a text-critical question

I have a question about how textual critics identify problems requiring explanation and then proceed to search for and land upon satisfactory solutions. I'm currently in the throes of reading Ehrman's extensive discussion of the textual issues surrounding Luke 22.19–20.¹ First, let me present the problem in very brief terms.

There are, essentially, two readings to consider, the "longer," more traditional, reading here is as follows:

19 "Then, taking [some] bread and giving thanks, he broke it and gave it to them, saying, 'This is my body, given for you. Do this in remembrance of me.' 20 And likewise the cup, after eating, saying, 'This cup is the new covenant in my blood, poured out for you. 21 But look! The hand of the one who is betraying me is with mine upon the table.'"

This longer text is attested "by all Greek manuscripts except D [= Bezae] and by most of the ancient versions and Fathers."² The "shorter, or Western, text"—that is, the reading attested in Bezae and in mss of the Old Latin, which is typically an expansionist tradition—omits vv. 19b–20:

19 "Then, taking [some] bread and giving thanks, he broke it and gave it to them, saying, 'This is my body. 21 But look! The hand of the one who is betraying me is with mine upon the table.'"

Ehrman rightly emphasizes how difficult it is to explain the omission of 19b–20 in the Western tradition if the longer reading is original. There aren't, apparently, any satisfactory contextual factors to support an accidental omission (e.g., homoeoteleuton or homoeoarchton). And, of course, why any scribe would intentionally omit (i) the description of Jesus' body as "given for you . . ." and (ii) any reference to cup at all presents insurmountable problems.

Ehrman provides other arguments in favor of the shorter reading, to be sure. But quite a bit of weight is put on the difficulty involved with explaining how vv. 19b–20 dropped out of the Western manuscript tradition if these verses were originally part of Luke's gospel. And I don't suppose I have anything to offer by way of explanation (though note that the UBS committee felt strong enough about including vv. 19b–20 in the text that they give it a [B] rating). But here's my question: How does the proposed originality of the shorter reading explain the presence of the shorter reading in some Western manuscripts? On the surface the answer seems obvious, but upon closer look the problems Ehrman seeks to avoid by accepting the Western reading loom just as large.

• Given the intractable problems explaining why any scribe would drop vv. 19b–20 from his exemplar, don't we still need to explain why Luke would omit this passage from his Markan source (on either the Two-Source or Farrer theories of synoptic relationships) or, less likely, from Matthew (assuming the Griesbach-Farmer hypothesis)? I can't find ready explanation why Luke would drop this passage, and this—it seems to me—balances evenly the difficulty of why a later scribe would omit it.
• Ehrman finds it significant that the vocabulary of vv. 19b–20 is non-Lukan and sounds an awful lot like Paul's version of the Last Supper in 1 Cor. 11.
  
  

  Interestingly, this "other tradition" (i.e., vv. 19b–20 in the longer text) is not only anomalous within Luke's Gospel itself, it also has very few connections with Luke's source, the Gospel of Mark. Instead, as has been frequently noted, the additional words practically mirror the familiar account of the institution preserved in Paul's first letter to the Corinthians. The fact that the words of the longer text of Luke are not precisely those of Paul should not be used, as it sometimes is, to argue that they could not have been added secondarily to the text of the Gospel. No one need think that a scribe referred to his manuscript of 1 Corinthians to check the accuracy of his interpolation into Luke. Instead, the addition has all the marks of a familiar narrative based on, or at least parallel to, Paul's account of Jesus' Last Supper.³

  
  I'm not convinced. If the best explanation for the manuscript data for which we must account is that a scribe inserted a Eucharistic tradition from another text into Luke's gospel, why should they have chosen 1 Corinthians? This becomes even more of a problem the more we stress the differences—and there are significant differences—between Paul's Eucharist and that found in Mark and Matthew. If a scribe felt the shorter, "original" text of Luke's version of the Last Supper required supplementing, why choose Paul's distinctive account instead of Mark's or Matthew's? Certainly we can't expect scribes in antiquity to be as aware of different traditions as we are (and esp. that Paul's account of the Eucharist conflicted with Luke's way of thinking about Jesus' death). But inasmuch as Matthew was the most popular, most widely read, and most frequently copied of the canonical gospels, I think we need to ask why a scribe would interpolate Paul's instead of Matthew's version of the Last Supper. The language of 22.19b–20 may be distinctive within Luke-Acts, but its Pauline rather than Matthean flavor creates a problem for a theory of interpolation. This falls short of proof for these verses' originality, but it does weaken Ehrman's argument in favor of the shorter reading.
  
[The following two problems belong together; I give the weaker one first.]

• If we accept the shorter reading because we simply can't imagine why a scribe would drop vv. 19b–20 (and let me say again, I can't imagine why), we actually need to suppose that at least some Christians of the West observed a Eucharistic liturgy of bread alone or that took the cup first and then broke the bread (see vv. 17–18). Either this, or we will have to suppose that the scribes responsible for the manuscripts bearing the shorter reading tolerated a striking discrepancy between their Eucharistic practices (supported by the accounts of Mark, Matthew, and/or 1 Corinthians) and Luke's gospel.
• The alternative to expecting a Lukan Eucharistic liturgy (according to the shorter reading) is to suppose that scribes in the second century and later [Bezae is a fifth-century uncial!] observed a traditional Eucharistic liturgy and preserved the original [= shorter] Lukan account of the Last Supper. This, in fact, is exactly what Ehrman argues. But how does the originality of the shorter reading have any explanatory power for the reading in Bezae and the Old Latin if we assume that these same scribes also observed a traditional Eucharist? We rightly scratch our heads and wonder how/why a scribe might have elided vv. 19b–20, but on Ehrman's explanation we still have to deal with scribes who believed the institution of the Lord's Supper occurred one way (see the traditional Eucharistic liturgy we are now positing for these scribes) but who believed, somehow, that the original text of Luke's gospel portrayed a significantly different account of that institution. In other words, it seems to me that the shorter reading still requires explanation in fifth century, irrespective of any judgment of originality. In yet other words, even the if the shorter reading is "original" and the longer reading "corrupt," the shorter reading would [ought?] have struck ancient scribes as corrupt. So why did they preserve this shorter, "original" reading?

In the end, I am (i) unable to answer Ehrman's objection that it's simply too difficult to explain the shorter reading if the longer reading is original, and (ii) unconvinced by his arguments that the shorter reading is, in fact, original. Instead, I think the unusual data requiring explanation is still the shorter reading: Why should any manuscript tradition omit vv. 19b–20? I don't know, but I don't think the problem is solved by supposing this is the original reading. Instead, I think it's a problem that Ehrman has to resort to reading Luke so far afield from one of his favorite sources and one of his favorite Christian characters; notice that he has no problem suggesting that Luke "understands the death of Jesus differently from both the predecessor of his Gospel (Mark) and the hero of his Acts (Paul)."⁴ But why should Luke have so willingly accepted (and reproduced) so much of Mark's gospel and expended so much effort broadcasting and endorsing Paul's reputation, if he held a fundamentally different view of so pivotal a Christian tenet as the death of Christ?!

So again, my question is: How does the proposed originality of the shorter reading explain the presence of the shorter reading in some Western manuscripts? Can anyone provide a reasonable answer?

[CLARIFICATION: Ehrman continues in this chapter to discuss the [in]famous "Western Non-Interpolations," a body of variants that he finds in service of anti-docetic polemics (see, e.g., pp. 255–56, n. 145). In many of these (e.g., Luke 24.12, 40, inter alia) the question I'm raising regarding his analysis of 22.19–20 would be inappropriate. The reason I find the shorter western reading's originality insufficient to explain its presence in Bezae and the Old Latin is precisely the use of the tradition in question (viz., the Lord's Supper) in the liturgical life of the church. Luke 24.12, of course, was not liturgically significant and so may be preserved more accurately (and succinctly) in the Western manuscript tradition.]

---

¹ See Bart Ehrman, The Orthodox Corruption of Scripture, 198–209.
² According to B. Metzger, A Textual Commentary on the Greek New Testament (second edition; New York: United Bible Societies, 1994 [1971]), 148.
³ Ehrman, Orthodox Corruption, 204.
⁴ Ehrman, Orthodox Corruption, 203.