Fragmentary manuscripts make up the majority of the surviving illuminated medieval material in existence today. The circumstances of how and why manuscripts are broken are as variable as their contents. However, their stories are linked by common themes of disaster and restoration, or perhaps more aptly, transformation. The manuscript fragment exists neither as a true representation of the original object nor as a facsimile. It is a relic requiring greater interpretation than a complete manuscript, and carrying with it complex records of its journey.

This essay examines three fragmentary manuscripts included in the State Library of Victoria's The Medieval Imagination exhibition, in particular the events surrounding their ruination and the way in which the meaning of these manuscripts has changed through both their perceived and physical states.

The Baillieu Library, University of Melbourne, owns four illuminated bi-folios from a book of hours produced in the early fifteenth-century century at the workshop of the Parisian illuminator known as the Boucicaut Master. The bi-folios have been badly damaged by water and the illuminations subsequently over-painted as part of a nineteenth-century restoration.

Nevertheless, these fragments remain significant as examples from the workshop of an influential master illuminator and as part of an early fifteenth-century Parisian book of hours that can be securely dated to 1408. Furthermore, alongside their changes in fortune these bi-folios have filled a series of distinct roles, from being perceived as items of affluence and piety, to decorative and artistic objects acquired as part of the Victorian infatuation with collecting, to their current status as items preserved for their informative and interpretive value.

The original manuscript was a small book of hours of high quality but without a clear indication of a patron or region, suggesting that the manuscript may have been produced without a commission. The manuscript originally consisted of 187 folios with 28 miniatures, and Christopher de Hamel has located 56 extant folios from it, including the Baillieu bi-folios and leaves at the Dunedin Public Library and the Princeton University Museum.1 The Baillieu Library bi-folios are from the Gospels of Mark, Luke and John and matins for the Hours of the Virgin. (Figs. 1-5)

The manuscript can be securely dated due to the presence of an inscription noted by Manion et al on folio 158v, Factum est anno m° cccc° viii° quo ceciderunt pontes parisiis,2 stating that this manuscript was completed shortly after the flood of Paris in January 1407-08 which destroyed the Petit Pont, the Grand Pont and the Pont Neuf. Christopher de Hamel notes the similarity between this inscription and one found on the Bodleian manuscript Douce 144, also credited to the workshop of the Boucicaut Master.3

The inscription may have had particular significance to the scribe as the Petit Pont joined the two areas where the Parisian manuscript community primarily lived and worked. As it was, this early association with disaster was to foreshadow future events affecting the manuscript.

The complete manuscript from which the Baillieu leaves were taken was first recorded in the collection of John Boykett Jarman. Jarman was a goldsmith and jeweller in London in the early nineteenth century who acquired an impressive collection of illuminated manuscripts, including such notable volumes at the Esdaile Missal and the Hours of Yolande of Flanders illuminated by Jean Pucelle. While in Jarman's collection, the manuscript was damaged during a freak storm in August 1846 that flooded the basement of Jarman's London house where his manuscript collection was stored. The manuscripts were stored in metal boxes which burst as the parchment inside swelled. The volumes remained in dirty water for three days until Jarman returned home from holiday to discover the damage. Approximately half of the sixty-five items in his collection were affected by water damage to varying degrees and the Hours of Yolande of Flanders and the Esdaile Missal were two of the most severely damaged.

After the flood, Jarman engaged the services of Caleb Wing, a skilled painter primarily employed to make accurate copies of artwork for publication. Wing commenced a campaign of restoring the miniatures that were most damaged by water. His work in this instance has been credited with a high degree of accuracy with respect to the original style and palette. Jarman's employment of Wing was not limited to the restoration of damaged works. Under Jarman's direction he continued to create new miniatures for those volumes that lacked embellishment, and also provided and extended decorated borders. Jarman's embellishment of his collection also included rebinding, often in a highly luxuriant fashion but sometimes with inaccurate incorporation of coats of arms.4

After Jarman's death in 1864 the manuscript was sold through Sothebys to Edward Arnold. It was purchased in Oxford in 1929 by Sir Walter Chester Beatty. It was while the manuscript was in Beatty's collection that it was broken and began to be disseminated.5 He sold six miniatures from the manuscript through Sothebys in 1932. Three miniatures from the Suffrages of the Saints, the Trinity, St Stephen and St Katherine were given as gifts and are in the collection of the Princeton University Museum.6 The remainder of the volume was sold after Beatty's death. The bulk of the remaining manuscript was purchased by Alan Thomas from whom the Baillieu Library purchased the bi-folios in 1974. A further three leaves became part of the Reed collection in Dunedin. While the manuscript was in the possession of Alan Thomas, Christopher de Hamel was able to borrow it for examination before its final dispersal.7

The ink of the text of the Baillieu Library bi-folios shows distinctive degradation, consistent with that of iron gall ink. Iron gall ink is naturally acidic, making it particularly suitable for use with parchment documents. Earlier inks relied largely on finely divided carbonaceous substances dispersed in oil or water. Carbon-based inks are very stable, but bind poorly to the smooth closed surface of parchment. Carbon inks are thus prone to flaking and have in some instances been removed almost entirely from a parchment surface, allowing the parchment to be re-used—as in the case of palimpsests. The acidity of iron gall ink was a distinct advantage, causing it to 'bite' into the parchment surface making it more stable and impossible to remove entirely. When iron gall ink is used on paper, it is common to see the ink 'eat' away the paper support leaving a burn-like impression. Sometimes components of the ink migrate and form a halo around the writing. Occasionally the paper fails completely where the ink is applied, causing the phenomenon known as 'lacing'. The Baillieu Library bi-folios exhibit lacing, which is uncommon in parchment.

It is unclear why lacing has occurred in this instance. It may be as a result of improperly prepared ink, which was too high in acid content. Or, the process of iron gall ink degradation may have been exacerbated by the water damage these leaves received in 1846. The latter scenario seems more likely, as an aqueous environment would have liberated chemically bound acidic material as well as swelling and opening the parchment network. The several days that the manuscripts remained wet would have allowed the acid to interact with the parchment causing the degradation. Other areas of the ink are blurred beyond legibility, confirming that the ink was not impervious to moisture.

In 2005, the Baillieu bi-folios were mounted using a method developed by Christopher Clarkson8 at the Bodleian Library, allowing them to be safely handled and displayed.9 As parchment is raw animal skin it remains reactive to fluctuations in environmental conditions, especially humidity. The mounting technique developed by Clarkson buffers these changes by attaching a series of linen threads to the edges of the parchment. The threads act to counter the expansion and contraction of the parchment. The mounted bi-folios were then encased in free-standing perspex envelopes.

The remains of this fragmentary manuscript, a large Latin Vulgate Gospel book completed in the late seventh or early eighth century,10 is divided between the collections of Corpus Christi College Cambridge and the British Library. Originally 110 folios, the larger part of the manuscript (Cotton Otho C.v.) in the British Library exists as 64 charred fragments as a result of the Cotton Library fire in 1731. The undamaged fragment at Corpus Christi (Corpus 197B) is 36 folios, comprising parts of the Gospels of John and Luke.11 A single fullpage illumination at the beginning of the Gospel of Saint John survives with a depiction of his symbol: the eagle. Part of a Canon Tables bound into the manuscript British Library Royal MS 7 C.xii., an eleventh-century copy of Aelfric's Homilies that belonged to Cardinal Wolsey (approx. 1471-1530),12 perhaps a third fragment of Otho C.v., but this has not been established conclusively.13

Why this manuscript was divided is unclear. It was most likely produced at a Northumbrian scriptorium and has clear associations with other large Insular gospel books including the Lindisfarne and Durham Gospels.14 Common inconsistencies in the Canon Tables suggest that this volume served as exemplar for or shared a model with the Book of Kells15. There are suggestions that this manuscript was in the collection of Saint Augustine's, Canterbury,16 a belief apparently held by the earliest known owner of the Corpus fragment, Matthew Parker (1504-1575), Archbishop of Canterbury under Elizabeth I. The manuscript bears a Latin inscription applied by Parker's clerk stating that 'This book was sent by Pope Gregory to Archbishop Augustine.'17 A similar note referring to the Cotton fragment appears in the 1696 printed catalogue of that collection.18 The manuscript had been broken by the early 16th century. Corpus 197B most likely came into the possession of Matthew Parker as part of the secularisation of monastic libraries resulting from the Reformation in Britain.19 It was through Matthew Parker's bequest that the fragment found its way into the Corpus Christi Collection.20

The largest portion of this manuscript and the most notable for its association with disaster are the 64 charred fragments of Cotton MS Otho C.v in the British Library. The Cotton Library is significant as the one of the founding collections of the British Museum; it was the most important private library of seventeenth-century England and the first lending library in Great Britain.21 The collection contained a great many significant manuscripts and documents, not least of which were two copies of the Magna Carta,22 a unique copy of Beowulf23 and many state papers. However, the collection is notorious for its decimation by fire in 1731, leaving a legacy of preservation needs. A Parliamentary Committee was formed to report on the condition of the collection after the fire.24 Subsequently, work on the Cotton collection occupied the attention of generations of Keepers of the Manuscripts at the British Museum over the next 140 years. Work continues in some form to the present day.25

Sir Robert Cotton (1571-1631), an avid bibliophile, developed his library as a scholastic and a political resource to feed his ambitions as an acolyte of the Jacobean administration. The political significance of the library is manifest in the fact that Sir Robert was arrested after the discovery of a seditious passage in his collection. Following his arrest the library was closed, and his papers searched and many confiscated.26 Despite its clear political value, Cotton ran his library as a lending institution, accepting almost all comers. The first disaster to affect the Cotton library is that an estimated ten percent was lost due to poor cataloguing and recording of loans within Sir Robert's lifetime.27

Upon Sir Robert's death in 1631, the library passed first to his son Sir Thomas, and then to his grandson Sir John Cotton. During Sir John's lifetime the library was transferred to the public as per Sir Robert's original wishes28 and in 1753 the Cotton Library became one of the founding collections of the new British Museum.29 In 1722 the Library was moved from its traditional home at Cotton House, which had fallen into disrepair, to Essex House. In 1730 it was moved again to Ashburnham House, ironically because Essex House was considered too great a risk for fire. It was at Ashburnham House in 1731 that a fire started in a chimney below the room which housed the manuscript collection,30 causing what has been described as 'perhaps the greatest bibliographical disaster of modern times in Britain.'31

The report to the Parliamentary Committee set up to investigate the fire stated that of the 958 manuscripts in the collection before the fire, 114 were lost entirely, and a further 98 damaged. However, these figures both over- and underestimated the final loss. A great many of the manuscripts recorded as lost, in fact survived in some form, though they may have been either unidentifiable or inaccessible at the time. In the final assessment only thirteen manuscripts were found to be entirely destroyed in the fire, though many of those surviving may only have existed as blackened lumps or numerous loose and charred fragments. Conversely, the figures for damaged manuscripts were greatly underestimated. The parchment of the manuscripts, being raw skin product, suffered not only from the heat and smoke but from the water used to extinguish fire. Many of the surviving manuscripts were badly embrittled and much textual loss was suffered as a result of subsequent handling.32

The salvage and conservation work began in earnest shortly after the fire. Paper manuscripts were disbound, the pages cleaned then washed with alum, and dried over a line. This was done rapidly by a large number of semi-skilled staff, to arrest the development of mould. However, when the pages were rebound the first of many errors in collation, which continue to vex scholars and caretakers of the collection to the present, occurred. Initially few of the parchment manuscripts were disbound as they were mostly able to be dried by leaving the volumes open and routinely turning the pages. Little could be done at this stage to address the burnt parchment. Exuded fat deposits were removed from the edges but otherwise the manuscripts were left largely as they remained after drying.33

After the initial salvage of the damaged material a great problem remained. Many loose fragments of paper and parchment existed, which were collated and assembled to a state as close to the appropriate order as could be determined. However, a lack of documentation at the time has further added to the continuing uncertainty regarding collation. The speed at which this work was undertaken (the bulk of the emergency conservation was completed in a three month period) no doubt contributed to the apparent lack of attention to detail.34

After this initial flurry of attention the collection floundered for over twenty years until the establishment of the British Museum in 1753. One of the first acts of the Trustees of the Museum was to inspect the collections. They found the Cotton Library dusty but otherwise unchanged, though once again poor record-keeping had caused some manuscripts initially recorded as undamaged to disappear into the general collection of fragments, from where they would not emerge for nearly a century.35

The collection was finally transferred to the Museum in 1757,36 where the trustees and caretakers found themselves at a loss for solutions to address the magnitude of the problem. The quantity of fragments alone was estimated as 'enough to fill a small cart.'37 Attempts were made to soften and open the burnt parchment fragments, often using experimental techniques on significant material with disastrous results. Several fragments of one of the treasures of the collection, the Cotton Genesis, were sent for an experimental treatment and were returned with the consistency of 'burnt biscuits'.38 The most successful and universally adopted technique for softening and opening the burnt parchment manuscripts was developed by the Keeper of Manuscripts, Josiah Forshall.

Forshall developed a technique that involved immersing the manuscripts in water. Once softened, the fragments were cut open and air-dried. The opened fragments were inlaid in paper to allow them to be handled without further losses to the edges, and were finally collated and rebound into volumes.39 As conservation did not yet exist as a discrete profession, almost all of the opening and flattening of the parchment was carried out by the Keepers of Manuscripts—first Forshall, then the true champion of the collection, Sir Frederick Madden.40 Madden's singular dedication to the recovery of the Cotton fragments spanned his forty year career as Keeper of Manuscripts to the collection and involved constant lobbying of the Board of the British Museum for funding for the project.

Some fragments of Otho C.V were collected in the fire clean-up but the majority was rediscovered by Forshall after they were opened and flattened. Further fragments were added and the manuscript arranged by Madden before they were inlaid and bound. Initially some fragments from a different manuscript were included but these were removed in 1855. In 1963 the inlaid leaves were rebound with the inclusion of guards to allow for the bulk of the pages.41

The Cambridge University Library has a series of richly illuminated borders credited to the workshop of Attavante degli Attavante. The borders were originally included in a large Papal Missal made for the second son of Lorenzo de'Medici, Giovanni de'Medici who became Pope Leo X. The borders contain the Medici device of a diamond ring with feathers and the motto 'Semper'.42 The borders would originally have decorated the margins of a large illumination. The 1714 Vatican inventory records three such Papal Missals.

These volumes were seized by France along with a large number of manuscripts, artwork and antiquities after the success of Napoleon's European campaigns. The fragments resurfaced in the first ever exclusive sale of manuscript cuttings in England in 1825—a sale which marked the beginning of the appreciation of manuscript illumination as an art form and set in motion the Victorian fascination for manuscript cuttings.

France's young public cultural institutions had benefited from the nationalisation of the cultural collections during the French revolution. As the collections of religious orders and the nobility were confiscated, they were deposited in the public museums and libraries. Collections swelled under the system, allowing the Bibliothèque Nationale to increase its holding fivefold.43 Under the Commission of Monuments the seizure of cultural property was mandated in an attempt to prevent its wanton destruction.44

The national collections were further enriched by the success of Napoleon's campaigns in Europe, which saw a vast number of treasures from wealthy families and religious collections in Italy, Germany, Belgium, Austria and the Netherlands transplanted to Paris, primarily to the Museum Français. These items arrived in convoys with much public fanfare.45 The first convoy alone contained more than sixty pieces of sculpture from the Vatican, nine Raphaels, frescoes, sculpture, jewels, exotic animals and natural history collections.46

In June 1796 the Pope signed a truce with France which included a promise of a further 500 manuscripts from the Vatican, as well as art and sculpture, with the Pope to pay the costs of transportation. The treaty of Tolentino in 1797 included a promise of more treasures from the Vatican plus the sum of three hundred million scudi. The peace negotiated at Campio Formio in October 1797 ended Napoleon's Italian campaign and granted Venice to Austria, though this did not see the end of the transportation of cultural treasure to France.47

The booty seized from Italy between 1796 and 1798 included sixteen hundred manuscripts, including the Papal Missals and all manuscripts from the Vatican dated before 900AD.48 The personal library of Pope Pius VI was confiscated in 1798; among its many great treasures were an illuminated copy of Dante's Divine Comedy (BnF MS ital 78) and Petrarch's Triumphs (BnF MS Ital 545), both of which remain in the Bibliothèque Nationale.49

Not all of this material survived to become part of the public collection—much of the precious metal was melted down and two public auctions were held in 1804 and 1811 to raise money for the continuation of the military campaigns. By the Bourbon Restoration in 1814 more than four thousand books had been removed to France from other parts of Europe.50

The justification for the wholesale looting of cultural material was convenient at best—from the stated need for exemplars for French artists to Napoleon's politic claim 'all men of genius… are French no matter in what country they were born'.51 After Napoleon's abdication in 1815 many of the owners of the confiscated material undertook campaigns for restoration with some success, though much material remained and even now forms core collections in both the Louvre and the Bibliothèque Nationale.

One of the most significant confiscations was the four bronze horses from St Mark's in Venice. These statues attributed to Lysippus were themselves looted from Constantinople in the 1204 during the fourth crusade. They were seized in 1797 and taken to Paris where Napoleon eventually had them installed at the top of the newly constructed Arc de Triomphe.52 In 1815, the Austrian army removed them under an armed guard and returned them to Venice.53

Of the location of the lost material, including the Papal Missals, little was known until a large number resurfaced in a series of sales of collections belonging to the enigmatic priest Abate Luigi Celotti in England. Celotti was the secretary and librarian to Count Giovanni Barbarigo from 1801. He purchased the libraries of many illustrious private and monastic collections in Italy and acted as a dealer in paintings, manuscripts and illuminations in the early nineteenth century.54

In 1825, Celotti held three auctions of manuscripts in England. The first two were through Sothebys and included the State Library of Victoria's Scriptores Historiea Augustae which also bears the Medici arms.55 The final auction which included the Cambridge borders was held through Christies and consisted entirely of manuscript cuttings. In this respect it was the first auction of its kind.56 The auction was titled A Catalogue of a Highly Valuable and Extremely Curious Collection of Illuminated Miniature Paintings, of the Greatest Beauty, and of Exquisite Finishing, Taken from the Choral Books of the Papal Chapel in the Vatican, during the French Revolution; and subsequently collected and brought to this Country by the Abate Celotti. Included were many large illuminations framed and presented as paintings, often with illuminated borders pasted around the edges. The presentation of the illuminations as paintings instead of mere historic curiosities was no doubt Celotti's intention. The choice of Christies over Sothebys was presumably based on Christies' greater popularity among art dealers.57 The Cambridge cuttings are believed to be the borders surrounding Crucifixion miniatures which comprised lots 53 or 51.

Celotti's sale of cuttings coincided with, and no doubt encouraged the popularity of manuscript cuttings in nineteenth century England. By presenting illuminations as monumental art, Celotti contributed to the appreciation of illumination as art. However, this shift in perception encouraged the isolation of the image from the text, which was the basis for the cutting up of many other manuscripts, a practice especially popular in Victorian times. Ironically many cuttings were eventually returned to a codex format by being compiled into albums.

Fragmentary manuscripts, such as those included in The Medieval Imagination exhibition at the State Library of Victoria, are difficult to interpret and can never provide the same information as an intact manuscript. Besides the obvious physical losses and lacunae, there is a significant loss of context. Evidence of damage can be visually distracting, and in addition, restoration work further obscures the original nature of the manuscript as it inherently requires reinterpretation by another individual.

In the case of Cotton Otho C.v. most of the text and all of the illuminations are lost. The visual impact is highly disrupted by the charring and water staining from the fire as well as the inlaying and binding. The collation has been disturbed and reconstructed. The illuminations of the Baillieu bi-folios are completely overpainted, the text largely illegible and the parchment discoloured and stained. The context is lost, as not only are the pages isolated from the complete volume, displayed flat and vertically, but also because they are bi-folios they are not sequential. The restoration of the illuminations, though credited with a high degree of accuracy, is largely a reconstruction. The Cambridge cuttings are so far removed from their original format as to bear no relationship to a volume.

Though the presence of damage and restoration stand as a barrier between the viewer and the original object, they also add another layer of context to the history of the manuscript.

Otho C.v. exists as an artifact of the collection and tragic decimation of the Library and ambitions of Sir Robert Cotton. The gathering of such a great collection was possible only due to the dissemination of monastic collections after the Reformation. The collection, though it provided substantial political power to its owner, was also highly accessible during his lifetime. The restoration of the Cotton Library was an epic event of in-depth conservation treatment and probably the largest and first of this type of project ever undertaken. The project extended over three centuries and spanned the careers of many individuals. The restoration also pioneered the development of new techniques for treating fire-damaged parchment.

The Cambridge cuttings exist in their current format due to the perception of cultural property as the spoils of war. Napoleon's seizure of cultural property during his European campaign was not the first of such operations. The thefts were explicitly justified at the time by the example of Rome's looting from Greece and the idea that this had permitted Rome's subsequent intellectual and artistic development.58 Parallels have also been made to the seizure and destruction of cultural artifacts by the Nazi regime.59

The inclusion of the illuminated borders in the Celotti art sale places them at a definitive point in the history of illuminated manuscripts, marking the beginning of Victorian interest in manuscript cuttings. Though this interest encouraged the appreciation of illumination as art alongside monumental art, it also led to the devastation of many more manuscripts in an attempt to isolate the images from the text.