Once upon a time, there was a letter named Ash. It was a combination of the letters A and E, smashed together. For centuries, Ash began names such Æneas, Æesop, and Æschylus, and was found in Cæsar, as well as less distinguished words such as archæology. After a distinguished career in mediæval English, Ash eventually retired. Ash is still employed, however, in the textual apparatus of the Nestle-Aland Novum Testamentum Græce: “mae” is the abbreviation used there to represent two manuscripts written in Middle Ægyptian. The first Middle Ægyptian manuscript, referred to as mae1 in the apparatus, is the Scheide Codex (described by Metzger in 1980), a copy of practically the entire Gospel of Matthew made in the 400’s. The second Middle Ægyptian manuscript, named mae2 in the apparatus, is Codex Schøyen 2650 – a manuscript which is surprisingly important considering how little attention it has received. (Other manuscripts, extant for other New Testament books, are also called mae; in the UBS4 apparatus they are called meg.)
Codex Schøyen 2650 was probably produced sometime in the 300’s. It is a prized item in the collection of Martin Schøyen. It was first described in detail in 2000 by Hans-Martin Schenke, who proposed that its text reflects a form of the text of the Gospel of Matthew that is drastically different from the canonical text – perhaps even the Hebrew text which some patristic writers suggest was the basis for the Greek text of Matthew. This tantalizing suggestion, however, was opposed by other scholars, including the late William L. Petersen and Tjitze Baarda. Baarda was gentle in his criticism of Schenke’s approach; Schenke died in 2002, and Baarda may have wished to adhere to the proverb, De mortuis nil nisi bonum.
Codex Schøyen 2650 was the subject of James Leonard’s 2012 dissertation, which he prepared with the full resources of Tyndale House at his disposal; his preparation also involved a year of research at the H. Milton Haggard Center for New Testament Textual Studies. His dissertation has since been published by Brill as Codex Schoyen 2650: A Middle Egyptian Coptic Witness to the Early Greek Text of Matthew’s Gospel. Leonard demonstrates with thunderous force that Schenke’s appraisal of the manuscript’s text minimized its close adherence to the Alexandrian Text (especially Sinaiticus and Vaticanus) and maximized readings which can be, and should be, regarded as points at which the translator resorted to paraphrasing, or which are the results of the Coptic dialect’s inherent inability to directly correspond to every nuance of the Greek text.
This is not to say that mae2 does not have more than its fair share of anomalies. Leonard selected three passages for very thorough analysis: , 12:3-27, and 28:1-20. In the first two passages, when mae2’s text initially appears to stray from the Alexandrian Text, there is almost always an explanation rooted either in the translator’s approach, in scribal mistakes, or in the limitations of Coptic syntax. But in Matthew 28, we encounter the following oddities:
● 28:1 – Mae2 adds, “when the stars were yet above” as a way of signifying that it was very early in the morning. (Mae1 also has a unique reading here.)
● 28:2 – Mae2 says the angel “took” the stone, rather than that the angel rolled away the stone.
● 28:2 – Mae2 includes not only the phrase “from the door” (agreeing with the Byzantine Text) but also “of the tomb” (agreeing with the Caesarean Text).
● 28:5 – Mae2 makes the unusual statement that when the guards shook due to their fear of the angel, “they arose as dead men.”
● 28:5 – Mae2 does not have the phrase “who was crucified.”
● 28:10 – Mae2 does not have the phrase “Do not be afraid.”
● 28:10 – Mae2 says, “Tell my brothers to return to me in
Galilee” instead of “Tell my brothers to go to Galilee.”
● 28:12 – Mae2 adds “of the people” after “elders.”
These do not add up to justification for the idea that the translator’s base-text was descended from a different source than the forms of Matthew 28 that are extant in Greek, but they do show that Egyptian translators were far from the models of precision that they are sometimes claimed to have been.
Leonard points out a distinctive characteristic of the translation preserved in mae2 that is particularly interesting: “Mae2 often compresses synonymous verb pairings to a single verb.” Leonard illustrates this by citing , where, in the Nestle-Aland compilation and in the Byzantine Text, two blind men “cry out and say.” The compilers of the Nestle-Aland text were so confident that this is the correct text that they did not even note that there is a variant-reading at this point. But in a few important manuscripts (C (which was corrected), L, and f13), the text only says that the blind men “cry out.” Mae2 agrees with this shortened reading. Leonard mentions Matthew 9:36, 11:1, , , , 26:4, 26:74, 27:2, and 27:48 as other passages which display this phenomenon in mae2.
This spurs a question which, although it was not even raised by Leonard, seems worth exploring: if scribes in
could thus shorten the text when translating,
could they not do the same thing when transcribing? The tendency to compress synonymous
verb-pairs displayed in mae2 could easily account for the short
Alexandrian readings one finds in passages such as Luke 24:53, where the
Byzantine reading has been regarded as a conflation.
Mae2’s text generally adheres closely to the text of B and Aleph even at some points where they have almost no other support: Mae2 does not have 12:47 (a verse which is exceptionally vulnerable to loss due to homoeoteleuton); mae2 agrees with B in 14:24 (where B’s text, though adopted in the Nestle-Aland/
In Codex Vaticanus, although the text is very strongly Alexandrian, Western readings pop up here and there, seemingly out of nowhere (such as in Matthew 27:24b). The same phenomenon manifests itself in mae2: the entire text of is absent from mae2, as it is from Codex D and several Old Latin manuscripts.
Along with Leonard’s many insightful observations about the readings of mae2 is a chapter about their possible impact on the Nestle-Aland compilation; Leonard proposes that at several variant-units where the evidence seems finely balanced, the weight of mae2 might tip the scales. These include variant-units in and , among others.
|Although the publisher is asking for|
about $130 for a copy,
less expensive copies
can be obtained second-hand
(at least, this was possible
when this post was written).
Leonard does not investigate the question of possible lector-influence upon the Alexandrian Text displayed in mae2, even though he provides some data which would be helpful in such an investigation. It is sometimes thought that where the Byzantine Text has the name “Jesus” where the Alexandrian Text does not, this is due to a Byzantine tendency to add the name “Jesus” at the beginnings of lections. In mae2 we see the opposite tendency: Jesus’ name is missing in and 17:8. Lector-influence may, however, be the cause of mae2’s inclusion of Jesus’ name in .
Now that James Leonard has provided the go-to resource for mae2, irrevocably establishing it as a strongly Alexandrian witness, the stage is set for the next logical step: a comparison of mae2 to its slightly younger relative, mae1. Although these two manuscripts represent the same Egyptian dialect (sometimes called Mesokemic), and their production-dates are relatively close, they disagree at many important points, such as 5:44 (mae1 has the longer form; mae2 agrees with B), 6:33 (mae1 has “of God”), 9:13 (mae1 has “to repentance;” mae2 does not), 12:47 (mae1 includes the verse), 17:21 (mae1 includes the verse), 24:7 (mae1 includes “and pestilences”), etc. That dissertation is yet waiting to be written. In the meantime we can celebrate that Leonard’s dissertation is available.
Codex Schoyen 2650: A Middle Egyptian Coptic Witness to the Early Greek Text of Matthew's Gospel - a Study in Translation Theory, Indigenous Coptic, and New Testament Textual Criticism, by James M. Leonard, is Copyright © 2014 Koninklijke Brill NV, Leiden, The Netherlands.