2 The Aramaic Background of the Seventy: * Language, Culture and History 3 JAN JOOSTEN, STRASBOURG 1 4 5 6 7 8 9 10 11 12 13 14 15 The thesis of the present paper is that the Jewish community among whom the Septuagint came into being was historically linked to the Egyptian Diaspora of the Persian period. Following some introductory remarks on the origin of the Septuagint, this thesis will be argued on the basis of two sets of evidence. The first part of the study is devoted to the influence of Aramaic on Septuagint Greek. As will briefly be rehearsed, this influence is rather pervasive. It can most naturally be explained on the understanding that the version came into being in a bilingual, Greek-Aramaic, milieu such as one would find, in the third century B.C.E., in Egypt. The second, more speculative, part tries to show that the Septuagint links up with several distinctive characteristics of Egyptian Judaism of the Persian period as known, particularly, from the Elephantine archives. 16 1. The provenance of the Septuagint version 17 18 19 20 21 22 Although scholarly consensus situates the origin of the oldest Greek version of the Hebrew Scriptures in Egypt,1 debate continues to rage around the question of the translators‘ provenance. Were they Egyptian Jews working with local knowledge and traditions, or had they recently arrived from Jerusalem, bringing their exegetical baggage with them? While the Egyptian origin of the Seventy, established with a wealth of arguments by Swete and * Reworked version of the Jeremie Lecture, presented before the Faculty of Divinity, University of Cambridge, on 25 May 2009. 1 In a study that appeared after the present paper was completed, Emanuel Tov has argued that some of the ―post-pentateuchal‖ translations contained in the Septuagint may have been produced in Palestine, see E. Tov, ―Reflections on the Septuagint with Special Attention to the Post-Pentateuchal Translations,‖ in Die Septuaginta – Texte, Theologien, Einflüsse (ed. W. Kraus, M. Karrer, M. Meiser; WUNT 252; Tübingen: Mohr, 2010) 3–22. Most of the features discussed in the present paper relate to the Pentateuch, in regard to which even Tov agrees that it was produced in Egypt. 1 2 1 2 3 4 5 6 7 8 9 10 11 12 13 BIOSCS 43 (2010) Thackeray at the beginning of the twentieth century, 2 still seems likely to most specialists, prominent experts of the Greek version, such as Arie van der Kooij or Emanuel Tov, have recently argued—in line with the traditional view expressed in the Letter of Aristeas—that the translators‘ home was in Palestine.3 The question requires a differentiated approach because the production of the Greek version is likely to have been complex. The biblical books were translated by different people and at various dates. 4 The phase of translation is to be distinguished from the phase of official publication. 5 Moreover, the process of translation itself may have involved several people: according to a recent study, the creation of a written translation in antiquity would typically involve at least two people, one reading the original text from a scroll and elucidating it, the other writing down the translation on another scroll.6 This model makes it possible to conceive in a practical way of 2 See H. B. Swete, An Introduction to the Old Testament in Greek (Cambridge: Cambridge University Press, 1914); H. St. J. Thackeray, A Grammar of the Old Testament in Greek (Cambridge: Cambridge University Press, 1909); idem, The Septuagint and Jewish Worship: A Study in Origins (London: British Academy, 1921). 3 A. van der Kooij, ―The Septuagint of the Pentateuch and Ptolemaic Rule,‖ in The Pentateuch as Torah. New Models for Understanding Its Promulgation and Acceptance (ed. G. N. Knoppers, B. M. Levinson; Winona Lake: Eisenbrauns, 2007) 289–300 , in particular 299–300; E. Tov ―Les traducteurs des écritures grecques et leur approches des écritures,‖ in Traduire la Bible hébraïque : de la Septante à la Nouvelle Bible Segond (ed. R. David, M. Jinbachian; Montréal: Canadian Bible Society/Médiaspaul, 2005) 103–49, in particular 122–25. 4 See the overview in M. Harl, G. Dorival, O. Munnich, La bible grecque des Septante: Du judaïsme hellénistique au christianisme ancien (Paris: Cerf, 1988) 39–110. 5 See S. Kreuzer, ―Entstehung und Publikation der Septuaginta im Horizont frühptolemäischer Bildungs- und Kulturpolitik,‖ in Im Brennpunkt: Die Septuaginta. Studien zur Entstehung und Bedeutung der Griechischen Bibel, Band 2 (ed. S. Kreuzer, J. P. Lesch; BWANT 161; Stuttgart: Kohlhammer, 2004) 61–75. 6 T. A. van der Louw, ―The Dictation of the Septuagint Version,‖ JSJ 39 (2008) 211– 29. Cf. Jerome, Prologue to Tobit: ―I do not cease to wonder at the constancy of your demanding. For you demand that I bring a book written in Chaldean words into Latin writing, indeed the book of Tobias, which the Hebrews exclude from the catalogue of Divine Scriptures, being mindful of those things which they have titled Hagiographa. I have done enough for your desire, yet not by my study. For the studies of the Hebrews rebuke us and find fault with us, to translate this for the ears of Latins contrary to their canon. But it is better to be judging the opinion of the Pharisees than to displease and to be subject to the commands of bishops. I have persisted as I have been able, and because the language of the Chaldeans is close to Hebrew speech, finding a speaker very skilled in both languages, I took to the work of one day, and whatever he expressed to me in Hebrew words, this, with a summoned scribe, I have set forth in Latin words.‖ Joosten: Aramaic Background of the Seventy 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 3 teamwork between Palestinian and Alexandrian Jews in the production of the Septuagint. In spite of these recent suggestions and possibilities, however, a number of facts continue to favor the Egyptian scenario. I. II. Linguistic features strongly indicate that the translators were Egyptian Jews. The closest parallels to the vocabulary and phraseology of the Septuagint come from Egyptian papyri. 7 Moreover, the fluency of the Septuagint‘s language indicates an origin in a Greek-speaking milieu, such as one would find in the Diaspora but not—at least not in the early third century B.C.E.— in the home country. The language of the Septuagint may be ―bad Greek‖ in the sense that it is non-literary, but in places it is surprisingly idiomatic.8 There are also some Egyptian loanwords, such as θῖβισ, ―basket‖ (Exod 2:3, 5, 6), ἄχει, ―reeds‖ (Gen 41:2, 18; Isa 19:7), οἰφ , ―ephah‖ (Lev 5:11; 6:13; Num 5:15), ςιςόη, ―lock of hair‖ (Lev 19:27).9 The translators show independent knowledge of things Egyptian, as can be observed, for instance, in the treatment of Egyptian proper names and geographical names, in the Joseph story and elsewhere. 10 7 To a certain extent this is an argument from silence—documentary papyri of the Hellenistic age happen to have been preserved almost exclusively in Egypt—but not wholly so. See A. Deissmann, Bibelstudien (Marburg: Elwert, 1895); idem, Neue Bibelstudien (Marburg: Elwert, 1897); idem, Licht vom Osten (Tübingen: Mohr, 1908); J. A. L. Lee, A Lexical Study of the Septuagint Version of the Pentateuch (SCSMS 14; Chico CA: Scholars, 1983). 8 See Lee, Lexical Study, 34–52 and passim. 9 For the first three, attested also in the papyri, see J.-L. Fournet, ―Les emprunts du grec à l‘égyptien,‖ Bulletin de la Société de Linguistique de Paris 84 (1989) 55–80, in particular 73, 68 and 71; for θῖβισ, see also Lee, Lexical Study, 115. The Egyptian origin of the word ςιςόη, has only recently been noted by M. Vahrenhorst, see his comment on Lev 19:27 in the forthcoming companion volume to the Septuagint Deutsch translation; for the Egyptological background of the word and its etymology, see H. De Meulenaere, ―Le nom propre Σις ισ et son prototype égyptien,‖ ChrEg 66 (1991) 129–35. 10 See S. Pfeiffer, ―Joseph in Ägypten. Althistorische Beobachtungen zur griechischen Übersetzung und Rezeption von Gen 39–50,‖ in Die Septuaginta — Texte, Kontexte, Lebenswelten (ed. M. Karrer, W. Kraus; WUNT 219; Tübingen: Mohr Siebeck, 2008) 313– 22. For similar indications, see P.-M. Bogaert, ―L‘orientation du parvis du sanctuaire dans la version grecque de l‘Exode (Ex., 9–13 LXX)‖ L’Antiquité Classique 50 (1981) 79–85; J. Joosten, ―To See God. Conflicting Exegetical Tendencies in the Septuagint,‖ in Die Septuaginta — Texte, Kontexte, Lebenswelten (ed. M. Karrer, W. Kraus; WUNT 219; Tübingen: Mohr Siebeck, 2008) 287–99; M. Rösel, ―Greek Bible and Hebrew Lexicography: Gesenius‘ use of the Septuagint,‖ forthcoming in a volume on Gesenius‘s Hebrew Dictionary, to be edited by S. Schorch. BIOSCS 43 (2010) 4 1 2 3 4 5 6 7 8 9 III. IV. Striking agreements between the Nash Papyrus—a unique Hebrew manuscript from Egypt dating probably from the second century B.C.E.— and the Septuagint text of the Ten Commandments suggest that a local text may have been used for the Greek version.11 Finally, some of the legal interpretations contained in the Septuagint fit the world of the Diaspora well, but contradict what we know about Jewish exegesis as practiced in the home country. A good example is the systematic incorporation of the donkey into laws on redemption of livestock, which affects three different passages in Exodus (Exod 22:29; 13:13; 34:20).12 10 11 12 Much of the exegetical knowledge incorporated into the Septuagint may have come from the land of Israel, but the translators appear to have been Egyptian Jews writing for a local Jewish readership. 13 2. Aramaic influence on the Septuagint 14 15 16 17 18 19 Further depth can be given to the view that the Septuagint is of Egyptian origin by paying attention to the Aramaic influence to which it testifies. For the most part, the discussion will be based on the Greek Pentateuch, the oldest portion of the Septuagint. Aramaic influence permeates the entire corpus, however, and some telling examples will be brought from other books as well. 20 21 22 23 24 25 2.1. Linguistic elements reflecting Aramaic influence A point of departure is provided by some linguistic traits that are well known.13 The Greek Pentateuch uses a small number of Aramaic loanwords that are not attested elsewhere in Greek—except in texts depending on the Septuagint. The feast of Pesach is referred to as πάςχα: the ending of the Greek word in alpha probably reflects the Aramaic status emphaticus.14 11 See I. Himbaza Le Décalogue et l’histoire du texte. Etudes des formes textuelles du Décalogue et leurs implications dans l’histoire du texte de l’Ancien Testament (OBO 207; Fribourg: Academic Press / Göttingen: Vandenhoeck & Ruprecht, 2004). 12 Z. Frankel, Ueber den Einfluss der palästinischen Exegese auf die alexandrinische Hermeneutik (Leipzig: Barth, 1851) 98–99. 13 See J. Joosten, ―The Septuagint as a source of information on Egyptian Aramaic in the Hellenistic Period,‖ in Aramaic in its Historical and Linguistic Setting (ed. H. Gzella, M. L. Folmer; Akademie der Wissenschaften und der Literatur – Mainz, Veröffentlichungen der Orientalistischen Kommission 50; Wiesbaden: Harrassowitz, 2008) 93–105. 14 See P. Walters, The Text of the Septuagint: Its Corruptions and their Emendation (Cambridge: CUP, 1973) 174. Joosten: Aramaic Background of the Seventy 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 5 Similarly, the Sabbath is called, ςαββ τα, corresponding to Aramaic ‫ׁבתא‬, but in this case re-analyzed as a neuter plural in Greek. 15 One also finds μάννα, ―manna,‖16 ς κερα, ―strong drink,‖ and γειώρασ/γ ωρα(σ), ―proselyte, convert to Judaism.‖17 From Greek Isa, the word παταχρα/παταχρον, ―idol,‖ reflecting Aramaic (and ultimately Persian) ‫פתכר‬, ―sculpture, image,‖ can be added. All these words—even ς κερα, ―strong drink,‖ which is practically limited to legal contexts18—clearly belong to the religious terminology of Judaism. To Greek ears, they must have sounded barbaric. Aramaic loanwords are only a handful in the Septuagint, but several other indications of Aramaic influence exist. An interesting case is that of gentilic adjectives. In the Septuagint, ethnic groups that are still around in the Hellenistic period are given their usual Greek name: e.g., Αἰγ πτιοσ, ―Egyptian,‖ Σ ροσ, ―Syrian,‖ Ἄραβεσ, ―Arabs,‖ Φο νικεσ, ―Phoenicians.‖ The peoples of old, however, known from the Bible only, are mostly referred to with gentilics ending in –aios. The examples begin early in Genesis: Χετταῖοσ, ―Hittite,‖ άεβουςαῖοσ, ―Jebusite,‖ Ἀμορραῖοσ, ―Amorrite,‖ and they continue through the entire corpus. Gentilics in –ites—evoking the usual form in Hebrew—are attested as well, but they are much less frequent: άςραηλ τησ, ―Israelite,‖ Λευ τησ, Ἀμμαν τησ, Μωαβ τησ. Thackeray, in his grammar of the Septuagint,19 compares the gentilics in –aios to forms like Ἀθηναῖοσ, ―Athenian,‖ θηβαῖοσ, ―Theban.‖ Such forms are rare, however, outside of the Bible, and invariably derive from city names ending in –ai: Ἀθ ναι, Θῆβαι. They cannot explain the remarkable frequency of -aios forms in the Septuagint. It is reasonable, therefore, to invoke influence from the Aramaic nisbe ending –ayya. Gentilics of the type Χαναναῖοσ, Ἐβραῖοσ reflect Aramaic ‫כנעניא‬, ‫עבריא‬. Although the forms sound Greek, they are really hybrids amalgamating Aramaic morphology with a Greek case ending. 15 In Greek, this word was re-analyzed as a neuter plural. The singular occurs nowhere in the Greek Pentateuch but is found twice in Isaiah and in books translated later. See A. Pelletier ―Σαββατα: Transcription grecque de l‘araméen,‖ VT 22 (1972) 436–47. 16 Num 11:6, 7, 9; Deut 8:3, 16 and elsewhere. In Exod 16, however, the Hebrew word is simply transcribed, μαν. See Walters, Text, 170–71. 17 See, on the attestation and variant forms of this word, Walters, Text, 33–34. 18 Outside the Pentateuch, the word is found also in narrative and poetic texts. 19 Thackeray, Grammar, 171. 6 BIOSCS 43 (2010) 1 2 3 4 5 6 Another indication revealing Aramaic influence is the fact that geographical names of the type Baal-X, systematically turn up as Beel-X in the Greek Pentateuch: 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 These transcriptions indicate that the translators thought of the Aramaic status constructus ‫ ִב ֵףל‬rather than Hebrew ‫ב ַףל‬. ַ As in the preceding instance, the Aramaic language interferes with the Hebrew: one should have expected the translators to adopt forms like *άεβους τησ and * Βααλςεπφών, after the Hebrew, but instead they use language that is tainted by Aramaic. The phraseology, too, of the Septuagint at times seems to go back to Aramaic models. One is accustomed to Aramaisms occurring in the Greek New Τestament, but it turns out there are similar cases in the Septuagint. A strong example is the way actions directed toward God are situated before him, or in his presence, in the Greek version, against the Hebrew Vorlage.20 The daughters of Zelophehad say, according to the Hebrew: ―Our father died in the wilderness; he was not among the company of those who conspired against the LORD in the company of Korah (‫ֹּף ִדים ַףל־יְ הוָ ה‬ ָ ‫) ָה ֵף ָדה ַה‬.‖ ‫ ‏‬In Greek, the latter phrase becomes: ―He was not in the midst of the gathering that conspired before the Lord in the gathering of Kore (τῆσ ςυναγωγῆσ τῆσ ἐπιςυςτάςησ ἔναντι κυρίου).‖ This ―distancing‖ use of prepositions meaning ―before‖ (in Greek, ἐνώπιον, ἐναντ ον and ἔναντι) is frequently attested in the Septuagint with verbs meaning ―to sin‖: one sins against a human being, but before God. The usage occurs more rarely with verbs meaning ―to say,‖ ―to speak,‖ ―to see,‖ ―to give,‖ ―to praise,‖ and ―to lie.‖ This type of language finds no parallels in original Greek writings, whether literary or documentary, except in texts that stand under the influence of the Septuagint. It recalls Targumic phraseology and has its background, as has been recognized by Targumic scholars, in the court language of the Persian period. In Official Aramaic texts one speaks ―before the King,‖ not ―to the King,‖ and one commits offenses ―before the King,‖ not ―against‖ him. This language was projected into the religious realm, as attested in Aramaic texts of both Jewish ‫ — ַב ַףל ְצפֹן‏‬Βεελςεπφών (Exod 14:2, 9; Num 33:7) ‫ — ַב ַףל ְפעֹר‏‬Βεελφεγώρ (Num 25:3, 5; Deut 4:3) ‫ — ַב ַףל ְמעֹן‏‬Βεελμεών (Num 32:38) 20 See J. Joosten, ―L‘Agir humain devant Dieu. Remarques sur une tournure remarquable de la Septante,‖ RB 113 (2006) 5–17. Joosten: Aramaic Background of the Seventy 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 7 and pagan origin; it has made a lasting impression on Jewish Aramaic religious phraseology. That Septuagint Greek too should show influence of this Aramaic usage is highly significant. While translating a Hebrew text, the Seventy used turns of phrase borrowed from Aramaic. 2.2. The process mediating Aramaic influence How did these words, forms, and phrases end up in the Septuagint? What is the process by which the Aramaic language came to taint the Greek style of the translators? The features discussed certainly do not find their origin in the Hellenistic culture of the translators. The Aramaizing elements are practically unattested outside the Septuagint. Moreover, they all belong, in one way or another, to a ―biblical‖ style typifying the Greek version, but of little use otherwise. They show, incidentally, that the Septuagint was aimed at a Jewish audience, not a Greek one as the letter of Aristeas would lead one to believe. If the Aramaic elements are not representative of Hellenistic Greek, neither can they be derived simply from the Hebrew Bible. Some factor must account for the interference of Aramaic. A few scholars have argued that the Septuagint was translated from an Aramaic version of the Hebrew Bible, an early Targum.21 This idea would indeed account for all the remarkable linguistic features listed above: if the translator of Exodus found Aramaic ‫ פסחא‬in his source text, instead of Hebrew ‫פסח‬, he might simply have transcribed this form into his translation, and so on. The hypothesis founders, however, upon the general character of the Septuagint, which gives ample evidence of having been translated from the Hebrew, not the Aramaic. If the Seventy disposed of a Targum—and the case in favor of this view is not as weak as one might think—they may have used it as an aid to understanding the Hebrew, not instead of the Hebrew.22 But such a model cannot explain our Aramaic features: it would be unnatural for a translator to take the Aramaic word ‫ פסחא‬from his Targumic tradition in order to render Hebrew ‫ פסח‬in his source text! 21 L. Delekat, ―Ein Septuagintatargum,‖ VT 8 (1958) 225–52. 22 For an example that may illustrate such use of a Targum, see M. Papoutsakis, ―Ostriches into Sirens: Toward an understanding of a Septuagint Crux,‖ JJS 55 (2004) 25– 36. 8 BIOSCS 43 (2010) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 The remaining option is to regard these features as reflecting the religious jargon of the community among which the version came into being. The Aramaic element, it seems, was not created by the translators, but adopted from the pre-existing speech habits of their community. In religious discourse—prayer, liturgy, and the teaching of Jewish law—their Greek phraseology consisted partly of expressions borrowed from Aramaic. The translators used ςαββ τα in reference to the Sabbath, not because ‫ שבתא‬was written in their source text, but because that was the way they always referred to the institution in question. Similarly, the other words of Aramaic origin such as γειώρασ and παταχρα were part of the Jewish Greek sociolect even before the Greek translation of the Bible was begun. Gentilics of the form Ἐβραῖος and geographical names like Βεελςεπφών must have been acceptable in this type of Greek. And turns of phrase like ―he sinned before God‖ instead of ―he sinned against God‖ may also have been a normal part of the religious parlance of the group in question. If this explanation of the Aramaic elements is the correct one, the balance of probability comes down squarely on the side of a background in the Diaspora, rather than the home country. Greek and Aramaic were spoken by Jews in both Palestine and Egypt, but the functional load of these languages was very different in the two localities. The crucial difference is that, while Greek came to be used as a vehicle of the Jewish religion in Egypt early on, in Palestine it remained confined to secular discourse for a long time. Evidence to this effect is sparse but unequivocal. 24 25 26 27 28 29 30 31 32 2.3. Hebrew, Aramaic and Greek: Palestine versus Egypt Let us first consider the linguistic situation in the land of Israel. 23 Throughout the Second Temple period, Hebrew continued to dominate here in religious discourse. The late biblical books, apocrypha such as Ben Sira or Jubilees, the Qumran texts, and early Rabbinic literature all show the predominance of Hebrew in this area of life. Aramaic was used in religious texts as well— think of Daniel and Ezra, Henoch, the Genesis Apocryphon—but it never displaced Hebrew from its dominant position. As to Greek, it appears to have had much trouble establishing itself as a religious language in Palestine. 23 See J. A. Fitzmyer, ―The Languages of Palestine in the First Century AD,‖ CBQ 32 (1970), pp. 501–31; J. Joosten, ―Hebrew, Aramaic and Greek in the Qumran Scrolls‖ forthcoming in J. J. Collins, T. H. Lim, The Oxford Handbook of the Dead Sea Scrolls (Oxford: OUP). Joosten: Aramaic Background of the Seventy 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 9 Although epigraphic remains attest the importance of Greek in public affairs, it is hard to point to a single instance of the use of Greek in religious literature by Jews in the Land of Israel during the Hellenistic period. 24 By the first century C.E., there were Greek-speaking religious communities in Jerusalem, as witnessed notably by the synagogue inscription of Theodotus and by the book of Acts. But this reflects a later situation. In the third century B.C.E., when the Greek Pentateuch was produced, it is very unlikely that Greek should have been regularly used in prayer and worship by Jews in Palestine. Even if was, it would be surprising to find that this type of Greek was deeply influenced by Aramaic: religious language would be far more likely to undergo influence from Hebrew. The linguistic situation in the Egyptian Diaspora was completely different. Hebrew was almost certainly no longer a living language here, although Scripture may have continued to be recited in the original language as is shown by the Nash Papyrus mentioned earlier. All available evidence indicates that the language spoken by Jews in Egypt from the fifth century onward, if not even earlier, was Aramaic. Aramaic continued to be used by at least some Jews for several generations beyond the onset of the Hellenistic period, as witnessed by a few documents and tomb inscriptions. 25 There is abundant proof that Aramaic-speaking Jews in Egypt maintained a form of their ancestral religion throughout the Persian period.26 Their normal language of religious discourse may be supposed to have been Aramaic. Although no specifically religious texts have been preserved, Aramaic documents from Egypt mention the Sabbath and the Pesach festival, as well as questions of ritual purity. 27 From the onset of the Hellenistic period, Greek 24 A probable exception is Eupolemus, see B. Z. Wacholder, Eupolemus. A Study of Judaeo-Greek Literature (Cincinatti – New York: Hebrew Union College & Jewish Institute of Religion, 1974). Although Eupolemus deals with religious matters, he is essentially a historian. His writings do no give evidence of the use of Greek in prayer or liturgy in Palestine in his time. 25 See the list in T. Muraoka, B. Porten, A Grammar of Egyptian Aramaic (Leiden: Brill, 1998) 1–2. For a recent re-edition of the most important texts, see A. Yardeni, Textbook of Aramaic, Hebrew and Nabataean Documentary Texts from the Judaean Desert and Related Materials. A The Documents (Jerusalem: Hebrew University, 2000) 384–441. 26 See, for example, B. Porten, Archives from Elephantine. The life of an ancient Jewish military colony (Berkeley: University of California Press, 1968). 27 For the Sabbath, see B. Porten, ‖The Religion of the Jews of Elephantine in Light of the Hermopolis Papyri,‖ JNES 28 (1969) 116–21; for Pesach, see P. Grelot, ―Etudes sur le 10 BIOSCS 43 (2010) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 became a language of religious discourse in Egypt, as witnessed by the Septuagint itself and by other writings arising in its wake: treatises by Jewish Hellenistic writers such as Demetrius and Aristobulus, additions to biblical books such as Daniel or Esther, and original creations such as the Letter of Aristeas. In the transitional period in the early third century, both Greek and Aramaic will have been in use among the Jewish community in Egypt. This is exactly the sort of situation where Aramaic could influence the Greek spoken by Jews, particularly in the religious domain. One may posit a kind of linguistic osmosis, where the older substrate language influenced the more recent superstratum, and vice versa. It is interesting to note that the interpenetration of Hebrew and Greek went both ways: in an Aramaic ostracon from Egypt dated to the third century B.C.E. and containing the Jewish proper name ‫מלכיה‬, we find the Greek loanword ‫ – פינך‬π ναξ, ―dish, plate‖ (TAD D7.57); conversely, in one of the Greek Zenon papyri (n° 59.762), also from the third century, one finds the Aramaic loanword ςαββ τα – ‫שבתא‬, ―Sabbath.‖ The Septuagint is a much more extensive corpus than Jewish papyri and ostraca from Hellenistic Egypt. But it appears to bear witness to the same linguistic situation. The idea that the Aramaic elements of the Septuagint derive from the interaction of Greek and Aramaic in the Jewish community in Egypt is not new.28 It has seldom been underscored, however, that Aramaic features in Septuagint Greek plead rather strongly for the Egyptian origin of the version. 23 24 25 26 27 28 29 2.4. Aramaic influence on the translational process On top of influencing the language of the Septuagint, Aramaic also strongly affected the translational process. 29 The Greek translators, of all biblical books, often understood Hebrew words of their source text in light of Aramaic. The phenomenon is complex, because late Hebrew itself adopted many forms and meanings from Aramaic. In some cases it is impossible to say whether the Seventy confused Hebrew and Aramaic, or whether they ―Papyrus Pascal‖ d‘Elephantine,‖ VT 4 (1954) 349–84 in particular 378 ; for ritual purity: B. Porten and A. Yardeni, ―Ostracon Clermont-Ganneau 125(?): A Case of Ritual Purity,‖ JAOS 113 (1993) 451–56. 28 See, for example, Delekat, ―Septuagintatargum.‖ 29 This aspect of the Septuagint is discussed at some length in J. Joosten, ―On Aramaising Renderings in the Septuagint,‖ 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) 587–600. Joosten: Aramaic Background of the Seventy 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 11 confused classical and post-classical Hebrew. The material is sufficiently abundant, however, to show that the translators had independent knowledge of Aramaic. Two basic procedures may be distinguished. In some instances, it appears the translator, struggling to understand a rare word, turned to Aramaic for help. Thus the hapax legomenon ‫מגְ ָבֹת‬, ִ purportedly meaning ―twisted cords,‖ in Exod 28:14, is rendered as καταμεμιγμ να ―mixed‖ in the Septuagint, probably on the basis of the Aramaic verb ‫גבל‬, ―to mix.‖ In this case, the ancient translator acts much like a modern exegete: rare words whose meaning has been forgotten can sometimes be understood better on the basis of cognates in other Semitic languages. More frequently, the Aramaic influence does not reflect the scholarship of the Seventy but seems to be due to inadvertence. In Prov 28:15, the Hebrew text reads: ―Like a roaring lion or a charging bear is a wicked ruler over a poor people‖; the Septuagint keeps the lion, but changes the bear into a wolf: ―A hungry lion and a thirsty wolf…‖ The Greek translation of Proverbs is notoriously free and not all of its renderings can easily be explained. In the present case, however, it seems very likely that the unvocalized Hebrew ‫ד ֹב‬, ―bear,‖ was read as Aramaic ‫ד)א(ב‬, ―wolf.‖ The Hebrew word for ―bear‖ is not difficult or rare. In other passages it is correctly rendered into Greek. And there is no exegetical reason to pair the lion and the wolf rather than the lion and the bear. The translator just happened to think of the Aramaic word first. Another example: in Num 3:10, the priestly formula ‫יּמת‬ ָ ‫וְ ַהזָ ר ַה ָק ֵרב‬, ―the non-priest who comes near to the sancta shall be put to death‖ is rendered into Greek ―the alien who touches (ὁ ἁπτ μενοσ) the sancta shall be put to death,‖ reflecting Aramaic ‫― קרב‬to touch.‖ Cases of inadvertent confusion of Hebrew and Aramaic are numerous enough to suggest that the translators knew Aramaic better than Hebrew. Aramaizing renderings do not contribute much to the question of provenance. Both the scholarly and the inadvertent use of Aramaic in translating the Hebrew source text find ample parallels in Palestinian sources such as the Qumran texts and early Rabbinic writings. However, such parallels may not be used to argue for a Palestinian background of the translators.30 Similar phenomena are expected in any setting where translators 30 As is done by Tov, ―Les traducteurs,‖ (n. 2) 123. 12 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 BIOSCS 43 (2010) can fall back on their intimate knowledge of Aramaic. If the Egyptian background of the translators is likely on other grounds, the Aramaizing renderings are fully compatible with it. In one or two cases, the Aramaic word underlying a Greek rendering is attested exclusively in Egyptian Aramaic. In Gen 21:33, where the Hebrew text states that Abraham ―planted a tamarisk tree (‫)אׁל‬,‖ the Septuagint has Abraham planting an ἄρουρα, a word meaning ―field‖ in classical Greek and designating a measure of land, 100 cubits square, in Egyptian papyri. The rendering may owe something to the translators‘ reservations to associate the patriarch with sacred trees, as is argued by James Barr.31 But this cannot be the whole story. Kutscher explained the divergence in reference to Egyptian Aramaic.32 The Aramaic word ‫אׁל‬, probably a loan from Akkadian (ashlu) originally meaning ―rope,‖ is used in reference to a measure of land in two Aramaic texts from Egypt. This usage seems to be found nowhere else. In later Aramaic dialects, the word occurs with the meaning ―rope,‖ or indicating a measure of length, but not in reference to area. If the distribution of the Aramaic term is not simply due to the hazards of attestation, the rendering in Gen 21:33 may indicate that the translators were familiar with Egyptian Aramaic specifically. But the evidence remains tenuous. 3. Wider cultural and historical connections Close analysis of the Aramaic element in Septuagintal Greek confirms the likelihood that the translation was produced among Egyptian Jews. The ―osmosis‖ of Greek and Aramaic, particularly in the religious domain, reflects the transition from one language to the other among the Egyptian Diaspora. Aramaic influence on the translational process is natural, too, in such a milieu. The early Hellenistic Age was a period of great upheavals. According to the Letter of Aristeas and Flavius Josephus, thousands of Jews migrated to Egypt or were forced to settle there. Alexandria was a new city, founded by the conqueror himself. This does not mean, however, that all Egyptian Jews of the early Hellenistic age—nor all Alexandrian Jews—had come straight from the home country. A prudent approach will be to submit that the Egyptian Diaspora of the third century B.C.E. was of mixed origin. A 31 J. Barr, ―Seeing the wood for the Trees? An Enigmatic Ancient Translation,‖ JSS 13 (1968) 11–20. 32 See E. Y. Kutscher, The Language and Linguistic Background of the Isaiah Scroll (Jerusalem: Magnes, 1959) 55; see also G. R. Driver, JThS 22 (1971) 548 - 552. Joosten: Aramaic Background of the Seventy 13 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 significant proportion of Egyptian Jews would appear to have descended from groups that had resided in the Land of the Nile for many generations. It is the presence of that local component that explains the Aramaic coloring of Septuagint Greek illustrated above. This conclusion has profound implications for the Septuagint as a whole. The Greek version, it turns out, goes back to a type of Judaism that is, at least in part, historically distinct from the Palestinian/Babylonian type standing at the back of the Hebrew Bible. Egyptian Jews entertained relations with Jerusalem—the Elephantine papyri give proof of this. There would have been exchanges and visits between the two groups. Nevertheless, the Egyptian Diaspora appears to have differed in its linguistic habits, its sociological make-up, its political allegiance and, no doubt, its cultural outlook from the Jewish community in the home country. If the Septuagint is the product of this other strand of Judaism, the implications could be momentous. In the second part of this paper, additional evidence will be sought confirming these suggestions. A number of peculiarities of the Septuagint, other than the language, can with some degree of probability be related to characteristic traits of Egyptian Judaism as it existed before the Hellenistic period. The source material at our disposal—the Septuagint and other JudeoHellenistic writings on the one hand, Aramaic documents going back to the Jews of Egypt on the other—is very limited and does not lend itself easily to comparative study. Nevertheless, a number of possible connections would seem to exist. 24 25 26 27 28 29 30 31 32 33 3.1. Egyptian Judaism in the Persian and Hellenistic periods The method applied in the following sections will be to seek for distinctive features of the Egyptian Diaspora that can be observed in both earlier Aramaic texts and later Greek ones, and also turn up in one way or another in the Septuagint. Although it cannot be proven in every instance that the various attestations of a given phenomenon are historically connected, they do create an impression of continuity among the western Diaspora over a long period of time. Sundry attestations, some of them far apart in time, are suggestive of trajectories. The fact that the Septuagint intersects with these trajectories is significant. 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 BIOSCS 43 (2010) 3.2. Soldiers with an interest in religious affairs Let us begin with a rather external point, namely the sociological profile of the groups concerned. Nobody doubts that the Jews of Elephantine were soldiers, settled there with the express intention of guarding the frontiers of the empire.33 Their being soldiers did not prevent them from being interested in religious affairs. They had a temple devoted to the God of the Jews with specialized personnel and a functioning sacrificial cult. When the temple was destroyed, they lobbied for it to be rebuilt. 34 With regard to the Diaspora of the early Hellenistic period, early historians and epigraphic material concur to show that many, perhaps most, Jews in Egypt also belonged in some way to the military.35 Now this similarity in social profile does not necessarily mean that the two groups are related. The colonies of Jewish soldiers of the Persian period may have disappeared in the mists of time, and the Jews serving in Ptolemy‘s army may have originated elsewhere. The resemblance could be due to accident. It is far more likely, however, that the similarity in sociological outline does indicate at least a measure of historical continuity. Surely the Jewish soldiers in service of the Ptolemaic kings had not learnt their trade in the Persian province of Yehud. A more probable scenario is that groups of soldiers who had served the Persian overlords of Egypt changed their allegiance to the new masters of the land. It is very difficult, of course, to relate this feature to the Greek translation of the Hebrew Scriptures—difficult, but perhaps not impossible. In a recent article, I have drawn attention to a few remarkable instances in the Septuagint where military vocabulary is used in a non-military context.36 The word παρεμβολ , ―camp‖, is a good example. Although the word expressed various meanings in Greek, from the Hellenistic period onward it specifically refers to a military camp or army (even an army in battle). In the Septuagint, the Hellenistic usage is well attested. In a few passages, however, the word refers not to a military camp but to a peaceful installation. When Jacob prepares to meet his brother Esau, after twenty years of separation, Genesis 32:8 says, in 33 B. Porten, ―The Elephantine Papyri‖ ABD 2 (1992) 445–55. 34 Porten, ―Elephantine Papyri.‖ 35 See A. Kasher, The Jews in Hellenistic and Roman Egypt. The Struggle for Equal Rights (TSAJ 7; Tübingen: Mohr-Siebeck, 1985) 38–48. 36 J. Joosten, ―Language as symptom. Linguistic clues to the social background of the Seventy,‖ Text 23 (2007) 69–80. Joosten: Aramaic Background of the Seventy 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 the Greek version: ―Jacob was greatly terrified, and was perplexed; and he divided the people that was with him, and the cows, and the camels, and the sheep, into two camps (παρεμβολ σ).‖ The use of the word παρεμβολ in reference to a peaceful encampment is remarkable, but it does not stand alone. Other military words are employed in non-military contexts in a similar way: the noun ἀποςκευ , ―military baggage, train‖, refers to the families of the patriarchs; and the verb ςτρατοπεδε ω, ―to march‖, describes Abraham‘s voyage through the desert. I have argued that this phenomenon reflects the professional jargon of soldiers. Soldiers would indeed refer to their own encampments or settlements as παρεμβολ ι, ―military camps,‖ to their families as ἀποςκευ , ―military baggage‖ (a usage found in the papyri), and to their travel with the verb ςτρατοπεδε ω, ―to march‖. If the translators were soldiers, it would be a small step for them to use the same terms in reference to the patriarchs. If this hypothesis is not excessively speculative, it would appear that the sociology of the group that produced the Septuagint is not unlike that of the Jewish settlement in Elephantine: they were soldiers, with access to a relatively high level of literacy and culture, and with a strong interest in religious affairs. 19 20 21 22 23 24 25 26 27 28 29 30 31 3.3. Judaism consisting of twelve tribes A more ideological feature potentially linking Jewish Hellenistic literature to the Elephantine corpus is the representation of the ―Jewish‖ people as consisting of twelve tribes. While Palestinian sources of the Persian and Hellenistic periods show clear understanding of the fact that the twelve tribes belong to the distant past, and that only Judah (and part of Benjamin) survived, Egyptian Jews appear to have conceived of the Jewish people as consisting of different tribes in their own time. Since the strongest testimonies to this view come from the Hellenistic period, the trajectory will be pursued backwards in time. A good starting point is the book of Judith. Although Judith has been regarded as a translation of a lost Hebrew original by most scholars until about ten years ago or so, recent research has shown that the book was 16 BIOSCS 43 (2010) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 probably written in Greek from the start 37. It belongs with the additions to Daniel and Esther and shows that Alexandrian Judaism did not just adopt writings considered authoritative in Jerusalem, but supplemented its religious canon with original writings as well. In regard to the issue at hand, it is interesting to note that the conspectus of Israelite history presented in Jdt 5 (told from the mouth of Achior, the Ammonite), passes over the schism and the Exile of the Northern kingdom completely. 38 The people settled around the Second Temple in Jerusalem descend directly from the people that came out of Egypt in the distant past. The land is called Judea, but the people inhabiting it represent different tribes. Judith herself belongs to the tribe of Simeon, as does Uzziah, the ruler of Bethulia. Other tribes are alluded to in two passages, although not by name. The remarkable view expressed in Judith could be written off as an oddity, but it does not stand alone. Apparently, it was held to already in the third century. Hecataeus of Abdera, a pagan historian of the time of Alexander the Great and Ptolemy I, whose words have been transmitted by Diodorus the Sicilian,39 provides an account of Jewish origins that diverges much from the standard one: 24 25 26 27 28 As in Judith so in this account, the Jewish People of the Hellenistic period are viewed as a direct continuation of the Exodus generation. The twelve tribes that once settled in and around Jerusalem are the ones that continue living in the land of Judea. The hierocracy characterizing their nation today is inherited from of old. The only discontinuity stems from their coming under Strangers of all sorts, having escaped from Egypt under Moses‘ guidance, settle in the land of Judea where they build the city of Jerusalem and the temple. Moses divides them into twelve tribes. They do not have a king but are ruled by a Chief Priest (archiereus). Later they come under foreign rule, first that of the Persians, then that of the Macedonians. 37 See J. Joosten, ―The Original Language and Historical Milieu of the Book of Judith,‖ Meghillot 5–6: A Festschrift for Devorah Dimant (2007),*159–*76. This study contains references to earlier literature. 38 In contrast, the picture in the book of Tobit is much more faithful to history. The hero is from the tribe of Naphtali, but his exile is traced back to the occupation of the Northern kingdom by the Assyrians, as it should be. Morever, the schism between the ten tribes and Judah is taken into account. As fragments from Qumran show, Tobit must have been written originally in HebrewHeb. or Aramaic. 39 See M. Stern, Greek and Latin Authors on Jews and Judaism (Jerusalem: Israel Academy of Sciences and Humanities, 1974) 1.26–28. Joosten: Aramaic Background of the Seventy 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 17 foreign rule. Hecataeus‘ knowledge is undoubtedly of Egyptian origin and must go back, directly or indirectly, to Jewish informants. What stands behind this portrayal is perhaps nothing else than the Pentateuch itself, complemented by knowledge of contemporary circumstances. What strikes one as bizarre, however, is that no other biblical sources seem to have been consulted—the period between Moses‘s death and the Babylonian Exile is a complete blank. The conception of the Jewish people as consisting of twelve tribes right up to the Hellenistic period is also very prominent in the Letter of Aristeas. In Aristeas, the people of the Law are systematically referred to as ―the Jews‖— the name Israel is never used. Nevertheless, the Jewish people are made up of twelve tribes. References to the twelve tribes are primarily linked to the translators who are sent to Alexandria: in paragraphs 32, 39 and 46, we find the phrase ―six from each tribe‖ and paragraphs 47–50 list the names of all the translators given by the number of the tribe; no tribal names appear. 40 The idea does not appear to be found in other writings, as opposed to the idea of a twelve-tribe Israel in ancient times, or at the end of days. How can this feature be related to the Elephantine papyri? The papyri do not devote much attention to historical matters. Although we do learn of the claim that the sanctuary in Elephantine existed already in the time of Cambyses, there is no information on the origins and provenance of the community. It has been stressed, however, by several specialists, most recently by Karel van der Toorn, that the ―Jews‖ of Elephantine very probably incorporated a substantial group tracing their descent to Israelites of the Northern kingdom.41 The prominence of the god Bethel and the peculiar syncretism surfacing in some of the texts, as well as some other details, are indications of this. The designation ―Jews‖ (‫)יהודיא‬, used often in the papyri, is to be taken in an inclusive sense. Indeed, the Elephantine group will have integrated Judeans at a later date. Furthermore, it may be submitted that Jews were a recognized group in the Persian empire in a way Israelites were not. ―Jews‖ of northern descent would therefore have an interest to stress their ―Jewishness.‖ 40 Thanks are due to Ben Wright with whom I have discussed this feature in The Letter of Aristeas in email correspondence. 41 K. van der Toorn, ―Anat-Yahu, Some Other Deities, and the Jews of Elephantine,‖ Numen 39 (1992) 80–101. 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 BIOSCS 43 (2010) If the ―Jewish‖ group of Elephantine incorporated families descending from the northern tribes, and if this type of Judaism was representative of the Egyptian Diaspora in general, this would explain at least partly the peculiar conception observed marginally in the Septuagint and more markedly in Judith, Hecataeus and the Letter of Aristeas. While the dominant group among Babylonian/Palestinian Jews was indeed made up of people whose genealogy tied them to the tribe of Judah, this was probably not the case among the Egyptian Diaspora. Although Judith is part of the Greek canon, the book is relatively late and not fully representative of the Septuagint corpus. In order to relate the Septuagint to the idea that the Jewish people consist of twelve tribes in the Hellenistic period, one would wish to find additional evidence. From the nature of the material, the conception of the ―Jewish‖ people as consisting of twelve tribes could not leave a strong imprint on the translated books of the Septuagint. The translators generally intended to provide a faithful rendering of the Hebrew text, not to give expression to their own ideas. In most of the translated books of the Greek Bible, the conception discussed here has left no trace. There may be one exception, however. Research in preparation of the French translation of the Septuagint of Hosea, published in the meantime in the Bible d’Alexandrie series, led to the identification of a recurrent phenomenon in this corpus. In all passages where the Hebrew text dissociates Judah from Israel, the Greek text reads differently and treats the two in the same way. Thus, in Hos 4:15, the Hebrew text has ―Though you play the whore, O Israel, do not let Judah become guilty ( NRSV).‖ But the Greek organizes the sentence differently: ―But you, O Israel, stop being ignorant, and you, O Judah stop going to Galgala (NETS).‖ Something of this kind happens four times in the Septuagint of Hosea, and once more in Micah. 42 To the translator of the Minor Prophets, it seems, the names Judah and Israel are synonymous and could not refer to two different groups. Could it be that the translator of Hosea was oblivious to the schism between Judah and Israel and to the exile of the ten Tribes? 42 See also Hos 1:6 (codex Vaticanus); 6:10; 12:1; Mic 1:5. Admittedly, most of the divergences in these verses can be explained as accidental misreadings of difficult verses. Nevertheless, the tendency is the same throughout. It appears, ten, that the assimilation of Judah and Israel does not represent a deliberate policy of the translator, but rather a more or less unconscious process. Joosten: Aramaic Background of the Seventy 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 19 3.4. The God Yaho43 One more detail linking Hellenistic Judaism to the Egyptian Diaspora in the Persian period is the peculiar form of the divine name that appears to have been in use in Egypt. The argument is again somewhat speculative, due to the scarcity of the evidence. One of the features connecting the Elephantine papyri to Northern Israel is the use of the divine name ‫יהו‬/‫יהה‬. The triliteral form of the divine name is clearly related to the Tetragram of the Hebrew Bible and of inscriptions of the monarchic period. Equally clearly, it represents a variant form that is never attested, as a free-standing form, in Judean Hebrew. 44 The only early attestation of the trigram outside of Elephantine occurs in an epigraphic corpus with northern connections. In the Kuntillet Ajrud inscriptions, dated to end of the ninth or the beginning of the eighth century B.C.E., the name ‫יהו‬ occurs twice, once side-by-side with quadriliteral ‫יהוה‬.45 Although Kuntillet Ajrud lies south of Judah, specialists agree that the famous inscriptions found there were inscribed by Israelites from the Northern kingdom.46 In this light, it appears that the Jews of Elephantine did not create the triliteral divine name themselves but brought it with them from the land of Israel. The Semitic triliteral form disappears after the fourth century. 47 But something very close to it turns up in Greek texts. According to Diodorus of 43 See more extensively J. Joosten, ―Le dieu Iaô et le tréfonds araméen des Septante‖ forthcoming in a Festschrift. 44 The use of the trigram in proper names is not to be confused with its use as a freestanding form. 45 See S. Ahituv, Asufat Ketovot Ivriyyot (Jerusalem: Bialik Institute, 1992) 153, 156. 46 See A. Lemaire, ―Date et origine des inscriptions hébraïques et phéniciennes de Kuntillet cAjrud,‖ SEL 1 (1984) 131–43; B. A. Mastin, ―The Inscriptions Written on Plaster at Kuntillet ‗Ajrud,‖ VT 59 (2009) 99–115. 47 In papyrus Amherst 63, IX 17, written in Aramaic in demotic script, the expression yaho toranana, ―Yaho, our bull is read,‖ according to a recent report, see M. S. Smith, ―Counting Calves at Bethel,‖ in ―Up to the Gates of Ekron:‖ Essays on the Archaeology and History of the Eastern Mediterranean in Honor of Seymour Gitin (ed. S. White Crawford; Jerusalem: Israel Exploration Society, 2007) 382–94, on p. 385. The God YHW also appears to be mentioned in an Idumean ostracon, see A. Lemaire, ―Les religions du Sud de la Palestine au IVe s. av. J.-C. d‘après les ostraca araméens d‘Idumée,‖ Comptesrendus des séances de l’Académie des inscriptions et belles-lettres 145 (2001) 1141–58, on p. 1152. These two attestations come from the very end of the Persian period. Much later, the form is found a few times in magical texts, see J. Naveh, S. Shaked, Amulets and Magic Bowls. Aramaic Incantations of Late Antiquity (Jerusalem: Magnes, 1985) 44, 50, 180. 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 BIOSCS 43 (2010) Sicily (first century B.C.E.), the Jews invoked their god as ΙΑΩ.48 The same Greek form is found in a number of ancient onomastica, in magical texts and in patristic testimonies.49 ΙΑΩ is triliteral like the Elephantine forms, and corresponds to the most probable pronunciation of the Semitic forms ‫יהו‬/‫יהה‬. Several scholars have postulated a historical connection between the Greek and Semitic forms.50 When Egyptian Jews adopted Greek as their main language, they continued to use the divine name they were accustomed to from of old, merely transcribing it into Greek. Admittedly, by the beginning of the Hellenistic period the tendency not to pronounce the divine name at all was gathering momentum, affecting the Diaspora as well as the home country.51 The name ΙΑΩ turns up, as an equivalent of the Hebrew Tetragram, in one of the earliest manuscripts of the Septuagint, 4QpapLXXLevb, of the first century B.C.E.52 Patrick Skehan, the editor of this text, suggested that the reading ΙΑΩ might be more original than the reading κ ριοσ found in practically all other manuscripts: ―This new evidence strongly suggests that the usage in question goes back for some books at least to the beginnings of the Septuagint rendering.‖53 Skehan‘s view is based on the notion that it is easier to imagine scribes changing ΙΑΩ into κ ριοσ than vice versa.54 Against Skehan, Pietersma has pointed to the fact that some early manuscripts of the Septuagint contain variants that were introduced in order to bring the version in line with the Hebrew text.55 But ΙΑΩ can hardly be explained in this light, unlike other reflexes of the Tetragram in early Greek manuscripts such as ‫י‬ written in square script or the infamous ΠΙΠΙ. 48 Bibl. Hist. I.94.2; Stern, Greek and Latin Authors, 171. 49 See the exhaustive review of the evidence in F. Shaw, The Earliest Non-Mystical Jewish use of Ιαω, Dissertation, University of Cincinnati, 2002. 50 See Shaw, Ιαω, 188, n. 141. 51 See M. Rösel, Adonaj – warum Gott „Herr‖ genannt wird (Tübingen: MohrSiebeck, 2000). 52 DJD 9. 53 P. W. Skehan, ―The Divine Name at Qumran, in the Massada Scroll and in the Septuagint,‖ BIOSCS 13 (1980) 14–44. 54 A view very close to that of Skehan has been argued more recently by E. Tov, ―The Greek Biblical Texts from the Judean Desert,‖ in The Bible as Book: The Transmission of the Greek Text (ed. S. McKendrick and O. O‘Sullivan; London: British Library, 2003) 97– 122, on pp. 112–14. 55 A. Pietersma, ―Kyrios or Tetragram: A Renewed Quest for the Original Septuagint,‖ in De Septuaginta: Studies in Honour of John William Wevers on his Sixty-Fifth Birthday (ed. A. Pietersma, C. Cox; Mississauga ON: Benben, 1984) 85–101. Joosten: Aramaic Background of the Seventy 1 2 3 21 Whether or not Skehan was right, the presence of ΙΑΩ in 4QpapLXXLevb shows that groups transmitting the Septuagint knew and used this peculiar form of the divine name. 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 In a recent Article, Arye Edrei and Doron Mendels have argued that, after the destruction of the Second Temple, the western Diaspora split off from Palestinian Judaism and from the eastern Diaspora.56 The fact that the oral law, the Passover Haggadah, formulaic prayers, and other normative texts of proto-rabbinic Judaism were never translated into Greek led to a rift between the practices of the western Diaspora and those of other Jews. Eventually, the ―biblical Judaism‖ of the western Diaspora disappeared, or was absorbed into the early Christian Church. The origin of this rift is traced back by Edrei and Mendels to the Hellenistic period and to the linguistic divide between Greekspeaking and Aramaic-or-Hebrew speaking Jews. In light of what has been said in the preceding sections, one might argue that the roots of this split stretch back before the Hellenistic period. Language would seem to be but one of the factors involved. The western Diaspora may have represented a distinct strand of Judaism—ethnically, sociologically, and culturally—from its inception. The textual basis for this conclusion is admittedly meager. Religious texts that can be traced back to Jewish groups in Egypt in the Persian period are very few. Arguing back from the Septuagint is possible, but perilous. The connection between the Septuagint and the Aramaic-speaking Judaism of the Egyptian Diaspora in Persian times runs along dotted lines that are hard to trace. The evidence is sufficiently suggestive to show that the linguistic facts do not lead into a blind alley. But the data do not allow one to build a solid and well-balanced theory of the Septuagint‘s antecedents. 28 4. Conclusion 29 30 The itinerary followed in this paper has consisted of a number of progressively more hesitating steps. On the basis of Aramaic elements 3.5. A split Diaspora? 56 A. Edrei, D. Mendels, ―A Split Jewish Diaspora: Its Dramatic Consequences,‖ JSP 16 (2007) 91–137. 22 BIOSCS 43 (2010) 1 2 3 4 5 6 7 8 incorporated in Septuagintal Greek—some of them well known, others only recently discovered—the argument has been advanced that the version originated in a bilingual milieu where both Greek and Aramaic were used in religious discourse. In the early Hellenistic period, such a milieu would have existed among Egyptian Jews, but hardly in Palestine. As has been argued, the Aramaic component of this milieu reveals a connection to the Egyptian Diaspora of the Persian period. Finally, a number of possible historical links between the Septuagint and earlier Egyptian Judaism have been explored. 9 10 11 12 13 14 15 If the hypothesis defended in this study is correct or even partly correct it will have profound implications for the study of the Septuagint. All kinds of peculiarities of the Greek version might be due not to idiosyncrasies of the translators, but to the background of the version in a type of Judaism whose historical roots are different from better-known strands of Palestinian Judaism.