Tuesday, December 20, 2005

Hengel on TC

I'm working through the volume edited by Charles Horton on P45. The essay by Martin Hengel, 'The Four Gospels and the One Gospel of Jesus Christ', is basically a summary of his book bearing the same title (London: SCM, 2000). It is an exceptionally useful distillation. One sentence that stood out was this:

'The text of the Gospels is the best-transmitted text in antiquity.' (p. 15)

I actually think that the text of some of the epistles might have been a better choice and better still might have been texts from the OT: proto-Masoretic mss from Qumran show a higher level of agreement with mss from about a millennium later.

Hengel's statement did, however, make me wonder how much scholars have actually compared the NT and Classical transmission processes from the perspective of their fidelity (rather than just observing that there are usually vastly more witnesses for the NT text).

What are the main investigations of this area?


  1. Well, given that so many documents of antiquity came down to us in a single manuscript (Beowulf comes to mind), the question of fidelity can't always even be addressed.

    There has been one study on the Historicity of Beowulf in the book "After the Flood" most of which can be googled.

  2. Is Hengel's statement primarily about quantity or quality?

  3. One problem with Hengel's cited comment is that it over-generalises to 'the text of the Gospels'. Our resources for the text of Mark (which are largely absent and/or problematic) are of quite a different character to those for the text of John.

  4. Homer is the best candidate for TC based on manuscript number. I googled "homer manuscript" and found some unreliable sites saying there are about 650 extant manuscripts. Wikipedia notes that one manuscript is particularly important in the study. If you know much about classical manuscripts, you can guess its background... 10th century. Most of the manuscripts that have survived did so through the Carolingian Renaissance (in this case shortly afterwards). Homer is also tricky due to the oral tradition behind it and the Ptolemaic redactions of the text.

    If one wandered into Latin, the Aeneid might provide a venue for exploration. Unfortunately, we only have a few manuscripts before the fifth century (if only they had read more Latin in Egypt!)

  5. Daniel,
    If you're referring to W.R. Cooper's book After the Flood then the work is notoriously misjudged in matters of history. Cooper did, however, make a good transcription of Tyndale's NT.

    Hengel's comment seems to be about quality of transmission.

    The thing is not really to do with numbers of mss. We all know about F.F. Bruce's argument The New Testament Documents: Are They Reliable? (picked up by Josh McDowell and now found round the world) that there are more mss of the NT than of Classical works (though in some ways of counting Homer might rival the Apocalypse). The question is rather of quality. Now I'm prepared to maintain that the quality of the NT transmission is pretty high; others would disagree. The question is, have there been any estimates of textual reliability for Classical works? How many letters per thousand in the Aeneid might be in serious doubt - and one what grounds? (Obviously 'doubt' needs to be defined.)

  6. This comment has been removed by a blog administrator.

  7. Regarding the quality of transmission of certain books of the OT, because the biblical mss from the DSS are still in all cases centuries after the originals, their similarity with the MT (inasmuch as such similarity exists) only proves fidelity of the MT to that text-form of a given book that was preserved therein. But it does not prove the fidelity of that text-form itself to the original. Isaiah is a good example of this, since 1QIsab is fairly close to the MT (much closer, for example, than a byzantine ms of Mark is to Sinaiticus). But then we also have 1QIsaa and several 4Q Isaiah scrolls and the LXX Vorlage, all of which show greater divergences from MT than does 1QIsab. And when the question comes up of what the original looked like, the case for the MT is not a homerun. All this is, of course, leaving aside arguments about the unity of the book, which have an unavoidable impact on even defining what is meant by the "original". We can certainly say that the OT text was preserved better than the NT text from the period of the 1st century AD until now. We just can't say (with certainty) that this applies to its earlier transmission history.

  8. I would think that up to the seventh century Homer manuscripts must outnumber NT manuscripts. Quantity, not quality (but at least it is possible to judge).

  9. Eric, You are right that we cannot have empirically derived certainty about the prior transmission of the Proto-MT. However, we could never have empirically based certainty in relation to any text that it has not been changed.

    I think Tov puts some Proto-MT mss back in the second century BC. Moreover, the lack of waw for the /o:/ vowel which derives from proto-Semitic /a:/, alongside the presence of waw for /o:/ from proto-Semitic /aw/ is most simply explained by positing that there were fixed rules for the transmission of spelling at least back to the time when proto-Semitic /aw/ and /a:/ with Canaanite shift became indistinguishable in pronunciation. This could have been well before the second century.

  10. It is true that the Masoretic text (and proto-MT) has a more archaic spelling than those DSS that differ from it the most. And I personally do agree that that is a point in its favor. But it isn't quite clear that a more ancient spelling or orthography accompanies a more ancient text.

    The numerous matres lectiones in 1QIsaa, for example, are certainly secondary, but the larger variants in this same scroll may occasionally be older than the MT, at least in theory. This is illustrated very well, I think, in the paleo-Hebrew scroll of Exodus, which exhibits a number of distinctly Samaritan readings. The orthography is older than the MT, but the readings are secondary.

    Regarding Tov, it is important to understand his position on TC properly. He does favor the proto-MT as the proper goal of TC. But he does not hold that it is to be taken as the original form of the text. He illustrates this with Jeremiah, where he argues that the short form of the LXX Vorlage is primary, but the secondary (for Him) proto-MT text is the proper goal of TC, since it is the text that underlies what we have recieved via the Masoretic tradition.

    I personally do feel very favorable to the MT in OT TC. I just think that the transmission histories of OT books prior to the Hasmonean period are so murky that very little can be said about them in comparison to Hengel's view of the textual transmission of the Gospels.

  11. Eric,
    I think we agree. The interesting question for evangelical textual criticism is what we should do when the data run out. Do all texts that come to us stand on an equal footing? What should we do about the pluriformity of OT texts in the 'pre-70' period? This is a subject that we need to thrash out here some time (perhaps after Christmas!).

    Since you raised the subject of the text of Jeremiah, it may be worthwhile considering that while Tov's approach, and that of J. Gerald Janzen, Studies in the Text of Jeremiah (HSM 6, Cambridge, MA: Harvard
    University, 1973) have often been held to be superior to that of Sven Soderlund, The Greek Text of Jeremiah: A Revised Hypothesis (JSOTSup 47; Sheffield: JSOT, 1985), Andrew G. Shead, The Open Book and the Sealed Book: Jeremiah 32
    in its Hebrew and Greek Recensions
    (Sheffield Academic Press, 2002) has argued that the differences between the G-text and MT cannot be explained by any simple textual process. Jack Lundbom also appears to have been producing some rather interesting work on the text, which again, suggests that the simple two-fold division of the texts of Jeremiah does not account for all the data. Unfortunately I've not had time to go through Lundbom's Anchor Commentary on Jeremiah. In the mean time I would maintain that the LXX Vorlage is not a completely known entity.

  12. Regarding the LXX Vorlage, I'd like to see more work on ms 88, which contains an alternate version of the Greek in at least the book of Daniel.