The Society of Biblical Literature was founded in 1880 to foster biblical scholarship.
Search SBL

Welcome to The SBL


SBL Events

For a list of events related to biblical scholarship, see our online calendar.

View the Calendar >>

SBL Forum Archive
<< Return to SBL Forum Archive "Archaic Mark" (MS 2427) and the Finding of a Manuscript Fake

Scholarly articles are supposed to be tidy affairs, with clearly defined claims and logically developed arguments, but their tidy organization often obscures the chaos of their development. In February 2006, I discovered that a certain remarkable and apparently medieval manuscript of the Gospel of Mark is merely a hundred-year-old copy of a critical edition first printed in 1860. I am presenting my case in detail at the 2006 annual meeting of the Society of Biblical Literature in the New Testament Textual Criticism section, but, in this SBL Forum piece, I would like to give a glimpse into the less than linear process that led me to this textual discovery.

"Archaic Mark" is the moniker given by Ernest Cadman Colwell to an unprovenanced, illuminated manuscript of the Gospel of Mark that has fascinated textual critics ever since it was acquired by the University of Chicago nearly seventy years ago from the estate of an Athenian collector. Also known by its Gregory number of 2427, this manuscript upon first impression appears to be an ordinary, medieval parchment codex of Mark. Appearances deceive, however, because 2427's text of Mark is not of the usual Byzantine text-type that so dominates the medieval transmission of the Greek text, belonging instead to a much older and interesting Alexandrian text-type. In fact, 2427's text is closer to the text of the famous fourth century Codex Vaticanus (B) than to that of any other manuscript.

Doubts about the authenticity of 2427 surfaced early and lingered. For example, Robert P. Casey voiced his suspicion in 1947 that the text could have been derived from a nineteenth century critical edition of the Greek New Testament, and, in the 1970s, Colwell himself endeavored to find such an exemplar for 2427 but to no avail. In 1989, the case in favor of authenticity turned for the worse when it was discovered that 2427's illuminations have been found to contain significant quantities of a modern pigment, Prussian Blue, first synthesized around 1704. Even so, that finding did not settle the question of its curious text because it is conceivable that the illuminations could have been retouched in a naïve attempt at restoration. Nevertheless, the lack of publication of 2427's text has stymied progress on this question. Apart from selected readings in the 1993 edition of the Nestle-Aland critical text (NA27), it was not until this past year that scholars worldwide had convenient access to its text, when Margaret M. Mitchell and Patricia A. Duncan of the University of Chicago published their collation of the text in Novum Testamentum and put high-resolution images of the manuscript online, in the hopes of stimulating research into Archaic Mark.[1] As it turns out, their hopes have not been in vain.

I had been aware that there was a controversy over 2427 because, every once in a while, a discussion about it would break out on TC-list, a textual criticism mailing list run by Jimmy Adair. I remember wanting to use 2427 in my study of the origins of the Caesarean text in Mark 6:45-8:26 (the Bethsaida section) that I did for the New Testament Textual Criticism section at SBL 2004, but I was hampered by the very selective list of readings in NA27. I needed a full collation. Then, I learned last summer (2005) that the images of Archaic Mark were put online, and I started collating the text. My interest in 2427's authenticity was piqued when I found what looked like a line omission in Mark 8:11. According to Paul Maas, Textual Criticism, line omissions in a manuscript are highly diagnostic of the manuscript's exemplar:

(a) If a witness J, exhibits all the errors of another surviving witness, F, and in addition at least one error of its own ('peculiar error'), then J must be assumed to derive from F. Sometimes a witness can be shown to depend on another surviving witness from a single passage, viz., if the peculiar error in the descendant is clearly due to the external state of the text in the surviving exemplar; e.g. . . . where in copying a prose exemplar a line has been omitted, destroying the logical unity, &c.[2]

Supposing that 2427 might be a corrected copy of B itself from the Middle Ages, I first checked a facsimile of B only to find out that B's line breaks cannot explain the omission I found. My next candidate was Westcott and Hort's edition of 1881, reputed to be close to B, but the lineation of that text too did not check out. Neither did either of Tischendorf's seventh or eighth editions. At this point, I had exhausted the immediately accessible editions to check and I realized that I would have to visit the libraries in downtown Washington, DC, to check my next candidates, Karl Lachmann's critical edition of the New Testament and Cardinal Angelo Mai's edition of B. I also wanted a full collation to identify all the suspicious omissions, and, since I had heard that Mitchell and Duncan were going to publish theirs in Novum Testamentum and I was already busy with getting my book, The Gospel Hoax, into print, I decided to wait until it was published.

When the collation was published this February, I read it and noticed two other possible line omissions at Mark 6:2 and 14:14. This reinvigorated my interest, so I decided to make one more attempt to find a possible nineteenth century exemplar for 2427 that could explain the line omission. This time, however, I was more systematic: I listed the line omissions and other interesting readings on a note card and went to visit the most comprehensive theological libraries in the area, including the Woodstock Theological Center at Georgetown University, the Library of Congress, and Mullen Library at the Catholic University of America.

It is a truism that you always find things in the last place you look, and it was true for me as well. After coming up short at Georgetown and the Library of Congress and looking at a dozen old editions at Catholic University, I finally found an edition that had the line breaks in just the right places to warrant further inquiry. This edition was Philipp Buttmann's 1860 edition of the New Testament for the Teubner classical library, based primarily on the text of Codex Vaticanus. Because I knew that Buttmann's efforts had preceded Tischendorf's most accurate transcription just a few years later, it became clear to me that Buttmann would have reproduced the inaccuracies of previous collators of the famous Vatican text. Accordingly, I then photocopied its text in Mark and proceeded to the next test, a detailed examination of its pattern of deviations from Codex Vaticanus.

In the Gospel of Mark, Buttmann's text departs from B at eighty-five variation units; 2427 agrees with Buttman's text more than eighty-one times, except in four cases where 2427 departs from both (usually with a singular reading). The dependence of 2427 on Buttmann's edition of B extends to Buttmann's selection of corrections. Significantly, many of 2427's readings in support of Buttmann's departures from B include his mistaken reliance on the inaccurate collations of B. Furthermore, of some 105 corrections in B, Buttmann followed the corrector in seventy-eight instances and the original hand in twenty-seven cases; the scribe of 2427 followed Buttmann's selection in all but five cases, all involving trivial differences such as modern spelling. This evidence led me to conclude that the exemplar of MS 2427 is the 1860 Buttmann edition of the New Testament or one of its stereotypic reprints.

My reflections on 2427 did not cease with the submission of a paper proposal to the New Testament textual criticism section in February 2006. I also e-mailed a copy of my proposed abstract to the textualcriticism mailing list (now run by Wieland Willker), and my results have now been confirmed by two textual critics who have looked into the matter. Nevertheless, I still had one mystery to solve—how is it that the scribe of 2427 had access to a critical edition that the great Ernest Cadman Colwell did not? I am saving the answer to this puzzle for my Tuesday morning session at SBL (S21-18).

Stephen C. Carlson, Fairfax, Va.

[1] The factual details in this paragraph are from Margaret M. Mitchell and Patricia A. Duncan, "Chicago's 'Archaic Mark' (MS 2427): A Reintroduction to its Enigmas and a Fresh Collation of its Readings," NovT 48 (2006): 1-35. Those wishing further background and reference should consult this useful article.

[2] Paul Maas, Textual Criticism (trans. Barbara Flowers; Oxford: Clarendon, 1958), 4.

Comments on this article? email:
Let us know if you would like your comments sent to the author or considered for publication as a letter to the editor. Please include your full name and, if you would like, your affiliation.

Citation: Stephen C. Carlson, " "Archaic Mark" (MS 2427) and the Finding of a Manuscript Fake," SBL Forum , n.p. [cited Aug 2006]. Online:


© 2014, Society of Biblical Literature. All Rights Reserved.