Wednesday, November 23, 2005

Conjectural emendation

In TC some of the insurmountable gaps between scholars have to do with a very different appreciation for some of the witnesses to the text of the NT (Codex B, the Majority text, vg, quotations in church fathers etc.). Transcriptional reasoning seems much less problematic, at first glance. The purest form of transcriptional reasoning is done by those who suggest conjectural emendations (cj) to the text.
I have noticed Evangelical scholarship tends to be quite reluctant to allow for cj.
Let me raise a question here: is cj compatible with a high view of Scripture? Personally I cannot see why not. In my article “Paul’s Use of Scripture in 1 Corinthians 1-4 and Conjectural Emendation in 4:6.” (Analecta Bruxellensia 9 (2004): 102-122) p. 108-115 I have suggested 10 criteria for evaluation of a conjectural emendation. I would appreciate your response. Here they are:
  1. The emendation does justice to the style or the idiom of the author, or at least more justice than the traditional reading.
  2. The emendation solves the problem in the text.
  3. The emendation does not introduce new difficulties or riddles.
  4. The extant readings are – either directly or indirectly – explicable as corruptions of the emended reading.
  5. Few early witnesses are available for the passage.
  6. The reconstruction of the original text has been contested in an early stage.
  7. The development from the conjectured original of at least one of the extant readings could have taken place in an early stage.
  8. The emendation requires only a minor intervention.
  9. Textual critics are not removing or softening elements in the traditional text that offend their logic, culture or ideology.
  10. The derivation of the traditional reading from the emended one may not require procedures that were not current in the earliest formative stage of the NT text.

Gie Vleugels


  1. Dear Gie,
    Looks like we're going to have fun! I hold a rather different view of conjectural emendation. I think that while in theory one might admit that a conjecture could be correct, in practice a conjecture should never be used by an editor and can certainly never function as part of the Bible. After all, how can something that we have made up act with authority over us?

    I'm glad that Bengel agrees, though I must admit that I find it hard to translate Monitum 6:

    Certe nulla unquam conjectura audienda est: tutius seponitur, quae forte laborare videtur, particula textus.

    I'd be interested to know what function you think a conjecture might play within an edition of the Bible?

  2. I would
    a. be very reluctant to accept a cj;
    b. never put one in the text of an edition.
    Accepting conjectures is the final consequence of the admission that the original wording of a NT text can be lost (X providential safeguarding). Anyone who does not accept this possibility, should reject all minority readings, since all of them were lost (at least for centuries) for a majority of the users of the Greek New Testament.

  3. Thanks for the clarification. I'm glad that you would have a more cautious editorial policy.

    The appeal to providential safeguarding should be dealt with thoroughly on our blog some time. Obviously this type of language tends to be used most often by TR advocates, and occasionally by Majority Text advocates. However, since all evangelicals believe in providence and that God has safeguarded the scriptures in some way then it seems that evangelicals who are not TR/Majority Text advocates need to explore what the possibilities are.

    The quotation from Tregelles that I cited a while back from the beginning of Pierpont and Robinson's new GNT said: 'As God in his providence has preserved Holy Scripture to us, so can He vouchsafe the needed wisdom to judge of its text simply on grounds of evidence.' Here Tregelles identifies 'providence' and 'vouching safe' (i.e. safeguarding) as things connected with his own position (which was not TR or Majority Text). Interestingly, I believe that Tregelles also accepted that conjectures could be made, but would not print them in an edition.

    All of which is to say that I see no reason why with your position you would not also want to talk of providential safeguarding. The alternatives to providential safeguarding are, after all, unprovidential safeguarding, unprovidential non-safeguarding, and providential non-safeguarding.

    Of course, we're going to have to get down to what providential safeguarding means in terms of mechanism and what its implications for the text are.

    We would, I'm sure, want to avoid repeated appeal to special 'intervention' with scribal activity. We also need to clarify the extent to which we can identify providence specifically (rather than generally) within history. Since we do not share God's omniscience we obviously need to avoid adopting a way of explaining things as if we understand all that His providence might be doing.

    Since God is not obliged to provide the Bible to all people of all times (even now most people on earth don't have a Bible) I do not see that we can really make any argument in relation to conjecture on the basis of the non-availability of scripture to a particular group.

    On the other hand, I think that while it is not problematic to say that the Bible was unavailable to most humans for most of its history, it is problematic to say that parts of the Bible were not available to anyone at all during part of its history.

  4. Dear Gie,

    Thanks for the reference to your article. It seems to me that there is no particular contradiction between 'a high view of Scripture' and the use of conjecture in resolving textual problems. If we can lose whole inspired texts (e.g. Paul's letter to the Laodiceans); then there doesn't seem to be any objection to the possibility that we have lost smaller parts of the true original texts.
    Of course, using conjecture will never provide as much confidence in a text as other methods; hence the need for caution as already discussed. One could argue that weighing transcriptional plausibility already involves some conjectures, not in creating readings but in postulating processes.
    Perhaps Eph 1.1 is a place where the best solution is probably a conjecture.

    On Tregelles. It doesn't strike me that he would be that keen on conjecture. What are you thinking of Pete?

  5. This notion of 'providential safeguarding' intrigues me.
    Some questions:
    1. Who came up with the idea? Why should I believe in it?
    2. Most of the Christians from the fifth to the fifteenth century that were copying and using a Byzantine text of the NT, took the (or a) LXX to be the inspired text of the OT. In the West Christians took the Vulgate for granted. If we believe in providential safeguarding, where was it in the Middle-Ages.
    3. Could conjectural emendation, like the discovery of a new manuscript with a variant that was not extant yet, be one of the providential means to safeguard Scripture?

  6. First, as Pete said, we haven't really defined "providential safeguarding" yet. But a text is just as truly safeguarded if it is hidden away in a library, unknown to the world, as it is if by way of conintuous use and copying -- remember the Josianic revival. But, as to your question, Gie, why should we believe in providential safeguarding. It is a corrolary to believing in providence generally. This is theological issue that we have to resolve in the realm of theological inquiry. For us evangelicals, this entails an investigation of the Bible for a proper view of providence. As I am fairly Calvinistic, I can't avoid believing that the preservation of the NT which has occured is a result of providence. But I tend to think that those of a more Arminian persuasion would still agree on this point.

    While I know that a long ongoing theological discussion on this blog could become an aggravation, I'm glad to see theology come up. It seems like that can be a unique aspect of a page devoted to *evangelical* textual criticism. TR and MT advocates have generally not avoided bringing their views of Scripture and providence into their arguments. But their opponents usually don't (although I have seen such things come up in Dan Wallace's articles in Maurice Robinson (despite the quote given earlier) also tends to argue for reasoned transmissionalism in a theologically neutral way.

    But if a given approach to TC is true, then it should be compatible with true theology. I would love to see a paper giving the *theological* basis for a Hortian approach to TC (and let nobody say there is no theological basis for it, as there is for everything).

  7. First, let me agree with Eric on the importance of theological discourse. It is the fact that theological reasoning is encouraged alongside textual study that gives this blog its distinctive character. I don't mind if theological arguments are long, provided progress is being made during their length.

    A general view of providence is pan-evangelical (at least for historic evangelicals).

    On Peter Head's post I'd agree that conjecture (e.g. about transmissional process) may be 'used' in textual reasoning, but we need to define 'use'. I maintain that no conjecture should now be printed as the text by a responsible editor of the NT.

    I said that no conjecture should 'now' be printed. The qualifier 'now' is important, because I would not necessarily have considered it irresponsible for editors early in the printing of the GNT to use conjectures. Such conjectures could be used because the manuscript base was so small that it may have been necessary to depart from the mss. We are now in a much better position, where none of the aporias in the NT is so grave that it is likely that any of the conjectures even approaches them in probability. Eph. 1:1 illustrates this: it is more likely that 'in Ephesus' is correct than any other name and it is also more likely that there was no phrase there than that any phrase we might come up with was original. Therefore the correct reading is one of the two in the mss.

    I have no reference for the suggestion that Tregelles engaged in conjecture but refused to give these a place in his edition. I read it somewhere, probably on the internet. And we all know what Peter Head says about authorities on the internet...

    PH does not distinguish in his post between an 'inspired' but lost letter to the Laodiceans and an 'inspired and canonical' text elsewhere. Theologians have not necessarily seen inspiration and canonicity as co-extensive. For instance, the prophets before the writing prophets were inspired, but left nothing for the canon.

    In answer to Gie:

    GF: Who first came up with the idea of providential preservation?

    PW: Not sure. It obviously goes back as far as the seventeenth century, Turretin, the Westminster Confession of Faith and probably the Formula Consensus Helvetica (1675). There are probably many earlier sources, but I don't know them.

    GF: Why should you believe in providential preservation?

    PW: We'd obviously need to define it before we could say whether you should believe in it. My earlier post merely illustrated that such language could be used by those working within an eclectic framework.

    The indestructability of God's Word is thematised in scripture in passages ranging from Jeremiah 36 to Matthew 5:18. NT writers see the OT as available to them through God's action, and most evangelical theologies agree that there is an analogy between OT and NT in this respect.

    The doctrine of the sufficiency of scripture (for salvation) necessitates a certain level of preservation of scripture.

    One way to produce a fairly useless doctrine of scripture would be to believe all the usual things but not to believe that scripture had been preserved. For instance you could believe in the Chicago Statement on Inerrancy and yet think that only 20% of scripture has been preserved to us. Would someone proposing this merely be guilty of bad scholarship, or would they not also be guilty of bad theology?

    If, say, I conjectured that the first two chapters of Luke were a non-inspired interpolation would I be guilty merely of bad scholarship, or would it be perfectly compatible with orthodox theology? I think that to reject a passage without manuscript evidence would be theologically problematic.

    GF: Most of the Christians from the fifth to the fifteenth century that were copying and using a Byzantine text of the NT, took the (or a) LXX to be the inspired text of the OT. In the West Christians took the Vulgate for granted. If we believe in providential safeguarding, where was it in the Middle-Ages.

    PW: The Greek texts that were used and the versions (Peshitta, Vulgate, Bohairic) were in no case very far from the original. I don't think we need perfection to argue that the scriptures were preserved in many forms by providence (here I'm linking providence closely with the evidence, as Tregelles did).

    GF: Could conjectural emendation, like the discovery of a new manuscript with a variant that was not extant yet, be one of the providential means to safeguard Scripture?

    PW: As an editor I might print something which came from a manuscript not yet discovered. My contention is not therefore with the theoretical possibility that a conjecture could be correct, but rather with the choice of current editors to resort to conjecture when they haven't even considered a fraction of the evidence currently available to us in the manuscripts.

  8. Please, allow me to select an emendation (suggested by Schlichting; cf. app. of NA27) that has impressed me, is theologically sensitive and seems to improve the logic of the passage (Romans 9:5-6).
    I’m curious what you think of it:

    Received text:
    οιτινες εισιν Ισραηλιται, ων η υιοθεσια και η δοξα καὶ αι διαθηκαι και η νομοθεσια και η λατρεια και αι επαγγελιαι, ων οι πατερες, και εξ ων ο Χριστος το κατα σαρκα, ο ων επι παντων Θεος ευλογητος εις τους αιωνας· αμην.

    Conjectural emendation (cj of Schlichting in app. of NA27))
    οιτινες εισιν Ισραηλιται, ων η υιοθεσια και η δοξα καὶ αι διαθηκαι και η νομοθεσια και η λατρεια και αι επαγγελιαι, ων οι πατερες, και εξ ων ο Χριστος το κατα σαρκα, ων o επι παντων Θεος ευλογητος εις τους αιωνας· αμην.

    Of course ων has a rough breathing (relativum) and introduces for the fourth time in the passage some valuable treasure of the Ισραηλιται.

  9. Just a practical note.

    In this blog we get some very interesting discussion in the comments. But it is harder to access than comments in the main blog.

    I should think we could make a discussion of Rom 9.5f into a main entry and then continuing discussion.

    We may need to think of a better way of managing the material. It is part of the team blog issue. Maybe we need a different sort of thing than blogger provides.

    2 lepta


  10. On Schlichting's conjecture.

    a) I'd like to know more about Schlichting and his ideas and motivations.
    b) I'm not too sure which problem this conjecture is solving.
    c) the conjectural reading doesn't strike me as particularly Pauline. Especially it doesn't follow well from EX WN used of Christ - if Paul hesitated to say that Christ is 'theirs' (WN O CRISTOS), one might doubt whether he would say this about God (especially in Romans - even allowing for the irony intrinsic to 9.4f).
    d) the doxological note doesn't seem so appropriate after WN O.


  11. Hello Peter Head,
    The words Ισραηλιται ... ων o επι παντων Θεος could refer to OT notions like Ακουε, Ισραηλ· Κυριος ο Θεος ημων (De. 6:4) and ο Θεος Ισραηλ (Judges 11:21, Jer. 46:25 etc.).
    I don't think that this formulation is 'less pauline' than the received text ο Χριστος το κατα σαρκα, ο ων επι παντων Θεος. Many commentators have been puzzled by the awquard formulation and untypical formulation.
    The conjecture of Schlichting was an effort to solve the problem.
    I am not convinced that he restored the text, but his suggestion is surely tempting.

  12. Gie said of Schlichting's conjecture on Romans 9:5: 'I am not convinced that he restored the text, but his suggestion is surely tempting.'

    So was the fruit! However, as there is no substantial difficulty in the existing text of Rom. 9:5 and no mss support for Schlichting I'll decline to have a bite. It wouldn't even get into the apparatus of an editio maior if I had anything to do with it.

  13. Note that as far as I have heard, plans for NA28 involve deleting ALL notes about conjectures.

  14. I find these notes on conjectures immensely interesting, and see no reason why they should be deleted.
    What about single readings?
    A single reading is often nothing but an old conjectural emendation.
    But of course laudamus Dominum.

  15. I think the aim of the NA28 in deleting such notes would be that it is a text based on the documentary evidence for the NT text, not for the history of the study of the NT. I gather as well that in researching the attributions it was found difficult to confirm the names noted in the Nestle apparatus. I hope that they may have an appendix listing conjectures, but perhaps we'll need a stand-alone resource.

  16. The reason I would not make space for conjectures in an editio maior is that to make space for these would be necessarily to take away space from other things. In an editio maior I would want to make room for information from the transmission history. This would include a positive apparatus, lists of manuscripts supporting various reading and information about the translation technique of the versions (as in the Biblia Hebraica Quinta.

    I think that you can make a good case that conjectures played an important role in the 16th century, but that they don't play such an important role now.