d. The very close and almost slavish version of the Pentateuch, by some Mauritanian Jew of the thir- teenth century, which Erpenius published at Leyden in 1622 — the so-called A rabs Erpenii.
e. The Samaritan Arabic version of Abu Sa'id. According to the author's preface affixed to the Paris MS. of this version (No. 4), the original of which is given in Eichhorn's Bibl. Biblioth. iii, 6, Abu Sa'id was induced to undertake it, parti}- by seeing the cor- rupt state to which ignorant copyists had reduced the version then used by the Samaritans, and partly by discovering that the version which the}- used, under the belief that it was that of Abu'l Hasan of Tyre, was in reality none other than that of Saadyah Hag- gaon. His national prejudice being thus excited against an accursed Jew, and the "manifest impiety" of some of his interpretations, he applied himself to this translation, and accompanied it with notes, in or- der to justify his renderings, to explain difficulties, and to dispute with the Jews. His version is charac- terized by extreme fidelity to the Samaritan text (i. c. in other words, to the Hebrew text with the differ- ences which distinguish the Samaritan recension of it),
ARABIC VERSIONS
352
ARAH
retaining even the order of the words, and often sac- rificing the proprieties of the Arabic idiom to the pres- ervation of the very terms of the original. It is cer- tainly not formed on the Samaritan version, although it sometimes agrees with it ; and it has such a resem- blance to the version of Saadyah as implies familiarity ■with it, or a designed use of its assistance ; and it ex- ceeds both these in the constant avoidance of all an- thropomorphic expressions. Its date is unknown, but it must have been executed between the tenth and thirteenth centuries, because it was necessarily pos- terior to Saadyah's version, and because the Barberini copy of it was written A.D. 1227. It is to be regret- ted that this version, although it would be chiefly available in determining the readings of the Samari- tan Pentateuch, is still unpublished. It exists in MS. at Oxford (one of the copies there being the one cited by Castell in the Appendix to the London Poly- glot), at Paris, Leyden, and at Rome, in the cele- brated Barberini Triglot (the best description of which is in De Rossi"s Specimen Var. Led. ct Chald. Estheris Additamenta, Tubingen, 1783). Portions only have been printed : the earliest by Hottinger, in his Promtuarium, p. 98 ; and the longest two by De Sacy, with an interesting dissertation, in Eichhorn's Bib!. Biblioth. x, and by Van Vloten, in his Specim. Philol *g. continens descrip. cod. MS. Biblioth. Lugd.- Bat. Partemque Vers. Sam. Arab.Pentat. (Leidse, 1S03).
f. A version of the Gospels, which was first print- ed at Rome in 1590, then in the Arabic New Testa- ment of Erpenius in 1G16, and afterward in the Paris Polyglot (the text of which last is the one copied in that of London). The first two of these editions are derived from MSS., and the variations which distin- guish the text of Paris from that of Rome are also sup- posed to have been obtained from a MS. The agree- ment and the diversity of all these texts are equally remarkable. The agreement is so great as to prove that they all represent only one and the same version, and thai one based immediately on the Greek. The diversities (exclusive of errors of copyists) consist in the irregular changes which have been made in every one of these MSS., separately, to adapt it indiscrimi- nately to the Peshito or Coptic versions. This sur- prising amalgamation is thus accounted for hy Hug : When the prevalence of the Arabic language had ren- dered the Syriac and Coptic obsolete, the Syrians and Copts were obliged to use an Arabic version. They therefore took some translation in that language, but first adapted it to the Peshito and Memphitic versions respectively. As the Peshito and Coptic versions still continued to be read first in their churches, and the Arabic translation immediately afterward, as a kind of Targum, it became usual to write their nation- al versions and this amended Arabic version in paral- lel columns. This mere juxtaposition led to a further adulteration in each case. Afterward, two of these MSS., which had thus suffered different adaptations, were brought together by some means, and mutually corrupted each other— by which a third text, the hy- brid one of our Arabic version, was produced. The age of the original Arabic text is uncertain ; but the circumstance of its adoption by the Syrians and Copts places it near the seventh century (Bertholdt's Einleit. i. 692 sq.).
g. The version of the Acts, of the Epistles of Paul, of the Catholic Epistles, and of the Apocalypse, which is found in both the Polyglots. The author is un- known, but he is supposed to have been a native of Cyrene, and the date to be the eighth or ninth century (Bertholdt, ibid.).
II. Arabic versions founded on the Sept.
a. The Polyglot version of the Prophets, which is expressly said in the inscription in the Paris MS. to have been made from the Greek by an Alexandrian priest. Its date is probably later than the tenth cen- tury.