Prayer of Manasse in Hebrew

Linguistic Observations on the Hebrew Prayer of Manasseh from the Cairo Genizah Wido van Peursen 1. Introduction Among the so-called magical texts from the Cairo Genizah edited by Peter Schäfer and Shaul Shaked, the Cambridge fragments T.-S. K 1.144, T.-S. K 21.95, and T.-S. K 21.95P constitute a manuscript containing various prayers, most of which have “a mystico-magical character.”1 Among these prayers we find a Hebrew version of the Prayer of Manasseh2 previously known in Greek and Syriac.3 There is no relationship with Manasseh’s prayer found at Qumran (4Q381 33 8–11) edited by Eileen Schuller4 and investigated by William M. Schniedewind.5 One of the very few studies on the Genizah text is an article by Reimund Leicht, in which he argues that this Hebrew text of PrMan is a tenth-century translation from a Greek text close to the text of the Codex Turicensis, but reflects also unequivocal influence from the Syriac versions. The picture that emerges from Leicht’s hypothesis is reminiscent of the model that various scholars in the late-19th and early-20th centuries advocated regarding the Hebrew text of Ben Sira from the Cairo Genizah,6 namely, that of a document written in the Second Temple 1 Abbreviations: BH = Biblical Hebrew; CH = Classical Hebrew (including BH and QH); MH = Mishnaic Hebrew; RH = Rabbinic Hebrew; QH = Qumran Hebrew; PrMan = Prayer of Manasseh; PrMan-Heb = The Hebrew text of the Prayer of Manasseh from the Cairo Genizah; DCH = D. J. A. Clines, ed., Dictionary of Classical Hebrew; HALOT = Koehler, Baumgartner, Stamm, eds., The Hebrew and Aramaic Lexicon of the Old Testament. We use “MH” if the distinction between the Tannaitic Hebrew and Amoraic Hebrew is applicable, and “RH” if that distinction does not apply. 2 Peter Schäfer and Shaul Shaked, eds., Magische Texte aus der Kairoer Geniza (3 vols.; TSAJ 42,64,72; Tübingen: Mohr Siebeck, 1994–1997), 2:27–78, PrMan on pp. 32 (text) and 53 (translation). 3 Also in other languages, but the other versions depend either on the Greek or on the Syriac text. 4 Eileen M. Schuller, Non-Canonical Psalms from Qumran: A Pseudepigrapic Collection (HSS 28; Atlanta: Scholars Press, 1986), PrMan on pp. 146 (Hebrew text), 151 (translation), 155–58 (comment); Schuller also published this text in the DJD series in 1998; for PrMan (text, translation and notes) see DJD 11:122–26. 5 W.M. Schniedewind, “A Qumran Fragment of the Ancient Prayer of Manasseh?” ZAW 108 (1996): 105–7, argued that the Qumran prayer represents an early extra-biblical tradition, which predates the Chronicler’s history, and was perhaps even part of the source to which the Chronicler refers. For a rather positive assessment of this view see Louis Jonker, “Tradition through Reading—Reading the Tradition: Reflections on Eep Talstra”s Exegetical Methodology,” in Tradition and Innovation in Biblical Interpretation: Studies Presented to Professor Eep Talstra on the Occasion of his Sixty-Fifth Birthday (ed. W. Th. van Peursen and J. W. Dyk; SSN 57; Leiden: Brill, 2011), 133–51, esp. 146–47; for a more critical assessment see Ariel Gutman and Wido van Peursen, The Two Syriac Versions of the Prayer of Manasseh (Gorgias Eastern Christian Studies 30; Piscataway, N.J.: Gorgias, 2011), 12. Schuller, Non-Canonical Psalms from Qumran, 31–32, 161–62, argues for the secondary attribution of the prayer to Manasseh; see also Schuller, “4Q380 and 4Q381: Non-Canonical Psalms from Qumran,” in The Dead Sea Scrolls: Forty Years of Research (ed. D. Dimant and U. Rappaport; STDJ 10; Leiden: Brill, 1992), 90–99, esp. 94–95; and Schuller, DJD 11:123. 6 Cf. W. Th. van Peursen, The Verbal System in the Hebrew Text of Ben Sira (SSL 41; Leiden: Brill, 2004), 20. Period,7 that survived in Greek and Syriac, and that in the Middle Ages, presumably the tenth century, was retranslated into Hebrew.8 For the Hebrew text of Ben Sira the retroversion theory could not stand the test of time, and after the discovery of the Masada text and the Qumran fragments, at most the partial retroversion theory could be maintained; that is, the view that the Genizah manuscripts of Ben Sira are basically the result of inner-Hebrew development, and that only some passages such as Sir 51:13–30 contain traces of a retroversion from Syriac (or Greek).9 For other books the scholarly discussion moved in the opposite direction. Thus M. Gaster considered the small Hebrew version of the story of Judith that he published in 1894 to be standing at the beginning of the literary and textual history of the book of Judith,10 and this view was in vogue for some time and entered, for example, the Encyclopedia Biblica edited by Cheyne and Black, in which Gaster wrote the entry on the book of Judith.11 In 1922, however, C. Meyer argued that the Hebrew text was a free retroversion from the Vulgate,12 and since then this view has become the majority view;13 although the minority view, giving priority to one or more of the extant Hebrew versions, did not completely disappear. It was advocated by A. M. Dubarle in his 1958 article.14 Similarly, the abridged Hebrew version of the Book of Maccabees was thought by its editor, Abraham Schweizer, to be original,15 a view that was refuted by C. C. Torrey.16 7 Note, however, that the date of origin of PrMan is not so easy to establish as some scholars have suggested and that a later date of origin cannot be ruled out; cf. Gutman and van Peursen, The Two Syriac Versions of the Prayer of Manasseh, 41–52. 8 Whether we should call PrMan-Heb a “retranslation” (back into Hebrew) or just a “translation” depends on the source of the Greek and Syriac versions. Only if we assume that these versions go back to a Hebrew original—which is far from certain (cf. Gutman and van Peursen, The Two Syriac Versions of the Prayer of Manasseh, 8–9 n. 11)—is it justified to speak of a re-translation. 9 Cf. van Peursen, The Verbal System in the Hebrew Text of Ben Sira, 19–23; van Peursen, “The Alleged Retroversions from Syriac in the Hebrew Text of Ben Sira Revisited: Linguistic Perspectives,” Kleine Untersuchungen zur Sprachen des Alten Testaments und seiner Umwelt 2 (2001): 47–95; and van Peursen, “Sirach 51:13–30 in Hebrew and Syriac,” in Hamlet on a Hill: Semitic and Greek Studies Presented to Professor T. Muraoka on the Occasion of his Sixty-Fifth Birthday (ed. M. F. J. Baasten and W. Th. van Peursen; OLA 118; Leuven: Peeters, 2003), 357–74. 10 M. Gaster, “An Unknown Hebrew Version of the History of Judith,” Proceedings of the Society of Biblical Archaeology 16 (1894): 156–63. 11 M. Gaster, “Judith, the Book of,” in Encyclopedia Biblica: A Critical Dictionary of the Literary, Political and Religion History, the Archeology, Geography and Natural History of the Bible (ed. T. K. Cheyne and J. Sutherland Black; 4 vols.; New York: The Macmillan Company, 1899–1903), 2:2642–46. 12 Carl Meyer, “Zur Entstehungsgeschichte des Buches Judith,” Biblica 3 (1922): 193–203. 13 Cf. Carey A. Moore, Judith (AB 40B; Garden City, N.Y.: Doubleday, 1985), 101–2. 14 A. M. Dubarle, “Les textes divers du livre de Judith,” VT 8 (1958): 344–73; see, however, also Dubarle, “Rectification: sur un texte hébreu de Judith,” VT 11 (1961): 86–87. 15 A. Schweizer, Untersuchungen uber die Reste eines hebraischen Textes vom ersten Makkabaerbuch (Berlin: Poppelauer, 1901); unfortunately, we were unable to consult a copy of this book, so we depend on the extensive review by Torrey. 16 Charles C. Torrey, “Schweizer’s ‘Remains of a Hebrew Text of 1 Maccabees,’” JBL 22 (1903): 51–59. Compare how Torrey (p. 53) linked up the discussion about the original Hebrew of 1 Maccabees with that Will PrMan-Heb from the Cairo Genizah undergo the same fate as the Hebrew text of Ben Sira from the Genizah? Will it also in the end—with our increased knowledge of Hebrew from the Second Temple period—turn out to be a genuine, original Hebrew document? Or should we rather consider it as a parallel to medieval translations into Hebrew of, for example, the books of Judith and 1 Maccabees? This question can be addressed from various perspectives. Leicht focuses on the textual affiliations of PrMan-Heb with the Greek and Syriac versions and draws upon linguistic observations to support his theory.17 My aim is to start with the Hebrew text in its own right and with its linguistic profile before proceeding to the larger texthistorical picture that emerges. Regardless of its origin, PrMan-Heb has cultural and literaryhistorical significance, being the only Hebrew witness from a Jewish context to a prayer otherwise only known from Christian transmission channels. 2. Rabbinic Elements Our analysis of the linguistic profile of PrMan-Heb will start with features that are typical of Rabbinic Hebrew.18 2.1 Orthography 1. 2. 3. 4. 5. The pronominal suffix attached to plural nouns or prepositions with a (pseudo-)plural ending is spelled ‫יי‬: 2a19 ‫“ אבותיי‬my fathers”; 2b8 ‫“ חטאותיי‬my sins,” ‫“ אשמותיי‬my transgressions”; 2b9+13 ‫“ עונותיי‬my iniquities”; 2b9+10 ‫“ חטאותיי‬my sins”; 2b12 ‫“ עליי‬on me” (contrast 2b16+17 ‫ ;)עלי‬2b15 ‫“ פשעיי‬my wrongdoings”; 2b16 ‫“ בחטאותיי‬in my sins”; 2b17 ‫“ בפניי‬before me,” ‫“ חטאותיי‬my sins.” The Niphal imperfect is written with a yod as vowel letter in the prefix: 2b5 ‫“ ותינחם‬and you relent.” Word-internal consonantal yod is written as ‫יי‬: 2b18 ‫ =“ תחייבני‬do not condemn me.”19 The Tetragrammaton is written with three yods in 2b6 and 2b18 ‫ײי‬. ‫ האלים‬in 2b18 ‫“ שאתה הוא ײי האלים לבני אדם‬because you are the Lord of the gods for the human beings” is at first sight somewhat peculiar. There is no exact parallel to ‫ײי‬ ‫ האלים‬in the Bible, the closest parallel being ‫ אל אלים‬in Dan 11:36. A more plausible explanation, however, is that it is a short form of ‫“ אלהים‬God”; the Genziah manuscript about the original Hebrew text of Ben Sira: “Since a part of the original Hebrew of Ecclesiasticus has recently come to light, we are prepared to hear of the recovery of the original text of other books of the Old Testament Apocrypha, now preserved only in translations.” 17 The studies about the Hebrew texts of Judith and 1 Maccabees mentioned above do not address linguistic arguments. 18 References are to the folio and line number of the Genizah manuscript. Thus 2a19 means: line 19 of folia 2a. 19 See also below, Section 2.2. in which PrMan-Heb is found contains also other abbreviated forms of ‫אלהים‬.20 2.2 Morphology The form of the Perfect 2 masc. sg. is ‫( קטלתה‬contrast BH ‫)קטלת‬: 2a20 ‫(“ עשיתה‬you who) made”; 2b1 ‫“ צויתה‬you commanded”; 2b7 ‫“ שמתה‬you put.” 2. After ‫ אל‬the full imperfect is used, rather than the short form in 2b17 ‫“ ואל יחרה‬and let (your wrath) not burn.”21 3. The Piel of the so-called hollow roots follows the pattern of the strong verb: 2b18 ‫(“ תחייבני‬do not) condemn me.”22 4. The pronominal suffix attached to the negation ‫ אין‬does not take the epenthetic nun: 3a1 ‫“ איני‬I am not” (contrast BH ‫)אינני‬.23 1. 2.3 Morphosyntax (conjunctions) ‫ ש‬is used as a relative (exclusively; there are no cases of ‫ )אשר‬in 2b5, 7, 9, 19, 20; and as a causal conjunction (cf. BH ‫ )כי‬in 2b16 ‫“ אבקשך שאני צריך אותך‬I seek you because I need you”; 2b18 ‫ שאתה הוא ײי‬... ‫ ואל תחייבני‬... “and do not condemn me … because you are the Lord …”; compare also ‫ ש‬introducing an adnominal prepositional phrase in 2b8 ‫“ מחול שעל שפת הים‬than the sand (that is) on the seashore” (contrast an adnominal prepositional phrase without relative in 2b3–4 ‫“ זעמך על החטאים‬your anger over sinners”). The introduction of an adnominal prepositional phrase with a relative is common in Syriac.24 2. The following compound conjunctions are attested: ‫“ מפני ש‬because” (2b8); ‫ועל ש‬ “because” (2b10; cf. BH ‫“ בעת ש ;)על אשר‬at the time when” (2a18; cf. ‫ בזמן ש‬etc. in 1. 20 21 Schäfer and Shaked, Magische Texte aus der Kairoer Geniza, 27. After ‫ אל‬the short imperfect is frequent throughout the CH period (including QH). In MH the short imperfect is used only in literary and elevated style, see Gideon Haneman, ‫( תורת הצורות של לשון המשנה‬A Morphology of Mishnaic Hebrew according to the Tradition of the Parma MS [De Rossi 138]) (TSHLRS 3; Tel Aviv: Tel Aviv University, 1980), 31–32; Mordechai Mishor, ‫“( מערכת הזמנים בלשון המשנה‬The Tense System in Tannaitic Hebrew”) (Ph.D. diss., Hebrew University of Jerusalem, 1983), 86–92; and van Peursen, The Verbal System in the Hebrew Text of Ben Sira, 92. 22 Cf. M. H. Segal, A Grammar of Mishnaic Hebrew (Oxford: Clarendon, 1927), 82–83 and Segal, ‫דקדוק‬ ‫( לשון המשנה‬A Grammar of Mishnaic Hebrew) (Tel Aviv: Dvir, 1936), 143. See also above, Section 2.1, on the orthography. 23 Cf. M. Pérez Fernández, An Introductory Grammar of Rabbinic Hebrew (trans. John F. Elwolde; Leiden: Brill, 1997), 19. 24 Van Peursen, Language and Interpretation in the Syriac Text of Ben Sira: A Comparative Linguistic and Literary Study (MPI 16; Leiden: Brill, 2007), 224. The examples given there include Gen 1:9 )‫ ܕ&ـ'(ـ' !ـ‬$#‫!ـ‬ *+,- “the waters (that are) under the sky” for MT ‫המים מתחת השמים‬. MH25); ‫“ אף על פי ש‬although” (3a1).26 2.4 Syntax 1. There is one nominal clause with the pattern ‫אין מי יקטול‬: 2b3 ‫“ ואין מי יעמוד לפני כוחך‬and there is none who can stand before your power.” This pattern is reminiscent of MH examples such as m. Menaḥ. 4:3 ‫“ אין לו מי יתירנו‬it has nothing which renders it permissible,”27 but the use of ‫( מי‬rather than ‫ )מי ש‬is remarkable.28 2. The pattern ‫ לקטול‬X ‫ אין לי‬is attested twice: 2b9 ‫“ ואין לי רשות להביט‬I have no authority to look”; 2b10–11 ‫“ ואין לי מצח להרים ראשי אליך‬and I do not have the insolence to raise my head to you.” Unlike the pattern ‫אין לקטול‬, which is well attested in LBH and QH, the pattern ‫ לקטול‬X ‫ אין‬is common in MH.29 2.5 Words and Phrases 1. PrMan-Heb contains a number of words that are infrequently attested or unattested in BH, but common in RH: 2b3 ‫“ סבל‬to carry”; 2b5 ‫“ ארוך‬long, lasting”30; 2b9 ‫רשות‬ “authority”; 2b15 ‫ עוה‬Qal “to transgress”31; 2b17 ‫ חרט‬Hitpael “to repent”; 2b17 ‫ חוב‬Piel “to condemn”32; 3a1 ‫“ זכאי‬worthy, deserving”; 3a2 ‫( נאה‬+ infinitive) “appropriate.”33 25 Cf. van Peursen, The Verbal System in the Hebrew Text of Ben Sira, 335, with references to Pérez Fernández, Introductory Grammar of Rabbinic Hebrew, 209–10, and M. Azar, ‫( תחביר לשון המשנה‬The Syntax of Mishnaic Hebrew) (Jerusalem: Academy of the Hebrew Language/University of Haifa, 1995), 117–18; but note that in MH the compounds with ‫ שעה‬and ‫ זמן‬are more common. 26 Cf. M. Jastrow, Dictionary of the Targumim, Talmud Babli, Yerushalmi and Midrashic Literature (2 vols.; New York: Judaica, 1886–1903), 99b. 27 Example from Azar, 89 ,‫תחביר לשון המשנה‬. 28 Cf. Pérez Fernández, Introductory Grammar of Rabbinic Hebrew, 41: “The interrogatives ‫ מִי‬and ‫ מָה‬can also have indefinite significance, particularly in the sequence ‫מה ש‬/‫“ מי‬whoever, whatever” … and especially when preceded and reinforced by ‫ כל מי ש( כל‬etc.).” 29 Cf. W. Th. van Peursen, “Negation in the Hebrew of Ben Sira,” in Sirach, Scrolls, and Sages: Proceedings of a Second International Symposium on the Hebrew of the Dead Sea Scrolls, Ben Sira, and the Mishnah held at Leiden University, 15–17 December 1997 (ed. T. Muraoka and J. F. Elwolde; STDJ 33; Leiden: Brill, 1999), 223–43, esp. 229. 30 Only three occurrences in BH: Job 11:8; 2 Sam 3:1; Jer 29:28; for its use in RH see J. Levy, Neuhebräisches und Chaldäisches Wörterbuch über die Targumim und Midraschim (4 vols.; Leipzig: F. A. Brockhaus, 1876–1889), 1:166a. 31 ‫ עוה‬Qal occurs in the Bible only in Esth 1:16 and Dan 9:5. 32 33 In the Hebrew Bible it occurs only in Dan 1:10, meaning “to make guilty.” See also Sir 11:18. Compare Sir 41:16, where the Genizah MSS B and C have ‫נאה‬, but the Masada Scroll ‫נאוה‬. Since in BH we find other forms with a waw, such as Ps 93:5 ‫( נַאֲוָה‬probably to be interpreted as a Niphal of ‫)אוה‬, I have argued elsewhere that the Masada text should be interpreted as a Niphal of ‫אוה‬, and that the Genizah manuscripts reflect a later stage in the history of the Hebrew language, in which the Niphal of ‫ אוה‬and the Qal of ‫ נאה‬have merged; cf. van Peursen, “Het Participium bij Ben Sira” (M.A. thesis, Leiden University, 1994), 35. For the purpose of the present study it suffices to observe that the form in Heb-PrMan (‫)נאה‬ agrees with the form in the Genizah manuscripts of Ben Sira as against the Masada Scroll. R. Leicht, “A 2. A word that is well-attested in both BH and RH but is used in PrMan-Heb with a sense that is typically RH is ‫עולם‬, with the meaning “world” (rather than “eternity”), in 2a19 ‫“ השליט בעולמו‬the ruler over his world”; and 2b2 ‫“ ישבחו כל העולם‬the whole world praises you”; also, in 2b17, “(in) this world” in contrast to ‫“ עולם הבא‬the world to come.”34 The biblical usage is attested as well, cf. 3a2 ‫“ לעולמי עולמים‬for ever and ever.” The introduction of “the world to come” in this prayer reflects acquaintance with Rabbinic discussions as to whether or not Manasseh will have a share in the world to come.35 3. Rabbinic idioms include, in addition to the above-mentioned ‫עולם הבא‬, 2a18 ‫עשה‬ ‫“ תשובה‬to do repentance” (e.g. m. Yoma 8:9; m. Ned. 9:3; m. ʾAbot 5:18).36 3. Non-Rabbinic Features The features discussed in Section 2 suggest a straightforward characterization of PrMan-Heb as a Rabbinic text, but there are also features that do not easily fit into a Rabbinic linguistic profile. 3.1 Orthography 1. Sin (rather than samek) occurs in the form 2b13 ‫“ השיגוני‬they have overtaken me,” for which we would expect in MH ‫הסיגוני‬.37 3.2 Morphology 1. The lengthened imperative is used in 2b16 ‫“ חוסה‬spare!”38 2. Unusual in RH are perfect forms of ‫“ יכל‬to be able,” as in 2b13 ‫יכולתי‬. In RH we find only the participle ‫ ;יכול‬the BH perfect structure ‫ יכֹלת‬,‫ יכֹלתי‬has been replaced in RH by perfect forms of the verb ‫ היה‬+ the participle ‫יכול‬.39 Newly Discovered Hebrew Version of the Apocryphal “Prayer of Manasseh,” JSQ 3 (1996): 359–73, at 366, refers to ‫ כי לך נאה‬in the Yishtebach of the morning prayer. For the construction of ‫ נאה‬+ infinitive see also van Peursen, The Verbal System in the Hebrew Text of Ben Sira, 270. 34 Cf. Levy, Neuhebräisches und Chaldäisches Wörterbuch über die Targumim und Midraschim, 3:655. 35 Leicht, “A Newly Discovered Hebrew Version,” 367, and Gutman and van Peursen, The Two Syriac Versions of the Prayer of Manasseh, 46–47. 36 ‫ תשובה‬occurs in the Bible with the meaning “return” (HALOT, 1800b). 37 Cf. Segal, A Grammar of Mishnaic Hebrew, 32; see Segal, 34 ,‫דקדוק לשון המשנה‬, on samek replacing sin, but there this verb is not mentioned. 38 Cf. Haneman, 31 ,‫ ;תורת הצורות של לשון המשנה‬Pérez Fernández, Introductory Grammar of Rabbinic Hebrew, 151; and van Peursen, The Verbal System in the Hebrew Text of Ben Sira, 92. 39 Pérez Fernández, Introductory Grammar of Rabbinic Hebrew, 114; see also Haneman, ‫תורת הצורות של לשון‬ 72–71 ,‫המשנה‬. 3.3 Morphosyntax (tenses) 1. The use of the imperfect for the present is attested in 2b5 ‫“ ותינחם‬and you relent”; 2b15 ‫“ פשעיי אני אדע‬I know my wrongdoings”; 2b19 ‫“ שיתחרטו‬who repent.” 3.4 Syntax (phrase structure) The use of the construct state agrees with the rules of BH, e.g., 2a18 ‫תפילת מנשה מלך‬ ‫“ יהודה‬Prayer of Manasseh, King of Judah”; 3a1 ‫“ כל ימי חיי‬all the days of my life”; 3a2 ‫כל‬ ‫“ צבאות שמים‬all the hosts of heaven.”40 Although it is unwarranted to call these examples, “non-Rabbinic,” it is worth noting that analytical genitive constructions with ‫ש‬, which are common in RH, do not occur. Furthermore, typically BH is the construction with an adjective in 2b4 ‫“ ישרי לב‬those whose heart is upright.”41 2. Another peculiarity is the discord or rather the constructio ad sensum in 2a19–20 ‫זרעם‬ ‫“ הצדיקים‬their righteous offspring,” but this type of disagreement is acceptable in both BH and RH.42 1. 3.5 Syntax (clause structure) The construction in 3a1 ‫“ איני זכאי להושיעני‬I do not deserve to save me (i.e. to be saved)” is peculiar in that the suffix attached to the infinitive is superfluous.43 The same applies to 3a2 ‫“ ולך נאה להללך‬to you, it is appropriate to praise you.”44 2. We find a nominal clause with ‫ אתה הוא‬in 2b4–5 ‫“ אתה הוא שהחסד עמך‬it is you with whom is mercy”; 2b6 ‫“ אתה הוא ײי אלהי הצדיקים‬you are the Lord, the God of the righteous”; 2b18 ‫“ שאתה הוא ײי האלים לבני אדם‬because you are the Lord, the God for the human beings.” According to Azar there is no example of ‫ אתה הוא‬in the Mishnah 1. 40 Other examples: 2a19 ‫“ אלהי אבותיי‬the God of my fathers”; 2b6 ‫“ רעת העם‬the evil over the people”; 2b6 ‫“ אלהי הצדיקים‬the God of the righteous”; 2b6 ‫“ טוב הצדיקים‬the good for the righteous”; 2b8 ‫“ שפת הים‬the seashore”; 2b10 ‫“ כבלי ברזל‬fetters of iron”; 2b13 ‫“ שערות ראשי‬the hairs of my head”; 2b18 ‫תחת תהום הארץ‬ “under the depths of the earth”; 2b18 ‫“ בני אדם‬the human beings”; 3a2 ‫“ לעולמי עולמים‬for ever and ever.” Note that we interpret 2b18 ‫ ײי האלים‬as an apposition: “the Lord God [‫ ;”]אלהים‬see Section 3.6. 41 For the use of this construction in BH see T. Muraoka, “The Status Constructus of Adjectives in Biblical Hebrew,” VT 27 (1977): 375–80. 42 Disagreement in number is attested with ‫ זרע‬as subject of a plural verb (Jer 31:3; Ezra 9:2; Neh 9:2); for the idiom used here compare Prov 11:21 ‫“ זרע צדיקים‬the offspring of the righteous” (but there the two words constitute a construct state connection rather than an apposition). For disagreement in number with an attributive adjective, Joüon and Muraoka, A Grammar of Biblical Hebrew, § 148a, give as an example Isa 9:1 ‫העם ההלכים‬. In MH, “with collectives, adjective agreement is according to semantic sense,” according to Pérez Fernández, Introductory Grammar of Rabbinic Hebrew, 81. 43 Cf. Joüon and Muraoka, A Grammar of Biblical Hebrew, § 124s, on the identification of the subject in infinitive constructions. 44 Leicht, “A Newly Discovered Hebrew Version,” 367, calls the use of the suffix pronoun here “rather clumsy.” and only one example of ‫( אני הוא‬m. Naz. 8:1).45 3. The use of ‫( מי‬rather than ‫ )מי ש‬for “someone who” in 2b3 is remarkable.46 3.6 Words and phrases 2b10 ‫“ מצח‬insolence” in 2b10–11 ‫“ ואין לי מצח להרים ראשי אליך‬and I do not have the insolence to raise my head to you,” is remarkable. It seems to have a meaning parallel to ‫ רשות‬in 2b9 ‫“ ואין לי רשות להביט‬I have no authority to look.” I have found no direct parallels for this use of ‫מצח‬, but it is reminiscent of its use in the Bible in combination with ‫( ָחזָ ק‬e.g., Ezek 3:7–8) or other idioms expressing the “strength of the ‫מצח‬.” 2. In 2b11–12 we find ‫“ להאריך אפך‬to make your anger endure.” In BH and RH this idiom is used for “to refrain from anger,”47 and the pronominal suffix attached to ‫אף‬, if present, refers to the (logical) subject of the verb. Thus we find Isa 48:9 ‫“ למען שמי אאריך אפי‬for my own name”s sake I refrain from anger”; Prov 19:11 ‫“ שכל אדם האריך אפו‬a man’s insight makes him patient”; similarly Sir 30:22 (MS B)48; cf. ‫“ אֹרך אפים‬patience” (Prov 25:15) and ‫“ ֶארך אפים‬patient” (e.g., Joel 2:13, quoted below). The use of this idiom in PrMan-Heb, in a meaning opposite to its usual meaning (provoking anger, rather than refraining from it) and in an uncommon syntactic construction (the logical subject of the verb is “I,” not the “you” referred to by the suffix attached to ‫)אף‬, is striking. 3. The participle ‫ מחננים‬in 3a1–2 ‫ כי לך מחננים כל צבאות שמים‬is problematic. Leicht translates with “because all hosts of heaven ask you for compassion,”49 but adds in a footnote that perhaps ‫ מחננים‬is a scribal error for ‫מרננים‬, since the other versions suggest a word like “to praise.”50 ‫ חנן‬Piel occurs only once in the Bible, in Prov 26:25, meaning “to make gracious, favourable (voice).” In the meaning “to ask for compassion” (cf. Leicht”s translation) the Hitpolel is used in BH. The Piel of ‫ חנן‬is not given in the RH dictionaries of Jastrow51 and Levy.52 4. 3a2 ‫“ כל צבאות שמים‬all the hosts of heaven” differs from the biblical idiom ‫צבא השמים‬, with the singular form of ‫צבא‬. For the plural in similar contexts, DCH (7:67) gives only two examples from Qumran Hebrew, one in a reconstructed text in 1QHa 9:12 (but not 1. 45 Azar, 80 ,‫תחביר לשון המשנה‬. The construction is common in Syriac—cf., e.g., Sir 36:22 $2‫ـــــ' ܗܘ ܐ&ـــــ‬/‫ܕܐ‬ 8‫ـ‬7‫ܕ‬6‫ـ‬3‫ـ‬4‫ـ‬5 “that you alone are God,” corresponding to [‫ כי אתה אל ]עו[ל]ם‬in the Hebrew text (MS B)—as well as in other Semitic languages; cf. Van Peursen, Language and Interpretation in the Syriac Text of Ben Sira, 304. 46 See Section 2.4 above. 47 HALOT, 88b. 48 Cf. van Peursen, The Verbal System in the Hebrew Text of Ben Sira, 249. 49 Leicht, “A Newly Discovered Hebrew Version,” 373; cf. Schäfer and Shaked, Magische Texte aus der Kairoer Geniza, 53: “denn alle Heerscharen des Himmels flehen zu dir.” 50 Leicht, “A Newly Discovered Hebrew Version,” 373 n. 45 (see also p. 365: “If ‫ מחננים‬in v. 15 is no scribal mistake this cannot be called a correct translation of the sources”). 51 Jastrow, Dictionary of the Targumim, Talmud Babli, Yerushalmi and Midrashic Literature, 484a-b. 52 Levy, Neuhebräisches und Chaldäisches Wörterbuch über die Targumim und Midraschim, 2:82b. anymore reconstructed in DJD 40), and one in 1QHa 5:25-26 ‫רקיע קודשך וכול צבאותיו‬ “your holy vault and [al]l its hosts,” where the suffix attached to a plural form of ‫צבא‬ refers to ‫רקיע‬. 4. Explanations for the Non-Rabbinic Elements The Rabbinic features of PrMan-Heb discussed in Section 2 make it basically a RH text. This calls for an explanation for those features, discussed in Section 3, that do not agree with this overall picture. This concerns BH features as well as peculiarities that are unusual both according to BH and according to RH standards. We will first focus on two phenomena highlighted by Leicht: biblical quotations and “etymological congruities” with the Syriac text. 4.1 Biblical Quotations PrMan-Heb contains many quotations from and allusions to the Bible. As Leicht puts it: “In many cases the translator adapts the verses he borrows from the Hebrew Bible rather than rendering his textual sources very exactly.”53 Some of the non-Rabbinic elements discussed above come from biblical quotations: 1. 2. 3. 4. 5. For the sin rather than samek in ‫ השיגוני‬and the form ‫ יכולתי‬in 2b13–14 ‫והשיגוני עונתיי‬ ‫“ ולא יכולתי לראות עצמי משערות ראשי ולבי עזבני‬and my iniquities have overtaken me and I cannot see; they have become more numerous [reading ‫ ]עצמו‬than the hairs of my head, and my heart has left me,” compare Ps 40:13 ‫והשיגוני עונתי ולא יכֹלתי לראות‬ ‫עצמו משערות ראשי ולבי עזבני‬. For the construct state of an adjective, ‫ ישרי לב‬in 2b4 ‫ובלא מספר חסדיך וצדקותיך לישרי‬ ‫“ לב‬and innumerable are your mercies and your righteous acts to those whose heart is upright,” compare Ps 36:11 ‫“ משך חסדיך לידעיך וצדקתך לישרי לב‬continue your mercy to those who know you and your righteousness to those whose heart is upright.” The use of the imperfect for the present tense in 2b15 ‫כי פשעיי אני אדע וחטאתי נגדי‬ ‫“ תמיד‬for I know my wrongdoings and my sin is always before me” can be explained from Ps 51:5 ‫כי פשעי אני אדע וחטאתי נגדי תמיד‬. Perhaps the construction of ‫ מי‬rather than ‫ מי ש‬for “someone who” in 2b3 ‫ואין מי יעמוד‬ ‫“ לפני כוחך‬and there is none who can stand before your power” comes from Nah 1:6 ‫“ לפני זעמו מי יעמד‬who can stand before his anger?”54 Yet, though the sequence ‫מי יעמוד‬ is identical, the syntactic structure in which these words function is different. Note also that the construction in 2b8 ‫“ מחול שעל שפת הים‬than the sand (that is) on the seashore” comes from Judg 7:12 ‫( מחול שעל שפת הים לרב‬contrast “the sand of the sea” in the Greek version of PrMan). 53 Leicht, “A Newly Discovered Hebrew Version,” 365. 54 Leicht, ibid., 365, erroneously refers to Neh 1:6. 6. Apart from these examples, which affect the linguistic profile of the text, other biblical quotations or allusions occur as well. Thus 2b5–6 ‫“ ותינחם על רעת העם‬and you relent the evil over the people” (with “people” instead of “human beings” in the Greek text) seems to be influenced by Exod 32:12 ‫“ והנחם על הרעה לעמך‬and relent the evil against your people” and Joel 2:13 ‫“ ארך אפים ורב חסד ונִ ַחם על הרעה‬slow to anger and abounding in love, and he relents the evil.” 2b19 ‫“ הראני ײי חסדך וישעך תתן לי‬show me, Lord, your mercy and give me your salvation” is a direct quote from Ps 85:8 ‫הראנו יהוה‬ ‫חסדך וישעך תתן לנו‬. 7. The lengthened imperative of the verb ‫“ חוס‬to spare” (2b16) occurs also in Neh 13:22 ‫“ וחוסה עלי כרב חסדך‬and spare me according to the greatness of your mercy” and Joel 2:17 ‫“ חוסה יהוה על עמך‬spare, o Lord, your people,” but since there are no other analogies with these passages, apart from words and idioms that we can expect in penitential prayers in general (cf. ‫ חסדך‬in Neh 13:22 and ‫“ חמל‬have compassion” in Joel 2:17), we do not consider the occurrence of the lengthened imperative in PrMan-Heb to be the result of a quotation. 8. Also the following examples are unsure: 2b10 ‫“ כבלי ברזל‬fetters of iron” which occurs (as an Aramaism in BH) also in Ps 105:18 and 149:8; 2b10 ‫“ וכבדו חטאותיי‬and my sins are heavy,” where we find the combination of ‫ חטאה‬and ‫כבד‬, which occurs also in Gen 18:20 2 ;‫וחטאתם כי כבדה‬b14 ‫“ נטיתי לבי‬I inclined my heart,” which occurs also in Ps 119:112 (contrast “I incline the knee of my heart” in the Greek). Biblical quotations do not explain all non-RH features. Thus the use of the imperfect for the present tense in 2b15 can be explained from Ps 51:5, but the same usage is attested in 2b5 and 2b19. The biblical elements should hence be described not only in terms of quotations from or allusions to biblical passages, but also as biblical language used in an otherwise RH text. 4.2 Close Similarity to Syriac Can we say more about the linguistic profile of PrMan-Heb than that it is basically a RH text that includes BH elements, partly in quotations, partly in a general tendency to employ biblical language? One factor should be included in the discussion, which links the linguistic analysis with text-historical considerations. Leicht has drawn attention to “the great number of etymological congruities between the Hebrew and the Syriac versions.”55 The “most obvious examples” that he mentions are the following:56 1. 2a19–20 ‫“ זרעם הצדיקים‬their righteous offspring” = SyrB *'(‫ܙܕ‬ ̈ ‫ܘܢ‬#$‫( ܙܪ‬cf. SyrA +,(‫ܘܢ ܙܕ‬#$‫)ܙܪ‬ 55 Ibid., 364. 56 We distinguish between SyrA, the version found in the Syriac Didascalia and biblical manuscripts, and SyrB, found in Melkite Horologia; cf. Gutman and van Peursen, The Two Syriac Versions of the Prayer of Manasseh, 24–25. 2b4–5, 2b18 ‫“ אתה הוא‬you are …” = SyrA+B ‫ ܗܘ‬-.‫ܐ‬ 2b7 ‫“ ושמתה תשובה לי‬you put repentance for me” = SyrA+B 68$\67 *‫ܬ‬45(‫ ܬ‬-2‫ܣ‬ ̣ 2b10 ‫“ כפוף אני‬I am bent” = SyrA+B +.‫ ܐ‬9:;< 2b16 ‫ואל תאבדני בחטאתי‬57 “and do not destroy me because of my sins” = SyrB 6‫ـ‬.?‫ ܬܘ>ـ‬+‫ـ‬7‫ܘ‬ AB@> ̈ (cf. SyrA ‫ܬܝ‬47EF ̈ D$ 6C(?>‫ ܬܘ‬+7‫)ܘ‬ 6. 2b18 ‫ואל תחייבני תחת תהום הארץ‬58 “do not condemn me under the depths of the earth” = SyrB +‫ـ‬$‫ܬܗ ܕܐܪ‬ ̇ Ï ‫ـ‬H-‫ >ـ‬6‫ـ‬C‫ـ‬5‫ـ‬:‫ـ‬H‫ ܬ‬+‫ـ‬7‫( ܘ‬cf. SyrA +‫ـ‬$‫ܗ ܕܐܪ‬̇ ‫(ـ‬-‫ـ‬H‫ܒܬܐ‬ ̈ 6‫ـ‬.?‫ـ‬K‫ ܘܬܐ‬6‫ـ‬C‫ـ‬5‫ܬܚ(ـ‬ ̇ +‫ـ‬7‫“ ܘ‬and do not 59 condemn me and banish me to the depths of earth”) 2. 3. 4. 5. Another example of possible Syriac influence that Leicht mentions is: 7. ‫ = צבאות שמים‬SyrA+B +‫ــــ‬:‫ــــ‬M‫ــــ‬K‫ܬ* ܕ‬4‫ــــ‬8‫ܚ(ــــ‬ ̈ , but he adds: “although this can be due to the 60 Hebrew language itself.” To this last example we can add the observation that the biblical idiom ‫ צבא שמים‬is rendered in the Peshitta with a plural in, for example, Deut 17:3 MT: ‫לכל צבא השמים‬, Pesh *‫ܬ‬4‫ـــ‬7̈6‫ـــ‬H O‫ـــ‬N‫ـــ‬7 +:MK‫ܕ‬.61 We admit that such “etymological congruities” should play a role in establishing the textual relationships between the various sources. For our linguistic analysis, however, it should be noted that the features of the Hebrew text that can be accounted for by the similarities with the Syriac text are not in themselves linguistically problematic. Thus the constructio ad sensum ‫ זרעם הצדיקים‬is indeed remarkable, especially in comparison with the singular form in the Greek, but fits well with the rules of BH and RH. Also the “peculiar structure”62 ‫ אתה הוא‬is reminiscent of biblical passages such as Isa 37:16 ‫“ אתה הוא האלהים‬you are God” (or: “it is you who are God”).63 Likewise, the use of the 2nd person perfect in the first 57 Leicht, “A Newly Discovered Hebrew Version,” has ‫ לא‬instead of ‫ אל‬in his list on p. 367, but the correct reading ‫ אל‬in his text on p. 370. In general, one can observe the alternation of ‫( לא תקטל‬e.g., 2b17 ‫לא תביא‬ “do not bring”) and ‫( אל תקטל‬2b16, 17, 18). 58 Here, too, Leicht, “A Newly Discovered Hebrew Version,” has ‫ לא‬instead of ‫ אל‬in his list on p. 364, but the correct reading ‫ אל‬in his text on p. 370. 59 Note the difference between the two Syriac versions, ignored by Leicht. 60 Cf. ibid., 364. 61 Cf. P. G. Borbone and K. D. Jenner, eds., The Old Testament in Syriac according to the Peshitta Version, Part V, Concordance 1: The Pentateuch (Leiden: Brill, 1997), 304b. 62 Cf. Leicht, “A Newly Discovered Hebrew Version,” 364: “The peculiar structure ‫אתה הוא‬, found in the Syriac version ‫' ܗܘ‬/‫ܐ‬, is very striking as well.” 63 This verse has played a major role in the linguistic study of the Hebrew tripartite nominal clause; see van Peursen, “Three Approaches to the Tripartite Nominal Clause in Classical Syriac,” in Corpus Linguistics and Textual History: A Computer-Assisted Interdisciplinary Approach to the Peshitta (ed. P. S. F. Van Keulen and W. Th. Van Peursen; SSN 48; Assen: Van Gorcum, 2006), 157–73, esp. 158–59; cf. F. Delitzsch, Biblischer Commentar über den Prophet Jesaia (Leipzig: Dorffling und Franke, 1866), 363: “‫הוא‬ in ‫ אתה הוא‬ist nachdrückliche Wiederaufnahme, also Verstärkung des Subj., wie 43,25. 51,12. 2S 7,28. Jer. 49,12. Ps. 44,5. Neh. 9,6f. Ezr. 5,11: tu ille (nicht tu es ille (Ges. § 121,2) = tu, nullus alius.” verses instead of the participle found in the Greek can well be explained by the influence of the Syriac, but as such the Hebrew text is not problematic at all. An exception can be made for 3a2 ‫ כל צבאות שמים‬because the plural in this construction is indeed unattested in the Hebrew Bible.64 Moreover, here, too, as in the case of biblical quotations, the “etymological congruities” do not result in consistently applied correspondences. Note, for example, that ‫אתה הוא‬, which twice corresponds to Syriac ‫ ܗܘ‬-‫ــــــ‬.‫ܐ‬, occurs also in 2b6, where SyrA+B has only -‫ــــــ‬.‫ܐ‬. Also, the “etymological congruities” cannot account for the most striking lexical peculiarities in the Hebrew text. Thus 2b11–12 ‫“ להאריך אפך‬to make your anger endure” cannot be explained by a translation error, since SyrA+B has the “normal” ‫ܟ‬-‫ـ‬M‫ـ‬R‫ـ‬7 ‫ܬ‬Q‫ـ‬P‫“ ܐܪ‬I provoked your anger.” The same applies to the lexical peculiarities mentioned above, in Section 3.6, such as the use of ‫מצח‬ “insolence” in 2b10. 5. Discussion and Evaluation 5.1 The Linguistic Profile of PrMan-Heb The linguistic profile of PrMan-Heb differs considerably from that of other Hebrew writings from the Second Temple period such as Ben Sira or the Dead Sea Scrolls. PrMan-Heb is basically a Rabbinic text reflecting Rabbinic grammar and lexicon. Also in its contents PrManHeb has an interesting Rabbinic element, because it reflects acquaintance with Rabbinic discussions about Manasseh’s share in the world to come. There are also some biblical forms and expressions, such as the occurrence of the lengthened imperative. Some of them occur in biblical quotations and allusions. The biblical influx may be due to the liturgical character of the text and does as such not change its overall Rabbinic appearance. There are also some linguistic oddities such as the construction with ‫“ מצח‬insolence” in 2b10, the use of ‫“ להאריך אפך‬to make your anger endure” in 2b11–12, and the Piel of ‫“ חנן‬ask for compassion” in 3a1. Leicht ascribed these oddities to the translator’s poor knowledge of the Hebrew language,65 and we think he is right. It is hard to find any other satisfying explanation for these features. Regarding the question about the origin of PrMan-Heb, this linguistic evidence is not decisive. If we postulate the composition of PrMan somewhere the late Second Temple period (but cf. n. 7 above), it is clear that the Genizah text is not the original composition, but this does not compel us to assume that it is a medieval composition (or translation), since it could also be the result of an inner-Hebrew development, reflecting, on the one hand, adaptations to later Hebrew usage and, on the other hand, the influence of Biblical passages with some traces of unsuccessful adaptations of the text. 64 See Section 3.6 above. 65 Leicht, “A Newly Discovered Hebrew Version,” 367. What we can observe, however, is that PrMan-Heb has a linguistic profile different from that of other texts from the Second Temple period that have been discovered in the Genizah. Thus in the case of Ben Sira, even though there are linguistic differences between the Masada and Qumran texts on the one hand and the Genizah texts on the other—and the latter contain some “late” features, such as the typically RH idiom ‫בית מדרש‬, bet midrash, in Sir 51:23 (MS B)66—we can even say of the Genizah texts that they reflect much more linguistic diversity than PrMan-Heb, and that they combine Standard BH, Late BH, and Post-BH elements, as well as many unique features. The same applies to the Damascus Document. In this case too, the discovery of the Dead Sea Scrolls revealed not only the way in which the text has been linguistically altered and updated in its transmission history,67 but also close linguistic affinities with Qumran Hebrew and agreements with the CD fragments.68 5.2 Textual Affiliations None of the linguistic oddities in PrMan-Heb mentioned above can be explained as translation errors from the Greek or the Syriac versions, because in each case these versions have a different reading that cannot account for the reading in the Hebrew text. There is one case where the Hebrew text is likely the result of a translation error because it seems to reflect a wrong rendering of the Greek text, but in that case the Hebrew text is not problematic in itself. In the Greek text of v. 13 we find αἰτοῦµαι δέοµενος “asking I request,” in which the obvious interpretation of δέοµενος is “asking”; however, the Greek verb can also mean “to need,” and that interpretation is reflected in Hebrew 2b16 ‫“ אבקשך שאני צריך אותך‬I seek you because I need you.”69 The textual affiliations with the Syriac text seem to be stronger, especially because of the “etymological congruities” between the Hebrew and Syriac versions to which Leicht has drawn attention. However, these are not as pervasive as, for example, those between the Hebrew text of Sir 51:13–30 in the Genizah MS B and the Syriac text,70 for which it has been suggested that 66 On this idiom see van Peursen, “Sirach 51:13–30 in Hebrew and Syriac,” 369–70; on the rabbinic flavour of the Hebrew text (and the Syriac text) of Ben Sira see further van Peursen, Language and Interpretation in the Syriac Text of Ben Sira, 115, with references. 67 E. Qimron did not include the Genizah manuscripts of the Damascus Document in his Hebrew of the Dead Sea Scrolls (HSS 29; Atlanta: Scholars Press, 1986), since “their text was distorted by the copyists of the Middle Ages and thus does not reflect the DSS language, especially in its phonology and morphology” (p. 15). 68 Cf. S. E. Fassberg, “The Linguistic Study of the Damascus Document: A Historical Perspective,” The Damascus Document, A Centennial of Discovery: Proceedings of the Third International Symposium of the Orion Center for the Study of the Dead Sea Scrolls and Associated Literature, 4–8 February, 1998 (ed. J. M. Baumgarten, E. G. Chazon, and A. Pinnick; STDJ 34; Leiden: Brill, 2000), 53–67, at 67: “The relationship of phenomena in the Damascus Document to features in late biblical Hebrew, mishnaic Hebrew, and the Hebrew of the Dead Sea Scrolls has been proven beyond doubt. Moreover, the Geniza manuscripts of the Damascus Document, once disparaged linguistically, are now recognized as medieval copies that still possess features of an earlier authentic type of Hebrew.” 69 Cf. Leicht, “A Newly Discovered Hebrew Version,” 364. 70 Cf. van Peursen, “Sirach 51:13–30 in Hebrew and Syriac,” passim. the Hebrew text is a retranslation from the Syriac. The divergences between the Hebrew and the Syriac texts are too large to assume that the Hebrew presents a rather literal translation from the Syriac. This means that the model that has been used in the comparison of the Hebrew and Syriac versions of Sir 51:13–30 cannot be applied in the very same form to PrManHeb. The agreements with the Syriac text in themselves still do not compel us to consider PrMan-Heb a retranslation from Syriac. If remarkable elements in the Hebrew text correspond to similar constructions in the Syriac text and the constructions in that case are less remarkable in Syriac, the “congruities” as such would be a strong argument for the dependency of the Hebrew text upon the Syriac. However, where the Hebrew text is somewhat extraordinary, as, for example, in 2b16 ‫“ ואל תאבדני בחטאתי‬and do not destroy me because of my sins,” the construction in the Syriac text (SyrB) A‫ـ‬B@ ̈ ‫ >ـ‬6‫ـ‬.?‫ ܬܘ>ـ‬+‫ـ‬7‫ ܘ‬is just as uncommon as the Hebrew construction. In such a case the agreements do not answer the question as to whether the Syriac text derives from the Hebrew or the other way round. 5.3 Text-Historical Reconstructions In Sections 5.1–5.2 we have seen that neither the linguistic profile of PrMan-Heb nor its textual affiliations with the Greek and Syriac versions provide decisive indications for its origin. We will now take a broader perspective and address the question as to how the linguistic profile and the textual affiliations can be projected into the textual history of PrMan. It is generally acknowledged that in the Greek transmission the text of PrMan in the Didascalia has priority over all other extant versions.71 It is also likely that PrMan entered the Syriac tradition through the Syriac translation of the Didascalia, and that from there it was introduced into biblical manuscripts (SyrA).72 The relation of this Syriac version (SyrA) with the version found in the Melkite Horologia (SyrB) is complex, but there is strong evidence that both derive from a Greek text, even if not from exactly the same Greek Vorlage.73 That PrMan-Heb stands between the Greek and the Syriac, i.e. that it is a translation of the Greek and was in turn the basis for the Syriac translation, is unlikely because of the interrelatedness of the Greek and the Syriac versions as part of the Didascalia. It follows that if we try to integrate PrMan-Heb within a reconstruction of the textual history of the Greek and Syriac versions, it should stand either at the beginning (as the source text of the Greek text), or at the end (as a translation from the Greek and/or the Syriac). If we assume that PrMan-Heb stands at the beginning of the textual history, this would require the postulation of a Hebrew text that on the one hand served as the source text of the Greek translation, and on the other hand underwent an inner-Hebrew development up to the version found in the Genizah. Apart from the fact that this postulation is highly hypothetical, because of the absence of any Hebrew manuscript corroboration of this reconstruction, it is 71 Gutman and van Peursen, The Two Syriac Versions of the Prayer of Manasseh, 8–9. 72 Ibid., 24. 73 Ibid., 201. problematic because of the agreements between the Hebrew and the Syriac, not only the “etymological congruities”, but also, for example, cases where the Hebrew and the Syriac versions have a perfect against a participle in the Greek text. It is hard to explain how these features have been retained (or rather: reintroduced) in the transition from Hebrew to Greek and from Greek to Syriac. Because of these text-historical considerations, I prefer the alternative reconstruction, namely that the Hebrew text is a retranslation from the Greek and/or Syriac. As indicated above, the correspondence between the Greek αἰτοῦµαι δέοµενος “asking I request” (v. 13) and the Hebrew ‫“ אבקשך שאני צריך אותך‬I seek you because I need you” (2b16) argues for the latter’s dependency on the Greek; the remarkable patterns of formal agreement argue for dependency on the Syriac. This agrees with Leicht’s view that the Hebrew text depends both on a Syriac text (more precisely, a text of the SyrB type) and on a Greek text (close to the text of the Codex Turicensis). As a consequence, the question as to whether the Greek version in the end goes back to a Hebrew original (cf. note 8 above) becomes irrelevant to our analysis of the Genizah text and cannot be answered on the basis of this text. 5.4 Historical Considerations The text-historical reconstruction does not answer the question of how the Hebrew text was written. Was there a Hebrew scribe who had both a Greek and a Syriac text in front of him? And if so, what made him decide to resort to either the one or the other? And why did he decide to translate this document and where did he get his sources from? In our study on the two Syriac versions of PrMan, we have proposed (following a suggestion that we received from James K. Aitken) that Christians who had converted to Judaism were responsible for the (re-)adaptation of Christian sources. Since these converts were acquainted with these sources and knew their languages, but probably learned Hebrew at a later age, this assumption may account for those peculiarities that can be explained from a poor knowledge of Hebrew. 74 The interactions between Jews, Christians, and Muslims in all areas of life and culture is well-documented in the vast collection of the Genizah materials. That the exchange also concerned religious texts and practices is apparent, for example, from some Syriac liturgical texts from the Genizah.75 The text under discussion is another piece of evidence of this exchange. The origin of the PrMan-Heb can be placed in the wider cultural context of Jewish translation activity in the Middle Ages in which the Jewish or allegedly Jewish sources that had been transmitted through Christian channels were rediscovered by Jews who translated 74 Ibid., 12–13. 75 S. P. Brock, “East Syrian Liturgical Fragments from the Cario Genizah,” OrChr 68 (1984): 58–79 and Brock, “Some Further East Syrian Liturgical Fragments from the Cairo Genizah,” OrChr 74 (1990): 44–61; cf. Leicht, “A Newly Discovered Hebrew Version,” 368 and Gutman and van Peursen, The Two Syriac Versions of the Prayer of Manasseh, 12. them into Hebrew. This activity gave rise to an abundance of Hebrew translations of all kinds of literature, as has been described in detail in the still classic work by M. Steinschneider.76 Thus, culturally, the Hebrew versions of the books of Judith and Maccabees discussed above provide better parallels to PrMan-Heb than the Genizah fragments of Ben Sira or the Damascus Document. Whether this is also true linguistically deserves further research. Above we noted the differences between the linguistic profiles of the Genizah texts of Ben Sira and the Damascus Document on the one hand, and that of PrMan-Heb on the other. A comparison of the linguistic profile of PrMan-Heb with that of the other mediaeval Hebrew translations is beyond the scope of the present study. Since until now the study of these translations has focused on textual affiliations and the quest for the original versions of these books, a linguistic description of them is still a desideratum. 6. Conclusions The linguistic profile of PrMan-Heb in itself does not prove its dependency on a Greek or Syriac text. The linguistic observations put forward to support this argument are not decisive. PrMan-Heb can be read as a Rabbinic text, reflecting Rabbinic language and ideas, with some passages that reflect biblical influence. It is only the textual affiliations and general textcritical and text-historical considerations that necessitate an explanation in terms of a retranslation from the Syriac or Greek rather than in terms of development within the Hebrew. Does this mean that our research has been useless because PrMan-Heb can only be positioned at the end of a long and complex transmission history? Certainly not! PrMan-Heb is a document that deserves to be studied in its own right, whether or not it reflects a Hebrew text from the Second Temple period, and whether or not it brings us back to the precursors of the Greek and Syriac versions that were available. We find here in a clearly Jewish environment, and with some adaptations to the Jewish context, a text that otherwise was transmitted only in Christian channels. As such it is a unique witness to PrMan, which adds an interesting chapter to the reception history of Manasseh and his Prayer. It also informs us about cultural and religious exchanges between Jews and Christians in the Middle Ages. 76 M Steinschneider, Die hebräischen Übersetzungen des Mittelalters und die Juden als Dolmetscher (2 vols.; Berlin: Kommissionsverlag des Bibliographischen Bureaus, 1893; reprint Graz: Akademische Druckund Verlagsanstalt, 1956); I am indebted to Prof. Wout van Bekkum for this reference.