tasteless Greek version probably of a Byzantine Jew, found in a manuscript in St. Mark's library at Venice, and called the Græca Veneta ; which indeed will ever remain interesting enough for the history of interpretation.16 Of the mediate versions, this of course is not the place to speak.

*

Let us delay for a while the progress of our discussion, in order to subjoin some remarks upon the use and value of these versions.

1. The first efforts of the interpreter must here, of course, be directed to obtain a correct understanding of the version itself, a task for which the ways are by no means so well broken, as one would have expected. At the very outset, the lexical helps yet extant for the Septuagint are in the highest degree imperfect. The authors of them, while they often give only an incomplete account of what the Greek translator meant in his frequently obscure expressions, and seem scarcely to have thought of any scientific arrangement of the significations of a word, merely write out from the concordance the Hebrew words for which each Greek word stands; busy themselves with conjectures, how the translator came to render so or so; and not unfrequently, in order to bring about a correspondence, force upon the Greek word the meaning of the Hebrew one, and vice versa.17 As helps for the understanding of difficult passages,

16 The manuscript contains the Pentateuch, the writings of Solomon, Ruth, Lamentations, and Daniel. The first has been published by C. F. Ammon, Erlang. 1790, 91, in 3 vols. 8vo. and the other books by Villoison, Strasb. 1784, 8vo.

* See the article VERSIONS in Calmet.-ED.

17 The most convenient concordance of the Septuagint is that of Tromm, pastor of Groningen, who died 1717. A. Trommii Concordantia Graeca Versionis vulgo dicta LXX Interpretum, Amst. et Traj. ad Rh. 1718, 2 vols. fol. It contains also the words from Aquila, Symmachus, and Theodotion, and from the Hexapla, ed. Montfaucon. The earlier one of Kircher, (see note 6,) is principally arranged in an inverted order, i. e. according to the order of the Hebrew words. The order of Tromm is followed by the lexicon of J. C. Biel, under the title: Novus Thesaurus philologicus, seu Lexicon in LXX et alios Interpretes et Scriptores apocryphos V. T. ed. Mutzenbecher, 3 tom. 8vo. Hag. Com. 1779. It was however already finished in 1745. To this work Schleusner, Bretschneider, and Kreyssig have published supplements. The more recent work of Schleusner, Novus Thesaurus philol. criticus, seu Lexicon in LXX cet. post Bielium et alios viros doctos edidit et conVOL. III. No. 9.

2

1

recourse may also be had to the Scholiasts and Glossists, among whom some have particular reference to the LXX, as Cyril and Olympiodorus; but more especially to the commentators and writers of Catene in the Greek church, Eusebius, Theodoret, Cyril, Chrysostom, Procopius, and sometimes Jerome.-For understanding the Chaldee versions, the lexicon of Buxtorf leaves very much less to be desired.-As to the Syriac version, the understanding of it, or of single difficult words in it, already made business for the Syriac-Arabian philologians of the ninth century, as we shall see further on; and hence we may easily conceive, that our lexicons, drawn as they are in a very imperfect manner from those works, as well as the Latin translation of the Syrian version in the Polyglotts,* cannot every where present us with certain and authentic information.18 The Arabian versions of Saadias are sometimes rendered obscure by the un-Arabic Hebraizing style of them;19 but still more is the Samaritan version here and there marked by important difficulties, which have their ground indeed in the deficiency of other monuments of the Samaritan language, but yet might doubtless for the most part be removed, by a careful comparison and comprehensive knowledge of the other Aramaean dialects.

2. But the interpreter of these versions will also not unfrequently meet with critical difficulties, which have their source in the still very imperfect condition of the texts. That of the LXX, as is well known, is perhaps in the worst state of all; and the various readings given in the Oxford edition of Holmes

gressit J. F. S. 5 tom. 8vo. 1820, 21, is merely an enlargement and remodelling of Biel, and not a new work; and the whole former mode of treatment, so ill adapted to a lexicon, is retained. A far more appropriate arrangement is presented by Böckel, in his specimen of such a lexicon, containing the letter Z: Novae Clavis in Graecos Interpretes V. T. ita adornatae, ut etiam Lexici in N. F. libros usum præbere possit etc. Specimen, Lips. 1820. 4to.

* This was made by two modern Syrians, who resided at Paris, Gabriel Sionita and Abraham Ecchellensis.

18 The Arabic version, which, not only in Job and the prophets, but also in the greater part of the historical books, is drawn from the Syriac, is here not unfrequently of exegetical use for this its nearest source. See Roediger de Arabicae libror. V. T. historicorum Versionis origine et indole, Halae 1828.

19 See the Allgem. L. Z. 1822, No. 155.

and Parsons, 20 as drawn from manuscripts and after-versions, lead us to the conviction, that after all the critical and uncritical revisions of this version, a restoration of the original text by means of the aids remaining to us, is scarcely possible; but yet that the Vatican Codex probably approaches comparatively the nearest to this. That also the other texts in the London Polyglott stand greatly in need of critical revision, and especially of a new collation of the manuscripts, is well known to all who have ever attempted such a revision;21 and it is therefore so much the more to be regreted, that the plan proposed some years since in England, of publishing a Polyglott corrected throughout from manuscripts, should have fallen to the ground.

3. In order to make the proper use of a version in particular instances, it is necessary for the student to obtain a complete view of its general character; so that he may not be led to regard mere arbitrary renderings, allusions to later times, or quid pro quo's, as authorities and sober explanations; and thus either build upon them as such, or at least feel perplexity in particular cases. It has for instance been thought strange, that the proper name ", Malachi, Mal. 1: 1, should be translated my messenger, and explained as referring to Ezra the scribe; while at the same time it has been overlooked, that the Targum of Jonathan very often treats proper names in this manner; e. g. 2 Sam. 17: 7. Is. 7: 3, 6.-In this particular respect, very much that is useful has been done by way of preparation in recent times.22

4. Since it is a traditional interpretation which lies at the basis of the versions, their value of course as historical testimony rises with their antiquity, and sinks with their modernness. Here too we may distinguish in a measure two lines of tradition,—that of the Alexandrine, and that of the Palestine Jews. The coincidence of both these voices, testifies to a direct and consistent tradition.

20 Vetus Testamentum Græcum cum variis Lectionibus.' T. I. ed. R. Holmes, Oxon. 1798. Tom. II-V. ed. Jac. Parsons. ib. 1820—27. Compare the Allgem. L. Z. 1816, No. 1 sq. 1832, No. 1 sq. Much may also be gained from the Hexapla-Syriac version; see Middeldorpf Cura Hexaplares in librum Jobi, Vratisl. 1817.

21 See Prof. Lee on the Syriac version, in the Classical Journal, No. XLVI, p. 245 sq.

22 See, among other works, those referred to above in the Notes 9-12, and others by Winer, Hirzel, Roediger, (note 18,) etc.

5. What the student has to expect from the versions, and that exclusively, is information as to the usus of words, especially as to their signification in a particular place, and as to their general meaning, which is often expressed by other words. What the versions can never furnish, and what from their nature they never ought to furnish, is a specification of the primitive significations and etymology of words,-a species of knowledge which in general we must not look for in those ages. As exceptions to this remark, however, we may name Aquila, a few of whose fragments only are extant, and the Venetian translator; both of whom attempted to let a glimmer of the etymology be visible in their translations; somewhat as Schultens has attempted it in modern times.

At the point where the versions cease, begins, in the history of Jewish and biblical literature, the grammatical investigation and culture of the Hebrew language, and the compilation of lexicons and philological commentaries; and the merit of having here first broken the way, belongs to the Jews who spoke Arabic, in Babylon, Spain, and on the coast of Africa. The information which I have elsewhere23 given on this subject, was not and could not be complete and authentic, because the most important works of all are yet unprinted; and therefore the following notices, written after having seen, examined, and in part made extracts from the manuscripts, may stand here as supplementary.

24

The earliest lexicographical attempt in respect to the Hebrew, of which we have any knowledge, comprises seventy difficult words, which the above-mentioned Saadias Gaon arranged together, explained briefly in Arabic, and compared with the Talmudic. The first complete lexicon, which lies in manuscript at Berlin among other places, was composed by Menahem ben Saruk (po 73 ), in the beginning of the eleventh century. The chief merit of this work, however, consists in the collection and orderly arrangement of the roots, of which it presents the first example; and it is an interesting circumstance, that the biliteral roots, (so he calls the verbs, , i,) the triliteral, and the quadriliteral, are all placed by themselves. For exegesis he rarely presents any thing of importance; and draws conjectures

23 Gesch. der Heb. Spache, p. 95 sq.

24 The manuscript lies at Oxford; Cod. Huntington. No. 373. Uri's Catal. Codd. Mss, orient. Bibl. Bodleianae No. 485. Nicoll's Catalogue p. 7.

mostly from the connexion, or leaves it for others to draw them.25 His labours are wholly unimportant in comparison with the excellent work, which the Spanish physician Rabbi Jonah (a) as the Jews call him, or Abulwalid Merwan ben Gannach according to his Arabian name,26 composed under the name of Book of Roots, and from which the afterwards more celebrated Kimchi has borrowed his best illustrations. In this lexicon, the traditional interpretation of the Jewish schools is indeed cited and used; but the author frequently breaks over the limits of this authority, and often makes a happy use of his own investigations and conjectures by the help of the Talmudic and Arabic,-the more happy indeed, because the latter was his native tongue; so that this rabbin may with strict propriety be regarded, as the forerunner of an Edward Pococke, Bochart, and Albert Schultens. Some few unimportant fragments have been published by Pococke and Schnurrer and to the more complete extracts which I have made from the original manuscript, I am indebted for several explanations given in my Hebrew lexicon, which will be recognized as a real addition to our lexical knowledge of the Hebrew. The explanations of

25 Abulwalid quotes him under the word, and cites also a treatise by Ben Librat, which is probably a criticism or confutation of the work.

26 The Jews who live among the Arabs have mostly two names, one Hebrew and the other Arabic; e. g. and Abú Sacharya; as also in the time of the Greek dominion they bore Hebrew and Greek names; e. g. Jonathan and Alexander Jannaeus; Salome and Alexandra. So also in France and Germany at present, the Jews have mostly each a Hebrew and German or French name; and this necessarily, according to the laws of Napoleon.

27 The copy in the Bodleian library at Oxford, (Uri's Catal. Codd. orient. 456, 457,) was brought by Dr E. Pococke from the East, and often used by him in his commentaries on Hosea and Joel; see his Theological Works, Lond. 1740, 2 vols. fol. It is therefore strange that A. Schultens, who must have been ignorant of Pococke's writings, should say of this work: Magno redemtum vellem opus quod sequentium Rabbinorum, quibus Arabica minus promta erunt, livor pressit. Superesse tamen dicitur, et spes est, cum tempore emersurum. Origg.Hebr. p. 290. At a later period Schnurrer made some extracts from it, but published very few of them; see his Dissertationes philol. crit. p. 46. -Some instances of the aid derived by the author from the work of Abulwalid, may be seen in his lexicon under the articles,,