The Bulletin of the Bezan Club

Posted on May 27, 2025 by Brent Nongbri

[[Update 5 June 2025: I managed to find two additional issues (8 and 9) and a copy of issue 6 that has all of its pages. So, now we lack only issues 10-12.]]

[[Update 26 June 2025: Thanks to Jean Putmans for obtaining a copy of issue 12! So, now only issues 10 and 11 remain to be found.]]

[[Update July 1 2025: Thanks to C.R. van Tilburg and Jean Putnams for sending scans of issues 10 and 11 and to John Muccigrosso for improving the quality of some of the files. The set is complete!]]

The Bulletin of the Bezan Club was a privately circulated journal that, to the best of my knowledge, consisted of 12 issues printed between 1925 and 1937. It was dedicated to studies about the problems of the so-called “Western Text” of the New Testament, but it contains many interesting articles on topics from palaeography to the complexities of the Diatessaron and other gospel harmonies. The issues that I have read offer a fascinating snapshot of one corner of New Testament textual criticism in the years between the two world wars.

The membership of the club, which never exceeded about two dozen names, includes many of the philological luminaries of the early twentieth century. To pick just a few:

  • J. Rendel Harris (1852-1941)
  • Kirsopp Lake (1872-1946)
  • Hans Lietzmann (1875-1942)
  • Elias Avery Lowe (1879-1969)
  • Daniel Plooij (1877-1935)
  • Theodor Zahn (1838-1933)

Rendel Harris introduced the club in this way in the first issue in 1925:

The Bezan Club has been formed by a group of scholars European and American, with the object of applying a combined effort to the elucidation of the mystery of what is called the Western Text of the New Testament; and since this Text has one of its most striking representatives in the Codex Bezae, the name of that manuscript has been made into a short appellation for the Club and its objects. This does not mean that the Codex Bezae and the Western Text are to be treated as nearly equivalent, but it does mean that no solution of the problem of the Western Text is possible, which will not at the same time elucidate the origin of the MS., which is its principal, and often its only Greek representative. It has long been recognised that the geographical term Western was a misnomer; that the evidence for such a Text was as much Eastern as Western, and that it would conduce to clarity of ideas, if the term were abandoned. But as this cannot be done without dislocating a mass of references and allusions, it may be sufficient to reiterate a well-known caution to beginners, and say “Western Text, or whatever that may mean.”

Many years ago, I needed to consult some of these discussions, but I could not locate any copies of the journal. I eventually wrote to Eldon Epp, asking if he knew where I might find copies. Some weeks later, much to my surprise, I received a package in the mail (in Sydney, Australia) containing neatly stapled xerox copies of issues 1-7 as well as a cumulative index (I came to learn that this kind of thoughtfulness and generosity is characteristic of Eldon). Eldon had collected and copied these over the years (some of these appear to have been E.A. Lowe’s personal copies).

These xerox copies travelled back to the US, spent some time in one of many boxes stacked in a garage, and eventually came to Norway with me. I recently needed to consult them again. As they still seem difficult to find online, I thought I should scan the hard copies in my possession. And so, with Eldon’s blessing, I have digitized these issues and posted them below. I am not aware of any copyright concerns, but if there are objections, please let me know.

I believe that there were 12 issues in total (1925-1937). If anyone has copies of the other issues, I would love to have them to post here to complete the set. [Update: As noted above, the set is now complete.]

Bulletin of the Bezan Club 1 (1925)

Bulletin of the Bezan Club 2 (1926)

Bulletin of the Bezan Club 3 (1926)

Bulletin of the Bezan Club 4 (1927)

Bulletin of the Bezan Club 5 (1928)

Bulletin of the Bezan Club 6 (1929)

Bulletin of the Bezan Club 7 (1929)

Bulletin of the Bezan Club 8 (1930)

Bulletin of the Bezan Club 9 (1931)

Bulletin of the Bezan Club 10 (1932)

Bulletin of the Bezan Club 11 (1936)

Bulletin of the Bezan Club 12 (1937)

Bulletin of the Bezan Club Index

Posted in Codex Bezae, J. Rendel Harris, New Testament, Palaeography, Textual criticism | Tagged , , , , , , | 17 Comments

7Q5 and Appeals to Authority, Part 1: Orsolina Montevecchi

Posted on May 23, 2025 by Brent Nongbri

I am fascinated by the Wikipedia entry for 7Q5, which seems to continuously bounce back and forth between being useful and informative to being goofy and borderline incoherent. 7Q5 is a tiny fragment of papyrus found in Cave 7Q at Qumran that contains an unidentified text in Greek. As I noted in an earlier post, it was a mistaken reading of a printed edition (not the manuscript itself) that led the Spanish scholar José O’Callaghan to conclude that fragment contained a portion of the Gospel According to Mark in the 1970s. This created a sensation because it is generally accepted that the manuscripts in the caves at Qumran predate the sack of Qumran in the late 60s CE. O’Callaghan’s error was pointed out immediately, but instead of admitting the slip, he doubled down, and his mistaken identification of 7Q5 has had a persistent afterlife.

The arguments against O’Callaghan’s proposal are compelling.1 Most importantly, O’Callaghan’s reconstruction both depended upon impossible or highly suspect readings of several letters and necessitated that one out of the mere nine undisputed letters on the papyrus must be a scribal error.

After a flurry of articles in the 1970s demonstrating the problems with O’Callaghan’s thesis, it largely (and justifiably) fell out of view, only to be revived in the 1990s by Carsten Peter Thiede (1952-2004).

In my earlier post on 7Q5, I pointed out that what was once a reasonably informative article on Wikipedia had become a confused collection of misinformation. The article was then cleaned up but has again become a jumble of decent scholarship and nonsense. One of the recent changes to the entry is a series of appeals to authority, especially that of Orsolina Montevecchi (1911-2009), an Italian papyrologist who endorsed O’Callaghan’s identification of 7Q5 as a fragment of Mark. However, these appeals to authority tend to be by way of hearsay. For instance, many appeals to authority come by way of Thiede. In a book co-authored with the journalist Matthew D’Ancona, Thiede presented Montevecchi’s view as decisive:

“In 1994 the last word on this particular identification seemed to have been uttered by one of the great papyrologists of our time, Orsolina Montevecchi, Honorary President of the International Papyrologists’ Association. She summarized the results of her analysis in a single, unequivocal sentence: ‘I do not think that there can be any doubt about the identification of 7Q5.'”2

This confident assertion made me wonder: What was Montevecchi’s actual reasoning? As far as I have been able to tell, in her massive bibliography, Montevecchi mentions 7Q5 just twice (I would be happy to be corrected if anyone knows of additional references) [[Update 24 May 2025: The count is up to three; see in the comments below]].

The first reference appears in her introductory textbook, La papirologia (1973). The preface to this book is dated May 1972. That is to say, a little more than one month after the first appearance of José O’Callaghan’s first publication on 7Q5 in the first issue of the 1972 volume of Biblica (which carried a print date in March of 1972). That is to say, this statement was composed before any of the rebuttals to O’Callaghan had had been published. At the conclusion of a list of the New Testament papyri that had been published to date, we find a single sentence:

“Furthermore, in the papyrus fragments 7Q5, 7Q6 (frag.1), and 7Q8, palaeographically datable to 50 BCE – 50 CE, have been recognized respectively, Mark 6:52-53 and 4:28 and James 1:23-24 (O’Callagahan, J. in Biblica 53 [1972], 91-100) [“Inoltre nei frammenti papiracei 7Q5, 7Q6, 1, 7Q8, paleograficamente datibili c. 50a-50p, sono stati riconosciuti rispettivamente Mc. 6, 52-53 e 4, 28 Iac. 1, 23-24 (O’Callaghan, J. in «Biblica» 53 1972, pp. 91-100).”]

As far as I know, that is all Montevecchi ever said about this papyrus in an academic publication. The other reference that comes up frequently is an interview conducted for the Catholic periodical 30 Giorni in 1994.

This article, which is the source of Thiede’s quotation endorsing the identification 7Q5, actually has an extended interview with Montevecchi, in which she outlines her reasoning for accepting the identification. It’s fascinating to see what she actually says.

Interviewer: Many excellent palaeographers do not agree with this identification.

Montevecchi: There are some difficulties because three words are missing from the text (epi tēn gēn = “to the land”) compared to the passage in Mark. We read in the gospel text handed down to us: “Having crossed the lake to the land.” But that “to the land” is superfluous. When one crosses a lake, one obviously goes to the other side. In fact, even though these palaeographers seem to ignore it, it’s quite common in the oldest texts of the Bible on papyrus to find the omission of some element not necessary for the understanding of the text. It is as if such words were added later, by way of explanation. Another source of opposition is the fact that there is an exchange of consonants, a tau (t) instead of a delta (d). But this is also a frequent error. Because texts were dictated, the writer transmitted errors of pronunciation. These are the only two objections, which are taken as an excuse to invalidate the identification of this papyrus, since they are the only variations from the text as it was handed down.3

Regardless of how good a papyrologist Montevecchi may have been or how important she may have been to the field, this is simply a very bad argument. Montevecchi neglects even to mention the most compelling counterargument to the identification–the fact that several of O’Callaghan’s readings of letters are wrong, doubtful, or impossible to verify. It is that fact, combined with the need to consider one of only nine undisputed letters as an error, combined with the need to posit the existence of the existence of an otherwise unattested textual variant (διαπερασαντες ηλθον εις γεννησαρετ in Mark 6:53), which makes the identification extremely doubtful if not impossible. Montevecchi was either not fully informed about the scholarly debate around the fragment, or she simply decided to try to defend O’Callaghan’s position by mischaracterizing the opposing arguments and ignoring the most damaging evidence against O’Callaghan’s identification. In the matter of 7Q5, the appeal to Montevecchi’s authority actually adds nothing of substance to the discussion.

  1. The most informed and thorough rebuttals to O’Callaghan (and later Thiede) include the following (in chronological order):
    Maurice Baillet, “Les manuscrits de la Grotte 7 de Qumrân et le Nouveau Testament.” Biblica 53.4 (1972) 508-516.
    C.H. Roberts, “On Some Presumed Papyrus Fragments of the New Testament from Qumran,” Journal of Theological Studies, n.s. 23 (1972) 446-447.
    Gordon D. Fee, “Some Dissenting Notes on 7Q5 = Mark 6:52-53,” Journal of Biblical Literature 92 (1973) 109-112.
    Stuart R. Pickering and Rosalie R.E. Cook, Has a Fragment of the Gospel of Mark Been Found at Qumran? (Sydney: Macquarie University Ancient History Documentary Research Centre, 1989).
    Robert H. Gundry, “No nu in Line 2 of 7Q5: A Final Disidentification of 7Q5 with Mark 6:52-53,” Journal of Biblical Literature 118 (1999) 698-707. ↩︎
  2. Carsten Peter Thiede and Matthew D’Ancona, The Jesus Papyrus (Weidenfeld and Nicolson, 1996), 56. ↩︎
  3. “È stata fatta qualche difficoltà perché nel testo mancano tre parole (epi ten ghen = verso terra) rispetto al brano dì Marco. «Avendo attraversato il lago verso terra”, si legge nel testo del Vangelo tramandatoci. Ma quel «verso terra» è superfluo: attraversando un lago, si va ovviamente dall’altra parte. In realtà, anche se questi paleografi sembrano ignorarlo, è abbastanza frequente nei testi più antichi della Bibbia su papiro, trovare l’omissione di qualche elemento non necessario per la comprensione del testo. È come se quelle parole fossero state aggiunte dopo, a mo’ di spiegazione. Un’altra fonte di opposizione è il fatto che c’è uno scambio di consonanti: una tau (t) invece di una delta (d). Ma anche questo è un errore frequente. Poiché i testi venivano dettati, chi scriveva riportava gli errori di pronuncia. Ci sono molti altri casi, nei papiri biblici di scambio di tau con delta. Queste sono le due sole obiezioni che sono prese a pretesto per invalidare l’identificazione di questo papiro, poiché sono le sole varianti rispetto al testo tramandato.” ↩︎
Posted in 7Q5, Dead Sea Scrolls, Orsolina Montevecchi | Tagged , , , , , , | 16 Comments

Pumice, Parchment, and Papyrus

Posted on April 28, 2025 by Brent Nongbri

In the latest issue of the Bulletin of the American Society of Papyrologists (2024, just out–table of contents here), I have an article on the use of pumice on papyrus (AAM here). Here is the abstract:

Papyrological handbooks published in the second half of the twentieth century suggest that pumice was used in antiquity to smooth the surface of papyrus in preparation for writing. Experimentation with modern papyrus and pumice calls this claim into question. The ancient literary references to pumice and papyrus, when they are properly understood, demonstrate that when pumice was used on papyrus rolls in antiquity, it was not used for the purpose of smoothing the writing surface. Pumice was instead used to smooth the frontes, the top and the bottom of the closed papyrus roll. This fact that was quite well established at the end of the nineteenth century, but it seems to have been largely forgotten over the course of the second half of the twentieth century.

People used pumice (Greek κίσηρις, Latin pumex) for many purposes in antiquity, including the preparation of parchment writing surfaces. Several ancient and medieval discussions of the production of parchment mention the use of pumice at different stages in the process. I’ve previously noted one of these, a few leaves preserved in a papyrus codex of the sixth or seventh century that contain instructions in Coptic for preparing parchment.

What interests me in this article are references to the use of pumice not on parchment, but on papyrus. Several Latin poets refer to the use of pumice on papyrus (all the references are collected in the article). Modern handbooks on papyrology sometimes claim that pumice was used to prepare the surface of papyrus for writing (thus Italo Gallo, Greek and Latin Papyrology, p. 7: “[the papyrus sheet] was then allowed to dry, and was finally polished with a pumice stone, an ebony tool, or a shell”). But doing a bit of experimentation with modern papyrus shows that if you rub even very fine pumice on papyrus, it tears the fibers, resulting in worse conditions for writing (smooth objects like bone or shell flatten the fibers and do create a better writing surface).

Homemade modern papyrus before rubbing with pumice (left) and after rubbing with pumice (right); image source: Brent Nongbri

In fact, the ancient references to the use of pumice on papyrus never say it was used to smooth the writing surface. Rather, it was used on the frontes, the top and bottom of the closed scroll. The pumice would grind off any stray fibers that protruded from the ends of the rolled up cylinder. This seems to have been one of the last stages of the manufacture of a bookroll, one that took place after the papyrus was already inscribed, in some cases even at the shops where books were sold.

The form of pumice used for this task seems to have been a hemispherical block that could be held in the palm of the hand (such a stone was found as part of what seems to be a writing kit in the so-called Tomb of the Scribe in the Vatican). This is very similar to hemispherical pumice stones found at Pompeii, often in bronze holders.

Pumice stone in bronze casing from Pompeii, inv. 7150; image source: Paul Roberts, Life and Death in Pompeii and Herculaneum (The British Museum Press, 2013), p. 132, fig. 142

When the context of these Pompeian hemispherical pumice stones is known, it is most frequently as part of cosmetics kits. This is interesting in light of the personification and sexualization of bookrolls that we sometimes find in Latin poets. The hemisphere of pumice offers us a case of a single tool that could be used on both the human body (for removal of hair and smoothing of skin) and the book (for tidying the frontes).

I enjoyed researching and writing this piece. It led me down several fascinating rabbit holes that I hope to talk about here in the near future.

Posted in Archaeological context, Book Trade in Antiquity, Voluminology | Tagged , , , , , | 1 Comment

The So-called Bust of Josephus

Posted on April 25, 2025 by Brent Nongbri

I recently had occasion to spend some time revisiting the works of the Jewish historian Josephus, which was a real pleasure. When working on Josephus, I occasionally come across the “bust of Josephus” that so often serves as an illustration of the historian, as it does on the Norwegian Wikipedia page for Josephus (albeit with an ambivalent caption):

image source: no.wikipedia.org

Eusebius mentions that there was a statue of Josephus in Rome (Eccl. Hist. 3.9.2), but there is no indication that this bust has anything to do with that statue. So, how and when did this bust become “Josephus”?

It turns out that there is an excellent chapter in a recent book that tells the fascinating story of this bust:

René Bloch, “Testa incognita: The History of the Pseudo-Josephus Bust in Copenhagen,” in R. Brody et al. (eds.), “A Vision of the Days”: Studies in Early Jewish History and Historiography in Honor of Daniel R. Schwartz (Leiden: Brill, 2023), 419-442.

The bust, which is now in the Ny Carlsberg Glyptotek in Copenhagen, was first identified as “Josephus” by Robert Eisler (1882-1949) in an article written in 1930. This point had already been noted by Steven Fine in 2013, but Bloch takes a deep dive into both the history of the bust itself and into Eisler’s interesting background. The identification was made on the flimsiest of grounds (Eisler believed the bust had a “Jewish-looking face”), and it is remarkable that the claim was ever given any credence.

But Bloch’s chapter is an engaging read full of wonderful detective work. I highly recommend it.

Some of it has been incorporated into the German Wikipedia article connected to this bust.

Posted in Josephus, Judaism, Sculpture | 7 Comments

More on the mikveh at Ostia and Other Jewish Materials

Posted on April 24, 2025 by Brent Nongbri

In March I noted the announcement that a mikveh, a Jewish ritual bath, had reportedly been found in Ostia, the port city of ancient Rome. I had missed an article from the Times of Israel that adds some details to the original press release. The article records an interview with Alessandro D’Alessio, the director of the Archaeological Park of Ostia Antica, who speculates a bit about the building that includes the pool structure:

“Our first hypothesis is that the structure was a luxurious private residence, but it could have also been a public building. …We uncovered at least six rooms. Two of them are equipped with an oven to cook. We also found a latrine and another room that was probably a kitchen. These rooms overlooked a courtyard facing south. A staircase suggests that there were at least two floors.”

It looks like some of the area has now been backfilled, so it’s difficult to see the details of the surrounding architecture.

Excavation of the pool at Ostia; image Brent Nongbri 2025

After noting the discovery of the lamp with Jewish symbols, D’Alessio added, “We found an additional two lamps, one very similar to the artifact from the bottom of the mikveh with the depiction of a menorah, the other one engraved with a Christogram [a combination of letters abbreviating the name Jesus Christ]. …We know that during that period, the Jewish and Christian communities in Rome were still very close; therefore, finding both symbols in the same environment is not surprising.”

It will be interesting to learn more precisely where the lamp with the Christian symbol was found in relation to the lamps with the Jewish symbols.

On a related point, back in August of 2024, I noted that the site museum at Ostia had reopened after a long closure. I also expressed some disappointment that none of the many Jewish artifacts from the site were on display. This has now been partially remedied with a small case containing the recently excavated menorah lamp along with two other decorated lamps. I say “partially” because the display contains almost no didactic material, and what little there is seems wrong. The display label does not say where the lamps were found and gives a date of “II-V secolo” for the whole group. Yet, we do know where the other two lamps (inv. 12495 and 12498) were found. They come from the synagogue excavations of the 1960s. They were found above a floor that post-dates the mid-fourth century and have been assigned by Letizia Ceccarelli to the latter part of the fourth century or the fifth century. I’m not aware of any lamps that exactly match the decoration of the lamp from the mikveh, but similarly shaped African lamps with a central decoration flanked by stylized palm leaves are also assigned to the fourth and fifth centuries.

Display of Jewish artifacts at the museum at Ostia Antica; image: Brent Nongbri 2025

The other item in the display case is described as an inscription of the first century CE. It may in fact be that old, but not necessarily. This is another relatively recently discovered piece of evidence for Jews at Ostia. The inscription was discovered by chance during construction work in 2006 in the Pianabella area southeast of Ostia. It identifies several people as Judaeans (Iudaei), including a certain Quintus Fabius Longor[um]. A Quintus Fabius Longus is known from the Fasti Ostiensis as a duovir and prefect in the 30s CE. The form of the name in the inscription, Longor[um], may indicate that we are dealing here with a freed person of the Fabii Longi, either in the first century or later.

The museum is also now displaying one of the decorated corbels from the synagogue (the other one has for many years been on a podium just outside the museum).

Display of Jewish and Christian artifacts in the museum at Ostia Antica; image: Brent Nongbri 2025

It’s helpful to be able to examine this corbel up close. Mary Jane Cuyler informs me that the reddish color on the menorah on the corbel is indicative of gilding, so these menorahs would have been a colorful addition to the torah shrine, which was itself decorated in different types of cut marble. Gold decorations also raise the question of reflection and lighting within synagogues, an issue discussed in a 2023 article by Karen Stern.

It’s nice to see this material on display.

Posted in Judaism, Ostia, Synagogues | Tagged , , | 3 Comments

Working with Manuscripts

Posted on April 4, 2025 by Brent Nongbri

It’s a nice moment when you receive the first copies of a book you’ve written. Yesterday, I had the pleasure of opening the box of authors’ copies of Working with Manuscripts, written together with my colleague Liv Ingeborg Lied.

This book emerged from our own experiences in trying to learn the ropes of working with manuscripts over the last couple decades. It’s a guide that addresses both the practicalities and ethics of studying manuscripts. We wrote it with graduate students in mind, but we hope it is widely useful for any scholar interested in including manuscripts in their research.

Thanks to the critical readers who offered feedback to some or all of the book, to the Research Council of Norway for supporting us (through the Lying Pen of Scribes project and The Early History of the Codex project), and to the editorial and production team at Yale University Press from making a nice looking and affordable volume.

Posted in Working with Manuscripts | 3 Comments

A Newly Discovered mikveh at Ostia

Posted on March 10, 2025 by Brent Nongbri

At a press conference this afternoon in the archaeological park at Ostia Antica, it was announced that a mikveh, a Jewish ritual bath, has been discovered in the center of the ancient city.

Ostia is well known to students of ancient Jewish life because of the late antique synagogue that was discovered there in the early 1960s. Just over a year ago, I posted about the publication of a statuette of Venus that was discovered in the synagogue.

The new discovery is in an entirely different sector of the city:

Plan of Ostia showing location of the synagogue near the ancient coastline and the mikveh in the center of the city; image adapted from a plan at Parco archeologico di Ostia Antica

While the synagogue was located close to the ancient coastline, the new discovery is inside the city walls in the central part of the town, not far from the large theater.

According to the press release, the mikveh was excavated in 2024. There are several photos from the excavation posted online:

View of the mikveh at Ostia from above; image source: Parco archeologico di Ostia Antica
View of the mikveh at Ostia during excavation; image source: Parco archeologico di Ostia Antica
View of the steps of the mikveh at Ostia ; image source: Parco archeologico di Ostia Antica

The following description from the press release gives the arguments for identifying the area as a mikveh:

“The peculiar characteristics of the room–such as the steps covering its entire width, the walls lined with hydraulic plaster, the presence of a well for capturing groundwater, the passage for communication with the adjacent room (possibly intended to house a pipe for adding water to groundwater), and again the discovery of the oil lamp with Jewish symbols at the bottom of the well–lead to an interpretation as a Jewish ritual bath (mikveh).”1

And in fact the Ostia Antica homepage also provides an image of the lamp with a menorah:

Lamp with menorah found at the bottom of the well in the mikveh at Ostia; image source: Parco archeologico di Ostia Antica

A couple preliminary observations: First, we seem not to have a date for the installation. The lamp and an accompanying glass vessel are both said to date from the 5th-6th century, so well into the “Christian” period in Ostia. The fill above this area also seems to have contained lamps decorated with the menorah, which are assigned to the 4th to 6th centuries. It will be good to learn what other evidence there is for assigning a date to this space. Second, there was a feature in the synagogue at Ostia that has often been described as a mikveh, but that identification has been challenged, and it must be said, the space in the synagogue looks nothing at all like the newly excavated area.

I’m sure there will be more to say as we learn more about this in the coming weeks. It’s an exciting discovery, and I very much look forward to the full publication of the area, so that we can better understand how this feature fits into the surrounding buildings and neighborhood.

  1. “Le peculiari caratteristiche dell’ambiente – quali i gradini estesi per la sua intera ampiezza, le pareti rivestite di intonaco idraulico, la presenza di un pozzo di captazione dell’acqua di falda, il condotto di comunicazione con l’ambiente adiacente (possibilmente destinato ad alloggiare una tubatura per l’aggiunta di acqua a quella di falda), e ancora il rinvenimento della lucerna con simboli ebraici sul fondo del pozzo – inducono a ipotizzarne una interpretazione come bagno rituale ebraico (mikveh).” ↩︎
Posted in Judaism, Ostia, Synagogues | Tagged , , , , , , , , , , | Leave a comment

The So-called Scriptorium at Bu Njem

Posted on March 8, 2025 by Brent Nongbri

It is common for historians of ancient Rome to state that writers did not use desks (As Theodor Birt put it, “In antiquity, people did not write on desks”).1 I have noted before on the blog that I am not sure this view is entirely accurate. In a well-known article, Bruce Metzger gathered some visual evidence for the use of desks in Late Antiquity, a relief from Ostia Antica, a relief from Portus, and a mosaic from Thabraca that all show writers at desks.

Another aspect of the discussion is the material evidence of so-called scriptoria in the Roman world. The most famous of these is a controversial room found in the ruins of Qumran. Another (the only other?) potential surviving example is a structure in the remains of a Roman military camp at Bu Njem in modern Libya.

The site was excavated between 1967 and 1980 by a team led by René Rebuffat (1930-2019). According to the excavators, the fort was established in 201 CE and used by Roman military garrisons until around 260 CE. It was occupied by squatters after that who left a few traces and then abandoned and buried by wind-swept sand. The fort thus survived to the present in remarkably good condition.

Among the discoveries at the site in 1971 was a space on the south side of the central facility contained benches along the walls and a raised rectangular block (un massif rectangulaire) in the center of the room, highlighted in green in the image:

Fort at Bu Njem, adapted from R. Rebuffat, “Bu Njem 1971,” Libya Antiqua 11-12 (1974-1975)

The excavators described this space as “a scriptorium” that existed in two phases. In the first phase that corresponds to the construction of the fort in 201, the room featured a flat rectangular table (109 cm long, 66 cm wide, about 60 cm high) and two short benches on either side of the table (the report describes the benches as 80 cm high, but this must be an error; the scale indicates they are about 40 cm high). In a second phase, additional benches were added to extend to the north and south walls on both sides of the room, and a triangular “lectern” (pupitre), about 10 cm high at its peak, was added to the top of the flat table. The authors provided a top plan and profile drawing of the area as they found it:

“Scriptorium” at Bu Njem, top plan and profile, adapted from R. Rebuffat, “Bu Njem 1971,” Libya Antiqua 11-12 (1974-1975)

The sloping triangular feature in the profile drawing would be the “lectern.” The dark grey feature above it represents a niche in the south wall that is also visible in dotted lines in the top plan above. The “lectern” was somewhat damaged when the excavators found it, as illustrated in a photo that accompanied the article:

The excavators describe the use of the room in the following way:

“The central block of the Bu Njem scriptorium is not a table but a double-sloped lectern (pupitre). It is likely that the flat table (built in 201) that preceded this development also served as a lectern, because it was much too low, and the slightest test shows the discomfort of the position to which it would have forced a writer. The position remained uncomfortable with the double-sloped lectern, and the extension of the benches shows that one could work without being in front of the block. These benches are not designed for squatting or kneeling: They were obviously set up so that a seated person can be comfortable, and their extension beyond the block also proves that they are not made to raise a kneeling or squatting person in front of the block. The only overall hypothesis is therefore that the lectern served to support documents being read by a reader and listened to by other people sitting on the benches. The reader could naturally write on his lap while reading, and the listeners could also write. The room then fully deserved the convenient name of ‘scriptorium’ that we have given it.”2

The excavators paint an interesting picture here, one in which the “desk” is not used for writing but for holding something in position for reading. The excavators seem to be describing the production of multiple copies of a text by dictation, but what sort of writing do they imagine is happening in this military setting? The written artifacts found at the site included more than 146 ostraca that date from near the end of the occupation of the site, ca. 250-260 CE, and the majority of which were found in the vicinity of the so-called scriptorium.3 But these are mostly daily reports, not really the kind of thing needed in multiple copies. There are inscriptions of a literary nature that have survived at the site, namely poems inscribed on stone and credited to two centurions.4 But again, drafts for inscriptions probably wouldn’t be needed in multiple copies. I’m not sure what kind of texts the excavators have in mind with this imagined scene of dictation.

So, this is a very interesting piece of evidence, but I’m unsure about the proposed use of this space. I think there is more to say, but the bibliography on this “scriptorium” seems very limited. I would be grateful for any tips on other relevant publications on this space. For a good recent overview of the study of Bu Njem, see Anna H. Walas, “New Perspectives on the Roman Military Base at Bu Njem,” Libyan Studies 53 (2022) 48-60.

  1. Theodor Birt, Die Buchrolle in der Kunst (Leipzig: Teubner, 1907), p. 209: “Im Altertum schrieb man nicht auf Pulten.” ↩︎
  2. “…le massif central du scriptorium de Bu Njem n’est pa un table, mai un pupitre à double pente. Il est probable que la table, plate qui a précédé cet aménagement et qui a été construite en 201, servait aussi de pupitre, car elle était beaucoup trop basse, et Ie moindre essai montre l’inconfort de la position à laquelle elle aurait contraint le scripteur. La position restait inconfortable avec le pupitre à double pente, et l’extension des banquettes montre qu’on pouvait travailler sans être en face du massif. Ces banquettes ne sont pas faite pour s’accroupir ou s’agenouiller: elles ont visiblement été réglées pour qu’un homme assis soit confortablement installé, et leur prolongation au-delà du massif est également bien la preuve qu’elles ne sont pa faite pour exhausser un homme agenouillé ou accroupi en face du massif. La seule hypothèse d’ensemble est donc que le pupitre servait à supporter des documents qu’un lecteur, lisait, et que d’autres personnage assis sur le banquettes écoutaient. Le lecteur pouvait naturellement écrire sur ses genoux en même temps qu’il lisait, et les auditeurs en tout cas écrire eux aussi. La salle méritait alors pleinement le nom commode de ‘scriptorium’ que nous lui avons donné d’abord.” ↩︎
  3. See Robert Marichal, Les ostraca de Bu Njem (Tripoli, 1992). Marchial published 146 ostraca, but he prepared others for publication, including alphabetic exercises, that have not yet been fully published. See the discussion here. ↩︎
  4. See J.N. Adams, “The Poets of Bu Njem,” Journal of Roman Studies 89 (1999) 109-134. ↩︎
Posted in desks, Ostraca, Scriptoria | Tagged , , , , , , | 4 Comments

A Mosaic from Thabraca with a Writer at a Desk

Posted on February 5, 2025 by Brent Nongbri

In earlier posts, I discussed two pieces of evidence for writing at desks in the late antique period, a relief from Ostia and a relief found at Portus. Another piece of evidence that Metzger mentioned in his well known chapter on desks was a mosaic floor panel from a church in Thabraca on the coast of Tunisia.1 It was published in 1906 by Paul Gauckler.2

The figure behind the desk has been identified as both a banker and a scribe. It is unfortunate that the inscription near the figure’s head is not better preserved (just a couple letters survive, probably the remains of [I]N PA[CE]). The figure holds a stylus and writes–in a way oriented toward the viewer of the mosaic rather than toward the writer–the letters MA. The woman below the desk is identified by the name Victoria.

These mosaic panels covered burials under the church floor. The panel with the writer and Victoria covered a lead coffin in which two skeletons were found. The panel is oddly oriented relative to the other panels and positioned directly in front of the apse, but off center:

Gauckler’s publication includes one photograph from the excavation, and fortunately, it happens to show this mosaic in context. The raised apse is visible on the left side of the image:

The mosaic is now in the Bardo Museum in Tunis, and the details are a bit easier to discern in a color image:

The date of this mosaic is ambiguous. Gauckler places the construction of the chapel in the age of Constantine and dates its abandonment to the end of the fifth century, which means that the production of the mosaics would likely fall into the second half of the fourth century or the first part of the fifth century. Yet, the evidence for this chronology is not clearly articulated, so those dates should probably be regarded as tentative.

  1. Bruce M. Metzger, Historical and Literary Studies: Pagan, Jewish, and Christian (Leiden: Brill, 1968), 123-137. ↩︎
  2. Paul Gauckler, “Mosaïques tombales d’une chapelle de martyrs à Thabraca,” Monuments et mémoires de la Fondation Eugène Piot 13 (1906) 175-228. ↩︎
Posted in desks | Tagged , , , , , | 1 Comment

Ehrman on Titles of the Gospels in Sinaiticus and Vaticanus: Fact Check

Posted on January 26, 2025 by Brent Nongbri

Over at his blog, Bart Ehrman has been posting some basic facts about different books of the New Testament. The last couple posts have been about the Gospel According to Mark, and yesterday’s post, which is publicly available, treats the question of the title of the Gospel According to Mark, and includes a short discussion of how the title shows up in our earliest complete manuscripts of Mark. Ehrman writes [[Update 28 January 2025: Ehrman has now adjusted the text of his post.]]:

“Our oldest two manuscripts (Sinaiticus and Vaticanus, for you fellow Bible nerds) come from toward the end of the fourth century (around 375 CE), and they have the shortest titles (“According to Mark”).  But in both cases, the titles were added by a later scribe (in a different hand).  We don’t really know how much later. So it’s impossible to know when the manuscripts began calling it this, except to say that the manuscripts that the authors of both these 4th century manuscripts used apparently didn’t have titles at all (since they lacked them until the later scribe added them). Interesting.”

But is this actually right? On papyrus rolls, titles typically appeared at the end of the text, and this practice carried over into codices. In the case of Codex Sinaiticus, the titles that appear at the ends of the works are widely regarded to be the product of the original copyists of the books. Here is what Milne and Skeat say in their detailed study of the codex:

“That those main titles are written by the same scribe as the immediately preceding text has never been doubted, and is so obvious from the most cursory inspection of the manuscript that it needs no justification here.”1

And in fact, the subscript title at the end of Mark in Codex Sinaiticus is not the short version but the longer version (ⲉⲩⲁⲅ’ⲅⲉⲗⲓⲟⲛ ⲕⲁⲧⲁ ⲙⲁⲣⲕⲟⲛ):

The end title of the Gospel According to Mark in Codex Sinaiticus; image source: codexsinaiticus.org

Some aspects of the script differ from that of the main text (the smaller epsilon and omicron raised off the base line, the mu with a curved belly), but other letters are identical with the script of the main text of Mark. There is no reason to doubt that these titles were part of the original production of the manuscript and that the differences in the script are simply decorative.

There are running titles (titles at the top of each page) in the gospels in Codex Sinaiticus that use the shorter version of the title, just ⲕⲁⲧⲁ ⲙⲁⲣⲕⲟⲛ, but these too are generally thought to be the work of the original copyists of the codex.

What about Codex Vaticanus? Here, we do indeed find the shorter version of the title at the end of Mark:

End title of the Gospel According to Mark in Codex Vaticanus; image source: DigiVatLib

But was this title a later addition or the work of the original producers of the codex? In this case, it’s a little harder to tell because almost all the letters in this codex have been re-inked by someone who was not as skilled as the original copyists at producing the biblical majuscule script. But again, the consensus view is that these end titles were the work of the copyists who produced the codex.

In a similar fashion to Codex Sinaiticus, Codex Vaticanus has running titles that use the shorter version of the title (ⲕⲁⲧⲁ ⲙⲁⲣⲕⲟⲛ), but these are also generally thought to have been added as a part of the production of the codex (and in fact, they constitute important evidence for the original script of the book because they have not been re-inked for the most part).

So, to summarize: The titles of the gospels in Codex Sinaiticus and Codex Vaticanus were most likely the work of the original producers of these books [codices] and attest to the use of both the longer version of the title (ⲉⲩⲁⲅⲅⲉⲗⲓⲟⲛ ⲕⲁⲧⲁ ⲙⲁⲣⲕⲟⲛ in Codex Sinaiticus) and the shorter version (ⲕⲁⲧⲁ ⲙⲁⲣⲕⲟⲛ in Codex Vaticanus) at the time when these books [codices] were produced.

  1. H.J.M. Milne and T.C. Skeat, Scribes and Correctors of the Codex Sinaiticus (London: The British Museum , 1938), 27 ↩︎
Posted in Biblical Majuscule, Codex Sinaiticus, Codex Vaticanus, Palaeography, Scripts | Tagged , , , , , , , , , , | 17 Comments