Few names in manuscript studies are as unstable as Codex Bobbiensis. The phrase is attached, in the catalogues and in the literature, to at least four different books that happen to share a place of origin, and to one further manuscript that shares only the name of the place. To ask what the Codex Bobbiensis is, in the definite article, is therefore not yet to ask a question about a manuscript at all. It is to ask which of several manuscripts has been made to stand for the others, and on what grounds.
This essay is about the palimpsests. Two volumes from the Bobbio library, both written over an earlier text, both surfaced again in the early nineteenth century, and both still carrying the ghost of the book that had been there before them: the Septuagint with its parallel Old Latin, now in Turin as B.I.1, or in the older Bobbio numbering 1 inf. 5; and, in a separate volume, the Hexateuch palimpsest, 1 inf. 7. These are the objects in which the practice of erasure was carried out so thoroughly that the erased layer survived by accident, and was then recovered by scholarship over a century and a half of patient work.
The reason the story belongs on this site is exact rather than decorative. What the Bobbio scribes did to a Greek Bible of the fifth century, they did in order to make an older text disappear beneath a newer one while keeping it – unreadably to the eye, but genuinely – underneath. That is very nearly what the Masoretes did, at a very different scale and for entirely different reasons, to a Hebrew Bible. The masorah is the record of a text’s earlier life preserved inside the frame of a later copy: the older spelling, the older form of the word, the older reading, written into the margin of the newer one so that a copyist checking his work against a checksum could see what the text had been before he began. The Bobbio palimpsest does the same thing in reverse and by accident. The text underneath was not kept on purpose; it was kept because the scraper could not quite obliterate it. What a discipline gains when it insists on recovering the erased layer is the question this essay is finally asking.
The Name, and Why It Is Dangerous
The first of the four is the one with which the name is most often associated, and it is the one that must be described with the most care, because it is a fourth-century book that survives only in fragments. Codex Bobbiensis, siglum k, Vetus Latina 1, is the Old Latin Gospel manuscript now in Turin, shelfmarked G.VII.15 in the Biblioteca Nazionale Universitaria. It is usually dated to the fourth century, and the Vetus Latina register describes it as copied in Africa in that century, with an exemplar thought to have been papyrus. What it preserves is roughly half a Gospel book: Matthew from 1:1 to 15:20, and Mark from 8:8 to 16:9. The register gives the order of the book as John, Luke, Mark, Matthew, although only portions of the latter two survive; the more dramatic Western sequence of John, Matthew, Mark, Luke is attested elsewhere in the Old Latin tradition, and belongs to no single witness. Its Mark text carries the shorter ending of the Gospel rather than the longer one, and the Vetus Latina’s account of the manuscript describes it as the only Greek or Latin gospel book to have that ending by itself. The Latin stands in the African type, and its text appears to antedate Cyprian of Carthage: the older witness of the African revision of the New Testament, and by some way the most important of the surviving Old Latin gospel books for the text of the fourth century.
The second is the volume at the centre of this essay. The Septuagint palimpsest, shelfmarked B.I.1 at Turin, or Bob. 1 inf. 5 in the older Bobbio numbering, contains a Greek Septuagint text of the Psalms with a parallel Old Latin version running alongside it, and it is a palimpsest: the parchment carries an earlier text beneath. It is usually dated to the fifth century, though sixth-century dates have been argued, and its condition is fragmentary – the Psalms do not survive there in continuous text, and what survives was itself written over something else. It should be said plainly that the two books are separate volumes with separate shelfmarks, and that the single label B.I.1 does not cover both the Septuagint and the Hexateuch. The Hexateuch palimpsest, Bob. 1 inf. 7, is the third of the four: a book of the first six books of the Bible in Latin, written over an earlier text, in the same Bobbio library and now in the same Turin collection.
The fourth is quite separate again. Vetus Latina 135, siglum J, is an Old Latin manuscript of Tobit, Esther and 2 Maccabees that came to the Biblioteca Ambrosiana in Milan with the Bobbio volumes that Cardinal Federico Borromeo acquired in 1606, and it is regularly cited under the same name. Nor is it the last of the near relations. Vetus Latina 53, Codex Bobiensis or Vindobonensis, is the Naples manuscript of Acts and the Catholic Epistles, copied in the sixth century and itself palimpsested at Bobbio in the eighth. The pattern is worth noticing: the name has been used less for a particular book than for a family resemblance – a group of early Latin and Greek witnesses that came out of one library, and that have been allowed to stand for each other in the secondary literature ever since.
There is a second reason for caution, which is chronological. The nineteenth-century scholarship that gave the name its currency was mostly about the Latin Gospel fragments, and the dates it fixed to them – Fleck’s edition of 1837, Tischendorf’s of 1847 and 1849, and the edition of Wordsworth, Sanday and White in 1886 – have a habit of migrating in the literature to the Septuagint palimpsest, to which they do not belong at all. The Septuagint and Hexateuch volumes had their own nineteenth-century history, with their own correcting and their own editors, and that history needs to be kept separate from the history of k if either is to be understood.
So the subject of this essay is declared. It is the two Bobbio palimpsests – the Septuagint with its parallel Old Latin, B.I.1, and the Hexateuch, 1 inf. 7 – read as what they are: books that were written over books. The Old Latin Gospel fragments at G.VII.15 will appear here as a neighbour and a comparison, because they were made in the same library and their editorial history is instructive, but they are a different object, and they will be given their own shelfmark and their own date whenever they appear.
Bobbio and the Dispersal
The story begins with an Irish mission. Bobbio, in the Trebbia valley south of Piacenza, was founded as a monastery in 614 by Columbanus, and the community that established itself there was Irish. The library that grew up around the house was an Irish foundation in an Italian monastery, and the scriptorium that served it was run for a time by Attala, who brought with him the manuscript discipline of the Irish schools. The library’s tenth-century catalogue lists some seven hundred codices – the figure is a standard one, and it is worth holding in mind that the catalogue counts volumes, not leaves, and that a great many of those volumes have since been dispersed, damaged or lost.
What happened to those volumes over the next thousand years is a story of confiscation. The monastery was suppressed by the French in 1803, and the library was broken up. But the dispersal had begun long before that, and the first and most consequential of the moves was made in 1606, when Cardinal Federico Borromeo acquired eighty-six volumes of the Bobbio library for the Biblioteca Ambrosiana in Milan, with the mediation of the Benedictine scholar Jean Mabillon. Twenty-six more went to the Vatican in 1618. The bulk of the remainder passed to Turin in 1824, and it is in Turin that the Bobbio collection is now, in the Biblioteca Nazionale Universitaria, where the Septuagint palimpsest is shelfmarked B.I.1, where the Hexateuch palimpsest is Bob. 1 inf. 7 in the older Bobbio numbering, and where the Old Latin Gospel fragments acquired the shelfmark G.VII.15.
It is worth being clear about what this sequence does to the subject of the essay. A Greek Septuagint of the fifth century, made in the eastern Mediterranean, at some point in the hands of a scribe in the western provinces, ends up in a northern Italian monastic library in the seventeenth century and then in a national library in the nineteenth, because a cardinal in Milan thought a library worth having. The chain of custody is not the chain of the text. What travelled from the fifth century to 1824 was not a manuscript so much as a set of circumstances that allowed a particular set of leaves to be scraped, written over, scraped again, and written over again, until it turned out to be the only place in the world where a certain lost text could still be read. The Book of Deer, which this site treats elsewhere, has a comparable trajectory, and a rather different ending. But the point here is the same: manuscripts are not found, they are inherited.
And the inheritance was damaged. On 26 January 1904 a fire in the Palazzo dell’Accademia delle Scienze in Turin, which housed part of the Bobbio collection, destroyed a large quantity of material. The extent of the damage across the Bobbio volumes was uneven, and some witnesses were badly hurt while others escaped, but the fire stands in the history of the collection as a real loss rather than a legend, and it is part of the reason the palimpsests were so hard to work on: what survived had to be read through damage, and what was recovered from beneath the writing had to be read through scraping as well. It is a reminder that the recovery of an erased layer is not simply a question of technique. It is a question of what is physically left to read.
The Palimpsest Technique, and What It Does to a Text
The word itself is an instruction. Palimpsestos is Greek for a scraped skin – palin, again, and psestos, rubbed or scraped – and a palimpsest is a parchment that has been used twice, the second use having been preceded by an erasure of the first. The word entered general scholarly use in the early modern period, in connection with the recovery of texts that had been overwritten rather than destroyed, and it acquired its sharpest meaning in the nineteenth century, when the work of Angelo Mai at Milan and of the German and Italian palaeographers who followed him demonstrated that whole libraries had been written over.
The physical process is straightforward and brutal. The old writing had to go, because parchment is expensive and because a library that was short of skins could not always afford new ones. The scraper – a knife, traditionally called a lunellum – was drawn across the surface of the leaf at an angle, cutting into the flesh of the parchment without cutting through it. The surface was then washed, usually with an alkaline solution, to remove the ink that remained and to open the pores of the skin. The leaf was then ruled again and written on afresh, usually in a smaller script and often with a different ruling pattern from the first, the two sets of lines running at angles to one another or crossing.
What survived was never the writing itself but the damage to it. The knife did not reach every part of the leaf, and where it did not, the old ink stayed in the hollows of the parchment; where it did, the ink that had pooled in the depressions was dragged up into the cut and remained legible as a shadow. In addition, because such parchment is thin, the ink of the undertext sometimes penetrated the fibres and can be read from the other side of the leaf as a mirror image. The result is a leaf that carries two texts at once, of which the older is a series of ghosts: partial letters, broken strokes, the impression of a word rather than the word. Some leaves preserve a great deal; others are almost clean. The variation depends on the pressure of the knife, on the absorbency of the particular skin, on how many times the leaf had been scraped before, and on how carefully the new scribe ruled his lines around the surviving traces rather than over them.
Two features of the Bobbio case make it exceptional. The first is that the Septuagint codex retains both layers, and that the two layers are not the same book. Beneath the Greek Septuagint of the Psalms, running in parallel columns with a parallel Old Latin version, there lies an earlier text of the same general kind – and both the Greek and the Latin of the upper layer were written over something. A palimpsest that had overwritten a palimpsest would be a curiosity; a palimpsest that preserves a Greek Septuagint and a Latin version together, both of them over a lost original, is a witness of the first rank to the way the Greek and the Latin circulated in the same room.
The second is the scale of the survival. Other palimpsests survive as isolated leaves or small groups; the Bobbio volumes survive as substantial books, with quires, with a structure, with the arrangement of the text in them. That means the recovery is not a matter of rescuing a few words from beneath a scraper, but of reassembling a book. And it should be said clearly that the Bobbio Gospel book usually called the Bobbio Missal, which is at Paris, is a different object altogether; it is an Irish Latin Gospel manuscript with its own history, and it has nothing to do with the Septuagint palimpsest beyond the fact that both came out of the same monastery.
There is a further consequence, and it is easy to forget. A palimpsest is not a damaged book. It is a book with two texts in it, one of which is accessible only through instruments – raking light, infrared photography, multispectral imaging, the trained eye of a palaeographer working from a high-resolution surrogate – and through the patient comparison of the surviving traces with what is known of the texts they belong to. The recovery of the erased layer is not a reading. It is a reconstruction, and its results are always in part conjectural, because the evidence is the evidence of accidents.
Angelo Mai, the Nineteenth Century, and the Two Editions
The Bobbio codices reached Milan in 1606, when Cardinal Federico Borromeo bought eighty-six volumes for the Biblioteca Ambrosiana, with the help of the Benedictine Jean Mabillon, who had no idea what he was bringing into the building. For two centuries the palimpsests sat in the Ambrosiana as what they appeared to be: ordinary Greek and Latin books, catalogued, listed and read as such, with the odd detail that some pages were hard to make out. The discovery of the Bobbio library’s true contents came earlier than the recognition of the palimpsest layer, but the date that matters for the palimpsests is 1818, when Angelo Mai, working in the Ambrosiana, recognised as a palimpsest certain leaves that had been catalogued as ordinary. He published the newly surfaced texts in 1828, and the effect was immediate. A fifth-century Septuagint of the Psalms with a parallel Old Latin version, and a Greek commentary on Job attributed to Chrysostom, emerged from books that had been sitting in a public library for two hundred years with their contents unread.
What followed was the long, unglamorous business of making the new text legible. The Latin required correcting – restoring the letters that the knife had taken and the eye had lost, and establishing a text that could be quoted – and the correcting was done by Karl Wilhelm von Haneberg and by Antonio Maria Ceriani, the first in the volumes of Biblische Studien whose run extends into the 1880s, and the second at the Ambrosiana itself. The precise year at which Haneberg’s contribution was made to the printed text varies in the accounts and is not worth insisting on. Giovanni Battista de Rossi’s studies of the Bobbio codices belong to the same campaign and belong to the same moment: the period when European scholars, with the help of the new art of photography and the new habit of collation, were beginning to establish that an enormous quantity of the written record of the early medieval West had been overwritten in the eighth and ninth centuries and was still in the parchment underneath.
And here a second narrative runs in parallel, which needs to be kept separate. The Old Latin Gospel fragments at Turin G.VII.15, siglum k, had a nineteenth-century career of their own. F. F. Fleck produced the first edition of k in 1837, and it was long judged inaccurate. Tischendorf edited it in 1847 and 1849, arranging the text in a way that proved inconvenient. The edition that established k as a working witness was that of John Wordsworth, W. Sanday and H. J. White, published at Oxford by the Clarendon Press in 1886 as number two in the Old Latin Biblical Texts series, with its title page giving the manuscript’s new Turin shelfmark. That 1837 date, which so often appears in general accounts of the Bobbio manuscripts, belongs to the Latin Gospel fragments and to Fleck. It has nothing to do with the Septuagint palimpsest, whose text was being recovered on a different track and by a different set of scholars at the same time.
The point of the parallel narrative is that the two recoveries converged on the same problem from opposite ends. For k, the difficulty was that a fragmentary fourth-century manuscript had to be rescued from the neglect of two centuries and given a readable text. For the palimpsests, the difficulty was that a whole text had to be lifted off the surface of a document that was itself the primary witness. One problem is the erasure of a book from the memory of scholarship; the other is the erasure of a book from the surface of a book. Both are the same problem – the loss of the first witness – and both were solved, in the end, by people willing to do the slow work of collation rather than the fast work of assertion.
It is worth being clear, too, about what a palimpsest recovery is not. It is not the recovery of a lost work in the sense that a lost work is lost – gone, with no trace and no possibility of reconstruction. It is the recovery of a text from a document that contains it, which is a rarer and in some ways a more instructive thing. The Chrysostom on Job is the extreme case. A commentary on the book of Job by the greatest preacher of the fourth century, one that had been quoted, paraphrased and used by the Latin Fathers, has no manuscript tradition of its own; what survives of it exists because a scribe in Bobbio needed a book to write over, and the scraper could not quite remove the ink. The text was not lost. It was hidden. And the discipline that recovered it had to be one that was prepared to look for what was not supposed to be there.
The Job Commentary That Was Only Hidden
Among the texts that came out of the Bobbio volumes in 1818 and 1828, one is stranger than the fifth-century Septuagint, because nothing like it was known to survive anywhere else. It is a Greek commentary on the book of Job, attributed to John Chrysostom, and it is known to us because it was written over: it survives in fragments recovered from the undertext of the Bobbio palimpsests, and it survives incompletely. It is not a lost work in the ordinary sense. A lost work is gone, and the question it raises is whether anyone can find it. This text was in a library, in a book, under a book, for perhaps eight hundred years, and the question it raises is a different and more exacting one: how much did the early medieval West read that we do not know they read?
The answer, for Job, is: some of it, and we would never have known. The recovery of the commentary was pioneering work – de Rossi’s studies of the Bobbio codices belong to the first generation of it – and it established a methodological point that has since become routine in the discipline but was, in 1818, close to revolutionary. A text is not lost if its material carrier survives. What is lost is not the work but the record of it: the library’s catalogue entry, the shelfmark, the cross-reference in a medieval reading list, the name of the monk who copied it. A library that was written over with care and thoroughness loses all of that at once, because the catalogue is kept on the same kind of surface as the books and is scraped in the same way.
It is worth extending the point, because it bears on how the Bobbio manuscripts are usually described. When the dispersal of the library is described as a series of losses – the monastery suppressed, the volumes scattered, the fire of 1904 – the emphasis falls on what was destroyed. But the more consequential loss had already happened inside the library itself, over several generations, in the ordinary course of a monastery that was short of parchment and long on books. The palimpsests are the survivors of that process, and they are survivors precisely because the erasure was imperfect. Everything else – the hundreds of works that were scraped cleanly, the commentaries, the theological treatises, the classical texts, the grammars – is simply gone, with no residue and no means of estimating the loss.
And that is what makes the recovered fragments so valuable beyond their own contents. A recovered text is a demonstration that the category of the lost is not a fact but an observation. When we say that a work is lost, we are usually reporting the state of our own libraries rather than the fate of the work, and the Bobbio palimpsests are standing proof that the two can diverge by several centuries and by an entire shelf of books. The Chrysostom on Job is a case in point of what was gained and, at a larger scale, of what was not. The commentary came back because a knife slipped. Everything that did not slip is beyond reach.
There is a second lesson here, and it is a methodological one. The recovery of a text from beneath a scraped surface requires a different discipline from the recovery of a text from a lost library, because the evidence is fragmentary and physical rather than narrative and documentary. The palaeographer must reconstruct a letter from the shadow of a stroke; the editor must then establish, by collation and by comparison with the printed and manuscript tradition of the work, which reading lies behind the shadow. The evidence is not a testimony that can be checked against a witness; it is a set of traces, and the discipline is an argument about probability. What emerged from the Bobbio undertexts in the nineteenth century was therefore not a text in the sense that a manuscript gives, but a text argued from traces – and the arguments have been refined ever since, by readers with better instruments and better languages.
What the Palimpsests Do to Septuagint Criticism
The two volumes are, in the end, witnesses to a question that has never gone away: how the Greek Septuagint and its Latin translations circulated together in the western provinces of the Roman empire. The Septuagint palimpsest holds the Greek text of the Psalms with the parallel Old Latin version running beside it, and the Hexateuch palimpsest holds a Latin Hexateuch written over an earlier text. Between them they give the study of the Latin Bible something that most manuscripts cannot supply at once: a Greek text and a Latin text set side by side on the same page, arranged one against the other by a scribe who wanted them to be read together. The arrangement is itself a datum. It means that the scribe believed the two belonged together – that the Latin was there to be checked against the Greek, or the Greek to be checked against the Latin, or simply that a house needed both and the skins were already to hand.
That permits a comparison that separate witnesses do not. The Latin of the Septuagint palimpsest stands in the Vetus Latina register among the psalters, which the register numbers in the range from 300 to 485, and it has to be asked of it whether it is a translation of the Greek text standing beside it or a translation of some other Greek text, and whether the two have a common ancestor. The parallel layout makes the question sharper rather than softer, because the moments where the Latin departs from the Greek stand side by side on the same leaf. A discipline that has to treat the Latin Old Testament largely as a scatter of separate witnesses has in these volumes something closer to a laboratory condition, and the comparison it has produced is one of the reasons the Bobbio texts have been collated and re-edited more than once.
The editions reflect a long collaboration. The nineteenth-century correcting of the Latin by Haneberg and Ceriani was followed, in the later twentieth and early twenty-first centuries, by a critical series issued under the title Bobbio. Biblioteca del Monastero di San Colombano and published at Munich by the Bayerische Akademie der Wissenschaften, whose volumes are edited by Alessandro Bausi and Alessandro Daneloni. Around that series stands the wider frame of the Vetus Latina Institute’s editions of the Septuaginta Vetus Testamentum, which provide the Greek text with a full critical apparatus and the sigla by which the Latin witnesses are identified and cross-referenced. The modern position is thus a collaborative one in which the Bobbio volumes are edited as what they are – part of a library, with a history and a neighbourhood – rather than as isolated items, and in which the Greek text and the Latin are printed with enough apparatus that a reader can see immediately where the two agree, where they do not, and where the undertext has anything to add.
It is worth asking, at this point, what the whole enterprise is for. It is not, in the first instance, about producing a text. Anyone can produce a text; the editions of the Bobbio manuscripts are the work of decades, and they exist to make a comparison possible that could not be made before: between the Greek of a fifth-century Septuagint, the Latin that a scribe set beside it, the older texts that lie underneath both, and the patristic commentary that lies underneath those. What is being assembled is a stratigraphy. Layer upon layer, each text with its own date, its own place, its own script, each one written over something and therefore recording the fact that something was there.
For a journal about the transmission of biblical texts, the resonance is exact. The Masoretic scribes did to a Hebrew Bible roughly what the Bobbio scribes did to a Greek one. They made the older text disappear beneath a new one while keeping it underneath – though the two cases differ in every respect that matters, and the difference is instructive rather than disqualifying. The masorah was written on purpose. It was a control system: an apparatus of checks that let a copyist verify his work against the work of earlier copyists without needing to understand the text he was copying, and that turned a codex from an isolated act of copying into a node in a chain of verification. The palimpsest was written by accident. Nothing in the Bobbio scriptorium was trying to preserve the text underneath; the text underneath survived because the knife slipped, and it survived illegibly, in a form no reader in the monastery could have used for anything.
And that is exactly where the analogy stops and the difference begins, and where the modern discipline earns its place. What the masorah did, the editor of a palimpsest now does deliberately. The recovery of an erased layer is the palimpsest’s masorah: an apparatus built after the fact, out of traces, by people who had no stake in the original and every stake in establishing what it said. The masorete wrote his checks beside the text he was copying, in the scriptorium, for the copyists who would come after him. The palaeographer writes his beside the text he is reconstructing, in the library, for the readers who will come after him. Both are the discipline of keeping the record of a text’s earlier life attached to the text itself, and both are the reason that a text can outlive the circumstances in which it was written.
Which brings the two traditions into a single line of thought. The Bobbio scribes erased a fifth-century Greek Bible to make room for a Greek Bible, and left the older one underneath in a state no one could read. The Masoretic scribes copied a Hebrew Bible with the earlier forms of that Bible recorded in its margins, in a state that anyone with the training could read. The difference between the two is that one erasure was a failure of materials and the other was a method. But the object they produced has the same shape, and it is the same shape for the same reason: a text is never only the text you can see. Underneath it there is always an earlier state, and the whole of textual criticism, whether it is done in the margin of a Hebrew codex with a knife and a pen or in a modern laboratory with a multispectral camera, is the business of deciding what that earlier state was and whether it is telling the truth.
The palimpsests are still doing that work. The Septuagint with its parallel Latin, and the Hexateuch beneath its own erased text, remain among the most consequential witnesses in the field, and the modern editions of them – Bausi and Daneloni’s volumes for the Bobbio series, set inside the Vetus Latina Institute’s Septuaginta Vetus Testamentum – are the current statement of what is legible in them. There is more in the parchment than anyone has yet extracted. That is the peculiar and lasting condition of the palimpsest: it is a document whose full content is known to be larger than the part that has been read, and every generation of readers since 1818 has found more in it than the last.
References
- Manuscripts of the Old Latin Bible — Vetus Latina Institute, University of Birmingham, master register, 2023, accessed 2 November 2026
- Old Latin Gospel Manuscripts — Vetus Latina Institute, University of Birmingham, 2023, accessed 2 November 2026
- Manuscripts of the Old Latin New Testament — Vetus Latina Institute, University of Birmingham, 2023, accessed 2 November 2026
- Concordance of Sigla for Old Latin New Testament Manuscripts — Vetus Latina Institute, University of Birmingham, 2023, accessed 2 November 2026
- The Latin New Testament: the project pages and registers — Vetus Latina Institute, University of Birmingham, accessed 2 November 2026
- The Latin New Testament: A Guide to its History, Texts, and Manuscripts — H. A. G. Houghton, Oxford University Press, 2023, open-access PDF, accessed 2 November 2026
- The Fourth Century and the Beginning of the Vulgate — H. A. G. Houghton, chapter 4 of The Latin New Testament, Oxford University Press, 2023, landing page, accessed 2 November 2026
- A Plain Introduction to the Criticism of the New Testament, 4th edition, volume ii — F. H. A. Scrivener, edited by E. Miller, London: George Bell and Sons, 1894, Project Gutenberg, accessed 2 November 2026
- Portions of the Gospels according to St. Mark and St. Matthew from the Bobbio ms. (k) now numbered G. vii. 15 in the National Library at Turin, together with other fragments of the Gospels from six mss. — John Wordsworth, W. Sanday and H. J. White (eds.), Old Latin Biblical Texts no. 2, Oxford: Clarendon Press, 1886, accessed 2 November 2026
- Manuscripts of the Greek Bible: An Introduction to Greek Palaeography — Bruce M. Metzger, New York: Oxford University Press, 1992, Internet Archive, accessed 2 November 2026
- New sets of images for Old Latin manuscripts — H. A. G. Houghton, Vetus Latina Institute, updated 2025, accessed 2 November 2026
- Early Readers, Scholars and Editors of the New Testament — H. A. G. Houghton, Gorgias Press, 2014, open-access PDF, accessed 2 November 2026
- The Latin Text of John in the Saint Gall Bilingual Gospels (Codex Sangallensis 48) — H. A. G. Houghton, in At One Remove: The Text of the New Testament in Early Translations and Quotations, Gorgias Press, 2021, accessed 2 November 2026
- Codici miniati e corali dei monasteri della regione piemontese, chapter on the Bobbio manuscripts — MuseoTorino and Regione Piemonte, scholarly volume, PDF chapter, accessed 2 November 2026
- Manoscritti appartenuti alla Biblioteca del Monastero di Bobbio — Val Trebbia project site, accessed 2 November 2026
- Il patrimonio culturale del monastero di S. Colombano in eta medioevale — Val Trebbia project site, 2026, accessed 2 November 2026
- From Bobbio to South Bend via Milan: The Modern Fate of an Early Medieval Library — University of Notre Dame, Medieval Studies Research Blog, 2020, accessed 2 November 2026
- Beihefte zur Septuaginta, the series in which the Vetus Latina Institute’s Septuagint volumes appear — De Gruyter Brill, accessed 2 November 2026
- The Book of Deer — masoretes.online, related essay, accessed 2 November 2026