Qumran, Qaraites and Controversies: Mosseri I.40
While some Genizah fragments are perennial favourites – and even have their own Wikipedia pages – others had a brief moment of fame before fizzling away and dropping out of sight. This month’s fragment is one of those latter, as the sparse publication record since its original discovery will show.
Beyond individual fragments, there are a few groups of manuscripts whose fame extends beyond the usual audience of Genizah scholars too. Among these, the ‘Second Temple texts’ are pre-eminent. These manuscripts, whose textual origins lie in the period between the construction of the Second Temple in the 6th c BCE and its destruction in 70 CE, have somehow been preserved in the amber of the Fusṭāṭ Genizah, despite largely dropping out of the historical record elsewhere.
Foremost among these are the fragments of the six Genizah manuscripts of Ben Sira – with copies a thousand years older subsequently turning up in caves 2 and 11 at Qumran, and at Masada. The first Ben Sira discoveries were followed rapidly, thanks to Solomon Schechter’s remarkable eye for the unusual, by the two manuscripts of the Zadokite Work, T-S 10K6 and T-S 16.311 . Later on, numerous fragments of its forebear, the Damascus Document, were found in Qumran caves 4 and 5. Added to these is the single parchment manuscript of Aramaic Levi in the Cambridge, Oxford and Manchester Genizah collections, which preserves a text probably composed in the 3rd c. BCE. Fragments were identified in Qumran caves 1 and 4.
What is the relationship between the Second Temple texts of the Genizah and the fragment under discussion here, Mosseri I.40? The first publication of the Mosseri fragment suggested a thematic and historical link to this group of Second Temple manuscripts. Since then, except for the occasional appearance in footnotes, the Mosseri manuscript had mostly been forgotten, until this year.
Mosseri I.40 is a small, torn piece of parchment with only a few lines of Hebrew text preserved. It was revealed by Israel Lévi way back in 1913 in an article called ‘Document relatif à la «Communauté des fils de Sadoc»’, drawing attention to its relationship with the Zadokite Work (Damascus Document) discovered only a few years before by Solomon Schechter: ‘L’intérèt que présente ce texte, formé d’un tronçon de parchemin, est tout entire dans la mention de «la Communauté des fils de Sadoc»’ (‘The interest presented by this text, formed of a section of parchment, is entirely in the mention of “the Community of the sons of Zadok”’, my translation; Lévi 1913: 24).
Recto of Mosseri I.40
Lévi’s short article recorded that line 3 of the text has the expression עדת בני צדוק, ‘the congregation of the sons of Zadok’, which had made him think of Schechter’s recent discovery. The term bene Ṣadoq occurs twice in Schechter’s Zadokite Work (ff. 2r–v of MS A, T-S 10K6): הכהנים והלוים ובני צדוק אשר שמרו את משמרת מקדשו, ‘The priests and the Levites and the sons of Zadok that kept the charge of His sanctuary’ (Schechter 1910: xxxiv) and ובני צדוק הם בחירי ישראל, ‘And the sons of Zadok are the chosen of Israel’ (Schechter 1910: xxxv).
MS A of the Zadokite Work (Damascus Document), folio 2v
Schechter’s ‘working hypothesis’ in his monograph ‘Documents of Jewish Sectaries’ was that the Zadokite Work was part of a ‘Zadok book’ that had been known to the Qaraite al-Qirqisānī. He did not come down firmly on the origins of the Zadokite sect, nor did he feel that they could be equated with the Sadducees, due to the poverty of ‘the present state of knowledge of the latter’s doctrines and practices’ (Schechter 1910: xxi).1 However, in his reply to the critical review of his book – which questioned its principal conclusions on the identification of the sect – by Adolf Büchler, Schechter stated that he ‘had serious misgivings about the determination of the period of the beginning of the Sect’ (Schechter 1913: 473). On the other hand, he gave strong hints as to the Zadokite Work’s antiquity, throwing a good measure of shade on Büchler at the same time: ‘Any student with something of a familiarity with ancient Jewish literature [i.e., not Adolf Büchler] would at once have been struck by the strange character of this text, and recognized that it belongs to a class of composition to which none of the Hebrew writings of the first ten centuries of our era offer a parallel’ (Schechter 1913: 473). As the Qumran discoveries nearly forty years later revealed, he was entirely correct in that assumption. Lévi, for his part, had stated in his article on the Mosseri fragment that, given the manuscript was written probably no later than the 10th century, ‘Il en résulte donc que la secte, née deux siècles avant l’ère chrétienne, n’était pas encore disparue douze cents ans plus tard’ (‘It follows that the sect, born two centuries before the Christian era, had still not disappeared twelve hundred years later’, my translation; Lévi 1913: 31). This was not a view shared by Solomon Schechter.
Schechter himself first mentioned Lévi’s manuscript in his short rejoinder to Büchler’s critical comments on his book, published as an ‘Announcement’ in the same issue of JQR as the review. Schechter wrote there that he would not be publishing his ‘refutation’ of Büchler yet, despite having formulated it already, as a rumour had reached him that ‘M. Israel Levi, of Paris, has discovered new pages of the Zadokite document which he is about to publish’, and he wished to wait to see this discovery and what bearing it might have on the Zadokite debate (Schechter 1913: 485). When his reply did come out in the next issue, he began by addressing the Lévi discovery: ‘The find of Professor Lévi … being admitted by the editor himself that it forms no part of the “Fragments of a Zadokite Work”, it need not be discussed in these pages’ (Schechter 1914: 449). But, being Schechter, he offered his opinion on the discovery anyway, ‘Personally, I think that the find represents a remainder of a liturgical piece of the well-known type of the Piyuṭ of a later period. In no case has it any bearing upon the “Zadokite” problem’, before giving Büchler both barrels over the next twenty-five pages (Schechter 1914: 449).
And that essentially ended Mosseri I.40’s fifteen minutes of fame, except for a few brief mentions, until this year when I came across ‘Some Observations on a Zadokite Fragment from the Cairo Genizah’ by Corrado Martone (2025: 135–46), which appears in the Festschrift for Mauro Perani, Habent sua fata fragmenta. This article is the first since Lévi to address in detail the fragment, offering a transcription, translation, and commentary on its theological context. The motivation for the article is that now, with the publication of the Dead Sea Scrolls completed, we are in a better position to assess the views of Lévi and Schechter against what we know of the Qumran background. Through his analysis, Martone has concluded that ‘much of the terminology and underlying ideology of this fragment has important features in common with what can be reconstructed concerning the ideology of the Qumran texts’ and, furthermore, ‘we cannot exclude the possibility that this fragment represents a further link in the chain linking Qumran literature to the Karaites by virtue of their shared claim to be part of the priestly lineage of the Sons of Zadok’ (Martone 2025: 145–6).
I certainly would not disagree with that conclusion, and would offer in support the persuasive view of Yoram Erder, the great expert on the relationship between Qumran and the Qaraites, who has been one of the few scholars after Lévi and Schechter to address Mosseri I.40, albeit briefly in a footnote (denoting its junior status amid the wealth of his other arguments): ‘The Karaites may well have realised the importance of the Zadokites from additional ancient texts available to them, as evidenced by a further minor passage from the Geniza which refers to “ʿAdat Benei Zadok”. See Lévi…’ (Erder 2017: 396 n. 493).
Given all this, why am I addressing the fragment myself here? Because I think this fragment is interesting, enigmatic, and deserves to be better known, and I have a few comments of my own to make on it, to supplement those of Lévi and Martone. The fragment can be seen here, on Cambridge Digital Library.
Cambridge University Library Jacques Mosseri Collection, Mosseri I.40
.1 ]כהן
.2 ]ת יודוני מבני [
.3 ]עדת בני צדוק [
.4 ]על חוק טמא וט[
.5 ]משפטיי ישט[
.6 ]קדישו וחול [
.7 ]ם יסובו דוכני[
.8 ]שיר יהללו או[
.9 ]יי עֹז לעמו יתן[
Notes
2. Here, I read differently from both Lévi and Martone, who read יורוני. The problem with that reading (and it’s fair to say that reš and dalet are not easily distinguished in this hand) is one of sense: who is the narrator of this text? ‘They will teach me’ is how Martone translates it in English (Lévi, ‘m’instruisent’), but that gives a problem with reading משפטיי, ‘my judgements’ later on: is God being taught here? That seems unlikely for the Omniscient One. Lévi avoids this problem by seeing the line with משפטיי as an embedded quotation; Martone prefers to read משפט יי, i.e., ‘the judgement of the LORD’. However, there is no problem if we see that this is in fact a ד (compare it to עדת in the following line, which we would otherwise have to read – intriguingly, but wrongly – as מערת) and it is יודוני, ‘they will give thanks to Me’, which is echoed later in the text by יהללו. Lévi thinks that the word lost at the beginning might have been עת. The short stroke preserved can really only be the foot of an ʿayin, and so the phrase בעת יודוני certainly seems likely, ‘when they give thanks to Me’.
Detail showing lines 2–3 of Mosseri I.40
3. We can speculate that it would have read מעדת, paralleling the מבני in the line above.
4. Lévi reconstructs וטהור, but Martone has slight doubts, on the grounds that the phrase is not found in BH, Qumran or ‘postbiblical literature’. It is a natural pairing, however, and in fact we can find a phrase such as this in a piyyuṭ, in a Maʿariv for Šavuʿot (T-S H3.18), which has חוק טמאה וטהרה, ‘a law of uncleanness and purity’, though I am not suggesting a direct link.
5. Lévi read this line משפטיי ישטפֿ (reconstructing the פ), translating it as a quotation ‘juge mes jugements’ (‘judge my judgements’), proposing a lapsus ישטפו for an intended ישפטו. Martone reads ]ישטפ, and also takes it as a misspelling, citing Ezekiel 23:45 in support (ואנשים צדיקם המה ישפטו אותהם משפט נאפות ומשפט שפכות דם כי נאפת הנה ודם בידיהן, ‘And the righteous men they shall judge them after the manner of adulteresses and after the manner of women that shed blood because they are adulteresses and blood is in their hands’).
Martone states that Lévi also takes משפטיי as a lapsus for משפטו (Martone 2025: 140), but Lévi actually says: ‘This line attests better to the use made by the author of this chapter (44) of Ezekiel, because it is a textual borrowing from verse 24 ועל־ריב המה יעמדו למשפט במשפטי ישפטהו. The pronominal affix of the first person is the incontestable proof of a quotation’ (Lévi 1913: 29; my translation and emphasis). Martone must have misread Lévi, as Lévi’s argument explicitly rests on it being the first person, not a lapse for the third. Martone suggests another possibility: to read the two yods at the end of משפטיי as an abbreviation for the divine name, ‘a judgement of God’ – which is how he gives it in his translation: ‘the ordinance of God will judge’ (Martone 2025: 135). Now, we do have the divine name abbreviated in the last line, ineffably so, as it’s embedded in a quotation from Psalms. However, there it may have been ייי, and we’ve just lost the first yod off the edge of the torn page. I think I would not expect two yods written close together like that, and very close to the end of the noun before them, without them being marked with a supralinear dot or stroke, as an abbreviation. So I don’t find Martone’s suggestion convincing. I think, especially given the יודוני in the first line of the text, that there is no problem with reading משפטיי, ‘my judgements’. I shall come back to this line below.
6. Neither Lévi nor Martone like this line as it stands. Though Lévi suggests יקדישו, he would have preferred to read קדש וחול or, even better, קדש לחול (Lévi 1913: 29). I think the text is too short to propose amending it without good reason, however. Martone (2025: 141) declares יקדישו ‘paleographically untenable’. However, I don’t think יקדישו is an impossibility – the scribe probably left a slightly larger space after the yod than expected, but not so different from יסובו in the line beneath. In this case, we might propose it said something like קדשיי אשר יקדישו contrasting it with the profane things, who get their own clause.
7. Both Lévi and Martone understand the last word as דוכן ‘(priest’s) stand’ – the raised platform where the priest pronounces his blessing – one taking it as plural – des estrades (Lévi 1913: 25), the other as singular – ‘the platform’ (Martone 2025: 135). There is a good parallel in the Tosefta (Pesaḥim 4:11) to the scene being depicted here and in the next line: הלוים עומדים על דוכנן וגומרין הלל בשירה (‘The Levites stand on their platform and read (to the end) the Hallel in song’). We might perhaps therefore reconstruct [הלוי]ם at the beginning of the line.
8. Though Martone thinks it imprudent to reconstruct the gaps, because the sequence of letters does not occur in the Bible or Qumran, given the Tosefta parallel, a reconstruction בשיר seems likely, and I’d go the whole way with בשיר יהללו אותי, ‘in song they praise Me’.
9. The last line is a quotation from Psalms 29:11, abbreviating the divine name, and marking a ḥolem on the defective עז.
As is clear, in reading יודוני in the opening line of the text, the character of it is changed, and a number of difficulties fall away. I am not competent to comment on the theological questions that this might lead to, but I’m reasonably happy that it produces a far more coherent text, which does not require the emendations proposed by both Lévi and, a century later, Martone. What about line 5, however, where both scholars assume the scribe of this fragment has mistakenly transposed פ and ט, writing ]ישטפ instead of ]ישפט? If the inspiration for this line is indeed, as proposed by Lévi, Ezekiel 44:24, then it seems perfectly acceptable (Lévi 1913: 29). However, it is a very short text and one that is carefully written. Do we have to accept that it is an error only because we cannot determine the meaning otherwise or find a comparable source?
Can we instead find a parallel that serves to explain the text as it stands, unchanged? I think potentially so, and an intriguing one. When I read this fragment, I thought about ‘The Genizah Psalms’, a unique text preserved in a manuscript in the Russian National Library, RNL Antonin 798, which is also very fond of using nouns from the root צדק, including the phrase בחירי צדק (1:21).2 I am familiar with the manuscript through David Stec’s meticulous edition of it (2013), but I now see that it has generated quite a lot of interest over the years.
RNL Antonin 798 was originally brought into the light by Alexander Harkavy in 1902. Since then Shmuel Safrai & David Flusser, Ezra Fleischer, Meir Bar-Ilan, Geert W. Lorein & Eveline van Staalduine-Sulman, Marc Philonenko & Alfred Marx, Menahem Haran, and James Charlesworth have all written articles, and both Charlesworth (2018) and David Stec (2013) have written monographs on the manuscript.3 Views on the origins of the text differ. Harkavy first proposed a medieval composition in the style of the Psalms; Fleischer, similarly, saw them as medieval pastiche. Bar-Ilan thought they were old, but not Second Temple old. Charlesworth and Stec have both made strong arguments that the texts originated in Palestine before the destruction of 70 CE, though they differ over their understanding of what is preserved in the manuscript itself (Charlesworth: one Psalm, number 156; Stec: four Psalms). My own view (for what it’s worth) is that the work has the character of an ancient composition, and the Antonin manuscript is old (by Genizah standards) and calligraphically written, suggesting a transmission history behind it, rather than a one-off, more recent or quotidian composition – which does not support Fleischer’s view (which itself echoes Schechter's early pronouncement on the text): Charlesworth’s and Stec’s Second Temple dating is more persuasive.
RNL Antonin 798 f. 2v. The NLI catalogue entry and link to the images on Ktiv is here.
So, with that established, let me suggest a potential interpretation of that problematic line in Mosseri I.40. Given the line’s underlying similarities with the text of Ezekiel, this is only offered as an alternative, for those who, like me, hate to amend the work of scribes without good reason. One could argue an underlying similarity with the themes of Antonin 798, however, too, as it refers often to ‘righteousness’ and, indeed, it contrasts it with שקר in a context familiar from Mosseri I.40 ובין טמא לטהור ובין צדק לשקר (1:4), ‘and between unclean and clean, and between righteousness and falsehood’ (Stec 2013: 34–5). So, for the problem of משפטיי ישט[פ], compare Genizah Psalms 4:24, כנהר שוטף הירביתה משפטך, ‘You have increased your justice like a flowing river’ (Stec 2013: 48–9). Given this parallel, could we read line 5 in our fragment as משפטיי ישטפו, ‘My judgements will flow’, possibly adding כנהר, ‘like a river’?
My proposed translation would therefore be:
1. [ ]priest
2. [… when] they give thanks to Me from the sons of […]
3. [… from] the congregation of the sons of Zadok […]
4. […] concerning a law of pure and [impure …]
5. […] My judgements will fl[ow …]
6. […] they will consecrate, and profane […]
7. […] they will surround the platforms of […] (or ‘My platform’)
8. [… with] song they will praise M[e …]
9. […] ‘May God give strength to his people’ […]
I think I have made a tenuous argument for this interpretation of line 5, which rests mostly on a faith in scribal infallibility (a fallacy if ever there was one). However, I am very grateful to Corrado Martone for choosing to examine this interesting text and bringing it back to our attention anew. His conclusion is cogent, especially when seen in the light that Erder has thrown on the connections between Qumran texts and Qaraite thought, as Martone notes. Futhermore, it has led me to consider again Antonin 798 and its connection to the Second Temple group of Genizah manuscripts, of which, in the footsteps of Stec and Charlesworth, we should absolutely consider it a part.
So, to conclude, I think we should consider the two fragments as part of the broader group of Second Temple texts in the Genizah Collection: Mosseri I.40 as an echo of that era’s ideology, and Antonin 798 as another astonishing survival of that era in the Fusṭāṭ Genizah.4
Bibliography
Charlesworth, J. 2018. Has Psalm 156 Been Found?
Charlesworth, J. 2018a. ‘Discovering Psalm 156 and Its Importance for Early Judaism and Christian Origins’, Anatomies of the Gospels and Beyond 164, 405–15.
Erder, Y. 1994. ‘The Karaites’ Sadducee Dilemma’, Israel Oriental Studies 14, 195–225.
Erder, Y. 2017. The Karaite Mourners of Zion and the Qumran Scrolls: On the History of an Alternative to Rabbinic Judaism.
Lévi, I. 1913. ‘Document relatif à la «Communauté des fils de Sadoc»’, Revue des Études Juives 65, 24–31.
Martone, C. 2025. ‘Some Observations on a Zadokite Fragment from the Cairo Genizah’, in Abate, E. et al, ‘Habent sua fata fragmenta’: Festschrift in Honour of Mauro Perani Offered by Friends and Colleagues, 135–46.
Schechter, S. 1910. Documents of Jewish Sectaries, vol. I.
Schechter, S. 1913. ‘Announcement’, Jewish Quarterly Review 3, 485.
Schechter, S. 1914. ‘Reply to Dr. Büchler’s Review of Schechter’s “Jewish Sectaries”’, Jewish Quarterly Review 4, 449–74.
Stec, D. 2013. The Genizah Psalms: a study of MS 798 of the Antonin Collection.
Footnotes
1 On the relations between the Bene Ṣadoq and the Sadducees, see the discussion on the Qaraites’ names for themselves by Erder (1994).
2 The term Beḥire Ṣedeq, ‘the Elected Ones of Righteousness’, is meaningful in Qumran texts. See Erder (1994: 212).
3 For a summation of the scholarship, see Charlesworth (2018a: 406–7).
4 I'm burying this in a footnote as purely speculative, but I believe we can distinguish between the Second Temple fragments in the Genizah through one aspect of their transmission history. All the Ben Sira and Damascus Document manuscripts in the Genizah are on paper, and therefore we cannot trace their transmission in the Cairo Genizah world back before the 10th century, when paper began to be used in Syria, Palestine and Egypt. To my mind that suggests they may have been imported via some group that arrived about that time, or in the decades before. Whereas Aramaic Levi, and now we should note also Antonin 798, are on parchment, and palaeographically earlier, suggesting a more lasting relationship with the Cairo Genizah world.
If you enjoyed this Fragment of the Month, you can find others here.
Contact us: genizah@lib.cam.ac.uk
The manuscripts in this article are part of the Cairo Genizah Collection in Cambridge University Library. To see more items from this collection visit: https://cudl.lib.cam.ac.uk/collections/genizah/1