque de l'Arsenal manuscript 25.

The Canon of the Mass originally carried four historiated initials. Three have been identified. Two of them are on a single leaf in Cleveland, and the third is at Oberlin. The fourth has not been found.

The Cleveland leaf is a gift in memory of the man who destroyed the book

Cleveland Museum of Art accession 1982.141 is a leaf from the Canon, ink, tempera and gold on vellum, sheet 28.5 by 20 cm, dated about 1300. It carries two historiated initials: a P, for Per omnia saecula saeculorum, showing a priest celebrating Mass, and a V, for Vere dignum et iustum est, showing Ecclesia and Synagoga. The museum's own description states that this leaf and others survive from a magnificent missal made for the use of Beauvais Cathedral, and that the two initials illustrate the preface to the Canon of the Mass.

The credit line reads: Gift of Mr. and Mrs. Milton Freudenheim in memory of Otto Ege.

That sentence is the whole subject of this record in eleven words. The leaf is in a public collection, catalogued, conserved, photographed and released for anyone to use, and it is there as a memorial to the man who cut it out of the book.

Boston bought its leaves directly from the man who cut them

Boston Public Library holds two leaves under MS pb Med. 110, formerly BPL MS 1538: parchment, 290 by 200 mm with a written area of 200 by 135 mm, Gothic bookhand in black ink with red rubrics, two columns of twenty-one lines, square Gothic notation on a four-line red staff, two-line initials in red and blue with white filigree. One leaf carries the Second Sunday after Easter; the other, Masses for saints' feasts between 29 August and 1 September.

The library's catalogue names its former owners in order: the Catholic Church, Robert de Hangest, Henri August Broelemann, Otto F. Ege, Philip C. Duschnes, the Cathedrale Saint-Pierre at Beauvais. The leaves were bought from Duschnes in 1950. A research library acquired the pieces from the dealer who made them pieces, eight years after he did it, and catalogued him as a previous owner. Nothing in this history was hidden. It was all in the front of the file.

122 of 309, and the reconstruction that will never be finished

Lisa Fagin Davis has been locating and reassembling the leaves since the mid-2010s, first as a case study for the Broken Books platform developed at Saint Louis University's Center for Digital Humanities, and now as a virtual reconstruction hosted on Fragmentarium. The reconstruction gathers images and metadata, pairs consecutive leaves by matching the gold offset one leaf leaves on the next, and rebuilds the quire structure. There are no folio numbers to work from.

The count has moved: more than eighty leaves identified by March 2015, ninety-three assembled that month across twenty-two states and six countries, the hundredth folio in June 2016, and 122 of 309 on the Fragmentarium record, where that figure has stood since at least December 2023. Holding institutions include Harvard's Houghton Library, the Cleveland Museum of Art, the Morgan Library, the Metropolitan Museum, the Beinecke, Boston Public Library, Oberlin, Waseda University in Tokyo, Oslo Cathedral School and the Australian National University.

One hundred and eighty-seven leaves are unaccounted for. That figure is DERIVED, 309 less 122, and it is not a statement that they are destroyed. Davis noted in 2015 that as many as thirty leaves were known but untraced, which is a different category again: photographed or catalogued once, and now nowhere.

A framed page on a wall at an estate sale in Maine

In September 2022 Will Sideri, a recent Colby College graduate who had taken a manuscripts course there, saw a framed page at an estate sale in Waterville, Maine, recognised the script, and called his former professor Megan Cook. Davis confirmed it as a leaf of the Beauvais Missal. Sideri had paid USD 75 for it. Davis estimated it was worth between USD 5,000 and USD 10,000.

The story was picked up by the Associated Press and travelled. Davis then wrote that in the months after publication seven more leaves came to light, in South Carolina, Delaware, Kansas, Connecticut, Tennessee, California and Ohio, every one of them identified by an owner who had read the story or one of its offshoots. She had never seen so many identifications in such a short time.

That is how a dismembered manuscript is now recovered: not by a dealer's ledger but by publicity reaching the people who inherited a framed page and never knew what it was. Meanwhile the trade continues in the other direction. Ege's set number 18 was offered as a boxed lot at Christie's in London on 14 December 2022, lot 19; it was bought by Martin Schoyen of Oslo, who gave it to the Oslo Cathedral School Rare Book and Manuscript Library in 2025. Its Beauvais leaf is number 15 in the box.

Nobody is proposing to put it back together

No museum in this record has deaccessioned or returned a leaf, and no published argument for physically reuniting the manuscript was found. The movement is all one way: accession, gift, catalogue entry. A leaf in a public collection is a leaf that is safe and a leaf that is staying where it is.

What replaces the codex is the reconstruction, and the reconstruction is a genuinely better research object than the book was in 1941, because it is open, searchable and used by anyone with a browser rather than by whoever can reach the shelf. It is also not the book. The calendar is gone. The donation inscription is gone. The two companion volumes were gone before any of this started. A digital reconstruction records what a manuscript said; it cannot restore what a manuscript was.