Wednesday, July 29, 2020

4QGenk: A Normalized Manuscript

4QGenk (4Q10) is a poorly preserved manuscript that contains 70 partial words. Among these partial words, there exist three variants (I am not counting differences of plene/defective spelling here). Each of these variants all function to replace an uncommon grammatical form of the MT with the more common form. This common denominator suggests to me that scribe of 4QGenk has taken liberties with his exemplar, or perhaps the scribe has copied from a manuscript that has taken these liberties, to normalize the grammar of the MT. 
Variants between MT and 4QGenk
Fragment, Line, and Verse
F1:L1 (Gen 1:9)
And let it appear
And let it appear
F2:L3 (Gen 1:14)
ולש֯[נים.  ]
and fory[ears]
and years
F5:L2 (Gen 3:1)

The difference persevered at Genesis 1:14 concerns a preposition, but the difference is minor since the scribe is simply making the grammar of the MT explicit. In the MT, one preposition governs two nouns – “days and years.” Although a preposition can govern more than one noun in a sequence in biblical Hebrew (the reading of the MT), the more typical form is to repeat the preposition after each noun (see GKC §119hh). The scribe has replaced the uncommon form of the MT with the more common form.

The same explanation applies to the addition of the interrogative heh in 4QGenk at F5:L2 (Gen 3:1). Interrogative statements can be expressed in Hebrew by an added particle (4QGenk) or by intonation (MT [See GKC, §150a]). Although both forms are possible, interrogative statements are most often marked with an interrogative particle (see GKC, §150c). 

Finally, the fragmentary nature of this manuscript complicates a certain understanding of the last difference. What is certain at F1:L1 (Gen 1:9) is that 4QGenk reads ותרא while Leningrad reads וְתֵרָאֶה. 4QGenhas omitted the final heh of the verb ראה. The form of the MT is a jussive (a tense of volition), but the form is uncommon. The typical jussive of a third yod/vav is an apocopated form: that is, the final heh is lost. It is not surprising given the other two variants in this manuscript that the scribe of 4QGenk provides this form: the more common form.

Tov labels this manuscript as non-aligned, and this categorization may give the impression that this manuscript provides positive evidence that the OT text existed in a state of diversity without any unity. This impression would be wrong. Rather, this manuscript suggests, in my mind, the presence of a stable text existing alongside a diversity of texts. Moreover, this manuscript demonstrates that some of the textual diversity preserved in the non-align category is the result of scribes normalizing uncommon forms. This tendency does not suggest the absence of an authoritative text in the Second Temple period; rather it supports its existence. 

*For more information about this manuscript, see my dissertation, A Comparison of the Non-Aligned Texts of Qumran to the Masoretic Text. It can be accessed on ProQuest. You may also view the presentation that I gave at the Sacred Words Conference.


  1. This comment has been removed by the author.

    1. I see that Kennicott also cites a Masoretic manuscript that has ולשנים in Gen 1:14.

  2. Eric, I looked at this variant and it looks like the Hebrew manuscript is numbered 650 by Kennicott. I looked up this manuscript, and unless I am mistaken, the reference refers to the Jerusalem Talmud. I was able to varify this in the English edition on Accordance. It is a reference to y. Berakhot 9:2.

    I think this may prove my point acutally. Notice that the overwhelming reading is to omit the preposition in the manuscript tradition. When you look at the Jerusalem Talmud, this verse is actually quoted 3 times. In the other two instances, the preposition is omitted.

    Its presence in 4QGenk, and in the Jerusalem Talmud, illustrates how easy it is to normalize the uncommon form of the MT (when I refer to MT, I am referring to Leningrad).

    Thanks for pointing this out, Eric.


  3. Hello Anthony,

    Seeing as the LXX reads καὶ εἰς ἡμέρας καὶ εἰς ἐνιαυτούς and not just καὶ εἰς ἡμέρας καὶ ἐνιαυτούς, wouldn't this indicate the LXX translator may've been working from a similar manuscript to 4QGenk, or is this just down to the translation technique of the LXX translator? (The other two variants aren't likely to result in a different translation to be able to demonstrate a possible manuscript correlation.)

    Furthermore, how do we know it's the copyist who has made these changes, and not just a reflection of the exemplar used? Are there any other uncommon grammatical occurrences in the extant 70 words which haven't been changed to more common ones, or is the manuscript just too fragmentary to give an answer?

    Thanks very much :)

    1. Stewalch,

      Thank you for your helpful observations. These decisions are always hard to determine, and of course, scholars often disagree on these issues. With that said, we should note A.T. Robertson’s discussion of prepositions in sequences. He says that “when several nouns are used with the same preposition the preposition is repeated rather more frequently than in the earlier Greek” (566). He nuances this position, but this statement seems to indicate that the normal Greek convention is to repeat the preposition – the reading of the LXX. The same is true in Hebrew – the reading of 4QGenk. Now, Wevers Notes on the Greek Text of Genesis would help us here, but I do not have a personal copy. Overall, this evidence about prepositions provides some warrant for seeing both scribes making these adjustments independent of one another.

      However, let’s assume that the LXX here is translating from a Hebrew text like 4QGenk. I think my point is still valid since agreement between the LXX and 4QGenk might only show that the LXX was translated from a resignified text.

      Your second question is also a very good one. I am not sure we can determine the answer here given how fragmentary this manuscript is.


    2. Hello Anthony,

      Thanks for your response. Regarding Robertson, I seem to read it differently to you: Robertson seems to be saying pre-NT Greek doesn't repeat the preps. as frequently as found in the NT, though he doesn't elaborate much on this. I found this somewhat helpful discussion, which though not exactly what would be needed to say for definite on repeated preps. in the LXX, does at least give some cause for thinking the repeated prep. in Gen 1:14 is more likely a reflection of the Hebrew Vorlage than not. The LXX Ezek. repeats the prep. 84% of the time the opportunity arises (which would be 100% of the time in the Hebrew), whereas Greek authors such as Thucydides and Luke were at the opposite end at 22%. One can't help think the substantial diff. between the translational Greek of LXX Ezek. and Greek-only authors with very little to no Semitic influence is a result of the frequent repetition in the underlying Hebrew. Obviously a full on study on the use of repeated prepositions with coordinated items in the LXX and other Greek literature would be much more helpful in determining one way or another. I do however agree that this doesn't invalidate your overarching point - I was more thinking of other ways to explain a difference, which doesn't implicate the copyist was doing some conscious editing of sorts. I'll say more when I respond to your other comment below :)

  4. Hello Anthony,

    Just checking the manuscript at Gen 1:9 ( how confident are you the copyist has changed וְתֵרָאֶ֖ה to ותרא, and hasn't in fact just omitted the final ה due to haplography, as הַיַּבָּשָׁ֑ה follows straight after? Also couldn't an early copyist have accidentally changed ותרא to ותראה due to dittography for the same reason הַיַּבָּשָׁ֑ה follows, and therefore that was the early change and 4QGenk actually preserves the original?

    The same also applies to Genesis 3:1: האשה precedes האף, so couldn't the copyist have accidentally added the ה before אף due to dittography, or it was omitted due to haplography to create the MT reading?

    I'm trying to see if your statement "This common denominator suggests to me that scribe of 4QGenk has taken liberties with his exemplar, or perhaps the scribe has copied from a manuscript that has taken these liberties, to normalize the grammar of the MT" is in fact a true reflection of the manuscript, or we are looking not at conscious decisions to change the text from uncommon forms to common ones, but possibly just the results of copyist errors.

    The only one that would look like a conscious change would be the variant at Genesis 1:14, but as per my other comment this could indicate an exemplar similar to the one used for the LXX was also the source of this reading.

    I haven't read your dissertation, so don't know whether you've already looked at the above as possible explanations for the 4QGenk readings. In any case, I don't think we can be as confident as you seem to be in your statements concerning the copyist, as "this manuscript demonstrates that some of the textual diversity preserved in the non-align category is the result of scribes normalizing uncommon forms" would be untrue if we're looking at readings that have come about due to copyist error, rather than conscious decisions to change the text.

    It's a shame there're no other DSS manuscripts which overlap in these places as 4QGenk, as we may've had more manuscripts with the same readings that could possibly indicate a different textual family.

    Thanks very much :)

    1. Great observation. No, I do not discuss these other explanations in my dissertation. Here are some of my thoughts:

      First, there are other non-aligned manuscripts that replace common forms for uncommon ones. Thus, the overarching point is still valid - some of the diverstiy derives from "normalization."

      Second, your explanation is possible. Scribes do at times omit one letter because of scribal error. This explanation is especially clear when the reading is corrected. With that said, in my experience with the non-aligned texts, normalizing uncommon forms seems to represent more of the data. Your explanation would be stronger if there was a scribal error that resulted from parablepsis of one letter or if the mansucript preserved a one letter correction. We have neither.

      Yes, we would have many more answers if the manuscripts were not so fragmentary.

    2. Finally, I would point out that Davila, the editor of 4QGenk in DJD, actually argues that the form of Gen 1:9 in 4QGenk agrees with the LXX. Davila, here, reads the form of 4QGenk as a vav-consecutive imperfect. Both forms are possible - the indicitive or the jussive. He bases his argument on reconstruction.


    3. Hello Anthony,

      Thanks for your second response. I certainly wasn't trying to suggest your overall point was invalid; I'm concentrating just on the text and copyist of 4QGenk, and looking at other ways to explain the readings contained therein. It could be all three are conscious decisions to normalise the MT text - but they also very much couldn't be, as copyist error not only explains at least two of the 4QGenk readings, but also could explain the MT text in the same two places. That we end up with two uncommon forms in the MT would actually, IMHO, show that the MT text is wrong in these cases. However most of the time commentators will draw on the maxim "prefer the harder reading", as we assume copyists would be more likely to stabilise a text they were copying, rather than create harder-to-explain readings. Nevertheless as the MT and 4QGenk readings aren't nonsense, this makes decisions such as this quite difficult. We do have evidence the 4QGenk copyist made errors on occasion - the erasure in Gen 1:27 demonstrates this quite well. Could that not also be the case when it comes to the omission or addition of one single letter?

      With regards to Davila's comments on Gen 1:9, reading what he says I see he actually argues 4QGenk here isn't reflecting the MT text that ends וְתֵרָאֶ֖ה הַיַּבָּשָׁ֑ה וַֽיְהִי־כֵֽן, which the LXX translates as καὶ ὀφθήτω ἡ ξηρά· καὶ ἐγένετο οὕτως, but rather represents the longer LXX addition and so 4QGenk's ותרא היב[שה] corresponds to the LXX's later καὶ ὤφθη ἡ ξηρά. One would then presume the MT's text which ends וְתֵרָאֶ֖ה הַיַּבָּשָׁ֑ה וַֽיְהִי־כֵֽן originally occurred earlier on in 4QGenk, which unfortunately is no longer extant. This would then be a counter point to your explanation 4QGenk has normalised the MT reading: it is rather a reflection of the LXX Vorlage and its longer text. Clearly the manuscript is too fragmentary to determine one way or another in this case, although we possibly may have further evidence in the Gen 1:14 reading. But again, we're not dealing with anything certain.

      One last comment (I seem to have a lot; apologies): I personally don't see how one can, with these three variants, state that either the Vorlage or the copyist of 4QGenk has possibly "taken liberties" with the text. Maybe it's a difference between UK and USA understanding (or more likely just mine), but when I read "liberties", I was envisaging something along the lines of reworked verses, or substantial paraphrases of the text; but rather we're discussing three very, very minor alterations of either the addition or omission of one single letter, two of which may in fact represent a slightly different Vorlage that was like the LXX as opposed to the MT (though even here, the MT and LXX *are* alike regardless). I see this sort of language when it comes to other copyists, especially NT manuscript ones of the early papyri, where commentators seem quite content with being able to use matter-of-fact statements about what a copyist has or hasn't done. I don't think we can be quite as certain as some appear to be when it comes to explaining variants as conscious edits of the copyist in question, and personally I think we should be giving copyists the benefit of the doubt when it comes to variants seen in a manuscript. We should be looking at all explanations for a reading, and not being quite so satisfied one explanation is enough, even if it may appear to be so. :)

      Either way none of the above takes away from your main point about the non-aligned texts being based on an MT-akin text. I'm very much enjoying the Blog posts, and I expect further ones will demonstrate your point even further. :)

  5. "Rather, this manuscript suggests, in my mind, the presence of a stable text existing alongside a diversity of texts." With all respect, it seems to me that what this particular MS shows is that some scribes were lax when it came to representing (what in by their day had become) archaic or unusual forms.

    "Moreover, this manuscript demonstrates that some of the textual diversity preserved in the non-align category is the result of scribes normalizing uncommon forms." Agreed.

    "This tendency does not suggest the absence of an authoritative text in the Second Temple period; rather it supports its existence." This seems to say too much, at least if we are dilating on 4Q10 as its relates to the MT. I do agree that there was an authoritative text in the Second Temple period, but to infer such specifically from an updated-spelling edition like 4Q10 seems like a stretch. But I think this is just a small part of a larger theory, so I quite possibly I'm not seeing the whole picture here.

    1. Stephen,

      Thank you for your comments. Have you looked at Andrew Teeter's book Scribal Laws? I have only looked at it briefly, but Peter Gentry's presentation at the Sacred Words conference discussed some of his main conclusions. He states that resignfied texts actually depend on a stable text. That is the idea that I am referring to here.

    2. Yeah, and actually I'm in general agreement with Teeter's approach to the nature of the complementary nature of interventionist texts (that is, updated and amplified texts, and presumably truncated ones as well, were meant to stand along side of, not replace, conservatively transmitted texts). But there's a lot more to the argument than saying, here we have a linguistically-updated text, so clearly that implies the existence of an authoritative, conservatively-transmitted text. I think my position overall would be pretty close to yours, but - and this could just be my mis-perception here - it seems like you're jumping to a conclusion based on incomplete evidence, making a MS like this one have greater significance than, in itself, it does.

    3. Stephen,

      I agree with you. Maybe I should clarify what I was trying to do in this blogpost. It is entirely likely that I was not clear. This blogpost fits into a wider blog series. In this series, I will investigate several non-aligned texts and hopefully show that good reasons exist for seeing many of these manuscripts as secondary and dependent on an MT-like text. My hope is to show that the textual diversity we see at Qumran does not necessarily indicate textual chaos.

      My discussion of 4QGenk is not as follows: here is a resignified text that depends on an MT-like text. Thus, an MT-like text is the stable text. Again, this is not my argument.

      Rather, my argument for this series is this: here are several resignified texts that appear to depend on an MT-like text. Thus, the evidence of the non-aligned texts suggests that an MT-like text was the stable text during the Second Temple period. I do look at all of the non-aligned texts in my dissertation. This series is an abbreviation of that work.

      Of course, I understand that the evidence is not limited to the non-aligned texts. There is a lot of other important evidence available. I have not yet dealt with the other evidence in detail, but I do address a few other items in my Sacred Words presentation. Some of the other topics I address there are the other Judean Desert manuscripts, Masoretic notations, the Letter of Aristeas, Josephus, and I believe a few others.

      Does this help clarify my argument?


    4. Yeah, that helps a lot, and, for what it's worth, makes me much happier with your line of argument. Thanks for engaging and clarifying!

    5. Stephen,

      I am glad. Thanks for your thoughtful interaction; I am really enjoying the dialogue. It helps sharpen my thoughts.


  6. Anthony, could you elaborate on what you mean y the phrase "authoritative text"?

    Certainly the texts contained in all of the manuscripts of Genesis among the DSS were regarded as authoritative by those who made and read them, regardless of the variants they each had that made them differ slightly from one another and from other witnesses that we now have for the text of Genesis.

    But I gather you mean something more than that, along the lines of some authoritative text that could have served them as a standard to decide which variants were right, and that this standard was very close to the same text as what we find in the Leningrad Codex. If you mean something like this, then this sounds a lot like the concept of an ecclesiastical or canonical text as advocated by some especially for the NT. And I wonder how much of this kind of a concept you meant to pack into that phrase "authoritative text."

    Personally, I'm very open to such an idea, probably more for the scriptures of Israel in the Second Temple period than for NT manuscripts much later on, in part because I tend to think that the Temple itself, where the Temple authorities had their own texts that were stored and read there, likely served as a unifying force affecting the copying and reading of texts in other places by those who occasionally worshiped in that Temple (as well as those whose forbears used to worship there but broke away, like may have been the case with the scribes and readers of the DSS).

    1. What I mean in my second paragraph above is that even the Genesis manuscripts among the DSS that differ the very most from the MT were still authoritative texts. And we might extend our perspective beyond that community to say that LXX Genesis was an authortative text for those diaspora Jews who read it, and the text of Genesis in the Samaritan Pentateuch was authoritative for the Samaritans.

      And then even more significant perhaps, if we return our perspective back again to the DSS, it's probably fair to say that even the text in the Reworked Pentateuch was an authoritative textual revision of Genesis, and Jubilees, which was more extreme than that, but still based on Genesis, was an authoritative text.

      I guess what I'm driving at is that, as comfortable as I am with the idea that a proto-masoretic text had some kind of preeminence in Second Temple Judaism, it's still also the case that variations from that standard were both tolerated and treated as authoritative, and in some quarters (particularly among sectarian quarters reflected in the DSS) this tolerance for textual revisionism alongside continued reverence for the proto-masoretic textform may have been remarkable.

      These are the kinds of caveats that seem necessary to me when using the phrase "authoritative text" in this context.

    2. Yes, I’d be happy to do that.

      I affirm that simply because a text is resignified does not mean it lacked authority. The Samaritan Pentateuch and the LXX were both likely copied from resignified texts and they remained authoritative in their respective communities (excerpted texts too may have been made from resignified copies). However, I am not sure that I agree with your statement that these texts were authoritative “regardless of the variants they each had.” Would you mind explaining this?

      In my mind, broadly speaking, Jews understood that a stable text existed amid a diversity of texts. That stable text was the temple text. The author of the Letter of Aristeas alludes to this idea. Notice that he claims people should trust his version because his text was copied from a text kept by the temple authorities. Josephus too makes this point. When he is discussing the unchangeability of the Law, he is not referring to the manuscripts at Qumran; he is referring to the texts of the temple (I believe Arie van der Kooij made this point). The fact that every LXX revision/recension updates its base text to fit a proto-Masoretic text too suggests this point. These scribes were not ok with diversity but updated their texts to fit the text they knew was superior. There is more evidence, but this is sufficient in my mind to at least strongly suggest that Jews of the Second Temple period were not ok with textual diversity.

      Thus, I would say that for various reasons textual diversity existed in the Second Temple period. Amid that diversity, Jews understand that the best text, the canonical text, was found in the temple.

      Does this help clarify my view at all?


    3. Thanks. I generally agree. But I'm uncomfortable with the way you use the word "canonical" for the Temple text (not just the delimitation of which books are scripture, but the particular readings of those books over against variants in other copies).

      What I mean by the phrase you quoted above is that it appears to me that early Christians and their Jewish contemporaries were very comfortable with textual variants in the copies of their scriptures. Minor divergences in wording from the copies at the Temple generally didn't make the copies they were using any less authoritative. And I think that the very explanations you gave for the differences between 4QGenk and the Leningrad Codex imply that the scribes who made those changes themselves had a view of the text that allowed for such changes to be made without a loss of authority.