Tuesday, September 29, 2015

Duplicating a Passage: Scribes getting in the way of the Apparatus

Minuscule 1573 writes the section Mt 25:22-23 twice. Not good.

This is the relevant section (images available here and here at the NT.VMR):

I don’t know who is responsible for crossing out the duplicated words, it may be the scribe, it may be a later reader / corrector.

What makes this duplication interesting is that the first time round, 1573 reads προσελθων και ο τα δυο ταλαντα (without δε, and only with Sinaiticus and Vaticanus), but the second time προσελθων δε και κτλ. with the rest of the Greek tradition. Legg cites 1573 as siding with the minority reading, but this is only correct when we look at the first occurence, since the second, crossed out version has the ‘normal’ text.

I am not sure how this could be put in an apparatus. Perhaps something like 1573primus and 1573secundus? Or should we just label the duplicated bit 1573dupl.? I assume this problem has been addressed in textual criticism somewhere, but I shouldn’t know where.


  1. Very interesting Dirk. Would you say the likely explanation is that the scribe was copying off of two exemplars simultaneously, and this is a case of cross-exemplar dittography?

    If so, I think maybe we should just view it as a record of two older witnesses. In crossing out one, someone was showing preference for one reading over the other, but seeing as we don't know who that was or what their grounds were, I really see no reason to give their text critical judgement any more status than, say, my own!

    Given that, i'm not sure i'd go for designators like "primus" or "secundus" sincce I think that implies a hierarchy that we probably should not (yet) b reading in. That is, we may later on various text critical grounds choose to rate one reading higher than the other, but I think the manuscript itself simply offers two readings.

  2. Dirk,

    A similar situation occurs in the Pericope Adulterae collation material, particularly in instances where Jn 8.12 might occur before and also after the PA, or in lectionaries that have differing Menologion forms of the PA in more than one location. My solution in such cases is to call the first occurrence "a" and the second "b". Thus in the present example, I would cite 1573a* for the first statement of the verses and 1573b for the second, as well as noting the crossed-out omission as 1573ac (the c being superscript). Perhaps not the most elegant solution, but one that worked for me.

  3. This comment has been removed by the author.

  4. If one were to represent this sort of thing in an apparatus, I’d recommend listing the I.D. number, followed by a superscripted “ditt” followed by “a” for the first occurrence and “b” for the second occurrence.

    (The inclusion of “ditt” is to convey that a repetition in the text is being referenced, rather than a corrector.)

  5. It seems to me that the exemplar of 1573 must most likely have had the de, and it was overlooked the first time by the scribe, since it otherwise wouldn't have gotten into the section in the second copying.

    So I would count 1573 with the majority.

    I also think it adds extra weight against the minority reading by showing how easy it is to omit that de accidentally. This point can't be conveyed in an apparatus, though.
    So I would

  6. This sort of thing is presumably quite common. For instance, Codex Bezae redupilcates Mk 1.34a. The second time, it reads perfectly normally καὶ ἐθεράπευσεν πολλοὺς κακῶς ἔχοντας ποικίλαις νόσοις καὶ δαιμόνια πολλὰ ἐξέβαλεν, but the first time it has the stylistic improvement καὶ ἐθεράπευσεν αὐτοὺς καὶ δαιμόνια πολλὰ ἐξέβαλεν. The thing that merited the Bezan improvement is of course that D b c e ff² g¹ q have πάντας τοὺς κακῶς ἔχοντας νόσοις ποικίλαις back at v32. An interesting variant – there's a good case to be made that Mark authorially had those νόσοις ⇔ ποικίλαις in both verses – but not easy to put in an apparatus.