The Ashburnham Pentateuch

The Ashburnham Pentateuch is a parchment codex of 142 folios, written in two columns and measuring 375 by 310 millimetres, now held by the Bibliotheque nationale de France under the shelfmark Nouvelles acquisitions latines 2334. Its main text is in a formal uncial hand of the late 6th or early 7th century. It once carried as many as 69 full-page miniatures illustrating the first five books of the Hebrew Bible; 18 survive today, at folios 1v, 6r, 9r, 10v, 18r, 21r, 22v, 25r, 30r, 40r, 44r, 50r, 56r, 58r, 65v, 68r, 76r and 127v, alongside a decorated frontispiece and three sets of table pages framed by painted arcades.

The book is fragmentary in a specific and telling way. Of the five books of the Pentateuch it was made to contain, Deuteronomy survives as a single leaf. Genesis, Exodus, Leviticus and Numbers are all present but with substantial gaps, and a 9th- or 10th-century reader left notes, in places using Tironian shorthand, recording exactly where leaves were already missing by then. The book that reaches the reader today is therefore not simply an old manuscript; it is the visible record of a long process of loss that had already been underway for centuries before the manuscript's more famous 19th-century misadventure.

The Latin text is a pre-Alcuinian version of the Vulgate, one that the manuscript's own cataloguers at the Bibliotheque nationale describe as related in places to the text tradition associated with Cassiodorus and in places to an older Italian version. This textual affiliation is itself a piece of evidence in the dating and localisation debate, since it ties the manuscript to text-critical traditions current in Italy in Late Antiquity, without settling where the physical book itself was copied and painted.

The surviving books run: Genesis (ff.2r-49v, opening with a table of contents), Exodus (ff.51r-90v, with a table of capitula under painted arcades), Leviticus (ff.91r-115r, missing its opening), Numbers (ff.115v-141v, with its own table of capitula) and, from Deuteronomy, a single leaf (f.142r). Scraped and rewritten passages in the same hand that supplied the missing leaves show that whoever restored the book in the 8th century worked directly on the damaged original, not on a fresh copy.

The main text is written in a disciplined biblical uncial, the formal book-hand of Late Antiquity, with rubricated titles alternating red and black ink. Thirteen folios (ff.3-4, 8, 37-38, 60-64, 122 and 129) are replacements copied in the mid-8th century, after the book had left its place of origin; their script mixes uncial and semi-uncial forms, and one leaf, f.32v, is written in Caroline minuscule, the reformed script of the Carolingian era, showing that the book continued to be worked on for centuries after it was first made.

A 2020 technical study of the pigments, using non-invasive UV-visible reflectance spectrophotometry, spectrofluorimetry, X-ray fluorescence and optical microscopy, found extensive use of Egyptian blue across the miniatures, an unusual choice for manuscript illumination of any period. More strikingly, the analysis found the pigment had been synthesised from a copper and zinc alloy in roughly a four-to-one ratio, rather than from the metallic copper or copper-ore sources documented in traditional Roman and earlier Egyptian manufacture of the same pigment. That is a genuine technical fingerprint, not yet matched to any single known regional workshop tradition, and it sits alongside the script and the text as a third independent body of evidence bearing on where the book was made.

The manuscript's date is usually given as late 6th or early 7th century, with the 8th-century Frankish restoration folios treated separately. Its place of origin is not settled at all. The Bibliotheque nationale's own catalogue entry lists the candidates without choosing among them: Spain, North Africa, northern Italy, or Rome, with a question mark against each.

Bezalel Narkiss, after years of study of the manuscript, argued for production in an imperial scriptorium at Rome in the 5th century, a full one to two centuries earlier than most other scholars accept, and associated it with the patronage of Galla Placidia; this fifth-century Roman hypothesis is a minority position. Others, on the basis of stylistic and palaeographic similarities to later Spanish manuscripts, argue for a Visigothic Spanish origin, typically in or around Seville, while allowing that the pictorial models the painters were copying could themselves have been North African or Egyptian. Depictions within the miniatures of both light-skinned and dark-skinned Egyptians have been cited on more than one side of this argument, as evidence read differently depending on the interpreter's starting assumption about where the artists were working. No single reading has displaced the others, and this record does not adjudicate between them; it reports the live dispute as a live dispute.

The manuscript's single most discussed image is its Creation page, at the front of Genesis, which originally depicted an anthropomorphic Trinity: paired figures representing the Father and the Son across a sequence of Creation vignettes, together with a winged figure representing the Holy Spirit, set in an undefined landscape. Jennifer Awes Freeman's 2022 study argues that this scheme once affirmed a strong, visually literal equality between Father and Son, at a moment when this manner of imagining the Trinity had only a few precedents in early Christian sarcophagus sculpture and no real continuation for centuries afterward.

At some point in the Carolingian period, by which time the manuscript had come to Tours, a corrector painted over parts of the image. The alteration is described as removing three figures representing the Son and one representing the Father, leaving, in the surviving state, three Fathers and one Son, an inversion Awes Freeman reads as a deliberate response to Carolingian anxieties about preserving the unity of the Trinity during a period of active Christological and image-theory debate following the Byzantine iconoclastic controversy. Neither the identity of the corrector nor the exact date of the repainting is recorded; what is documented is the change itself, and its consistency with a known current of Carolingian theological concern.

In 1953-54, Joseph Gutmann argued in the Jewish Quarterly Review that the Ashburnham miniatures preserve echoes of midrashic exegesis not found in the biblical text itself, and concluded that the pictorial conception underlying them was fundamentally Jewish, descending from a lost tradition of Jewish biblical illustration related in spirit to the wall paintings of the Dura-Europos synagogue. Kurt Weitzmann, whose own work on the transmission of biblical picture cycles gave the comparison real authority, treated the Dura-Europos connection with caution rather than adopting Gutmann's thesis outright.

Dorothy Verkerk's 2004 study argues instead for a Roman origin and reads the cycle as a teaching tool aimed at instructing catechumens in the Christian community, pushing back against what her reviewer, Herbert Kessler, describes as "the tendency to look through the Paris manuscript to lost Jewish pictorial prototypes." Kessler is careful to note that Verkerk's target on this point is not really Weitzmann, who had already distinguished the Ashburnham cycle from the Dura-Europos paintings rather than equating them. The dispute is not resolved in the literature consulted for this record. It is presented here as a live disagreement between named scholars, not as a settled fact in either direction; folkloric and exegetical elaboration of the Hebrew Bible was shared by Jewish and Christian communities in Late Antiquity, and that overlap is itself part of why the argument has not been closed by either side.

The manuscript had reached Frankish Gaul by the 8th century, when the 13 replacement leaves were copied in to repair losses from the original quires. By the following century it was at Tours, where it remained for the whole of the medieval period. It was at Tours, not at its place of origin, that the Creation page was later reworked, and it was at Tours that a 9th- or 10th-century reader added the marginal notes recording missing folios.

The manuscript's presence at Tours left a mark beyond its own pages. Andre Grabar's 1957 study in Cahiers archeologiques traces Romanesque wall paintings in the former church of Saint-Julien at Tours back to the Ashburnham Pentateuch's Tabernacle imagery, arguing that the manuscript itself served as the model consulted by the medieval muralists. The manuscript is also described in the literature as a model consulted for the group of Carolingian-era Touranian ('Tours') bibles, the great illustrated Bibles produced in the Tours scriptoria in the 9th century, connecting this one small, damaged, largely un-illustrated-looking Late Antique codex to one of the major manuscript-painting enterprises of the Carolingian renaissance.

Guglielmo Libri Carucci dalla Sommaja was an Italian-born mathematician who became a naturalised French citizen in 1833 and, over the following decade, built a career inside French academic and political institutions, holding a chair at the College de France and a post at the Sorbonne. In 1841, helped by his connections to the statesman and historian Francois Guizot, he secured appointment as the French government's Inspector-General of Libraries, a role charged in part with cataloguing valuable books and manuscripts in provincial collections across the country.

That official position is what makes Libri's thefts a case study rather than an isolated crime. He used the access it gave him, combined with a performance of poor health that let him work alone and unsupervised in archive rooms, to remove items that were never returned. Documented losses attributed to him include material from the Bibliotheque inguimbertine at Carpentras, among it the bulk of Descartes's surviving letters to Father Marin Mersenne, and the mutilation of several volumes from the Peiresc collection, from which an estimated two thousand leaves disappeared. Estimates of the total scale of what he removed from French libraries run into the tens of thousands of items.

The Ashburnham Pentateuch was stolen at Tours in 1842, during the years Libri held his inspectorate. The Bibliotheque nationale's own catalogue entry states plainly that Libri took it at Tours in that year; it is one of the best-documented single items in his catalogue of thefts precisely because its recovery, decades later, depended on proving exactly this fact.

A secondary, non-scholarly source consulted for this record adds a more specific claim: that the manuscript had been kept, through the medieval period, in the library of the cathedral chapter of Saint-Gatien at Tours, and had been transferred, along with other confiscated ecclesiastical holdings, to the civic Bibliotheque municipale de Tours during the French Revolution, and that it was from this municipal library that Libri took it. This record has not been able to corroborate the Saint-Gatien detail or the Revolutionary transfer in the scholarly or institutional sources consulted, and flags it here as unverified rather than presenting it as established fact.

As the French monarchy fell in the 1848 revolution and a warrant was issued for his arrest, Libri was tipped off and fled to London, shipping eighteen large trunks containing an estimated 30,000 books and manuscripts ahead of him. In London he found a protector in Antonio Panizzi, Principal Librarian of the British Museum, and successfully framed the accusations against him to English opinion as anti-Italian prejudice rather than a substantiated criminal case.

On 22 June 1850, a French court convicted Libri of theft in absentia and sentenced him to ten years' imprisonment. He never returned to France to serve it; he lived in England on the proceeds of selling his collection, two large sales in 1861 alone reportedly netting him over a million francs, before returning to Italy in 1868 and dying near Florence the following year. The same non-scholarly source that describes the Saint-Gatien provenance also claims that Libri sold the manuscript to Lord Ashburnham under a false attribution, representing it as having come from Grottaferrata, near Rome, rather than from France. This record could not corroborate that specific claim in any other source consulted and flags it as unverified; it is included here only because, if true, it would be directly consistent with Libri's documented general practice of falsifying records to cover the origin of what he sold.

In 1847, Bertram Ashburnham, 4th Earl of Ashburnham, bought the Ashburnham Pentateuch from Libri as part of a much larger purchase, one that eventually brought roughly 2,000 manuscripts from Libri's holdings into the Ashburnham library at Ashburnham Place in Sussex. It is this purchase that gives the manuscript its English name.

Whether the 4th Earl knew, or should have known, that a substantial part of what he was buying had been stolen from French and Italian public collections is a genuine point of historical debate, and the sources consulted for this record do not resolve it. What is documented is the outcome rather than the Earl's state of mind: decades later, French and Italian investigators were able to identify large numbers of manuscripts in the Ashburnham collection as items missing from specific French and Italian public libraries, a task that would not have been straightforward, and arguably would not have been necessary, had the collection's origins been openly declared at the time of purchase.

Bertram Ashburnham, 4th Earl, died in 1878. His son, the 5th Earl, opened negotiations to sell part of his father's manuscript collection, and Leopold Delisle, the Bibliotheque nationale's leading manuscripts scholar, spent much of the following decade building the documentary case that a large part of what the Ashburnham family held of French origin, including this manuscript, had been stolen from French public libraries by Libri and by the bookseller Jean-Baptiste Barrois. Delisle later published this material in full, first in a report to the French Minister of Public Instruction and Fine Arts, and then, once the purchase was complete, in his Catalogue des manuscrits des fonds Libri et Barrois of 1888, whose preface sets out the whole history of the affair.

The Ashburnham Pentateuch itself was purchased from the London bookselling firm Trubner, acting as intermediary, on 23 February 1888, and entered into the Bibliotheque nationale's manuscripts register under the number 8070(12). This record has not found a published figure for the price paid, either for this manuscript individually or as its share of the wider 1888 purchase, in the sources consulted, and records that gap rather than estimating a figure.

The Ashburnham Pentateuch was one item within a much larger, internationally scattered problem created by Libri's thefts. Manuscripts he had stolen in Italy and later sold to Lord Ashburnham were the subject of a separate recovery: in 1884, the Italian government purchased essentially the whole of this Italian portion of the Ashburnham manuscripts, all but about 100 of them, and deposited it in the Biblioteca Medicea Laurenziana in Florence, where it remains catalogued today as the Fondo Ashburnham. The French recovery negotiated by Delisle in 1888 was a separate transaction, covering manuscripts of French origin (100 further Libri manuscripts and 66 from Barrois), of which the Ashburnham Pentateuch was one.

This distinction matters for accuracy. Other Libri-Ashburnham manuscripts went to different institutions at different dates, under different negotiations, for different and in some cases still-undocumented prices; none of that later history belongs to this specific manuscript, whose documented path runs from Tours, to Libri, to Ashburnham, to the Bibliotheque nationale de France in 1888, and no further.

The manuscript's current binding, a 19th-century half-binding in Russian leather over wooden boards, postdates its arrival in France and is itself an artefact of that period's conservation practice rather than of the manuscript's medieval history; its spine lettering, giving a 5th-century date, reflects an estimate later scholarship has revised. A digitised facsimile is available through the Bibliotheque nationale's Gallica platform, and the library also holds microfilm substitute copies for consultation, with access to the original manuscript restricted and requiring the director's authorisation.

The manuscript has been included in a sequence of major exhibitions at the Bibliotheque nationale itself (1954, 1972 and 2007) and beyond, including the Metropolitan Museum of Art's 1977-78 exhibition Age of Spirituality, edited by Kurt Weitzmann, and the Musee de Cluny's 2016-17 exhibition on the Merovingian centuries. The 2020 pigment study represents its most recent published scientific examination found in the sources consulted, and Jennifer Awes Freeman's 2022 monograph is its most recent dedicated scholarly study.