Cambridge University Library

December 2009

A Ketubba in Palimpsest (T-S K23.3)

Some time ago Mark Scarlatta, PhD student at the Faculty of Asian and Middle Eastern Studies and a volunteer in the Genizah Unit, showed me a manuscript that could not be easily categorised. It was one of three palimpsests—vellum manuscripts that have been washed or scraped of their original text and then reused—grouped under the classmarks T-S K23.1 and T-S K23.3. Unusually, all these palimpsests are Hebrew-Hebrew, in the sense that both layers of text are written in Hebrew characters (the languages are: Hebrew, Judaeo-Arabic and Aramaic). Most palimpsests in the Genizah have Hebrew overwriting, but the underscript is usually something quite different like Greek, Latin, Georgian, Christian Palestinian Aramaic—languages used by the Christian, rather than Jewish, community, hence making it quite acceptable for the Jewish scribe to reuse the parchment. After examining the manuscript under an ultra-violet lamp, however, we were able to establish that the underscript is in Jewish Aramaic, and that the partially-erased text is a Jewish legal document, the dowry list from a ketubba.

Several features mark it as unusual. First of all, we have no other examples of ketubbot in the Genizah that have been reused as palimpsests, although the reuse of ketubbot in other ways is not uncommon. In this case, the original parchment has been cut, turned sideways and then folded to form a bifolium. Thus the upper script is written transversely in relation to the lower script, and much of the original ketubba has been cut away. The upper script consists of a magical text listing names of angels (mentioned by Margaliot, 1966; Naveh/Shaked, 1985; Schiffman/Swartz 1992) and continuing on the verso. Secondly, with regard to the content, the formula [ואעלת מבית אב[והי introducing the dowry list is found not at the beginning but in the second half of the document, following the description of the boundaries of a property (a house or an apartment: 'to the north...', 'to the east...,' etc.) being given to the bride. Normally a property would be given to her as a part of her dowry by her father, but here, as in the copy of the betrothal deed found in T-S 16.181 (a register of the Babylonian congregation of Damascus, 932–933, published by Friedman 1981), it appears to be an additional marriage gift from the groom, complementary to the mohar ('advance payment'), that is why a description of it precedes the dowry list.The characteristic concluding formula סימן שריר ובריר can be deciphered at the end of the second leaf and strongly suggests that the document is a ketubba, probably from the 10th or beginning of the 11th century. The name of a certain Bar Ṣulḥ b. S..., who owns an apartment or rooms 'to the east' (line 7: ...מדנחה בתוי בר צלח בר ש), is mentioned in the document. He might be a relative of Joshua B. Ṣulḥ, who signed his name as one of the witnesses in the marriage register of T-S 16.181. The word טימיתיה ('its cost/value,' l. 9), borrowed from Greek τιμή into Palestinian Aramaic of the Byzantine period, suggests the influence of the Palestinian tradition of Aramaic.

It is surprising that such an interesting and rare document has so far remained unearthed. But perhaps the fact that it was 'buried' in a palimpsest accounts for it. Hopefully in the near future it will be wholly deciphered and studied by scholars.

I am grateful to Avi Shivtiel, Dotan Arad and Judith Olszowy-Schlanger for their help in deciphering and identifying the document.

[*For further notes please see below]


T-S K23.3 with the Ketubba underscript showing through clearly on the blank page


M. A. Friedman, Jewish marriage in Palestine (2 vols; 1980–1), vol. ii pp. 396–429
M. Margaliot (ed.), Sefer ha-Razim, a newly-recovered book of magic from the Talmudic period (1966), pp. 52, 104
J. Naveh and S. Shaked, Amulets and magic bowls: Aramaic incantations of late antiquity (1985), p. 166
L. H. Schiffman and M. D. Swartz (eds), Hebrew and Aramaic incantation texts from the Cairo Genizah: selected texts from Taylor-Schechter Box K1 (1992), pp. 67, 175

Mila Ginsburskaya, University of Birmingham

Readers are invited to send comments to The Taylor-Schechter Genizah Research Unit is not under any obligation to acknowledge or to publish comments.

* We are grateful to Prof. M. A. Friedman and Dr Amir Ashur, both of Tel Aviv University, for their comments on the first draft of this page. Dr Ashur has pointed out that the name צלח might have been pronounced Ṣalaḥ. Both Prof. Friedman and Dr Ashur suggest that in place of סימן שריר ובריר, an unusual expression, we should read הימן שריר ובריר, with הימן being a variant form of מהימן found in Palestinian Aramaic sources, see M. Sokoloff, A Dictionary of Jewish Palestinian Aramaic (1990), p. 162. On further examination of the image, Prof. Friedman felt that further comments were warranted, and we are extremely grateful for his preliminary appreciation of the text, which is reproduced below in full.


More on the Palimpsest-Ketubba (T-S K23.3)
by Mordechai A. Friedman

Old discarded ketubbot written on vellum were often targeted for recycling because of the fine writing material and the blank verso. Several examples can be found among the 10th or early 11th-century Palestinian-style ketubbot, including cases where the vellum was cut in strips and sewn together to form part of a pamphlet [see note 1]. Usually only the blank side was used, but T-S K23.3, the right side of a sheet of vellum reused for writing a magical text, provides an interesting example where the blank portion did not satisfy the writer, and he also wrote (perpendicularly) over half of the original document, thus forming a palimpsest. This is an exciting find, and I congratulate Dr Mila Ginsburskaya for identifying and describing it.

As Ginsburskaya noted, the original document contains the description of the boundaries of a property, followed by [ואעלת מבית אב[התה [see note 2], the formula introducing the bride's dowry list. From this sequence she correctly concluded that the groom presented the property to the bride. The more common practice in Geniza ketubbot was for the bride's family to provide the young couple with living quarters, but the groom presented it to the bride also in a betrothal deed from the Babylonian congregation of Damascus, written in 933 [see note 3]. This and the concluding formula הימן שריר ובריר [see note 4] strongly suggest, noted Ginsburskaya, that the document is a ketubba, probably from the 10th or beginning of the 11th century

A closer examination of the text substantiates this suggestion. The preserved portion opens in the middle of the formula for the bride's acceptance of the groom's proposal. The description of the property boundaries is preceded by a phrase stating that this was a mohar gift (from the groom to the bride), as in the 933 document from Damascus. Other comparisons can be found with that manuscript, and the newly-identified text is likely to come from the same milieu. Towards the end of the text we find a truncated but explicit declaration by "the groom" that he accepts responsibility for the dowry.

The groom's name was צלח. In his comments on a letter from North Africa, Goitein took this spelling of the writer's name as a defective form of the more common צאלח (Ṣāliḥ) [see note 5]. Several examples of the triliteral spelling can be cited from texts both from North Africa and from Egypt-Eretz Israel-Syria. Those from the first area include someone who sent questions to Hai Gaon and an accomplished scribe, who signed his name [see note 6]. The latter, besides three occurrences in this text, include the signatory of a Palestinian-style ketubba from Tinnīs, Egypt, and the father of a signatory to the 933 Damascus document. [see note 7]. It is unlikely that all of these are defective spellings. In my earlier study, I transcribed the name: Ṣulḥ. Moshe Gil transcribed it: Ṣalaḥ [see note 8]. In any event it appears to be an addition to the Jewish (Arabic) onomasticon [see note 9]. The document is written in distinctively Jewish Palestinian Aramaic, which merits a separate analysis. Some words are in Hebrew and Arabic. A glance at the digital image, which I had at my disposal, reveals how unintelligible the text is. The following transcription is more often than not speculative. It is offered as a tentative, rough copy, which I hope will be corrected and supplemented by others with or without the assistance of better tools.


1. והוות לצלח בר [...]

2. לה מהר חולקיה [...]

3. בדרתה דאברהם [ה...] בר חות[ם(?) ...]

4. צפונה [דמיגו(?)] תרעא אתריה [...]

5. [ ] רשות הרבים ולעל [...]

6. מן בייתה(?) אבו[הי](?) ומן דרו[מה ...]

7. [מ]דנחה בתוי בר צלח בר ש[...]

8. וכון(?) [...] צפונה דאיוב ב[ר ...]

9. רשות הרבים וטימיתיה ח[...]

10. דינרין ואעלת מבית אב[התה ...]

11. עקר(?) [ ] רג [...]

12. דינרין(?) [...]

13. [ ] תלתה(?) [...]

14. [...]

15. [...]

16. צנדוק [ ] י דנ' [...]

17. חמשין ותרין ואתנון(?) [...]

18. עלי אנה צלח חתנה ביריבי [...]

19. דאנה קני ועתיד למקני ב[ין מטלטלין ...]

20. ביניהם(?) [...]

21. [...]

22. הימן שריר ובריר [...]

23. [. ]א[...]


(1) And she become [the wife] of Ṣulḥ b. […] (2) [and he gave/wrote] her as mohar the portions/his portion […] (3) in the house of Abraham […] b. Ḥota[m(?) …] (4) the north [inside] the gate of his place […] (5) […] the public thoroughfare and above […] (6) from the room/apartment of his father(?) and to the sou[th … to] (7) the east the rooms/apartments of Bar Ṣulḥ b. Š[…] (8) and the windows(?) […] of Job/Ayyūb b. […] (9) the public thoroughfare, and its value […] (10) dinars. And she brought in from her parents' home […] (12) dinars(?) […] (13) three(?) […] (16) a chest […] 10 dinars […] (17) fifty two. And they stipulated(?) [… Undertaken] (18) by me, I Ṣulḥ the groom b. […] (19) which I possess and which I shall acquire, wh[ether movables …] (20) between them(?) (22) trustworthy, binding and valid. (23) [.]A (apparently the remnant of a signature) […]

Mordechai A. Friedman


[1] For examples of pieces sewn together see M. A. Friedman, Jewish Marriage in Palestine (2 vols.; 1980–1; below: JMP), vol. II, plates no. 12 (Bodl. MS. Heb. a. 3, fol. 32v) & no. 39 (T-S 8.133+16.210).
[2] I have restored אבהתה('her parents') rather than אבוהי; see JMP, I:308.
[3] See JMP, I:298–9; II:401 (T-S 16.181[A]).
[4] Following the reading as corrected by Amir Ashur.
[5] S. D. Goitein, Tarbiz 37 (1978), 158 (Or. 1080 J 154).
[6] Questioner: Harkavy, Teshuvot ha-Geonim, 24; scribe: T-S 10 J 21.2 (Ben-Sasson, Sefunot 5[20], 67–8).
[7] Tinnīs: Bodl. MS Heb. b. 3, fol. 28 (JMP, I:11); 933: T-S 16.181 (JMP, II:422, already noted by Ginsburskaya). See also T-S 12.147, line 23; verso, line 5 (M. Gil, Palestine during the First Muslim Period (634–1099) (Hebrew), II:544-5).
[8] I in the aforementioned places in JMP. Gil, In the Kingdom of Ishmael (Hebrew), I:175n120 (the North African questioner).
[9] Since more often than not Ṣulḥ is combined with the father's or son's Hebrew name, it too could have been considered Hebrew, which would suggest a different vocalization.

Notes to the transcription

Line 4: For the possessive without ד- in the Palestinian-style ketubbot (including the 933 document), see JMP, I:75. Cf. lines 6–7.
Line 5: This Hebrew phrase, also in line 9, is used in an Aramaic context in the boundary description in T-S 16.181(D), line 18 (JMP, II:422) as well.
Line 7: I am unable to decipher this word (בתוי) from the digital image and copy it from Ginsburskaya's description, where this line is transcribed.
Line 8: = וכוין?
Line 16: 'A chest', this appears to be the Arabic ṣandūq.



T-S K23.3 rotated to more clearly show the text of the underlying ketubba