The Dead Sea Scrolls: Psalms, Lamentations, Numbers
by James Duguid | April 17, 2026
There is a new rotation of scrolls on exhibit at the Museum of the Bible in DC. I thought I’d put together a new post about the biblical manuscripts in particular, looking at interesting text critical readings. For an introduction to the topic, seemy first post here. To see all my posts so far, click here.
11Q5 Psalmsa
11Q5 Psalmsa is by far the most interesting psalm manuscript from the Dead Sea Scrolls because it is so different from the psalter that has come down to us today. For starters, the order of the psalms is different. On top of that, there are a number of psalms not found in our canonical tradition. The apparent order runs Psalms 101-104; 109; 105; 146; 148; 121-132; 119; 135-136; 118; 145; Psalm 154 (known from Syriac); Plea for Deliverance; Psalm 139; Psalm 137-138; Sirach 51:13ff; Apostraphe to Zion; Psalm 93; Psalm 141; Psalm 133; Psalm 144; Psalm 155 (known from Syriac); Psalm 142-143; Psalm 149; Psalm 150; Hymn to the Creator; 2 Samuel 23; David’s Composition; Psalm 140; Psalm 134; Psalm 151 (known from the Old Greek). As you can see, some of the extra psalms are known from Greek and Syriac translations, while others are only attested here at Qumran. One can only imagine what the rest of the psalter would look like if the whole book had been preserved. A document like this can only serve as fodder for scholarly theories, but there is a good argument to be made that the version in the Masoretic Text is older than these other editions – see Drew Goodacre’s discussion here.
A book like the Psalter would have undergone several updatings over time, as new songs needed to be incorporated. This manuscript makes it clear that this updating could also involve re-ordering of the previously existing psalms to fit a new editorial vision. This in turn suggests the order of the psalms was important to readers at this time, though it can be difficult to guess exactly why each psalm has been put where. There are also many interesting differences in readings for individual psalms in this manuscript.
Another feature of note for this manuscript is that the name of the Lord is written in a different script than the rest of the manuscript. The body of the text is written in the usual Aramaic block script you are probably familiar with in Hebrew text. But the name of the Lord is in the old paleo-Hebrew script, as a signal of its sacredness. This means that it stands out in a way even someone who does not read Hebrew can identify.
The fragments that are here in DC contain Psalm 146; 148; 121; 124-131. One important reading in this span is Psalm 129:3 (I have bolded words with significant textual differences):
Over my back the wicked have plowed, they have made their furrows long.11Q5 Psalm 129:3
Where 11Q5 Psalmsa has “wicked,” the Masoretic Text has “plowers,” the cognate noun for the verb. The Vulgate, Targum, and Peshitta agree with the Masoretic Text. But the Old Greek agrees with 11Q5 Psalmsa. The Greek translators Aquila, Theodotion, and Symmachus all correct the Greek to agree with the Masoretic Text. Thus 11Q5 Psalmsa provides important support for the reading in the Old Greek.
4Q111 Lamentations
The fragments that are here in DC contain Lamentations 1:1-18 and 2:5. As I reviewed this portion of the manuscript, I was developing a low view of its quality. Where there are textual differences, 4Q111 Lamentations is usually without support by other manuscripts and clearly in error (Frank Moore Cross says in his edition, “The scribe of 4QLam is often careless”). But I did find an important variant in Lamentations 1:12. I have bolded words with significant textual differences.
4Q111 Lamentations 1:12Is it nothing to you, all who pass by on the w[ay? Look and se]e if there is pain like my pain, which he has done to me, when the Lord terrified me in the day of the anger of his wrath.
4Q111 Lamentations has “terrified me” where the Masoretic Text has simply “inflicted.” Cross mentions that the Old Greek and Peshitta agree with 4Q111 Lamentations in having the object “me” (Targum Lamentations also has it), though I would register some doubt here, as this object might be required for smooth reading in translation. In other words, it might have been added by the translator even if it wasn't in their Hebrew source text. Still, the reading of 4Q111 Lamentations is three letters longer than the Masoretic Text, and an abbreviation is the more likely error here. The verb used in 4Q111 Lamentations is a rare verb, and this is the only time it is used in this stem, whereas the verb in the Masoretic Text is also found in verse 5 – it seems more likely a scribe would err by conforming the rare word to the common one. So the reading in 4Q111 Lamentations is a good candidate to be the original reading. This is a reminder that even if a poor scribe makes many mistakes, the manuscript he is copying from may contain rare readings of value.
4Q27 Numbersb
4Q27 Numbersb is the most important witness to Numbers from the Dead Sea Scrolls, with substantial portions of Numbers 11-36 preserved.
The fragment that is here in DC contains Numbers 18:25-19:6. Here is a translation of Numbers 18:30-31 from this fragment, which concerns the command for the Levites to offer to the Lord a tithe of the tithe they receive. I have bolded words with significant textual differences.
30 [And you shall say to them,] when you offer up its best portion from it, your contribution shall be reckoned to you as pro[duce from the threshing floor and as produce] from the winepress. 31 You may eat them in every place, you and your house, [for it is wages for you] in return for your work in the tent of meeting.4Q27 Numbers 18:30-31
- Where 4Q27 Numbersb has “your contribution will be reckoned to you,” the Masoretic Text has “it will be reckoned to the Levites.” The Old Greek and the Samaritan Pentateuch agree with the Masoretic Text here, as do all the Targums and the Syriac Peshitta. But the Vulgate has the words “to you” instead of “to the Levites,” agreeing with Numbersb. Meanwhile, only 4Q27 Numbersb has the words “your contribution.” Of course, in context Moses is speaking to the Levites, so “you” and “the Levites” would have the same referent. In fact, the words “your contribution shall be reckoned to you” already occurred in verse 27. Probably this reading arose by a scribe accidentally conforming this verse to the one he had written just before. The fact that it is found in the Vulgate means that this reading was still around in Hebrew manuscripts in the 4th century when Jerome made his translation.
- 4Q27 Numbersb has “from the winepress,” while the Masoretic Text has “of the winepress.” The difference is only the addition of an explicit preposition. But the Old Greek also seems to have the preposition, as does Targum Neofiti. The Samaritan Pentateuch, Targum Onqelos and Pseudo Jonathan, the Vulgate and the Peshitta agree with the Masoretic text. Again, this phrasing conforms to verse 27.
- 4Q27 Numbersb has “you may eat them,” the Masoretic Text has “you may eat it.” The Samaritan Pentateuch, Targums, and Peshitta agree with the Masoretic Text. The Vulgate agrees with 4Q27 Numbersb. But Greek manuscripts are split, with some important manuscripts supporting either reading. Probably this means that the Old Greek agreed with 4Q27 Numbersb, and later Greek translations corrected towards an early form of the Masoretic text.
These differences don’t change the meaning of the command, but they do illustrate a point: the variant readings represented in 4Q27 Numbersb still exist in later translations of the Bible. This confirms that many of these variants in translation do indeed go back to differences in ancient Hebrew manuscripts. Thus, we should not rely solely on the Masoretic text, but should give some weight to these kind of variants. Of course, when we have evidence from the Dead Sea Scrolls that supports these variants, we can give them greater weight. Still, the Masoretic Text is likely to represent the original meaning for most of the variants here, illustrating that it is a high quality tradition even if it may have some textual errors.