A. Aramaic

The Aramaic version of our book demands our careful attention. This is the case not simply because, as a result of the pronouncements of Neubauer and Bickell,7 it has been popularly regarded as one of the most genuine representatives of the original form of the book, and is still supposed by some scholars to be the ‘Chaldee’ text used by Jerome,1 but also on account of the problem of the Aramaic dialect in which it appears, and finally on account of the subsidiary evidence it supplies in favour of the antiquity and originality of Rs. It was first published by Neubauer in 1875, being the fifth part of a MS.2 in the Bodleian Library, and is headed והוא כתוב במדרש רבה דרבה. This Midrash Rabbah of Rabbah is identified by Neubauer with the B’reshith Rabbah major of Martini, which in turn is identified by Zunz,3 but not by Neubauer, with the Bereshith Rabbah of R. Moses had-Darshan.

The linguistic characteristics of this version were first subjected by Nöldeke to a scientific criticism and treatment in the appendix to his epoch-making essay, to which we shall have reason constantly to refer. His investigations at once showed how optimistic was the supposition of Neubauer and Bickell4 that it usually represents the oldest and most genuine form of the original work even when it differs from Rs and Rv. Nöldeke, on the contrary, came to the conclusion that its dialect was Palestinian, and intermediate between the so-called Babylonian Targums and the more modern Palestinian dialect of the Palestinian Talmud, Midrashim, and Targums, and that consequently this extant form of the Aramaic goes back only to c. a.d. 300.

Nöldeke himself, in framing this hypothesis, recognized many of its difficulties, and endeavoured to explain them by the supposition that the text has suffered considerably from errors, mutilations, and grammatical and syntactical alterations of ‘ignorant and careless’ copyists who had no knowledge whatsoever of the influence of grammatical rules, dialectic variations, or the earlier and later forms of the language other than their own vernacular.

May not many of these characteristics that Nöldeke regards as secondary and scribal corruptions be as primary as those he regards as the most original? Are they not too numerous to allow of the supposition that they are less characteristic than those of the earlier dialects which also survive? Dalman, in fact, is probably correct in his supposition that the linguistic peculiarities in our MS. did not appear earlier than the seventh century in circles which were influenced by both Talmuds and by the more ancient Targums.

The Source of Ar. The extant Aramaic text goes back to an earlier Aramaic version,5 but the question still remains as to whether—through that earlier Aramaic text—it is to be regarded as a translation and redaction from a Semitic or a Greek original. In Neubauer’s judgement ‘the pure Semitic idiom of the Chaldee text does not admit for a moment the possibility of its being a translation from a non-Semitic text.6 Dalman writes: ‘Possibly a source in the style of the Targum of Onkelos might have been utilized and might have been the text known by Jerome, but it is also probable that the Aramaic text is a translation from the Latin.’7 Nöldeke, on the other hand, rightly argues that the language of Ar. certainly does not prevent the recognition, in parts of it, of even a fairly literal, though never a slavish, rendering of the Greek B (= Rs).’ Thus, in spite of Bickell’s explanations,8 the forms ראגיש (רגאש, רגאיש, ריגש) רגש presuppose the translation from the Greek ᾽Ραγαῖς or ᾽Ραγαῖς, whereas a Semitic original would have resulted in the appearance of רג or רגא.9 Similarly אגבתניס is a transcription of Ἐκβατάνοις, 10 whereas a Semitic original would have known the Hebrew form אחמתא. For תיגרין see note to 6:2 (1). טובי represents the dative Τωβεί. אשר 1:2 in M = Ἀσσήρ (= Ἀσσώρ?), whereas a Semitic original would have preserved the correct חצור, just as עשאל=Ἀσιήλ, which in LXX regularly represents יחציאל. Supposed misreadings in Ar. of a Semitic original and the absence of the dog in Ar. and M11 are equally futile (see notes to 6:3, 16) as evidence of the translation of Ar. and M from a non-Greek original. The forms and partial omissions of Aḥiḳar in Ar. and M are also emphatically in favour of a Greek original.

Type of Ar.’s Text. In Neubauer’s judgement Ar. ‘agrees for the greater part with the Sinaitic text, and consequently with the Itala. However, the Chaldee text has sentences which are to be found sometimes in one, sometimes in another.’ Nöldeke, as mentioned above, believes that the original Aramaic was a translation from a Greek MS. of Rs. Ar. is, therefore, a not unimportant witness to Rs as the most original text extant, while to some small extent it illustrates among Aramaic-speaking peoples a gradual evolution of the text on lines somewhat parallel to those which culminated in the publication of Rv. On the other hand, the fact that it was translated from a Greek MS. is far from supporting the theory that the book was originally written in Hebrew or Aramaic (see § 6). The use of the third person throughout is paralleled in the Vulgate.