Wednesday, September 20, 2006

Orr on Lundbom on LXX Jeremiah

[PJW: I have received the following contribution from Peter Orr]

As a undergraduate I have been recently exposed to the wonderful world of text-criticism. One of the cases I have spent a little bit of time looking at has been that of Jeremiah in the LXX and MT. I thought your readers might be interested in a summary of some of Jack Lundbom’s work on this subject.

Lundbom has argued that haplography has affected the LXX of Jeremiah to an extent that has not been recognised before and that severely questions its quality as a text. In a 1999 article with David Freedman he examined Jeremiah 1-20 and argued that there are around 50 cases of haplography in LXXV. When he came to write his commentaries on Jeremiah 21-52, he imagined he would find more expansion in the MT since the section contains more prose. However, he actually found an additional 278 cases of haplography in the LXX. This gives a total of 330 losses in LXX due to haplography - ‘an extraordinary number by any measure’.

Lundbom does acknowledge that some haplographies are clearer than others. So he points out that in Jeremiah 5:15 of the MT we read: גוי איתן הוא גוי מעולם הוא גוי לא־תדע לשנו. If we retrovert the LXX it reads: גוי לא־תדע לשנו. This seems to be a clear case where the scribe’s eye skipped from the first to the second גוי. However, in Jeremiah 1:15 we read: לכל־משפחות ממלכות צפונה in the MT, while in LXXV משפחות is missing. It is not altogether clear if this is a case of the secondary addition of the text under the influence of Jeremiah 25:9 where the word משפחות appears in a similar context; a conflation of two variant readings since משפחות and ממלכות are interchangeable elsewhere (cf. Jeremiah 10:25 and Psalm 97:6) or in fact a case of haplography where the scribe skipped from the מ of משפחות to the מ at the beginning of ממלכות. Lundbom argues that while each case should be investigated individually, haplography should be the default explanation since expansion is ‘complex and difficult both to pin down and explain, since it requires an inquiry into the mental operations of a scribe’. With haplography, we have ‘a well-established and perfectly objective text-critical method [by which] a considerable number of differences between the two texts can be explained’. In many of these cases, such as the above, there may be no way to settle the issue, but Occam’s razor should prevail and ‘parablepsis is a better solution than its alternative’.

Far from having a superior text, Lundbom concludes bluntly that ‘the LXX translator(s) of Jeremiah had the misfortune of working from a “bad Hebrew Bible”’ and that we should be ‘glad that more Jeremiah passages were not quoted by writers of the NT, for whom the LXX Bible was normative’.

Lundbom has recently published a summary of his work: ‘Haplography in the Hebrew Vorlage of LXX Jeremiah,’ Hebrew Studies 46 (2005): 301-320.
I was wondering what your readers think of both his conclusion regarding Jeremiah in the LXX and his contention that we should prefer the ‘objective’ explanation of haplography by default.


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

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

  3. THird times the charm
    : )

    Is Hebrew more prone to copyist mistakes than, e.g., Greek? Repetition is a key component of the language, and some of the letters look very similar (more so than I would say in Greek):
    ב כ
    ו ג ן נ
    ם מ
    ד ר

  4. I may be prejudiced but I find old uncial Greek manuscripts more prone to misreading than Hebrew manuscripts. Hebrew keeps the words separate and the eye is able to take in the whole word better than in Greek. Without word boundaries it is easy to make a 'false start' in Greek and then discover a few syllables later that you need to reread.

    My eyes don't clearly distinguish the letters that you've listed in the morning paper without glasses. But I have usually finished reading the paper before I consider going to another room to get glasses. So it is possible to read without vowels and with fuzzy dalet/bet/nun's. This would be less true in first temple times, though, with less מלא-spelling.