with the Syriac original, but sometimes follows a reading different from that of the common printed text.]"

2. There are two versions of the Psalms-one, that of the Syriac edition of the Psalms printed at Mount Lebanon, (§ 64;) the other, an unprinted Arabic Psalter in the British Museum."

[Under the direction of Sergius Risius, the learned bishop of Damascus, a Psalter in the Syriac and Arabic languages was printed, at Kasheja, in the valley of Mount Lebanon, in the seventeenth century, for the use of the Syrian Christians. The age and author of the version are unknown; but it evidently follows the Peshito as its original. Sometimes it differs from the Syriac, in minute peculiarities, and even in important readings; but this only shows that attempts had not been made to reconcile the two, before this was printed. The Arabic sometimes inclines to the Greek.

There is an independent Arabic Psalter, in manuscript, in the British Museum, which follows the Syriac, as its original. Its age and author are not known. It differs entirely from that printed at Lebanon, in th: position of its words, and in the explanation of difficult places. It appears to have been altered after the Hebrew text. Perhaps the author had, also, the Hebrew original before him, when he translated.]

с

3. There are several Arabic versions of the [Syriac] Pentateuch; but some of them have not been printed, and some are unknown."

[Eichhorn, § 290.]

Döderlein, On the Arabic Psalters, in Eichhorn's Repert. vol. ii. p. 159, 170, sqq.

[Eichhorn, § 291, 293.]

Asseman, l. c. vol. ii. p. 309. Abulfarag, Hist. Dynast. p. 355. Schnur

rer, De Pentat. Arab. Polyglot., in his Dissertatt. p. 203. Paulus, Spec. Vers.

§ 66.

V. ARABIC VERSIONS.

1. FROM THE JEWISH-HEBREW TEXT.

1. We are still in possession of a translation of the Pentateuch," and of Isaiah,' from the hand of Rabbi

Pent. Arab. p. 36, sqq. curate for critical use.

[These versions are little known, and are too inacSee Paulus, Com. crit. exhibens...... Spec. Vers. Pent. septem Arabicarum nondum editarum; Jen. 1784, 4to. Eichhorn, § 294, b.

See, also, Hottinger, Thes. phil. p. 270. Walton, Prol. xiv. 19. Grabe, Prol. ad LXX. ch. iii. § 5. White's Letter to the Bishop of London, p. 56. Kennicott, Diss. Gen. ed Bruns, § 84, and Eichhorn, § 294, c. for the Arabic translation of the hexaplary Syriac version, by Hareth Ben Senan.]

This has been printed in Pentat. Heb. Pers. Arab.; Constant. 1516, fol. See Wolf, 1. c. vol. ii. p. 354. Le Long, ed. Masch, vol. i. p. 393, sqq. Adler, Bibl. krit. Reise, p. 221.- in Paris Polyglot, vol. vi.—in London Polyglot, vol. i. (with the various readings of the Constantinople and Paris editions, vol. vi.) See O. G. Tychsen, in Eichhorn's Repert. vol. x. p. 95, on the sources whence the MSS. of the Arabic version in the Polyglots have been derived. See, also, his essay, Whether R. Saadias Haggaon is the author of the Arabic version in the Polyglots, in Eichhorn, 1. c. vol. xi. p. 82. He thinks Abu Said is the author, and not Saadias. Hottinger doubts that the versions in these Polyglots are the same. See his Diss. de Heptaplis Paris.; Tig. 1649, 4to. (Analect. Hist. Theol.; Tig. 1653, 8vo.) Smegma Orient.; Heidelberg, 1659, 4to. p. 93, sqq. Other doubts arose from misunderstanding the preface to the Paris Polyglot. (See Michaelis, Or. Bib. vol. ix. p. 153, sqq.) Schnurrer removed these doubts, by republishing that preface. See his Diss. de Pent. Arab. Polyg., (Tub. 1780, 4to.,) and in his Dissertations.

R. Saadia Phijumensis Vers. Jesaia Arab. cum aliis speciminibus Arab. bibl. e MS. Bodlej. nunc primum ed. atque glossar. perpet. instruxit, H. E. G. Paulus, Fasc. i. ii.; Jenæ, 1790, 1791, 8vo. See Eichhorn, Allg. Bib. vol. iii. p. 19, sqq., 456, sqq.

There are traces of an Arabic version of Job and Hosea, by Saadias. See Eichhorn, on the extent of Saadias's Arabic version, l. c. vol. ii. p. 181, sqq. Gesenius found his version of Job at Oxford, and copied it. See his preface to Isaiah, p. x.

[Traces of Saadias's version of Job, says Eichhorn, may be found in Cod. 40, of the Arabic MSS. in the Bodleyan Library. The preface begins as follows: Hæc interpretatio est liber directionis derivatus ex Jobo juxta interpretationem præfecti synagogæ ac magistri Saadia. Kimchi cites him on

Saadias Gaon, who died A. C. 942. These versions are executed in an explanatory, paraphrastic style, harmonizing with the explanations of the Targums and the rabbins. They are a fine monument of the rabbinical philology and knowledge of the Bible in the tenth century.

b

[According to Wolf, Rabbi Saadias was a native of Pithom-whence he is often called Pithumensis-a city in the Egyptian province of Fagum. He enjoyed such a reputation for his learning, that, in the year 927, he was invited to take charge of the academy at Babylon, then in a declining state. But, two years after, he was obliged to flee for his life. He lived in concealment the next seven years, and wrote various works. Perhaps the Arabic versions of the Pentateuch and Isaiah were composed at this time."

Eichhorn thinks there were Arabic versions before the seventh century, though the first version which is known to us belongs to the tenth. Erpenius and Pococke think Saadias translated the whole of the Old Testament. But they ground their belief on the fact that there are manuscripts containing the whole of an Arabic version of the Old Testament, part of which, certainly, proceeded from Saadias.

His version is made directly from the Hebrew; but

Hosea vi. 9. Eichhorn, § 282. This MS. is written in Hebrew characters. Erpen was so confident Saadias translated the whole Old Testament, that he promised to publish the entire version. See Walton, Prol. xiv. 15.]

See accounts of this famous Jewish scholar in Wolf, 1. c. vol. i. p. 832, sqq. Gesenius, Heb. Sprache, p. 96.

• See more respecting the critical and exegetical value of this version in Carpzov, Crit. sac. p. 646, sqq. Gesenius, Com. üb. Jes. vol. i. p. 90, sqq. On the affinity between the Paris and London impressions of the version of the Pentateuch, see Eichhorn, § 281, sq., and Adler, l. c. p. 149.

[Wolf, l. c., who gives a list of his works.]

it dilutes and expands the original till all its poetic beauties are lost. But many of the paraphrastic passages, perhaps, could not have come from Saadias, and we can never be sure we possess the version in the original form which proceeded from his hand. The transcribers have used great freedom with this, as with all the Arabic versions; and some manuscripts may approach nearer to the Hebrew text," where the printed copy is paraphrastic. Like other versions, it has been interpolated in the course of time. Some of the manuscripts have been altered systematically, to say nothing of the errors which carelessness has introduced. Even between the two printed copies of Saadias there are many discrepancies." Wherever God appears or acts, according to the edition of Constantinople, an angel of God appears or acts, according to the text of the Polyglots.

C

It has likewise been altered from the Arabic version of the Samaritan Pentateuch. It is a characteristic of the Samaritan version to render the word God by angel of God in certain connections, and this peculiarity often occurs in the Paris impression, which renders it probable that these passages have been altered by a later interpolator, and made to conform to the Samaritan. There is, besides, that general agreement between the two manuscripts of Saadias and the fragments of the Samaritan-Arabic version, which is usually found be

a

[See a comparison of a small portion of the text of the Polyglots, with that of the Florentine and Roman MSS., in Eichhorn, § 280, and in Adler, Bib. krit. Reise, p. 175. The Florentine MS. follows the Heb. very closely.]

[E. g. Deut. xxxiv. 10: "Novit eum Deus presentem," in the Polyglots; but in the Constantinople edition it is "quem Deus allocutus est sine medio," (immediate.) So, in Gen. xviii. 13, in the former it is, "and the prophet of God said to Abraham;" but in the latter, "and God said to Abraham."] [Eichhorn, $281.]

с

www

tween interpolated manuscripts. The text of the Paris and London Polyglots has been altered more than that of Constantinople, which, however, has not escaped untouched."

Only a single manuscript is known of Saadias's version of Isaiah; the subscription refers it to him, and the style agrees very closely with that of his version of the Pentateuch; so that no one can doubt that both proceeded from the same hand. Its value is inconsiderable; but it is always agreeable to know how one of the best Jewish scholars of the tenth century understood Isaiah. Here and there his version seems to have something peculiar to itself.]'

2. There is a version of Joshua and the following passages, namely: 1 Kings xiii.-2 Kings xii. 16, and Neh. i.-ix. 27, which is printed in the Polyglots."

3. There is also a version of the Pentateuch, made in the thirteenth century by an African Jew. It has been edited by Erpen." [In general, this version follows the masoretic text step by step, and a real variant is an extremely rare phenomenon in it. But if the author had an ancient Hebrew text before him, his version would be valuable in criticism to show what readings he followed; for he usually applies himself so closely to the letters that Erpen thought a Latin version was indispensable to his edition. To favor this literalness, the translator indulges in expressions that are foreign to the

[Eichhorn, § 283, b. See, also, Notice sur Rabbi Saadias Gaon et sa Version Arabe d'Isäie, et sur une Version Persane MS. de la Bibliotheque royale, suivie d'une Extrait du Livre Dalalat Al Hazerin, en Arabe et en Français, &c., par Salomon Munk; Paris, 1838. See, also, Berlin Jahrbücher, for April, 1840, p. 633, sqq.]

Rödiger, 1. c. lib. i. ch. iii.

• Pentateuchus Mosis, Arabice; Lug. Bat. 1622, small 4to.