Thursday, February 22, 2018

Matt 18.11 as a Test of the ‘Internal’ Consistency of the Byzantine Priority Position

28
NOT the Silver-haired Assassin
Matt 18.11 It is one of a dozen or so verses that are completely missing (relative to the KJV) from modern English Bibles. It follows precedes Matthew’s version of the parable of the lost sheep and fits quite well in that role. It reads thus in the KJV: “For the Son of man is come to save that which was lost.”

This verse is demoted to the apparatus in eclectic texts and the explanation usually given is that it was brought over from Luke 19.10 almost verbatim. That “almost” is what I want to address here. Compare:
  1. omit — 01, 03, 019* f1, f13, 33, e, ff, syr.sin, etc.
  2. ἦλθεν γὰρ ὁ υἱὸς τοῦ ἀνθρώπου σῶσαι τὸ ἀπολωλός — 05, 032, Byz, lat, syr.cur.pesh.hk, etc.
  3. ἦλθεν γὰρ ὁ υἱὸς τοῦ ἀνθρώπου ζητῆσαι καὶ σῶσαι τὸ ἀπολωλός — 019mg, 579, 892c, etc.
  4. ἦλθεν γὰρ ὁ υἱὸς τοῦ ἀνθρώπου ζητῆσαι καὶ σῶσαι τὸ ἀπολωλός — Luke 19.10 (no vll. in NA)
On external grounds, the external argument from the Byzantine prioritist is straightforward. There is no serious split in the Byz manuscripts at this point and therefore reading 2 is original at Matt 18.11.

But how does the same Byz prioritist explain the corruption that is reading 3 on internal grounds without appeal to harmonization and thus showing an inconsistency? After all, if ζητῆσαι καί is an addition due to Luke 19.10 then surely the whole verse can (and should) be explained the the same way. 

It seems to me that this shows an “internal” inconsistency on the part of the Byzantine priority position. Internal evidence is accepted when it supports the conclusion made from external evidence. But the same evidence is rejected when it does not.

As always when I talk about the Byz text, I shall wait to be corrected by our resident Byzantinist, the Silver-haired Assassin himself!

28 comments

  1. Or alternatively, a Byz prioritist could explain it like this: ζητῆσαι καὶ in Variant 3 is due to harmonization to Luke 19:10. On the other hand, the fact that the majority of Byz MSS (variant 2) DON'T have this addition in Matt shows that maybe variant 2 was actually the original Matthean text. It wouldn't be the first time when the Synoptic Gospels had very similar sentences. Maybe because of this similarity between Matt and Luke at this point, a copyist who knew Luke well could inadvertently add that Lukan phrase into Matthew's text.

    Florenc (Lenci) Mene (not a Byz prioritist)

    ReplyDelete
  2. Jonathan Clark Borland has produced some detailed work on this variant; those with access to the Academia site can download it:
    https://www.academia.edu/6085076/THE_AUTHENTICITY_AND_INTERPRETATION_OF_MATTHEW_17_21

    ReplyDelete
  3. (Wait; nevamind; that's a different passage. (Go to sleep, Jim!)

    ReplyDelete
  4. "Internal evidence is accepted when it supports the conclusion made from external evidence. But the same evidence is dismissed when it does not."

    Can't this be fairly said to apply to every position? I don't know of any major position (Alexandrian priority, Byzantine priority, reasoned eclecticism, thoroughgoing eclecticism) that applies every principle they use to every passage.

    ("Death is swallowed up in controversy," 1Cor 15, is certainly more difficult. It has enough external support to be plausible to an Alexandrian-priority or eclectic critic, but I don't know of any who actually advocate it.)

    ReplyDelete
  5. Here's a quotation from Robinson's "The Case for Byzantine Priority" appendix to RP2005): "That scribes engaged in some harmonization or assimilation to parallel passages or contexts can be demonstrated repeatedly within the pages of a critical apparatus... Yet, one must carefully guard against the assumption that verbal identity where parallels exist is presumptive evidence against authenticity. Merely because harmonization or assimilation could occur at a given location, one must not assume that scribes would harmonize whenever possible. Nor is scribal harmonization when it does occur more characteristic of the Byzantine-era scribes than any other. Once more, transmissional aspects remain the primary basis for decision. The apparatuses demonstrate that most of the numerous cases of harmonization or assimilation did not perpetuate in any great quantity. While scribes did harmonize at various places, and that frequently enough, the vast majority of scribes did not accept or perpetuate such alterations to any significant degree. Even if parallel locations were known from personal familiarity with scripture, most scribes would not adopt or add to the text that which was not in the exemplar before them. Harmonization simply did not occur on the grand scale. It would be a transmissional absurdity to assume numerous "harmonization-prone" scribes adopting a few dozen harmonizations into their Byzantine MSS while failing to continue the process in hundreds of other places where scribes had produced more plausible and attractive harmonizations--none of which were incorporated into the main stream of transmission... When harmonization or assimilation did occur, it was sporadic" (548-49).

    It's not that BP says "aha!" when it sees an apparent harmonization/assimilation in one case and then closes its eyes and hums loudly when it sees another (Byzantine) case. Apparent harmonization/assimilation might in any given case be a secondary scribal move or it might be the autographic wording; transmissional probability is under BP theory given more weight than the traditional supposition that most scribes harmonized where they could, and this weighting applies to both apparent Byzantine harmonizations/assimilations and to apparent non-Byzantine harmonizations/assimilations.

    ReplyDelete
  6. Firstly, I want to point out what is sadly lacking in this conversation: Praise God that Jesus is the "Son of Man [who] came to seek and save the lost!"

    Secondly (and to the point of the post!), Wouldn't a BP need to argue as to why haplography would occur in this passage, thereby explaining the absense of this clause in different manuscript traditions? I do not see homoeoarkton or homoeoteleuton. As beloved of a theological statement (and true!) that this clause is, I find it hard to argue that it would be in the original.

    ReplyDelete
  7. I see the exact same evidence pointing the other way.

    I don't think Byzantine priorists have any need to deny that corruptions by way of harmonization do occur. There's nothing about the hypothesis that reading 3 arose as a harmonization to Luke that causes a problem for their view. It would merely be that, by their model, scribes had exemplars with reading 2 and altered that to reading 3 to agree with the longer version in Luke.

    On the other hand, given that reading 2 does have such stronger attestation than reading 3 in Matthew, and there is no variant of this sort at all in Luke, it's hard to explain why scribes would alter a text of Matthew that lacks 18:11 by adding a version that agrees with Luke, but that isn't word-for-word the same. The fact that reading 2 doesn't exactly match Luke but is still so much better attested than reading 3 fits the Byzantine priorist theory quite well.

    It seems to me that the problem the Byzantine priorist faces here is not about how a scribal tendency of harmonization fits this data, but rather how omission of the verse (regardless if it were version 2 or 3) would come about if it were original.

    ReplyDelete
    Replies
    1. My theory is that an early redaction deleted this verse on the theory that it was an interpolation; it is found, after all, in neither Mk 9:38-50 nor Lk 9:49-50. Hellenistic textual criticism was on the lookout for supposed interpolations. Why many scribes would add this verse in this location, however, is difficult to answer.

      Alternatively, homoeoteleuton based on the last letter alone of the verse is possible.

      Either explanation would struggle to explain reading 1 if it was widely attested, but as it is support is limited and largely focused, it would seem, on areas of scholarly activity where the luxury for and interest in edited texts existed.

      Delete
  8. I agree with Florenc that the explanation of harmonization doesn't have to be thrown out the window here. If Byzantine scribes had wanted to harmonize and had access to Luke, then why would they have harmonized only partially? The minority support for reading 3 makes it clear that some scribes would not be satisfied with a partial similarity, but given the widespread support for reading 2, it seems that most scribes were okay with it. The "almost" has a good deal of explanatory power here.

    As for the question of explaining reading 1 from reading 2, I agree that haplography does not lend itself as an explanation here. I'll start by defending the longer reading on the grounds that harmonization from some other gospel to Matthew was less likely than harmonization the other way around; we would normally expect to see Luke conformed to Matthew in a situation like this. Harmonization the other way certainly did occur sometimes, but it's worth noting that it was atypical.

    More importantly, the placement of the verse in Matthew may have been a problem to some scribes. While the OP states that "It follows Matthew's version of the parable of the lost sheep and fits quite well in that role," this isn't quite correct. I agree that this verse fits well with the parable, but it doesn't follow it; it precedes it. The γαρ of v11, in its usual function, would prompt the reader to consider a connection to what was just said. But what does despising little ones or the place of their angels have to do with the statement of v11? It seems like v11 was intended to introduce the material that follows rather than conclude the material that precedes. An early scribe may have obelized or omitted the verse on stylistic grounds, with subsequent scribes leaving it out of their copies.

    What I wonder is, do we see scribes taking any simpler approaches to resolve the problem of v11? What about a simple change from γαρ to something like δε? What about transplanting the verse from between v10 and v12 to after v14, where it seems more fitting as a conclusion? I'd be interested to see if the manuscript evidence supports any of these possibilities. I'd also like to see how the manuscripts that include v11 handled paragraphing around it. I notice that 01 starts a new paragraph at v12, and 03 adds a space before it begins v12.

    ReplyDelete
  9. Of course Byzantine prioritists admit harmonization as a cause of corruption. My point was that in this case, you can’t keep your transcriptional cake and eat it too. The fact that the widely attested form is missing two of eleven words from the parallel in Luke doesn’t convince me at all that harmonization is not the obvious and immediate cause. The lack of transcriptional explanation for the short reading in Matt is the clear proof in my view.

    Thanks for the correction on the placement of v. 11 Joey. Its addition there is just as easily explained though, per Metzger: “The reason for the interpolation was apparently to provide a connection between ver. 10 and verses 12–14.”

    ReplyDelete
    Replies
    1. It seems to me that BP theory would be eating the cake if it appealed to harmonization as the decisive consideration in this or any similar instance. Such is not the case.

      Delete
    2. That would certainly heighten the inconsistency, but it’s there if harmonization is appealed to at all to explain the longest reading. Maurice has been quite clear that he thinks internal arguments support the Byz position. My point here is that they do ... except when they don’t! In this case the problem happens to be evident in the very same variation.

      Delete
    3. Is it not the case that reasoned eclectics appeal often to the shorter reading, yet at other times appeal to other factors - let's say external evidence- to support a longer reading? In these latter cases, it's not that the shorter reading tenet has been abandoned, but rather subordinated to a weightier or more pertinent factor.

      Delete
    4. Hence the name “eclectic.” But the Byz priority claims that the internal evidence is taken seriously. My problem with this claim is that internal evidence is never able to challenge the unified Byz reading in practice. Hence, in my view, internal evidence is superfluous in that approach except where Byz is split.

      Matt 18.11 is a good example. The internal evidence supports the Byzantinist’s externally-based rejection of reading 3 but that same evidence goes against his externally-based acceptance of reading 2. In both cases, the only real factor is external evidence; internal is superfluous.

      Delete
    5. Robinson can correct me, but I think the label he would give internal evidence where the Byzantine witnesses are not significantly divided is "confirmatory." An original reading can in general at least be expected to have internal points in its favor. If internal evidence were constantly against Byzantine readings, meanwhile, I would see reason to revisit that interpretation of the historical and other data which leads me to assign the weight to the Byzantine consensus text that I do.

      Delete
    6. In other words, internal evidence is a “yes man.” ;) It has no real influence.

      Delete
    7. In any one instance, yes.

      I do have indisputable proof that internal evidence supports the Byzantine reading and explains the others, but, to paraphrase a mathematician, the space alotted on the blog comment feature is too small. ;)

      Delete
    8. Well, no, actually. In all instances where it might actually challenge the method’s starting assumptions. That’s why this example is more than a trivial exception, in my view. It is a single illustration, yes. But what it illustrates is the larger problem.

      Delete
    9. Actually I didn't mean to concede this verse as an exception.

      For my part, I think it is not a problem to make internal evidence play second fiddle where external evidence is strong.

      Delete
    10. Sure. My point though is that it’s not playing fiddle at all.

      Delete
    11. For the record, my wife looked at this and her reaction was something like, "You guys are so cute with your fiddles and cake."

      Delete
  10. Metzger's explanation is certainly a possibility, but why add a connection here when the passage is already smoother without it? Is v11 is absent, then the parable of the sheep becomes less about saving what is lost and more about the Father's compassion for "these little ones," which ties vv12-14 neatly back to v11.

    I'll admit, of course, that this would also make a text without v11 more likely on intrinsic grounds. But I think it applies to transcriptional likelihood equally, and a passage that includes v11 still makes enough sense to have been original to the author of the gospel.

    ReplyDelete
  11. I think most text critics (including Byzantine priorists) would admit that multiple transcriptional proclivities were active on multiple scribes at any given unit of variation, and that often the proclivities of one or another scribe may be in conflict even in the same variation unit. There is no ultimate transcriptional panacea to solve every textual problem, although Bengel's proclivi scriptioni praestat ardua comes close. Although I cannot speak for him, I think that Robinson, as Hort and in many respects even editors using the CBGM to produce a resultant text, does not in the final stage arbitrarily judge any single variation unit in isolation from his judgments on all the others. E.g., his judgment on the passage in question in Matt 18:11 is not divorced from his judgment in Matt 16:2-3 on the quality of many of the same witnesses that omit there (i.e. ℵ B f13 sy-s sa mae bo-pt; Origen), or even especially from the various alterations in 18:10 to ἐν οὐρανοῖς (B 33. 892), in 18:12 to ἀφείς and the non-inclusion of καί (B L Θ f13 892) and (after ἐννέα) to the non-inclusion of πρόβατα (B Θ f13 mae). So is the charge of inconsistency a result of confusion over Robinson's method, or that he uses transcriptional criteria in way inconsistent with, say, Metzger's use of them (i.e., constantly dealing with conflicting scribal proclivities at any variation unit), or something else? Some clarity in this regard may assist in the discussion.

    ReplyDelete
  12. I note that IGNTP records the following witnesses as missing ζητῆσαι καί in Luke 19.10: 1187, 2757, Marcion, AM, Lc, Sat. This is obviously due to homoioteleuton (-σαι to -σαι) and the same mistake would explain why reading 2 in Matt 18.11 omits them too.

    ReplyDelete
    Replies
    1. It's not clear to me how the omission of ζητησαι και before σωσαι can be explained by homoioteleuton. A skip from one ending to the other would result in the omission of the latter phrase (και σωσαι), not the former. Homoioarcton would be a more appropriate mechanism for the omission of the first two words, but it doesn't seem to be applicable here.

      Alternatively, the witnesses to the reading without ζητησαι και in Luke 19:10 might have harmonized to the same reading in Matthew 18:11, meaning that the Byzantine reading in Matthew was already circulating at the time of Marcion.

      Delete
    2. Assimilation is possible but probably not in Marcion of all people given his rejection of all but Luke.

      Delete
    3. Agreed, but that still leaves the possibility that he inherited the reading in Luke from an even earlier exemplar.

      Delete