Matt. 20:22 (traditional text)
αποκριθεις δε ο ιησους ειπεν ουκ
οιδατε τι αιτεισθε δυνασθε πιειν
το ποτηριον ο εγω μελλω πινειν η
το βαπτισμα ο εγω βαπτιζομαι βαπτισθηναι
λεγουσιν αυτω δυναμεθα
και λεγει αυτοις, το μεν ποτηριον μου πιεσθε
και το βαπτισμα ο εγω βαπτιζομαι βαπτισθησεσθε
το δε καθισαι εκ δεξιων μου και εξ ευωνυμων μου
ουκ εστιν εμον δουναι αλλ οις
ητοιμασται υπο του πατρος μου
But Jesus answered, "You do not
know what you are asking. Are you able to drink
the cup that I am about to drink?
- or the baptism that I am baptized, to be baptized with ?"
They said to Him, "We are able." and he saith unto them, of my cup Ye shall indeed drink,
and with the baptism that I am baptized with be baptized:
but to sit on my right hand, and on my left,
is not mine to give, but it shall be given to them
for whom it is prepared of my Father.
INCLUDE LINE: C, W, X, Δ Σ Φ f13, 33, 579, Byz, Maj (Majority of MSS), f, h, q, Sy-P, Sy-H, bo(pt) Lect
OMIT: א, B, D, L, Z, Θ, 085, f1, 788(f13), 22, pc,
Lat, Sy-S, Sy-C, sa, mae-1+2, bo(pt)
B: no umlaut
There is obviously plenty of opportunity for Haplography here, but we must strongly suspect some deliberate and unfortunate editing is really what occured.
A 3rd or 4th century Alexandrian editor probably looked at the first incredibly awkward clause, and jumped to the conclusion that it was in fact a marginal note of an alternate reading that had been inserted into the main text: That is, he read the "-or, ..." (20:22) as a textcritical comment by a previous copyist.
With this conclusion before him, he came to the second line, and took the "-and, ..." (20:23) as again a headsup for an alternate reading. Since the first half of each speech by Jesus makes plain sense, this editor adopted the first clause in each verse, and deleted the second clause as a secondary (incorrect alternate) reading. This editor's logic here is flawed however, since "and" is part of the reading itself, not a marginal note.
This novel solution to the plainly difficult and wordy text of Mark (as copied by Matthew) was attractive because it completely smooths out the awkwardness of the baptism references, including perhaps some sense of anachronism from their presence in Mark's original.
But the origin of these two clauses is not some imagined marginal gloss, but simply the text of Mark, which needs no explanation. It is Mark's usual clumsy translation/Semitic Greek, and its flavour has the usual (slightly anachronistic and back-viewing) features of a Markan speech of Jesus.
Another reason for rejecting the thinking and efforts of some Alexandrian editor here, is the second instance (20:23). This has the triple coincidence of homoioarcton (similar beginning), homoioteleuton (similar ending), and same length of clause (lucky alignment). From this we must strongly suspect that the original cause of the confusion and variation was indeed an accidental Haplography error in Matt. 20:23. This led to some Alexandrian editor examining the preceding verse and jumping to the idea that this was a case of alternate readings conflated from a marginal note.
Naturally this novel solution and the variant text created did not have the power to influence the entire transmission stream this late in the day (c. 2nd to early 3rd cent.), and it became a discontinued minority reading of the Alexandrian texttype.
Willker notes the lines originate from the parallel passage in Mark (Mk 10:38-39), and says; "Very probably copied from Mark (so Weiss). The support is not very good.".
But this is pointless. Matthew, even more than Luke, tends to copy Mark verbatum, even when its wordy or awkward, limiting himself to deleting unimportant details, not the speeches of Jesus.
Once we acknowledge that Matthew has here copied everything other single word verbatum from Mark, there is no strength nor need to claim this was some kind of later interpolation/harmonization with Mark by a copyist. The natural assumption is that Matthew simply copied Mark.
When Willker says "support is not very good", he has already rejected the traditional text (majority/Byz) used by Christians for centuries as "not very good". This bias permeates his entire commentary.
Westcott/Hort delete the verses, favouring again Aleph/B, and Nestle, UBS-2 follow. UBS says nothing in the apparatus, and so once again almost all the "modern versions" (English Translations) drop the verses without perhaps even knowing they have been hoodwinked again.
Another two clauses have been quietly deleted from the Bible without any documentation, notice, or justification.