How Marginal Comments Can Corrupt a Text in Transmission

A leaf from P.Mich.inv. 6238 (P46) showing the

end of Romans and the beginning and title of Hebrews

P.Mich.inv. 6238 / Recto | Advanced Papyrological Information System (APIS UM) | University of Michigan Library Digital Collections (umich.edu)

A number of years ago Michael W. Holmes wrote a book chapter in which he argued that several variations in the text of Romans in P46 were evidence of marginal comments entering into the body of text. This occurred, Holmes argued, when a copyist confused a marginal comment for a correction to be entered into the body of text (Holmes pg. 202-205).

"What now appear only as variant readings in the text originally were, I suggest, comments in the margins about the text—the earliest “commentary” (in the sense of activity, rather than genre) on the text of Romans." (Holmes pg. 205)

Though Holmes gave no examples from antiquity, there are two instances in which an ancient author mentions this phenomenon occurring during the copying of a text. 

The second century physician Galen in his commentary on Hippocrates "Epidemics" gives two explanations for a corruption in the text of Hippocrates. The first explanation Galen gives is that Hippocrates gave a parenthetical statement in order to point out the difference between two types of patients currently being discussed in the work. Galen makes the following comment for the second.

“The other explanation is something you often observe in many books: we sometimes write two expressions for the same thing, putting it in two different ways, and place one of them in the text body and the other in the margin of the book so as to pick the better of the two at leisure when we want to edit the book. It seems that Hippocrates did the same in this place. The first editor of this book then copied both expressions together into the book’s text. We then did not look at and consider it again, pause and fix this error. Many people passed on the copy of the book, and it remained uncorrected.” (In Hipp. epid. comm. I, 1.36) (Vagelpohl pg. 167)

Here Galen describes a practice in which the author of a work gives an explanatory note in the margin of their "autograph." When the text is initially edited and copied, the scribe would think that the marginal note was meant to be entered into the text as a correction. The result is that the book was then widely copied and circulated with the error uncorrected.

Another instance of this phenomenon is mentioned by Jerome. In about 403 CE Jerome received a letter from two men asking for an explanation as to why there were so many differences between the Septuagint LXX these men were using, and the Latin Psalter Jerome had produced many years before in about 383 CE. Along with this letter, these men sent a long list of passages where there were differences in the text between the LXX and the Latin Psalter. For one of the variations in the text, Jerome gave the following explanation.

"I wonder why some rash fellow has thought that the note: “the correct form is not καταπαύσωμεν, as some think, but κατακαύσωμεν, that is, incendamus,” which was placed by me for the guidance of the reader into the margin, should be put into the body of the text." (Ep. 106.46)

Apparently, in the first edition of Jerome's Psalter he had placed an explanatory comment in the margin. A later copyist confused this entry as a correction to be entered into the body of text. Thus, all subsequent copies of his Psalter had this variation transmitted in the text.

______________________________

Corpus Medicorum Graecorum 5.10.1. English translation taken from, Galen, Commentary on Hippocrates’ Epidemics Book I: Parts I-III (Translated by Uwe Vagelpohl. Corpus Medicrum Graecorum Supplementum Orientale V1. Berlin: Walter De Gruyter, 2014).

Holmes, Michael W. “The Text of P46: Evidence of the Earliest ‘Commentary’ on Romans?” In New Testament Manuscripts: Their Text and Their World (ed. by Tobias Nicklas; Leiden: Brill, 2006) 189-206.

Translation of Jerome's Epistle 106 by Michael Metlen, The Journal of English and Germanic Philology, Vol. 36, No. 4 (Oct., 1937), pp. 515-542.

The Church Fathers and The Useful life of Ancient Books

I recently finished reading through George Houston's, Inside Roman Libraries: Book Collections and Their Management in Antiquity, and was struck by the wonderful insights that Houston discovered in his detailed study of Roman libraries. In many ways these insights illuminate the world of Christian book collections and manuscripts as much as they do Roman libraries.
This is especially true of Houston's observations on the useful life of papyrus book rolls. In the conclusion to Inside Roman Libraries, Houston wrote;

The identification of such collections, and of the manuscripts within them, provides new evidence on an old question: how long did a papyrus roll last? The evidence from our collections indicates that a usable lifetime of about 100 to 125 years was common and can reasonably be considered the norm; a small but significant number of manuscripts were still usable some 300 years after they were first created; and on rare occasions a manuscript might last, it seems, for half a millennium. (p. 257)

Houston's observations are helpful in understanding some of the references made by various Church fathers to ancient manuscripts consulted during their studies.

Irenaeus's Ancient Copies
Writing at the end of the second century, Irenaeus of Lugdunum, in his Against Heresies, discussed at length an important variant that was present in some manuscript copies of Revelation in circulation in his day. Irenaeus wrote that the number of the beast, 666, was "found in all the most approved and ancient copies" yet some manuscripts contained the number 616 (Hear. 5.30.1). If the average life of a papyrus book was 100 to 125 years, then it is possible that Irenaeus was referring to manuscripts that were 75 or more years old. It may be surmised then, that the manuscript copies that contained the 616 variant reading were not as old in Irenaeus's estimation. The Greek phrase used by Irenaeus here (preserved by Eusebius in Hist. eccl. 5.8) was ἐν πᾶσι τοῖς σπουδαίοις καὶ ἀρχαἰοις ἀντιγράφοις. The Greek word ἀρχαἰοις (ancient) used to describe the age of the manuscripts of Revelation that read 666 was the same word used by Irenaeus to describe Papias as "a man of the early period," or in the Greek, "ἀρχαἰος ἀνήρ" (preserved by Eusebius in Hist. eccl. 3.39). According to Eusebius, Papias was at the peak of his career near the beginning of Emperor Trajan's reign at around 100 CE (Hist. eccl. 3.39). Thus, in Irenaeus's mind, copies of Revelation that dated to the beginning of the second century would have been "ancient." And these copies would have been well within the 100 to 125 year useful life of papyrus manuscripts observed by Houston.

Tertullian and the Autographs
Writing in the early third century, Tertullian in Carthage North Africa wrote;

Come now, you who would indulge a better curiosity, if you would apply it to the business of your salvation, run over to the apostolic churches, in which the very thrones of the apostles are still preeminent in their places, in which their own authentic writings are read, uttering the voice and representing the face of each of them severally. (Praescr. 36; ANF 3:261; emphasis mine)

This has been a much disputed passage with discussions going at least as far back as Charles Hodge at Princeton University in the early 1800s.
The Latin word used by Tertullian here for "authentic writings" is "authenticae litterae." According to the Latin dictionary available on Perseus, this word means; "that comes from the author, authentic, original, genuine . . . the original writing, the original." It seems then that Tertullian was telling his readers that he thought the original apostolic documents were being read in the apostolic Churches in the early third century. It is difficult for me to accept that the original writings of, say, Paul were still existing at Rome or Corinth, but it may have been that very early copies of these writings were still being read in these Churches. If we consider the useful life of papyrus documents given by Houston, of 100 to 125 years with an upper limit of 300 years, then it is entirely possible that copies of Paul's letters dating to the later half of the first century were still being read in some Churches.

The Colophon of Pamphilus in Sinaiticus
In the fourth century manuscript of the Christian Bible, Codex Sinaiticus, there is preserved at the end of 2 Esdras and again at the end of Esther a colophon stating (from the end of 2 Esdras);

Collated against an extremely old copy corrected in the hand of the holy martyr Pamphilus, which copy at the end has a signature in his own hand, reading thus:
Copied from and corrected against the Hexapla of Origen
Antoninus collated
I, Pamphilus, corrected. (Parker, Codex Sinaiticus, 81)

According to Milne and Skeat, these corrections, and the corresponding colophon, date to the 6th or 7th century (Metzger, Text of the New Testament, 4th ed., 67). The Greek phrase in the colophon at the end of 2 Esdras reads "αντεβληθη προϲ παλαιω τατον λιαν αντιγραφον," the  Greek word παλαιω already means "old" but the adverb "λιαν" was added which means "very" or "extremely." Thus this manuscript must have been considered very old at the time and was valued for its textual pedigree. If we consider that Houston noted examples of papyrus book rolls still in use 300 and even 500 years after they were fabricated, then it is feasible that there was a 300 to 400 year-old copy of the Greek Bible corrected against the Hexapla by Pamphilus still present at the library of Caesarea.
Origen's Hexapla, and Pamphilus's copies were still present in Caesarea in the 5th century when Jerome wrote;

Pamphilus the presbyter, patron of Eusebius bishop of Caesarea, was so inflamed with love of sacred literature, that he transcribed the greater part of the works of Origen with his own hand and these are still preserved in the library at Caesarea. (Jerome, Vir ill. 75; NPNF 2:377)

The Second/Third Century New Testament Text
If Pamphilus's manuscripts were still available for reference in the 6th century, 300 years after they were copied, it may be representative of other major Christian centers. The Churches of Rome, Corinth, Antioch, Smyrna, Ephesus, Philippi, and others, may have had very old manuscripts of Christian writings available for reference and copying for several hundred years. Perhaps Irenaeus had sent a letter to Ephesus or Smyrna asking them to check their ancient copies of Revelation.
Manuscripts of the New Testament such as P66 and P75 give evidence of a high-quality textual pedigree and careful copying and, due to their extensive repairs, show evidence of a long useful life. Because of the apparent longevity of ancient books it is possible that the extant manuscripts of the New Testament preserved from the late second and early third century may have been copied from exemplars 100 years older and actually represent the state of the text at around ca. 100 CE and not just at ca 200 CE.

(For more discussion on explicit references to manuscripts in Church fathers, read here.)

Bibliography
Eusebius of Caesarea. The Ecclesiastical History, Books I-V. Translated by Kirsopp Lake. Vol 1. Loeb Classical Library. New York: G. P. Putnam's Sons, 1926.

Gamble, Harry Y. Books and Readers in the Early Church: A History of Early Christian Texts. New Haven: Yale University Press, 1995.

Holmes, Michael W., ed. The Apostolic Fathers: Greek texts and English translations. Grand Rapids: Baker Books, 1999.

Houston, George W. Inside Roman Libraries: Book Collections and Their Management in Antiquity. Chapel Hill: UNC Press, 2014.

Metzger, Bruce Manning, and Bart D. Ehrman. The Text of the New Testament: Its Transmission, Corruption, and Restoration. 4th ed. New York: Oxford University Press, 2005.

Parker, D. C. Codex Sinaiticus: The Story of the World's Oldest Bible. London: British Library, 2010.